Environmental Protection
Office of Water
Regulations and Standards
Washington, D.C. 20460
EPA 440/5-85-001
veson
•*"*" £•
Nonpoint Source
lutioi
-------
PERSPECTIVES ON
NONPOINT SOURCE
POLLUTION
Proceedings of a National Conference
Kansas City, Missouri
May 19-22,1985
U.S. Environmental Protection Agency
1985
-------
Review Notice
This report has been reviewed by the U.S. Environmental Protection Agency and approved
for publication. Approval does not signify that the contents necessarily reflect the views and
policies of the Environmental Protection Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendations for use.
EPA 440/&S&001
U.S. Environmental Protection Agency
Office of Water Regulations and Standards
Washington, D.C. 20460
Copies of this publication are available for $8.50 postage and handling
($20 for foreign orders) while supply lasts from the North American
Lake Management Society, P.O. Box 217, Menifield, VA 22116; (202)
833-3382. A list of conference attendees is also available from the
same source.
-------
FOREWORD
A cursory scan of this book will tell the reader that this is no ordinary proceedings of a
conference. Indeed, the papers published here reflect a conference that was neither
ordinary nor routine. These papers are very dissimilar—in length, in format, in tone.
But they are alike in one vital way: each represents a slightly different perspective on
the problem of nonpoint source pollution of our Nation's water.
And that was the intent: to gather together all those organizations and individuals
concerned with this problem and to draw from them the most practical ways to deal
with it.
As Assistant Administrator for Water at the U.S. Environmental Protection Agency
Jack Ravan conceived this conference as an integral part of the Agency's approach
to nonpoint source pollution. The other three components are discussed in this
volume: ASIWPCA's survey of the States (being completed this fall); the Federal
Nonpoint Source Task Force (which concluded its initial policy and strategy develop-
ment at the end of 1984); and the Chesapeake Bay study now underway by the
National Association of Conservation Districts.
Ravan's instructions for this conference were fourfold: (1) focus on drawing infor-
mation from everybody involved with the problem; (2) find out how people throughout
the country, at the local level, perceive the nonpoint source problem and how they
believe it should be handled; (3) make the information flow from the grass roots
upward—the Federal role was to listen and learn and exchange information, not to
dominate; and (4) make it practical.
The conference steering committee took this charge very seriously, designing a
structure that stimulated this flow of information. The program committee fleshed it
out, using both submitted and invited presentations.
The result was essentially a practical dialog. Of course, it cannot be fully covered in
these presentations, but they will serve to remind participants of the equally valuable
informal exchanges that took place during the week in Kansas City.
Four basic themes evolved as the conference developed.
Practical, affordable solutions not imposed by Federal authority but worked out
at the local level was the message of the keynoter, Congressman Pat Roberts of
Kansas' First District.
-------
The knowledge exists, participants reiterated throughout the sessions. Putting it
to work is the next step.
Nonpoint source pollution is best solved at the local level, concluded Robert I.
Broadbent, Assistant Secretary for Water and Science at the Department of the
Interior, as he moderated the closing plenary session. Broadbent's conclusion was
drawn from two days of listening to sessions and to individuals—and is repeated
throughout this volume. Discussions of new State programs in Missouri and Maryland
reinforce this belief.
Cooperation is the key. Ravan sounded that note at the opening plenary—and it
continued to be apparent throughout the conference. Nearly 40 organizations came
together to cosponsor this conference—many others were represented among the
attendees. In the real-life capitals of this country, these organizations often find them-
selves at odds[ many had never talked over their mutual concerns.
This conference certainly began such a dialog. And, as a session chairman, Roger
Bellinger, observed, there was evident a willingness to talk, a maturity "that may allow
us to truly begin solving our water quality problems."
If a mature optimism and diversity of perspective were its hallmarks, then this
conference's true success can be measured only by how we move forward from here.
We have the information—in this volume, from this conference—but the communica-
tion established must translate into working together to improve the quality of our
Nation's waters.
iv
-------
CONTENTS
FOREWORD.
in
THE POLICY PERSPECTIVE—A LOOK TO THE
GRASS ROOTS
Welcome to Conference
LeeM. Thomas, Adminstrator, U.S. Environmental
Protection Agency
Keynote Address
U.S. Representative Pat Roberts
Nonpoint Source Pollution—A Problem for All.
John R. Block, Secretary of Agriculture
A Congressional Viewpoint on Nonpoint Source
Pollution
U.S. Representative AHan Stangeland
3
5
PERSPECTIVES ON NONPOINT SOURCE
POLLUTION
Nonpoint Source Pollution—The Illinois Approach .
Larry A. Worries
The Fertilizer Industry's Perspective on Nonpoint
Pollution
G. W. Garrett
Perspectives on Nonpoint Source Pollution Control:
Silviculture
John M. Bethea
Nonpoint Source Control: A Continuing Challenge .
John Spencer
Perspectives on Nonpoint Source Pollution Control:
A Conservation View
Benjamin C. Dysart III
A Livestock Industry Perspective on Nonpoint
Source Pollution Control
Lester Coy
11
13
14
16
19
MONITORING AND ASSESSMENT TECHNIQUES
The St. Albans Bay Watershed RCWR A Case
Study of Monitoring and Assessment 21
John C. Clausen
Land Use Monitoring and Assessment for Nonpoint
Source Pollution Control 25
Richmond B. Hopkins, Jr., and John C. Clausen
Appropriate Designs for Documenting Water
Quality Improvements from Agricultural NPS
Control Programs
J. Spooner, R. P. Maas, S. A. Dressing,
M. D. Smolen, and F. J. Humenik
Monitoring for Water Quality Objectives in
Response to Nonpoint Source Pollution
K. W. Thomson and G. D. Haffner
Use of Bioassays to Determine Potential Toxicity
Effects of Environmental Pollutants
S. A. Peterson, W. E. Miller, J. C. Greene, and
C. A. Callahan
LEGAL ASPECTS OF NONPOINT SOURCE
POLLUTION
An Overview of the National Nonpoint Source
Policy
Amy Marasco, Claire Gesalman, Vivian Daub, Carl
Myers and James Meek
Intergovernmental Coordination: Feast or Famine?.
Robert J. Massarelli
30
35
38
The Basic Legal Issues
James T. B. Tripp
Compelling On-1 he-uround impiementaiion of
Measures to Control Nonpoint Source Pollution ..
Hope M. Babcock
47
51
55
60
-------
Controlling Nonpoint Sources of Pollution—The
Federal Legal Framework and the Alternative of
Nonfederal Action
Richard R. Greenfield
STATE NONPOINT SOURCE PROGRAMS
Funding Nonpoint Control Projects in Missouri.
John Howland
State of Maryland Nonpoint Source Control
Implementation Program
Kenneth E. McElroy and Marie C. Halka
63
The Wisconsin Nonpoint Source Program.
John G. Konrad
INSTITUTIONAL/FINANCIAL ASPECTS OF
NONPOINT SOURCE CONTROLS
Bridging the Gap Between Water Quality and
Nonpoint Source Activities: A Continuum of
Institutional Arrangements
Bart Hague
69
71
76
The Utah Agriculture Resource Development Loan
Program
James A. Paraskeva
79
85
Developing Nonpoint Source Control Strategies for
Big Stone Lake; Two Approaches 88
Gay/en F. Reetz, Timothy Bjork, Patrick J. Mulloy,
David R. German, Steven A. Heiskary, and Donald
Roberts
Nonpoint Source Pollution of Reservoirs: What TVA
Is Doing About It 93
Larry R. Clark
Comprehensive Protection for Two Multipurpose
Reservoirs in Central North Carolina—EPA's
National Nonpoint Source Policy Can Work
Edward A. Holland and Alan W. Klimek
96
GROUND WATER QUALITY
Ground Water Contamination by Aldicarb Pesticide
in Eastern Suffolk County, Long Island 101
Julian Soren and W. G. Stelz
Nonpoint Source Contamination of Ground Water
in Karst-Carbonate Aquifers in Iowa 109
George R. Hallberg, Robert D. Libra, and Bernard
E. Hoyer
An Interdisciplinary Approach to Shallow Ground
Water Contamination in North-Central Montana .... 115
Jane M. Holzer, Jeff D. Farkel, Brian J. Harrison,
and Glenn A. Hockett
Nonpoint Source Impacts on Ground Water Quality
in Major Land Resource Areas of the Southwest.... 121
S. J. Smith., J. W. Naney, and W. A. Berg
Monitoring the Effects to the Ground Water System
Attributable to Agricultural Practices 125
Clark Gregory Kimball
LAKE QUALITY
Urban Nonpoint Source Impacts on a Surface
Water Supply 129
William W. Walker, Jr.
Nonpoint Source Pollution Control Funding for
Lake Restoration: A Case Study at Carlisle Lake ... 138
Barry C. Moore, William H. Funk, and Richard
Bainbridge
Why Scofield Reservoir is Eutrophic: Effects of
Nonpoint Source Pollutants on a Water Supply
Reservoir in Utah 142
Doyle Stephens
Trophic State Response to Nonpoint Pollution
Control: Application of Coupled Microcomputer
Models to the Great Lakes 147
Martin T. Auer, Thomas M. Heidtke, and Raymond
P. Canale
A Project to Manage Agriculture Wastes Has
Improved the Quality of Vermont's Lake Parker .... 153
Richard J. Croft
ESTUARINE QUALITY
Urban Runoff Pollutant Inputs to Narrangansett
Bay: Comparison to Point Sources 159
Eva J. Hoffman
Chesapeake Bay Nonpoint Source Pollution 165
Joseph Macknis
The Influence of NPS Pollution in Florida Estuaries:
A Case Study 172
Joe Ryan and J. H. Cox
Nonpoint Pollution Control in Small Bays of Puget
Sound 177
Bob Saunders
Shellfish Sanitation in Oregon: Can It Be Achieved
Through Pollution Source Management? 180
John E. Jackson
STREAMS AND RIVERS
Monitoring Changes in Agricultural Runoff Quality
in the LaPlatte River Watershed, Vermont 185
Donald W. Meals
Nonpoint Source Pollution in the Rice Creek
Watershed District—The Results of 10 Years of
Water Quality Monitoring
Pefer R. Wlllenbring, Eugene A. Hickock, and
William D. Weidenbacher
Pesticide Monitoring in Kansas Surface Waters:
1973-1984
Michael K. Butler and Joseph A. Arruda
Effects of Intensive Agricultural Land Use on
Regional Water Quality in Northwestern Ohio.
David B. Baker, Kenneth A. Kreiger, R. Peter
Richards, and Jack W. Kramer
191
196
201
LIVESTOCK WASTE MANAGEMENT
What Do You Do with a Regulation? ..
Mary Burke
209
VI
-------
A National Perspective for Livestock Waste
Management 211
Ronald A. Michieli
Ecopsychorrhea 213
Tom Hovendon
Controlling Water Pollution from Nonpoint Source
Livestock Operations 215
John M. Sweeten and Stewart W. MeMn
Application of New Technologies to Livestock
Waste Management 218
J. B. Martin, Jr., B. G. Ruffin, J. O. Donald, and
L L Behrends
NONPOINT PROGRAMS: THE STATUS
Nonpoint Source Control Programs 221
Charles L Boothby
The Status of Silvicultural Nonpoint Source
Programs 223
George G. Ice
The Association of State and Interstate Water
Pollution Control Administrators 227
Robbi J. Savage and Linda Eichmiller
ECONOMICS OF NONPOINT SOURCE
POLLUTION
Economics: Nonpoint Source Pollution Impacts .... 229
Sandra S. Batie
Economics of Nonpoint Source Pollution Control:
Lake Tahoe, California/Nevada 232
David S. Zlegler
Controlling Agricultural Runoff: Government's
Perspective 234
Richard S. Magleby and C. Edwin Young
Soil Erosion as a Nonpoint Source—A Farmer's
Perspective 237
flobert Warrick
Gross Erosion Rates, Sediment Yields, and
Nutrient Yields for Lake Erie Tributaries:
Implications for Targeting
Daw'd B. Baker, Kenneth A. Kreiger, R. Peter
Richards, and Jack W. Kramer
251
Watershed Water Quality Programs: Lessons
Learned in Illinois 256
Tom Davenport and John Lowrey
Prairie Rose Lake Rural Clean Water Program
Project 259
Ubbo Agena, Monica Wnuck, and C. Merle Lawyer
Agricultural Sources of Nitrate Contamination in a
Shallow Sand and Gravel Aquifer in Eastern South
Dakota 264
Jeanne Goodman
AGRICULTURAL ISSUES: WESTERN
EXPERIENCE
Agricultural Issues: The Nebraska Perspective..... 269
Roger E. Gold
Hydropolitical Solutions to Complex Nonpoint
Salinity Pollution Problems in the Colorado River
Basin 273
Jack A. Barnett
Accumulation of Sediment, Nutrients, and Cesium-
137 in Prairie Potholes in Cultivated and
Noncultivated Watersheds 274
Dan B. Martin
Irrigated Agriculture and Nonpoint Source Pollution
in the San Joaquin Valley of California 276
William R. Johnston
URBAN ISSUES: RUNOFF
Bellevue Experiences with Urban Runoff Quality
Control Strategies 279
Pam Bissonette
AGRICULTURAL ISSUES: EASTERN AND
SOUTHERN EXPERIENCE
Nonpoint Source Pollution: SCS Perspective 239
Ernest V. Todd
Nonpoint Source Pollution from Plant Nutrients .... 241
0. P. Englestad and K. S. Brady
Nonpoint Source Pollution: Managing Nutrients a
Key to Control 244
George Wolff
Agricultural Land Treatment Project Planning for
Off-Site Phosphorus Reduction 245
Francis M. Keeler
AGRICULTURAL ISSUES: MIDWESTERN
EXPERIENCE
Identifying Critical NPS Contributing Watershed
Areas 247
Kent W. Thornton and Dennis E. Ford
The Effects of Carbonate Geology on Urban
Runoff 281
floger P. Betson and Jack D. Milligan
Implementing an Urban Nonpoint Source Control
Strategy 285
David F. Lakatos and Alan Cavacas
Urban Storm Water Quality Management: The
Florida Experience 289
Eric H. Livingston and John H. Cox
URBAN ISSUES: CONSTRUCTION NONPOINT
SOURCE POLLUTION
Hampton Roads Water Quality Agency Nonpoint
Source Program 293
Paul E. Fisher
Problems and Progress in Urban Soil Erosion and
Sedimentation Control: A Bicounty Perspective ..
Peter G. Thum and Gerald A. Paulson
VII
-------
URBAN ISSUES: HYDROLOGIC MODIFICATION
AND SEPTIC TANKS
National Perspective on Environmental Constraints
to Hydroelectric Development 301
S. G. Hildebrand, M. J. Sale, G. F. Cada, and
J.M.Loar
Perspectives on Septic Tanks as Nonpoint Source
Pollution 304
B. L Carlile
Hydrologic Modification: Compounding the Impact
of Nonpoint Source Pollution 306
A. David McKinney
RURAL ISSUES: COAL MINING AND
ABANDONED LAND RECLAMATION
Acid Mine Drainage: Surface Mine Treatment and
In Situ Abatement Technology
Frank T. Caruccio and Gwendolyn Geldel
Coal Industry Perspectives on Nonpoint Source
Pollution
Virginia P. Leftwich
Trends in Post Mining Land Uses—Are We Doing
Our Children Justice?
Don Eagleston
Factors and Treatment of Abandoned Acid Mine
Lands for Controlling Nonpoint Source Pollution..
V. P. (Bill) Evangelou and W. O. Thorn
307
311
313
. 314
Rural Issues: Noncoal Mining and Abandoned
Land Reclamation 337
Gary Uebelhoer
Noncoal Mining and Reclamation (Current and
Abandoned Operations) in the Tennessee River
Basin 340
Jack A. Muncy
Phosphate and Peat Mining in Florida 342
Carol J. Fall
Water Quality Problems Caused by Abandoned
Metal Mines and Tailings 344
John Ford
SALINITY: A NONPOINT SOURCE PROBLEM
Managing Headwater Areas for Control of
Sediment and Salt Production from Western
Rangelands 347
William L. Jackson, Eric B. Janes, and Bruce P. Van
Haveren
Salinity: Nonpoint Source Problem in the Colorado
River Basin 352
Al R. Jonez
Continuous Salinity Monitoring in the Colorado
River Basin by the Utah Bureau of Water Pollution
Control 356
Roy D. Gunnell
Salinity Control in the Grand Valley of Colorado 359
Frank R. Biggie and Larry N. Kysar
RURAL ISSUES: IMPACT ON SMALL
COMMUNITIES
Monitoring the Managers: A Community Enterprise . 317
Harry Manning
Southeast Minnesota's Karst Topography Leads to
Ground Water Pollution from Nonpoint Sources .... 319
Loni Kemp
RURAL ISSUES: SILVICULTURAL NONPOINT
SOURCE POLLUTION
U. S. Department of Agriculture's Perspective on
Silvicultural Nonpoint Source Water Quality
J. Lamar Beasley and Warren C. Harper
Implementing the Public/Private Nonpoint Source
Management Partnership: A State Forestry
Perspective
floger L. Davis and Robert L Miller
The Forest Industry's Perspective of 208.
Dale J. McGreer
321
325
330
Controlling Nonpoint Source Pollution from
Silvicultural Operations: What We Know and Don't
Know 332
George W. Brown
NONCOAL MINING AND ABANDONED LAND
RECLAMATION
Crushed Stone Quarries and Land Reclamation.... 335
F. A. Renninger
LAND USE ISSUES: MANAGEMENT AND
ASSESSMENT
Practical Guidelines for Selecting Critical Areas for
Controlling Nonpoint Source Pesticide
Contamination of Aquatic Systems 363
R. P. Maas, M. D. Smolen, S. A. Dressing,
C. A. Jamieson, and J. Spooner
A Method for Prioritizing Water Quality Problem
Areas
R. A. Young, C. A. Onstad, D. Bosch, and
W. P. Anderson
368
The Impact of Nonpoint Source Fecal Loading on
Backcountry Waters in Grand Canyon 374
Brock Tunnicliff and Stanley K. Brickler
The Use of Wetlands in Treating Nonpoint Source
Pollution 380
Peter R. \MllenbringandWilliamD. Weidenbacher
Antidesertification of Riparian Zones and Control
of Nonpoint Source Pollution 382
Quentin D. Skinner, Jerrold L Dodd, J. Daniel
Rodgers and Michael A. Smith
CASE STUDIES
Highway Runoff Drainage Impacts 387
Byron N. Lord
Rock Creek Rural Clean Water Project: The
Experiment Continues 391
Michael J. Neubeiser
VIII
-------
Regulating Nonpoint Sources of Pollution from
i Timber. Harvesting—A Case History of the
California Experience 397
Carton S. Yee
Agricultural Nonpoint Source Studies in
Southeastern Watersheds: Field Monitoring and
Farmer Surveys 402
F. J. Humenik
Vermont's LaPlatte River Watershed Project:
Lessons Learned 408
Jeffrey D. Mahood
MAKING DECISIONS ABOUT NONPOINT
SOURCE POLLUTION
Point/Nonpoint Source Trading Program for Dillon
Reservoir and Planned Extensions for Other Areas 413
Tom Elmore, John Jaksch, and Donna Downing
Optimizing Point/Nonpoint Source Tradeoff in the
Holston River Near Kingsport, Tennessee 417
Mahesh K. Podar, John A. Jaksch, Stuart
L Sessions, John C. Grossman, Richard J. Ruane,
Gary Mauser, and David E. Burmaster
Protecting Tillamook Bay Shellfish with Point/
Nonpoint Source Controls 425
John E. Jackson
Point/Nonpoint Source Interface Issues in
Wisconsin 426
Bruce Baker and Steven Skavroneck
DATA AVAILABILITY AND NEEDS
A Data Management System to Evaluate Water
Quality Impacts of Nonpoint Source Pollution
Control 429
M. D. Smolen, S. A. Dressing, R. P. Maas,
J. Spooner, C. A. Jamieson, A. D. Newell, and
F. J. Humenik
Development of a Nonpoint Source Data Center 433
Claire M. Gesalman
Water Quality Data And Urban Nonpoint Source
Pollution: The Nationwide Urban Runoff Program... 437
Diane Niedzialkowski and Dennis Athayde
The RFF National Data Base for Nonpoint Source
Policy Assessments 442
Leonard P. Gianessi
WATER QUALITY CRITERIA AND STANDARDS
Bacterial Water Quality and Shellfish Harvesting
Elaine A. Glendening
447
Evaluation of Nonpoint Source Impacts on Water
Quality of Forest Practices in Idaho: Relation to
Water Quality 455
Stephen B. Bauer
Illinois Agricultural Soil Erosion Control Standards:
A Useful Tool for Nonpoint Source Pollution Control. 459
Harry Hendrickson, George Deverman, and Jim
Pendowski
Ground Water Quality Standards.
U. Gale Mutton
464
SOURCES AND FATES OF MATERIAL
INFLUENCING WATER QUALITY IN THE
AGRICULTURAL MIDWEST
Management Practices to Reduce Farm Chemical
Losses with Agricultural Drainage 467
James L Baker
The Fate of Materials Exported by the Big Blue and
the Black Vermillion Rivers into Turtle Creek
Reservoir, Kansas 471
J. R. Shuman, G. R. MarzoH, and J. A. Arruda
The Interaction of Biological and Hydrological
Phenomena that Mediate the Qualities of Water
Draining Native Tallgrass Prairie on the Konza
Prairie Research Natural Area 478
J. Vaun McArthur, Martin E. Gurtz, Cathy M. Tate,
and Frank S. Gilliam
Implications of Airshed Processes and
Atmospheric Deposition of Nonpoint Pollutants 483
One L Loucks
CROSS BOUNDARY NONPOINT SOURCE
POLLUTION: THE IMPLICATIONS
Great Lakes Pollution from Land Use Activities..
Norm Berg
Irrigation Return Flows and Salinity Problems in
the Colorado River Basin
Mohamed T. EI-Ashry
Agricultural Nonpoint Source Pollution in the
Midwest
Robert D. Walker
CONTRIBUTED PAPERS
The Effects of Carbonate Geology on Urban
Runoff: Water Quality Aspects
Jack D. Milligan and Roger P. Betson
Using In-Stream Monitoring Stations To Evaluate
Pollution from Urban Runoff
Leland L Harms
Conservation Service Field Office Program
Delivery by Hydrologic Areas
Stephen F. Black
.. 487
495
497
499
502
506
Agricultural Land Improvement and Water Quality
in South Central Minnesota 508
Henry W. Quade
IX
-------
SPONSORS
U.S. Environmental Protection Agency
U.S. Department of Agriculture
Soil Conservation Service
Forest Service
U.S. Department of Interior
U.S. Department of Transportation
Federal Highway Administration
U.S. Army Corps of Engineers
Tennessee Valley Authority
North American Lake Management Society
COSPONSORS
National Association of State Departments of Agriculture
National Association of Conservation Districts
The Fertilizer Institute
American Forestry Association
The American Society of Mechanical Engineers
American Water Resources Association
Trout Unlimited
American Society of Civil Engineers
Environmental Policy Institute
Land Improvement Contractors of America
American Farm Bureau Federation
Society of American Foresters
The Wildlife Society
National Association of Home Builders
American Forest Institute
Wildlife Management Institute
National Corn Growers Association
The Conservation Foundation
Chemical Manufacturers Association
Agricultural Research Institute
Forest Farmers Association
National Stone Association
National Association of Water Research Centers and Institutes
American Society of Agricultural Engineers
National Wildlife Federation
Soil Conservation Society of America
International Association of Fish and Wildlife Agencies
National Cattlemen's Association
American Farmland Trust
Natural Resources Council of America
Resources for the Future
American Fisheries Society
The National Grange
Association for State and Interstate Water Pollution Control Administrators
National Milk Producers Federation
American Institute of Biological Sciences
National Woodland Owners Association
-------
National Audubon Society
National Agricultural Chemicals Association
Guest Speakers:
The Honorable Pat Roberts
U.S. House of Representatives
Kansas—1st District
The Honorable Arlan Stangeland
U.S. House of Representatives
Minnesota—7th District
The Honorable Frank W. Naylor, Jr.
Under Secretary for Small Community and Rural Development
U.S. Department of Agriculture
Steering Committee
Charles Boothby, Executive Vice President, National Association of Conservation Districts
Robert Broadbent, Assistant Secretary for Science and Water, Department of Interior
(represented by Wayne Marchant and Tom Fair)
Frank X. Browne, F.X. Browne Associates, Inc., representing the North American Lake
Management Society
James Giltmier, Washington Representative, Tennessee Valley Authority
J. B. Grant, Executive Secretary, National Association of State Departments of Agriculture
Robert J. Johnson, Tennessee Valley Authority, Past President of North American Lake
Management Society
Peter C. Myers, Assistant Secretary for Natural Resources, U.S. Department of Agriculture
(represented by Walt Rittall)
Jack Ravan, Assistant Administrator for Water, U.S Environmental Protection Agency
(represented by Carl Myers and Dan Burrows)
Senator Paul Trible, Virginia (represented by Libby Whitely)
Program Committee
Robert J. Johnson, chairman
Tennessee Valley Authority
and Past President of North American Lake Management Society
Brenda Dawson, co-chair
The Fertilizer Institute
Edwin H. Clark II
The Conservation Foundation
Melinda Cohen
National Association of State Foresters
Al Duda
Tennessee Valley Authority
Hal Hiemstra
National Association of State Departments of Agriculture
Patricia Hill
National Paper Institute
Maureen Hinkle
National Audubon Society
Larry Isaacson
Federal Highway Administration
xi
-------
Henry Peskin
Resources for the Future
Mark Rey
National Forest Products
Richard Stump
U.S. Forest Service
Conference Coordinator Proceedings Editor
Judith F. Taggart M. Lynn Moore
JT&A, Inc. JT&A, Inc.
Washington, D.C. Washington, D.C.
Conference Logistics Rapporteurs
Wayne Poppe Joe Greenburg
Tennessee Valley Authority Scott Carr
Chattanooga, Tennessee Rhodes Resource Center
Local Arrangements Clemson University
Lynn Kring Conference Liaison
U.S. Environmental Protection Morris Kay
Agency Administrator, Region VII
Kansas City, Missouri U.S. EPA
Kansas City, Missouri
Session Chairs and Co-Chairs
Opening Plenary: Jack Ravan, Regional Administrator, Region IV, U.S. Environmental Pro-
tection Agency, moderator
Panel Discussion I: Cecil Andrus, former Secretary of Interior, moderator.
Monitoring and Assessment Techniques: Thomas Grizzard, Occoquan Monitoring Lab, Oc-
coquan, Va.; Al Cassell, Vermont Water Resources Research Center, Burlington.
Legal Aspects of Nonpoint Source Pollution Control: Richard Wedepohl, Wisconsin Depart-
ment of Natural Resources, Madison; Robert J. Johnson, Tennessee Valley Authority, Knox-
ville.
State Nonpoint Source Pollution Control: Frank X. Browne, F.X. Browne Associates, Inc.,
Lansdale, Pa.; Ralph Hazel, Region VII, U.S. Environmental Protection Agency, Kansas City,
Kans.
Institutional/Financial Aspects of Nonpoint Source Controls: Wesley D. Seitz, University of
Illinois, Urbana-Champaign; John G. Konrad, Wisconsin Department of Natural Resources,
Madison.
Contributed Papers: Greg Knauer, Booker Associates, St. Louis, Mo.
Ground Water Quality: Robert Schoen, U.S. Geological Survey, Reston, Va.; Paula Magnu-
son, Geraghty and Miller, Newtown, Pa.
Lake Quality: T. Al Austin, Iowa State Water Resources Research Institute, Ames; William W.
Walker, Jr., Consultant, Concord, Mass.
Estuarine Quality: Virginia Tipple, Chesapeake Bay Program, Annapolis, Md.; Carol Jolly,
Puget Sound Action Program, Seattle, Wash.
Streams and Rivers: David B. Baker, Heidelberg College, Tiffin, Ohio; Lee Mulkey, U.S.
Environmental Protection Agency, Athens, Ga.
Livestock Waste Management: Bart Cardon, University of Arizona, Tuscon; Ron Michieli,
National Cattlemen's Association, Washington, D.C.
Nonpoint Runoff Programs—The Status: Peter Wise, Great Lakes Program , Chicago, III.
xii
-------
Economics of Nonpoint Source Pollution: Edwin H. Clark II, The Conservation Foundation,
Washington, D.C.
Agricultural Issues—Eastern and Southern Experience: Stuart Smith, Maine Agricultural
Commissioner, Augusta, Me.
Urban Issues—Runoff: Martin Wanielista, University of Central Florida, Orlando.
Rural Issues—Coal Mining and Abandoned Land Reclamation: Donald Graves, University
of Kentucky, Lexington.
Workshop: Land and Water 201 Project, Tennessee Valley Authority: Porter Russ, Tennes-
see Valley Authority, Knoxville.
Salinity—A Nonpoint Source Problem: David H. Robbins, Hill & Robbins, Denver, Colo.;
Jack Barnett, Colorado River Basin Salinity Forum, Bountiful, Utah.
Agricultural Issues—Midwestern Experience: Tom Davenport, Region V, U.S. Environmental
Protection Agency, Chicago, III.
Urban Issues—Construction Nonpoint Source Pollution: David Moreau, North Carolina Wa-
ter Research Center, Raleigh.
Rural Issues—Noncoal Mining and Abandoned Land Reclamation: Roger Bellinger, Tenne-
see Valley Authority, Chattanooga, Tenn.
Rural Issues—Impact on Small Communities: Beth Ytell, Great Lakes Rural Network, Fre-
mont, Ohio.
Land Use Issues—Management and Assessment: Eben Chesebrough, Massachusetts Divi-
sion of Water Pollution Control, Westboro, Mass.; Robert Kirschner, Northeastern Illinois
Planning Commission, Chicago, III.
Agricultural Issues—Western Experience: Dennis Grams, Nebraska Environmental Control,
Lincoln, Neb.
Urban Issues—Hydrologic Modification and Septic Tanks: James Ruane, Tennessee Valley
Authority, Chattahooga, Tenn.
Rural Issues—Silvicultural Nonpoint Source Pollution: Fred Haeussler, President, Society of
American Foresters, Union Camp Corp., Savannah, Ga.
Case Studies: Byron Lord, Federal Highway Administration, McLean, Va.
Making Point/Nonpoint Abatement Tradeoffs: Carl Myers, U.S. Environmental Protection
Agency, Washington, D.C.
Data Availability and Needs: Henry Peskin, Resources for the Future, Washington, D.C.;
Michael Smolen, North Carolina State University, Raleigh, N.C.
Water Quality Criteria and Standards: Alan Klimek, North Carolina Water Resources Re-
search Center, Raleigh; Ron Jarman, Oklahoma Water Resources Board, Oklahoma City.
Sources and Fates of Materials Influencing Water Quality in the Agricultural Midwest: G.
Richard Marzolf, Kansas State University, Manhattan; John Howland, Missouri Department
of Natural Resources, Jefferson City.
Cross Boundary Nonpoint Source Pollution—The Implications: Mohamed EI-Ashry World
Resources Institute, Washington, D.C.
Closing Plenary: Nonpoint Source Pollution Programs—Where We're Going: Robert Broad-
bent, Assistant Secretary for Water and Science, Department of the Interior.
xiii
-------
The Policy Perspective
A Look to the Grass Roots
WELCOME TO CONFERENCE
LEE M. THOMAS
Administrator
U.S. Environmental Protection Agency
We have made steady progress toward attaining the Na-
tion's water quality goals since the passage of the Clean
Water Act of 1972. Much of this forward movement has
been accomplished by controlling industrial and municipal
point sources. Further achievement will require acceler-
ated implementation of nonpoint source management pro-
grams in addition to our ongoing point source control ef-
forts.
Many States and local governments have already taken
steps to address their nonpoint source challenges. Given
the nature of nonpoint source pollution, State and local
management is a key. Only at this level does enough flexi-
bility exist to make site-specific and source-specific deci-
sions that really work.
Of course, EPA and other Federal agencies have an
important role as well. Our nonpoint source pollution con-
trol program is getting increasing attention as we imple-
ment recommendations of our interagency Nonpoint
Source Task Force established a year ago. The Task
Force's national policy provides direction for future initia-
tives by Federal, State, and local agencies, and, most
importantly, by the private sector.
We intend to incorporate nonpoint source concerns into
all aspects of water management. It is EPA's job to provide
national coordination and oversight, give practical assist-
ance for nonpoint program development, and promote in-
novation. We are intensifying our efforts in each of these
areas. We will continue to work with other Federal agen-
cies, such as the Department of Agriculture, to better use
their existing programs to address nonpoint source needs.
In a report to Congress in 1984, EPA summarized what
is known about nonpoint source pollution, concluding that
it is among the leading causes of the Nation's remaining
water quality problems. Specifically, the report said that in
six of 10 EPA regions nonpoint sources are the principal
remaining cause of water quality problems. Half of the
States say that nonpoint pollution is a significant source of
their difficulties, and virtually every State reports some
kind of water quality problem related to these sources.
Research suggests that lakes, reservoirs, and estuaries,
like Chesapeake Bay, are particularly vulnerable to non-
point pollutants.
The report identified agricultural operations as the most
pervasive nonpoint source in every region. Nonpoint
source impacts from urban areas, mining, forestry activi-
ties, and construction sites also deserve attention.
As you well know, managing nonpoint source pollution
is not easy, institutionally or technically. Nonetheless, ef-
fective steps can be taken to control it. The basic ap-
proach under the Clean Water Act for managing point
sources—technological controls for classes of discharg-
ers—is not appropriate for nonpoint sources. Instead, flex-
ible site-specific and source-specific decisionmaking is
the key to success.
States must take the lead in managing nonpoint
sources because they have the adaptability, perspective,
and intimate knowledge to develop such site-specific solu-
tions. They can easily reach individual landowners and
operators and help them change the way they manage
their land.
Experts at this level are in the best position to determine
which surface or ground water problems are related to
nonpoint sources, establish which waters will receive pri-
ority attention, determine what type of control strategy is
needed, and evaluate progress.
Substantial, cost-effective water quality improvements
have been made by carefully targeting control activities.
Targeting schemes need to identify the principal sources
of nonpoint pollutants as well as determine which water-
bodies are most likely to benefit from intensive work.
Recent studies indicate that off-site impacts of erosion
cost the Nation an estimated $6 billion a year, with over $2
billion accounted for by cropland erosion alone. These
costs include:
• waterways polluted by sediment and agricultural
chemicals,
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
• destruction of breeding grounds for fish,
• increased expenses for dredging harbors and treat-
ing wastewater,
• higher riverbeds leading to greater flooding, and
• reservoirs and lakes silting up more quickly than an-
ticipated.
Although it takes resources to address nonppint prob-
lems, direct and indirect costs are clearly associated with
nor coming to grips with this problem.
An area of growing concern is toxics, from pesticides
and other chemicals applied to the land, entering ground
water. A new study of pesticides in drinking water drawn
from ground water, now in the design phase, will provide a
national picture of the extent of the problem. We are also
working on policies to reduce this potential threat to drink-
ing water supplies.
A new report to be completed this fall by the Association
of State and Interstate Water Pollution Control Administra-
tors will provide an important baseline of data on State
nonpoint problems and the status of current State man-
agement efforts. This information will be used to better
assess nonpoint problems and as a basis for policy deci-
sions.
The challenge for this conference is to exchange know-
ledge of the nature of nonpoint problems and what control
approaches work. I am hopeful that this conference will be
a turning point for nonpoint source management—that the
exchanges of ideas here will result in a surge of aware-
ness and commitment to nonpoint implementation efforts
at all levels.
On a final note, I want to emphasize that nonpoint
source control is at the top of EPA's agenda; it is clearly
identified as a priority issue in EPA's Agency Operating
Guidance for FY 1986-87. We are committed to work with
States to incorporate nonpoint control measures into their
water quality programs. This issue must receive attention
at all levels of government; but a more aggressive ap-
proach at State and local levels, in concert with the private
sector, is absolutely essential for substantial progress.
With that commitment we will eventually get a grip on this
persistent and growing problem and begin to bring it un-
der control.
-------
KEYNOTE ADDRESS
PAT ROBERTS
U.S. House of Representatives
First District, Kansas
I noticed from the program that this conference's co-spon-
sors are a very diverse group. I think that is very appropri-
ate and in keeping with the purpose of the conference—to
gather practical information from the grass roots level. Too
often in Washington we create laws and programs without
listening to those back home who know better.
I thought today I would try to bring you up to date on
what is happening in Washington in regard to the nonpoint
source pollution issue and make a comment from my per-
spective on what needs to be done. In terms of prece-
dence the budget and the farm bill come first.
But let me give you a little background. I represent a
district that produces more wheat than any other State.
My district, the First District of Kansas or as we call it "the
Big First," is larger than the State of Virginia. From the
time that our pioneer forefathers brought "Turkey Red
wheat" to Kansas in the 1870's, we have excelled in pro-
ducing hard red winter wheat. In addition to wheat, how-
ever, my district is number one in the production of grain
sorghum and the cattle industry is a vital segment of our
economy. As a matter of fact, I noticed the other day a
"Washington Post" story on the 20 counties in the Nation
that are most dependent on farming as a source of in-
come. In that list of 20 counties, I have the privilege of
representing five. Needless to say, the difficult times we
are experiencing in farm country have had a severe im-
pact on my district and the high plains.
As a footnote, my district during the "Dirty 30's" was
always on the move. One day it might be blowing into
Nebraska and the next day back into Oklahoma. That was
back in the days before we called the Kansas wind a
nonpoint source of pollution. Perhaps air coming from
Washington is a point source!
Our number one priority this year is writing a farm bill
that will put a profit back into agriculture. Without profit
any rural management plan be it local, county, State or
Federal will not be successful! The very existence of the
farmer-stockman, agribusiness and main street rural
America is threatened today by the continuing problems of
the budget deficit, low commodity prices, and the high
interest rates. I won't go through the long list of problems
that have plagued the farmer. Instead I am going to try to
outline some solutions.
First the budget. Two weeks ago the Senate passed a
budget resolution that does represent a ray of hope. The
budget package calls for $56 billion in cuts in 1986 and
about $295 billion over the next 3 years. It effectively
freezes defense spending, and provides for a one year
cost of living adjustment freeze on Social Security, Vet-
erans, and Military/Civilian Retirement. It does not call for
tax increases.
The package has a long way to go. It is our turn now in
the House, but it is a good start towards reducing the
deficit, bringing interest rates down, and keeping the
economy ona steady path.
Specifically for agriculture, the budget was very posi-
tive. It added $3.5 billion back to earlier agricultural
budget proposals. It provides for a 2 percent matching
interest rate buy-down for credit-strapped farmers, re-
stored some funding to soil and water conservation, pro-
vided $1 billion per year for farm credit guarantees and a
new export incentive program using a billion dollars worth
of CCC stocks to counter foreign subsidies and get our
grain competitive in world markets. While this budget is
not the across the board freeze I have been supporting, it
is a major step in getting our Nation's fiscal house in order.
SEEDS OF RECOVERY
Let me turn now to the farm bill. Because of the budget
battle, work on the farm bill has been delayed. The farm
bill we write will have to be budget responsible. Given the
limited budget, one of the major hurdles we face is how to
be competitive and regain export markets without bank-
rupting a whole generation of farmers.
In spite of the tight budget, there seems near unani-
mous support in Congress this year for some type of long-
term land retirement program to take highly erodible crop-
land out of production. I predict that the farm bill signed
into law later this year (I hope it gets done this year) will
have a long term land retirement program that will take up
to 20 million highly erodible acres out of production for 10-
to 15-year periods. I also predict that the farm bill will
contain strong "sod buster" language to end the current
policy of rewarding farmers who plow up fragile land.
These two programs will go a long way in reducing soil
and water erosion and hopefully in controlling nonpoint
sources of pollution from agriculture.
This is not to say that agriculture's role regarding non-
point pollution will not be significant or without contro-
versy. In their 1984 report to Congress, EPA identified agri-
cultural operations as the most pervasive nonpoint source
in every region of America. As a result of this report and
our substantial gains in controlling point source pollution,
attention has once again focused on nonpoint pollution as
a problem that must be addressed.
It is the opinion of this member of Congress that the
most effective control of nonpoint source pollution can
best be accomplished with Federal help at the State and
local level. States must take the lead in managing non-
point sources because they have the adaptability, per-
spective, and knowledge to develop appropriate site-spe-
cific solutions. The last thing the farmer needs in these
difficult economic times is a massive new set of Federal
regulations to tell him how to control runoff. Let's get the
problem solved—let us not repeat EDB.
On May 2, the Senate Environment Committee rejected
a provision during consideration of the Clean Water Act to
set specific limits for pesticides, fertilizers, and other point
source pollutants. However, the Senate Committee did
adopt a provision to require States to establish a manage-
ment program and authorized $300 million in grants to
help States set up the programs. Action on the Clean
Water Bill is still pending in the House.
Farmers are faced with the challenge of surviving in a
very competitive industry. In an effort to reduce costs, the
use of conservation and minimum tillage is on the rise.
However, this has the downside risk of increasing pesti-
cides and herbicides use to control what tilling used to
control. One of the best nonpoint pollution controls is pro-
moting sound conservation practices.
I have always felt that the farmer is the true conserva-
tionist. But in these perilous economic times, the farmer is
often forced to choose between building terraces and pay-
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
ing the mortgage. I urge this conference to bear in mind
throughout your discussions that the farmer is undergoing
a cost-price squeeze bear hug! Any new policies for con-
trolling nonpoint pollution must not impose heavy financial
burdens on the agricultural community. You must keep in
mind the cost versus the benefits of nonpoint control.
And, you cannot expect the farmer to bear the entire
cost of controlling nonpoint sources of pollution while most
of the benefits will accrue to society as a whole. If we must
have expensive new control methods, society must share
in the cost. With the record budget deficits we have in
Washington, the money may not be there to help. Again,
that is why we must make this effort one of a partnership.
Once again, thank you for the invitation. My final
thought is best summed up by a statement from EPA Ad-
ministrator Lee M. Thomas:
Nonpoint pollution ... must receive the attention it de-
serves at all levels of government, but a more aggressive
approach at State and local levels is absolutely essential
for substantial progress. With that commitment, the Na-
tion will eventually get a grip on this persistent problem
and ensure that continued progress is made towards
meeting our water quality objectives.
-------
NONPOINT SOURCE POLLUTION—A PROBLEM FOR ALL
JOHN R. BLOCK
Secretary
U.S. Department of Agriculture
Washington, D.C.
On May 20, 1927, a young American aviator took off from
a New York airfield—alone—and headed for Paris,
France. That historic flight across the Atlantic by Charles
Lindbergh helped set the stage for a technological revolu-
tion in space that still goes on today. But equally impor-
tant, it reminds us that American know-how, American
initiative, and American success, are based on one impor-
tant ingredient: the determination that makes Americans
willing to take a chance.
I prefer to think of those chances as challenges. And
when it comes to nonpoint source pollution, I suspect that
many of those challenges will require as much determina-
tion as Lindbergh needed if we are to really succeed in our
role as stewards of our natural resources.
It's no secret that we have serious water quality prob-
lems all across this country. The sources of these prob-
lems cut across every segment of our society, including
agriculture and governmental policies. But nothing can be
gained by each of us pointing a finger at someone else.
Likewise, nothing will be accomplished toward correcting
these problems by complaining that someone else is not
doing enough. Rather, the true measure of success will
come only after we have cast aside such judgmental
temptations and have joined together to make maximum
use of our limited resources.
The Department of Agriculture is celebrating the 50th
anniversary of the soil and water conservation movement
this year. Over the years, we have faced many challenges
as stewards of our Nation's soil and water resources. And
we are understandably proud of the accomplishments
made through USDA programs in meeting those chal-
lenges.
Nonpoint source pollution control is one specific chal-
lenge that has come to the forefront in recent years. Con-
trary to what some may believe, we have not shied away
from this challenge in the Department of Agriculture. We
have been providing financial and technical assistance—
as well as a proven educational delivery system—all
along. Those efforts are part of our mission. We call it
conservation. And we shall continue to fine tune our ef-
forts and adjust to meet new goals as they are estab-
lished.
Certainly, preserving and protecting the quality of our
water resources is now, and shall continue to be, an im-
portant part of this overall effort. We know about the chal-
lenges. We know about the limitations on available re-
sources. And, we know that we need your
cooperation—and your ideas—to implement a workable
strategy that will contribute to raising the quality of our
water supplies.
The President's 1982 National Soil and Water Conser-
vation Program established nine priorities for the use of
USDA soil and water conservation program funds. Our
commitment to solving water problems is second only to
erosion control. In 1984, our USDA conservation agencies
spent $66 million to improve water quality alone. We are
indeed committed to improving the quality of our Nation's
water supplies—within the limits of our financial resources
and our traditional responsibilities.
Currently we are looking at the off-site effects of soil
erosion, particularly as it concerns water quality. We are
also funding special studies to look at nonpoint source
pollution relationships to ground water quality. We have
our work cut out for us. Where State and local officials
have identified water quality to be more important than
gross soil erosion, we stand ready to target our resources
into nonpoint source pollution from agriculture.
Winston Churchill once said: "You can always count on
the Americans to do the right thing...after they've tried
everything else." Well, I think this is the time and the place
to prove Mr. Churchill wrong. Let's not wait until we've
each tried everything else.
Let's begin working closer together now, joining forces,
to find out what does and does not succeed; and then let's
draw upon that combination of good, old-fashioned Ameri-
can determination and modern technology to solve our
water quality problems.
-------
A CONGRESSIONAL VIEWPOINT ON NONPOINT SOURCE
POLLUTION
ARLAN STANGELAND
U.S. House of Representatives
7th District, Minnesota
I commend the organizers of the conference on nonpoint
source pollution for scheduling 3 days of intense study
and discussion of what is a growing and increasingly visi-
ble problem. With the mechanisms for point source pollu-
tion largely in place, although certainly not without flaws, a
major thrust is needed to address nonpoint source pollu-
tion. The beginnings of that thrust are reflected to a signifi-
cant degree in the 1985 amendments to the Clean Water
Act, which are receiving committee action in Congress
now.
The Senate Environment and Public Works Committee
has reported its version of the Clean Water Act amend-
ments, and the Water Resources Subcommittee of the
House expect to mark up our own bill with full committee
action to follow. The bill is designed to significantly im-
prove the effectiveness of the Nation's water pollution
control program, and nonpoint source pollution control is a
very important part of this legislation. And well it should
be.
We have been at this procedural point before, of course,
with a very similar legislative vehicle. I am hopeful that this
year, unlike last year, clean water legislation will reach the
floor in both houses. However, while the Senate commit-
tee has indeed reported its bill, filed its report, and sched-
uled a tentative date for floor consideration, some 20
holds have been placed on the bill as reported, mostly
because of disagreements over the allotment formulas.
Despite our disappointment in not finalizing Clean Wa-
ter Act amendments last congress, I think it is fair to say
that the time spent on this issue has been time well spent.
As a quick summary, our subcommittee has developed
over the past 3 years an extensive record on possible
amendments to the Clean Water Act. Following the sign-
ing into law of the Municipal Wastewater Treatment Con-
struction Grant Amendments of 1981, the subcommittee
held 5 days of hearings on the Clean Water Act in 1982;
approximately 2,400 pages of testimony were received.
During the 98th Congress, we held an additional 15 days
of hearings on the same subject, receiving more than
3,700 pages of testimony.
With that very substantial hearing record, we reported a
bill and brought it to the House floor, where it passed on
June 26,1984, by the overwhelming bipartisan vote of 405
to 11. However, the Senate failed to bring its bill to the floor
before adjournment.
This year, Chairman Jim Howard has introduced H.R. 8,
of which I am a cosponsor, and which could reauthorize
the Clean Water Act into the next decade, including a
number of significant new programs and improvements in
many existing ones. Although this bill is quite similar to
that which our committee reported and which the House
passed in the last congress, substantial changes have
been made. In fact, as the result of weeks of study, includ-
ing 2 days of hearings, our subcommittee has developed
new language taking into account recommendations of
the Administration and affected interest groups as well as
provisions in the Senate bill.
Consequently, the bill we will be marking up in subcom-
mittee reduces the authorizations for the Construction
Grants Program from the $3.4 billion annually contained in
last year's bill to $2.4 billion per year in FY 1986-90, re-
taining the Federal share at 55 percent rather than raising
it to 65 percent as the House bill proposed earlier.
Grants for State water pollution control revolving funds
would be cut from $1.6 billion annually to $600 million
annually for FY 1986-90. Then, when the construction
grants cease, $6 billion would be authorized for the revolv-
ing fund program over the next 4 fiscal years.
NPDES permits would continue for a maximum of 5
years, except in those cases where nontoxics are in-
volved, or only insignificant amounts of toxics and no ad-
verse effects on the environment. For these cases, per-
mits could be for 10 years, but quality standards would still
apply.
The 4 percent set aside of construction grant funds for
rural areas would be increased. States that have 25 per-
cent or more of their population in rural areas will be able
at the Governor's request, to use from 4 percent to 7.5
percent of their State allotment under the Program for
Alternatives to Conventional Treatment. The Senate bill
would make no change in the current 4 percent set aside.
Fundamentally different factors (PDF), at a facility, vari-
ances from the best available technology based on the
presence of fundamentally different factors from those
considered by EPA in developing the best available tech-
nology (BAT) effluent guidelines, could continue to be
granted, but only in those cases where the facility involved
furnished information to EPA during the rulemaking or did,
not have a reasonable opportunity to do so.
Of course, when we get to subcommittee and full com-
mittee markup, amendments could be added to our sub-
committee's preliminary deliberations on the bill. We have
a number of new programs with large price tags in this bill,
and even though we have pared funding back in a number
of programs, some programs may not survive in confer-
ence. Moreover, even if this legislation authorizes funding
at higher levels, the budget process might impose addi-
tional limits on the appropriation committee's ability to ap-
propriate funds above current levels.
The legislation now before our committee, like its prede-
cessor in the 98th Congress, reflects what has become
the conventional wisdom that the uses desired for our
Nation's rivers and streams will not be achieved without
control of nonpoint sources of pollution. We have not lost
sight of the fact that the 1972 act had as one of its goals
the achievement by July 1, 1983, wherever attainable, of
fishable and swimmable water quality in all of the rivers,
lakes, and streams of this Nation. In the past, the primary
thrust to achieve this goal has been through the discharge
of pollutants from point sources. We are rapidly learning,
however, that point sources are not the whole problem,
and unless the problem of nonpoint pollution is solved,
many rivers and lakes will not be able to meet this fish-
able-swimmable goal.
Nonpoint source pollution is an enormous problem for
our farmers, just to cite one example, both in terms of the
loss of billions of tons of topsoil and the degradation of
water quality in nearby streams and lakes. Millions of
acres of productive farmland are removed from cultivation
each year because of eroded soils. By the same token, the
-------
THE POLICY PERSPECTIVE—A LOOK TO THE GRASS ROOTS
herbicides, pesticides and nutrient-rich fertilizers that flow
in streams along with the eroding topsoil destroy aquatic
life. It poses a strong land management challenge, and
one which must be met.
This problem underlines the urgency of seeking meth-
ods of controlling nonpoint source pollution to provide the
desired environmental benefits without placing intolerable
operational cost burdens on the agricultural community.
In many areas, throughout this country, nonpoint
sources are the major cause of water pollution. In fact,
estimates are that more than half of all the pollution in the
Nation's streams comes from nonpoint sources. More
specifically, the Environmental Protection Agency testified
a few years ago that of the Nation's 246 river basins, 68
percent were affected wholly or in part by agricultural run-
off, 52 percent by urban stormwater runoff, and 30 per-
cent were by mine runoff.
The threat posed by nonpoint sources, as well as point
sources, makes it clear that we need a balanced approach
to the problem of water pollution control in general. H.R. 8
underlines the point well at the outset by expressing that
the National policy plans for the control of nonpoint source
pollution be developed and implemented in an expeditious
manner, so that the goals of the act may be met through
the control of both point and nonpoint sources of pollution.
In other words, as a National policy, we should control
point and nonpoint sources in a balanced manner.
And once nonpoint source pollution is given its proper
priority, it is important that the States play a role in the
planning and implementation of the required nonpoint
abatement measures. Land use management has tradi-
tionally been a State role, and, while the Federal Govern-
ment has a legitimate interest in addressing certainly a
National problem, the States should be permitted to de-
velop their own programs and management practices.
In H.R. 8, we give them that responsibility. We require
States to set up programs to take a look at the problem of
nonpoint source pollution, to examine the courses of
action that might be taken and the alternatives available to
deal with the problem.
The bill provides some important funding authorizations
for programs dealing with the control of nonpoint sources.
It reauthorizes the existing section 208 areawide planning
and clean lakes programs and provides some major new
initiatives.
One such new initiative is a program of grants to States
to control nonpoint sources of pollution, for which $150
million would be authorized through 1990. States would
be required to develop and implement nonpoint source
pollution control plans on a watershed-by-watershed ba-
sis, with the Federal Government providing grants of up to
50 percent to States to implement their plans.
The Federal share could rise to as high as 60 percent if
a significant number of nonfederal and nonstate interests
in a watershed agree to participate in nonpoint source
pollution control measures. In developing and implement-
ing its plan, a State would be required, to the maximum
extent practicable, to use local, public, and private agen-
cies and organizations of expertise in control on nonpoint
sources of pollution.
In a similar fashion, the bill reported by the Senate com-
mittee provides 75 percent grants to assist in the imple-
mentation of approved management programs. The Sen-
ate bill authorizes somewhat lower funding levels: $70
million for 1986, $100 million for FY 1987, and $130 mil-
lion for 1988. In addition, the Senate bill contains a new
set aside of 1 percent of a State's allotment or $100,000,
whichever is greater, for the purpose of carrying out a
State's nonpoint source pollution program.
The cause of nonpoint source pollution control is certain
to be advanced further by a significant change our bill
makes in the discretionary funding provided under the
Construction Grants Program.
The 20 percent of a State's annual allotment is now
available at the Governor's discretion for otherwise ineligi-
ble categories and is specifically available for control of
nonpoint source pollution, including innovative and alter-
native approaches.
Another important nonpoint source provision in H.R. 8
authorizes $100 million annually through 1990 in grants to
States for priority projects designed to control nonpoint
sources of pollution that contribute to the degradation of
water quality in lakes. In addition, the Clean Lakes Pro-
gram would be made applicable to saline, as well as to
fresh water lakes. Federal funding could provide up to 70
percent of the cost of a project implemented under this
provision. This amounts to a major expansion of grant
authority for restoration of the water quality of lakes.
Our House bill also addresses the problem of acid depo-
sition in our lakes and streams. It authorizes $25 million
per fiscal year for each of the FY1986 through 1990 for
grants to States to carry out approved methods and proce-
dures to restore water quality, which has deteriorated be-
cause of high acidity. We also provide $25 million annually
over the same period for a demonstration program to re-
store the biological integrity of acidified lakes and water-
sheds through liming. The purpose would be to determine
the effectiveness of liming in reducing the acidity of lakes
and watersheds, and in restoring their biological integrity.
The bill also extends the Rural Clean Water Program at
a level of $50 million per fiscal year. This program is ad-
ministered by the Department of Agriculture and provides
valuable assistance to farmers to control pollution runoff
from agricultural land.
As a means of improving the water quality of estuaries,
the bill adds a new provision to the Clean Water Act
authorizing the EPA administrator to convene an interstate
management conference where he or she determines that
control of point and nonpoint sources of pollution is
needed in more than one State. The provision is founded
on the definition of the term "estuarine zone," which is
intended to include an entire basin of watershed.
The management conferences would develop a com-
prehensive master plan for the estuary, coordinate the im-
plementation of that plan by participating States, recom-
mend corrective actions to be taken against the most
serious point and nonpoint sources of pollution, and fi-
nally, monitor the program's effectiveness.
To fund these management conferences, H.R. 8 as in-
troduced would have authorized.$195 million over the next
5 years. The bill we expect to mark up will reduce this
amount to $75 million. The Senate's bill has a similar pro-
vision but at an even lower funding level. Under our bill the
Federal grants to participating States or interstate agen-
cies would amount to 55 percent of a State's or agency's
cost of implementing the master plan for each fiscal year.
Estuaries have all too often been the dumping grounds
of much of our National waste. Estuarian habitats are dis-
appearing, and we need to act swiftly to protect these
natural ecosystems while there is still time to act. The
provisions in H.R. 8 help meet that need by providing
important protections for our estuaries.
The problem of ground water contamination from both
point and nonpoint sources is also addressed in H.R. 8. It
authorizes $150 million for each of 1986-88 to provide
grants to public water system operators and units of local
government, to make alternative water supplies available
to users whose water from nonpublic water systems is
made unfit for consumption because of ground water con-
tamination.
The grants can be used for providing these alternative
water supplies on a temporary basis, and for permanent
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
remedies, including drilling new wells and installing new
pipes.
The Federal share will be 50 percent, with an annual
maximum grant of $2 million per applicant. EPA will report
to congress each year on progress made under the grant
. program.
The Agency would also be authorized to make grants to
assist States in carrying out ground water quality protec-
tion activities as part of a comprehensive nonpoint source
pollution control program; $7.5 million each year for 5
years is authorized for this purpose. The activities eligible
for the grant program include research, planning, ground
water assessments, demonstration programs, enforce-
ment, technical assistance, and education and training to
protect the quality of ground water and to prevent contam-
ination of ground water from nonpoint sources of pollution.
The importance of controlling ground water contamina-
tion has been given special emphasis with the establish-
ment of a National Ground Water Commission, which ac-
tually was approved as part of the Resource Conservation
and Recovery Act passed in the 98th Congress and
signed into law last November.
Although the administration has not recommended
funding, the 18-member commission would be responsi-
ble for inventorying the Nation's ground water resources
and the extent of contamination, projecting the future
availability of usable ground water, examining methods for
the abatement and containment of ground water contami-
nation and for aquifer restoration, and assessing the roles
of government (State, local and Federal) in managing
ground water quality and quantity
All in all, our proposed bill is another strong response to
the need to preserve and enhance the quality of our Na-
tion's precious water supplies. Of special importance to
those at the nonpoint conference, it expands the scope of
its coverage to address many of the issues raised by non-
point source pollution.
In many respects, it is a new beginning, but a strong
beginning, as we seek to develop the most cost-effective
and politically feasible ways of dealing with this problem.
I think that the clean water bills under consideration in
the House and the Senate are coming together. The Sen-
ate clearly has a major problem to iron out concerning a
new and highly controversial allotment formula based on
logarithms and logarithms cubed. If they can resolve the
allotment formula and agree to take the bill to the floor, I
believe the chances of enacting meaningful reauthoriza-
tion legislation—legislation that will introduce new direc-
tions in the field of nonpoint source pollution—would be
greatly enhanced.
-------
Perspectives on Nonppint
Source Pollution
NONPOINT SOURCE POLLUTION—THE ILLINOIS APPROACH
LARRY A. WERRIES
Illinois Director of Agriculture
Springfield, Illinois
The 208 Water Quality Management Plan, completed in
January 1979, designated the Illinois Department of Agri-
culture as the lead agency in dealing with nonpoint source
pollution. As major strides are made in reducing industrial
and municipal waste pollution of our rivers, streams, and
lakes, progress in controlling nonpoint pollution appears
to have stagnated and perhaps even been set back.
Today, we are told that soil erosion is worse than in the
1930's. After 50 years of the conservation movement in
this country, we surely should have been able to make a
difference. We have made a difference. The soil erosion
problem is not as severe as it would have been if left
unattended.
Several reasons account for our current depressing sit-
uation, including a lack of financial resources, existing
farm crises, duoculture agriculture, and existing conser-
vation philosophy.
In response to the national clamor to clean up our rivers,
lakes, and streams, the Federal and State governments
have devoted vast financial resources to eliminating in-
dustrial and municipal waste for those waters. At the same
time, funding for soil conservation has been sorely ne-
glected. In fact, the Federal government's role in soil con-
servation has been slowly shrinking for the past two dec-
ades, and State government participation in the effort to
fight erosion has been miniscule.
During the late 1970's and early 1980's, the agricultural
community has put into production considerable amounts
of highly erodible land. These lands were brought into
production in response to the boom years of the 1970's
and the need to produce more to lessen income decline
during the 1980's.
The dependency of farmers on the duoculture of corn
and soybeans has eliminated, in a lot of areas, the old
system of crop rotations. This process has increased ero-
sion.
Frequently, conservationists say, "I want to leave my
land in better shape than I found it," or "I need to maintain
production so that I can contribute to feeding the world."
While very noble, this does not explain the benefits of
conservation to the urban populace. We need to empha-
size off-site benefits such as improving water quality, and
extending the life of lakes by reducing sediment, by reduc-
ing ditch-cleaning activities, by reducing dredging, and so
forth.
Having discussed some of the problems confronting us
in battling nonpoint source pollution, I will describe what
we are doing in Illinois. Initially, assistance to soil and
water conservation districts by State government was lim-
ited essentially to technical and educational assistance.
However, in FY 1985, the State provided approximately
$2.2 million for the districts to operate programs and em-
ploy technical staff. Also, the State has increased funding
for soil surveys from $200,000 in FY 1981 to $614,000 in
FY 1986. As you can see, the State has made a financial
commitment to soil conservation that will increase in the
future.
Since water quality became a national issue and the
208 Water Quality process has been completed, nonpoint
sources of pollution have received much attention. As a
result, the Illinois General Assembly passed the State Soil
Erosion and Sediment Control Guidelines. The State's
guidelines set forth the following:
January 1,1983: soil loss at or below 4T
January 1, 1988: land 5 percent or less slope—T, land
greater than 5 percent slope—2T
January 1, 1994: land greater than 5 percent slope—
1.5T
January 1, 2000: all land at T
Additionally, the State program required each of the 98
districts to prepare and enact their own set of guidelines
by April 18,1982; however, these could not be less restric-
tive than the State guidelines.
Unfortunately, many viewed these standards as the plan
to reach T by 2000. However, the standards are goals or
benchmarks to be achieved by 1988, 1994, and 2000.
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Consequently, the Department of Agriculture requested
that each of the 98 districts prepare a T-by-2000 Plan to
provide a road map for reaching T by 2000.
These local plans have been synthesized to form the
State T-by-2000 Plan. The plan is divided into two parts:
(1) statistical, and (2) action. The statistical part of the plan
details information on technical staff years needed, edu-
cation staff years needed, cost of required resource man-
agement systems, and the number of acres needing treat-
ment. The action part of the plan details programs to be
instituted and members of the local conservation team
responsible for various parts of implementation.
The plan has brought to light some startling information
concerning manpower and financial resources needed to
reach T by 2000 in Illinois. Some of the findings are as
follows:
° Contrary to popular belief, conservation tillage is not
a panacea for our erosion problems. Only 199,315 ha
(498,288 acres) of the 4.64 million ha (11.6 million acres)
needing treatment can be treated solely by conservation
tillage.
° In addition to existing Soil Conservation Service and
district personnel, 197 new staff positions are needed to
provide the necessary technical assistance to reach T by
2000.
0 $1 billion needs to be expended on enduring prac-
tices by 2000 to meet T.
° 72,392 farms will need some conservation practices.
0 1,186 Cooperative Extension Service staff years will
be needed for educational purposes between now and
2000.
As a result of the T-by-2000 planning effort, the Illinois
General Assembly is considering a 5-year $20 million cost-
share program. These monies will be cost-shared with
farmers on the construction of conservation practices.
The program will be divided equally between a traditional
Agricultural Conservation Program approach, where
every county receives an allocation, and a watershed ap-
proach, in which lakes supplying water will receive the
highest priority. The purpose of the watershed approach is
to improve water quality and extend the life of lakes used
as water supplies.
The SCS in Illinois is developing a conservation plan
based on hydrologic units to complement the State's ef-
fort. The hydrologic unit approach will not only allow us to
enumerate on-site benefits but it will enable better docu-
mentation of off-site benefits.
The SCS response comes on the heels of efforts in
Washington to seriously shrink the budget commitment to
the Soil Conservation Service. The logic of cutting funding
when the needs are obviously becoming greater is beyond
me. To cripple the efforts of SCS by reducing funding to
the levels proposed by the Reagan administration seems
to me a serious mistake.
I believe that most everyone in Washington realizes the
problem, with the possible exception of the people at the
Office of Management and Budget. Details are pending on
the mandatory soil erosion control levels recently intro-
duced as amendments to the Clean Water Act in the Sen-
ate. Point source pollutants have faced such regulations
for more than 15 years. However, voluntary compliance
has been our preferred approach for nonpoint source pol-
lution control. Obviously, we have not accomplished what
we might have hoped. If we reduce our commitment to this
effort instead of increasing it, we will face failures more
frequently
Even though these mandatory soil erosion control
amendments have been offered, I am told the support will
not be ample to see enactment in 1985... we may have 1
or 2 more years to demonstrate the ability to voluntarily
comply before we face a serious threat of mandatory regu-
lations. We will avoid such regulations only if we become
very serious in addressing this problem soon.
A few weeks ago, the Conservation Foundation re-
leased a study called Eroding Soils ... the Off-farm Im-
pact. The report estimated that the degradation of water
as a result of soil erosion costs the Nation $6 billion each
year. According to the report, fishery yields have been
reduced, recreational opportunities have been hindered,
drinking water supplies and quality have diminished, and
navigational channels have experienced heavy sedimen-
tation.
I do not feel that agriculture is the sole offender. As the
Conservation Foundations' report explained, runoff from
construction sites, mining operations, and other areas are
strong concerns. But, we cannot deny that a tremendous
amount of the concern falls within agriculture. We cannot
begin to address such a problem without adequate sup-
port.
I would challenge anyone to do whatever is within his or
her means to generate support for controlling nonpoint
source pollution... at whatever level. It is my strong belief
that a much stronger commitment is necessary from the
Federal level; however, we should not rely on the Federal
government to carry the entire burden. State and local
governments should also become involved. The situation
is serious and the needs are immediate.
10
-------
THE FERTILIZER INDUSTRY'S PERSPECTIVES ON
NONPOINT POLLUTION
G. W. GARRETT
Alliance Agronomics, Inc.
Mechanicsville, Virginia
ABSTRACT
The fertilizer industry, through its national association,
The Fertilizer Institute (TFI), is taking a very positive role
in helping reduce the amounts of nutrients that make their
way into our Nation's waterways. TFI has launched an
extensive educational campaign to make the fertilizer in-
dustry aware of nonpoint pollution problems and to en-
courage voluntary use of agricultural best management
practices (BMP's). Industry views the nonpoint pollution
issue as a legitimate concern, despite the lack of informa-
tion pinpointing sources of the problem. Broad educa-
tional efforts already reaching retailers, producers, and
growers throughout the country have highlighted the
need for action to curb possible nutrient losses. In addi-
tion to BMP's, actions include judicious applications of
fertilizer (based on soil test recommendations), soil con-
servation measures, and proper timing of field opera-
tions. Reduced nutrient losses are also seen as a means
to help maintain a viable and efficient farm economy,
since sound management of nutrients can maximize pro-
ductivity and ease farmers' financial burden. By working
with farmers to help them design fertilizer management
programs, industry can help farmers increase profit per
acre while at the same time reduce the amounts of nutri-
ents lost to the environment. Continued educational ef-
forts with State and regional industry groups, national
and local legislation, and other concerned groups will
help to find responsible solutions to the nonpoint runoff
problem.
My fertilizer retail business serves customers who raise
crops in the Chesapeake Bay watershed, an area obvi-
ously affected by nonpoint runoff, itself an issue of con-
cern to me and to my farm customers. As a businessman,
I take very seriously the potential loss of nutrients through
runoff, because such losses are costly. My customers
can't afford to pay for nutrients not efficiently used by their
crops. They look to me for sound fertilizer management
recommendations, and certainly I must give those recom-
mendations in the context of a shared concern about run-
off.
I could tell about the importance of fertilizer—that its
continued use is essential to assure an abundant food
supply for our Nation—that 30-50 percent of the food and
fiber produced by U.S. farmers is attributed to the use of
fertilizer. But I think you already know that. The fact is, we
will continue to need fertilizer to meet present and future
demands. We will also, however, need to be aware of the
potential effects fertilizer may have on the environment,
and to employ those techniques that keep fertilizer on our
fields and out of our streams, lakes, and rivers.
Among the best tools that the fertilizer industry and its
customers can use to reduce nutrient runoff, and reduce
plant nutrient losses, are Best Management Practices
(BMP's). These practices seek to reduce water runoff and
soil erosion from farmland. Practices such as conserva-
tion tillage, soil testing, timing and placement of fertilizer
application, strip cropping, cover crops, terracing, and
buffer strips are highly effective in cutting losses of plant
nutrients. The fertilizer industry has, for a number of
years, been promoting research on the use of BMP's to
increase efficiency of fertilizer use and reduce nutrient
losses. The future of essential crop production, farm pro-
duction efficiency, and soil and nutrient conservation de-
pend on the ability of our Nation's farmers to expand their
use of sound management practices. The fertilizer indus-
try's position, therefore, is to support and encourage the
voluntary adoption of BMP's for agriculture.
Our industry leaders have realized that we can play a
positive role in reducing losses of nutrients to the environ-
ment. This role is not a defensive one, not one designed to
dispute someone else's facts on the sources of nutrients
found in our Nation's waters. Our role is not to bury our
head in the sand and hope that the issue of nonpoint
pollution or nutrient losses will somehow go away. As an
industry, we have from the beginning taken an active lead-
ership role to stem the runoff problem, regardless of the
source. We actively share in this very real concern.
The board of directors of The Fertilizer Institute adopted
a Plant Nutrient Use Resolution this past year stating that
our industry policy will be to support the judicious use of
plant nutrients, with three considerations in mind:
1. protection of the environment
2. enhancement of farming profitability
3. improvement of food and fiber productivity.
To carry out this directive, The Fertilizer Institute has
undertaken a massive educational effort to inform indus-
try, farmers, Congress, and the public about nutrient run-
off, and about BMP's. The first such information was dis-
tributed through a cover story in Fertilizer PROGRESS
magazine. Over 30,000 readers, most of whom are fertil-
izer retailers, receive this publication. The article ex-
plained the consequences of nutrient runoff to fertilizer
businesses and stressed the voluntary use of BMP's to
reduce nutrient runoff. More than 10,000 reprints of this
article have been disseminated since the story appeared
last June.
To complement the article, PROGRESS also began a
regular series called Best Management Practices, which
details various BMP methods for dealers to share with
farm customers. Reprints of this series are also distributed
widely.
A further step has been the Institute's publication of a
brochure on clean water, explaining the nutrient runoff
issue, our industry's efforts to curb the problem, and a
brief description of the fertilizer industry. Already, we've
distributed over 15,000 copies to State fertilizer associa-
tions, fertilizer dealers, Cooperative Extension Services,
and others throughout the Nation. The Fertilizer Institute
has also participated in EPA's Nonpoint Task Group; we
served as a major cosponsor of USDA's first National Con-
ference on Conservation Tillage; we are cosponsoring
EPA's National Nonpoint Pollution Conference; we're
members of the Conservation Tillage Information Center;
and we are sponsoring our own Nutrient Use and The
Environment Symposium. All these efforts position our in-
dustry as a leader, rather than as a reactor, to this issue.
Why would the fertilizer industry care to help solve the
dilemma of nutrient losses? There are several reasons.
First, the industry (and its association, The Fertilizer
11
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Institute) has always based its case, regardless of the is-
sue, on facts. It's a fact, for example, that fertilizers are an
essential input, that fertilizer is the farmer's single best
production investment, and that fertilizers do not carry
with them the same problems so often associated with
pesticides. In the case of nonpoint, runoff, however, the
facts are scarce. Elevated levels of nitrates have been
found in some lakes and streams, but who or what caused
it? Livestock? Decaying leaves? Human or industrial
wastes? Or farmland runoff?
Since the answers don't yet exist, we as an industry feel
an obligation to act now rather than wait for those an-
swers. If runoff occurs, and if we may be one of the
sources, we feel our business obligations extend to civic
and ecological responsibility as well. We want and expect
clean water as much as anyone.
Second, we feel that nonpoint pollution can be stopped
only if all of us, every potential contributor—and I stress
the word potential—would voluntarily take steps to curb
the problem. My industry believes that we're all served
best when we work together in the same direction, rather
than at cross-purposes.
Finally, the fertilizer industry has another reason for be-
ing involved. My business—and the entire industry—can
only be as strong and viable as the customers we service.
You are all aware of the tremendous financial challenges
faced by today's farmers. Commodity prices are weak,
interest rates high, land values falling, and credit ex-
tremely tight. Farmers can't afford to put excess fertilizer
on their land, can't afford to apply nutrients in a manner
exceeding the crop's ability to use them, and certainly
can't afford to see his rich topsoil wash away along with
the valuable nutrients it contains.
Businessmen like me can and do help the farmer re-
duce his fertilizer losses by making recommendations
based on soil and tissue testing, by encouraging his un-
derstanding of the agronomic aspects of soil conserva-
tion, and by reminding him of BMP's that help reduce
runoff potential.
The fertilizer industry is determined to continue its re-
sponsible role to help solve nutrient runoff problems. Fer-
tilizer producers and retailers are getting involved with
legislation and planning on national, State, and local lev-
els in an effort to do just that.
12
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
CONTROL: SILVICULTURE
JOHN M. BETHEA
Director of the Florida Division of Forestry
Tallahassee, Florida
Considering the other land uses affected by the nonpoint
source issue, silviculture has one important inherent ad-
vantage—the long rotation age of a forest. This means
there are lengthy periods between major disturbances in a
managed forest. When forest soils are disturbed, opportu-
nities occur for erosion, sedimentation, and nonpoint
source pollution. As a result of this and several other fac-
tors, silvicultural activities generally do not contribute as
much nonpoint source pollution to the Nation's waterways
as other land uses.
Silvicultural activities cause nonpoint source pollution
problems that are mostly localized, and of short duration.
Infrequent disturbance means there are only limited op-
portunities for erosion and sediment production from for-
est land. One possible exception to this involves an activ-
ity associated with forest land management—forest roads
and access systems—which can be a considerable
source of nonpoint pollution throughout an entire rotation.
Because of this potential for nonpoint pollution, and given
the fact that most streams naturally drain forested water-
sheds, foresters as land managers have the opportunity
and inherent responsibility to protect our valuable water
resource.
Silviculture is a unique land use in many ways. Like
agriculture, silviculture encompasses a broad land base;
but as previously outlined, intensive activity on forest land
is much more infrequent. Small private landowners, gov-
ernment, and corporations, both large and small, own and
manage forest land. In consideration of these diverse
ownership patterns and varying potential pollution prob-
lems, different areas of the country have taken specific
approaches to controlling silvicultural nonpoint sources.
With the support of the Environmental Protection Agency,
the South, Northeast, and Midwest have generally elected
to control nonpoint source pollution through nonregulatory
programs, while the West has leaned toward the regula-
tory route. Neither approach appears to be right or wrong;
both can work with appropriate conditions and good man-
agement.
Since our experience in Florida has been with the non-
regulatory approach and the Florida Division of Forestry is
the designated management agency for implementing the
silvicultural element of the State Water Quality Plan, we
should review the concepts behind these programs more
closely.
Nonregulatory means something more than voluntary.
Nonregulatory implies an expectation that landowners and
land managers will follow best management practices
(BMP's). Encouraging compliance with a nonregulatory
program requires a sustained effort.
Creative, new approaches to nonpoint source pollution
problems should be developed if nonregulatory programs
are to succeed. States using a nonregulatory approach
should support an initial training phase, as most have
enthusiastically done, followed by a continuing implemen-
tation process.
George Reinert, Chief, Bureau of Resource Planning, Florida Divi-
sion of Forestry, presented this paper at the conference.
If a nonregulatory approach is to succeed for silvicul-
tural activities, then the years following the initial training
phase are critical. These crucial years will demonstrate
whether the States are willing to make a commitment to
continue the sustained effort necessary to achieve suc-
cessful implementation. Certainly the lack of Federal as-
sistance available to the States in these continuing imple-
mentation efforts is a clear problem, but not a legitimate
excuse. The resource involved—water—deserves our
high priority for obvious reasons.
Educating landowners and land managers in BMP's
can have a positive and rippling effect to encourage par-
ticipation in eliminating nonpoint pollution problems. For
example, in Florida's Panhandle it is generally acknowl-
edged that roads and access systems represent a poten-
tially significant source of sediment from forest lands. To
help address this economic and environmental problem,
our agency cooperated with the forest industry to develop
a Forest Road Demonstration Area on industrial land in
the Central Panhandle.
After an appropriate site was selected, Florida Division
of Forestry personnel worked with Southwest Forest In-
dustries land managers on the road layout and design.
Division staff then guided the road construction crew to
include appropriate erosion control features. The area
was logged, and the site prepared and planted to illustrate
the benefits of a good access system to typical silvicultural
operations. During the past year we have used this area
for field workshops aimed at a variety of groups:
1. Society of American Foresters chapter meeting,
2. Division Foresters and other personnel,
3. Forest industry personnel,
4. Personnel from State regulatory agencies, and
5. Other individuals and small groups.
We are convinced of the value of demonstration areas in
a nonregulatory program; these will continue to play a
major role in the Florida implementation effort.
Communication techniques such as brochures, slide
series, films, displays, public meetings, road signs, and
compliance surveys can be used to inform landowners of
nonpoint source pollution and encourage participation in
eliminating these problems.
Regional meetings involving agency personnel respon-
sible for implementing nonpoint source pollution control
could help generate new ideas needed to keep these pro-
grams active. Also, implementing agencies need to work
to achieve a satisfactory, sustained level of effort.
Failure to satisfactorily implement nonregulatory silvi-
cultural pollution control plans will result in the develop-
ment of regulatory programs for more States. Regulatory
programs in the South will be expensive, difficult to en-
force considering the widespread small private owner-
ships, and not foolproof by any means.
The forestry community has the opportunity to test the
nonregulatory approach. Most forestry interests realize
that a program of this kind requires a long-term commit-
ment its success will depend en whether these same
forestry interests have the determination to sustain the
effort that is needed over a period of years.
13
-------
JOHN SPENCER
Seattle Metro
Seattle, Washington
There continues, in our part of the country, a major inter-
est in controlling nonpoint sources of pollution. Evidence
continues to mount on the significant role such sources of
pollution play in the bacterial and toxic contamination of
marine and freshwater bodies in Washington State.
Closed shellfish beds, silted salmon spawning grounds,
fishkills in rural streams, and toxic hot spots in urban bays
all indicate a growing need for nonpoint source pollution
control. In response, our legislature has passed several
new laws; most notable is a new authority for managing
Puget Sound.
My perspective is heightened as a result of recent
actions involving Puget Sound. It has become clear that a
major cleanup of contaminated areas will not happen until
both point and nonpoint sources of pollution are con-
trolled, particularly the nonpoint sources. Cleanup is un-
likely to occur until major sources are under control be-
cause no one wants cleanup projects ruined by
recontamination, and natural "cleansing" may obviate the
need for very expensive cleanup.
The Washington State Legislature has just gone
through a debate over pollution control priorities, much
like the one Congress waged in 1972 when adopting a
uniform national treatment standard. Given our limited
funds, we debated the priority of funding nonpoint source
and point source controls and cleanup actions. Controlling
sources emerged as the undisputed thrust of our State's
pollution control efforts for the remainder of the 1980's.
Nonpoint sources of pollution are undoubtedly the most
difficult to control because the variety of sources is large
and their dispersion extreme. But more importantly, they
are difficult to control because the controls involve chang-
ing individual and corporate behavior patterns. In agricul-
ture it involves changing how a farmer cultivates and irri-
gates his land; in the city it involves the homeowner's
habits when changing oil in the car, or disposing of house-
hold chemicals and solvents. In either case, government
becomes directly involved in how people carry out individ-
ual actions. This situation does not lend itself well to what
we have come to accept as the mode for environmental
protection—namely, permits and structured compliance
monitoring. EPA and State regulators are having to em-
ploy more innovative ways to search for pollution sources
and control them. (We call it search and control.)
The emphasis on deadlines, enforceable provisions in
permits and orders, and technology-based enforcement
standards led to great frustration during the late 1970's
when State and local governments were developing
areawide waste management plans. Most planners and
environmental agency directors found themselves trying
to introduce and implement cooperative actions, best
management practices, and educational programs within
a legal and political framework of environmental protec-
tion based on permits, schedules, enforcement orders,
fines, and penalties. This confused the public and created
mistrust among land managers and resource agency per-
sonnel. My own observations were that it took as much
energy and time for personnel in various government
agencies to understand how these BMP's would work as it
did for the public.
Unfortunately, once the momentum was established to
deal with nonpoint sources of pollution, our Nation's econ-
omy faltered, trimming our efforts toward nonpoint control.
Nonetheless, a great deal was accomplished and we are
well positioned to continue the task of controlling nonpoint
sources.
Major steps have been taken to control nonpoint
sources of pollution in Washington, including but not lim-
ited to forest practices, dairy wastes, urban stormwater
runoff, and construction activities. With the rebuilding of
our environmental programs I believe the progress made
in these areas will be applied more widely, and will result in
action to control other serious nonpoint sources, such as
failing septic tanks, leachate from contaminated industrial
areas, runoff from small noncommercial farming activities,
and illegal dumping of commercial and household waste.
It is worth highlighting two areas of progress: forest
practices control and urban stormwater runoff. Forest
practices regulation meant bringing control to that part of
an industry known for its rugged individualism, high risk,
and economic boom and bust conditions. Many legislative
debates and protests were waged over legislation aimed
at regulating forest practices, and both sides threatened
serious litigation over the regulation issue.
In the end, a workable program was created to achieve
best management practices in harvesting timber from
State and privately-owned lands, and to relieve the indus-
try from double jeopardy under the dual requirements of
meeting best management practices and water quality
standards. (The standards are not enforceable if approved
best management practices are employed.)
The success of these BMP's was shown in subsequent
field audits that found water quality violations occurred
most often where best management practices were not
used or not enforced. This was a milestone for nonpoint
source control in Washington. Despite some continuing
arguments over various BMP's for forest practices, the
real success was in developing a control program that
works for an industry of this size and nature.
Urban stormwater runoff control in the Seattle area is a
good example of the value of 208 planning. Many consider
208 planning a bust. But King County, our largest urban
county, is now formally considering creating a stormwater
utility to control and finance projects for drainage and wa-
ter quality improvements. King County has benefited from
the areawide planning done under section 208 of the
Clean Water act. The County has already implemented
critical stream reach designations where more stringent
land development codes are imposed.
Puget Sound is an example of where State and local
authorities are cooperating to control water pollution. In
Washington we have just enacted a law creating a plan-
ning authority to prepare an areawide management plan
for Puget Sound and its adjacent marine waters. The goal
is to bring our State and local government agencies (re-
source management as well as regulatory) together in a
coordinated effort to control all sources of pollution to the
Sound, particularly fecal coliform and toxics contamina-
tion from nonpoint sources. Moreover, the authority is to
develop a perspective for managing Puget Sound, or
more precisely, a management plan with priorities for con-
trolling both nonpoint and point sources of pollution. This
plan will dramatically affect local land use programs as
well as EPA and State agency compliance efforts.
14
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
This Water Quality Authority, more than anything I can Major difficulties must be overcome, particularly as non-
point to today, illustrates the strength of public concern point source control affects land use decisions and individ-
over the diffuse and varied sources of pollution. The public ual actions. Innovations such as those used in developing
understands that until nonpoint sources are controlled, we best management practices will have to be made. But, like
will not see measurable improvements in the Sound's con- forest practices, other equally difficult nonpoint sources of
laminated areas. pollution can be controlled.
15
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION CONTROL:
A CONSERVATION VIEW
BENJAMIN C. DYSART III
National Wildlife Federation
Washington, D.C.
INTRODUCTION
As immediate past president of the National Wildlife Fed-
eration, the Nation's largest conservation organization at
4.5 million members, I can give some conservation per-
spectives. I would articulate the same positions if I were
here as an environmental engineer, as a researcher, or as
a university professor—all of which I am.
Our resources—precious soils, forests, lakes and
freeflowing streams, energy and other mineral resources,
scenic vistas, wild flowers, and fish—are interrelated in
the watersheds. Through managing our watersheds, we
also manage our aquatic wild and natural resources wher-
ever they occur, out in the Big Hole basin in Montana, in
the Cache River Basin in Arkansas, in Alaska, on the
Great Plains, in little trout streams in the East like the
Thompson in the Carolinas and Penn's Creek in Pennsyl-
vania, or in the urban areas that dot our national country-
side.
PUTTING NONPOINT POLLUTION
IN CONTEXT
The Council on Environmental Quality's 1979 Annual Re-
port stated that nearly all drainage basins were affected in
some locations by pollution from agriculture and urban
runoff. In 1982, six States reported nonpoint sources as
the primary causes of water degradation.
This year, 15 of 20 States reporting progress towards
the Clean Water Act's fishable/swimmable goal listed non-
point pollution as a significant pollutant source in their
remaining problem waters. Finally, nonpoint sources are
being identified as the major contributors to pollution in
large waterbodies such as the lower Great Lakes and
Chesapeake Bay as well as the bays, surface water, and
ground water of Long Island, New York.
Since 1974, hundreds of millions of dollars have been
spent on studies, plans, and demonstration projects. Our
failure to substantially reduce nonpoint source pollution in
the last decade is attributable to our failure to implement
what we've learned, or in many cases what we have all
suspected.
The perception has prevailed that the problem of non-
point pollution is enormous and a solution politically and
technically difficult to devise and put in place. That per-
ception is, I believe, not entirely correct. If we examine two
of the major sources—agriculture and urban runoff—we
find that many effective techniques are already known and
used on a small scale and that, by applying the better
techniques to the critical areas nationwide, we can sub-
stantially reduce nonpoint source pollution.
AGRICULTURAL NONPOINT POLLUTION
In terms of mass, sediment is the major water pollutant
from agricultural activities. Approximately six billion tons
are lost from farmlands each year. This soil loss seriously
affects not just productivity of the land, but also the quality
of the waters into which most of the soil flows and the
bottom habitat in our aquatic ecosystems.
Scientists estimate that 75 percent of the sediment en-
tering streams, rivers, and lakes comes from cropland ero-
sion. If you don't believe this is serious, then I suggest you
look at a stream that gets the runoff from a soybean field
without adequate conservation practices in place, and
then think about the millions of acres of prime bottomland
overflow hardwood habitat that's been cleared and put
into row crops in recent decades in the Mississippi flyway.
In Illinois, about 2 bushels of soil wash from cropland for
each bushel of corn produced in the State. The Illinois
Natural History Survey has found that the backwater lakes
along the Illinois River are half filled with mud because of
siltation from neighboring farms, produced in the last 15
years. The Government Accounting Office has reported
that soil losses resulting from poor agricultural practices
are 25 percent worse today than in 1934.
It's important to realize, however, that less than 3 per-
cent of the land contributes over a third of the total annual
siltation loss. Controlling erosion from a surprisingly small
number of areas, and some of them rather small in size,
can result in most of the pollutant reduction needed to
protect water quality and habitat for fish and wildlife. This
applies to small river basins, as well as to the Nation as a
whole.
And we have to remember that, in addition to soil, we
have lots of toxic materials, nutrients, and other materials
adsorbed onto the soil particles. The nonpoint source pol-
lutants that wash off the land can be loaded with all sorts
of bad actors.
Some nonpoint programs and individual projects have
been real bright spots, but some have had very modest
water quality accomplishments. These programs—208,
Model Implementation, Rural Clean Water, and Clean
Lakes—have resulted in plans and management practices
on a relatively small scale. A nationwide program with
appropriate management practices is needed now to pro-
vide substantial water quality benefits in all the areas
where nonpoint pollution is a critical water quality and
aquatic habitat problem.
URBAN RUNOFF
The other major source of nonpoint pollution is urban run-
off. Pollution in urban runoff includes air pollutants that
have settled in streets, erosion from construction sites,
salt and other deicing chemicals, litter, and animal refuse.
Of the bacterial loading in the bays of Long Island, 95
percent comes from urban runoff and has resulted in the
closing of many shellfish areas. Most of the lakes selected
for study under the Clean Lakes Program are urban park
lakes adversely affected by urban runoff.
Similar to agricultural nonpoint pollution, most of the
urban runoff pollution comes from limited areas, such as
the industrialized and highly urbanized sections of a city
Appropriate management practices targeted at those ar-
eas can control the pollution. Such practices include more
frequent street cleanings, use of porous pavement, and
suitably designed and maintained sedimentation and
catchment basins to reduce the amount of soil and ad-
16
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
sorbed toxic and other nonpoint source pollutants carried
in runoff.
The National Urban Runoff Program is studying 28 met-
ropolitan regions, with Baltimore identified as having the
most contaminated runoff among them. The few areas in
Baltimore that contribute most of the pollution are the
heavily industrialized and urbanized sections.
HISTORICAL PERSPECTIVE
With Congressional passage of the Clean Water Act and
other laws in the early 1970's, sanitary engineers and
environmental scientists took aim at problems identified
now well over a decade ago. Today, I believe we all can
take well-deserved credit for the remarkable progress in
addressing some of the most conspicuous water quality,
resource, and pollution problems.
But today, as a Nation and as resource management
professionals, we find ourselves facing some very difficult
water quality challenges—much tougher than we faced a
decade ago. Part of this is due, in my opinion, to our
having focused too much on "treating wastes" as op-
posed to achieving or maintaining desired water quality in
our streams and lakes, and looking out for the rest of,
indeed most of, our environment—which is outside the
chainlink fence of the wastewater treatment plants. Many
of my colleagues, the water pollution control experts, have
too long concentrated on treating wastewater that comes
to us in a pipe, rather than on the larger scale and much
more important issue, albeit much more complex and
fuzzy, of achieving the water quality goals mandated for us
by society through the political process.
A common mindset must have been passed on to us,
because this problem is prevalent. For example, it led us
to devote tremendous effort and funding to nutrient re-
moval in municipal waste treatment plants and to chang-
ing our laundry detergents. But at the same time, the
water pollution experts in our regulatory agencies, design
firms, and research and development shops paid compar-
atively little attention to controlling the massive quantities
of nutrients entering our streams, lakes, and wetlands
from other diffuse sources, such as fertilizer runoff.
From today's perspective, it's hard to understand why
we tried to achieve a very low suspended solids concen-
tration in municipal wastewater treatment plant effluents
but, at the same time, pretty much ignored the more mas-
sive quantities of turbidity-causing and habitat-smothering
solids entering the surface waters as nonpoint source pol-
lutants. The game plan seemed to be to insist on rational
design and high tech for wastes in pipes, but accept low-
tech rules of thumb and conventional wisdom for control-
ling nonpoint source pollutants.
The GAO's 1977 report telling us we weren't going to
meet the Nation's water quality goals if we didn't come to
grips with nonpoint source pollutants helped focus our
attention on this so-called "new" water pollution problem.
Nonpoint pollution is probably the most important water
quality issue today, especially considering the interrela-
tionship with the management—and frequently misman-
agement—of hazardous wastes, ground water contamina-
tion, abandoned toxic waste dumps, and the like. Non-
point pollution contributes not only conventional pollutants
like sediments, oxygen-demanding wastes, coliform bac-
teria, and nutrients, but toxics like heavy metals, pesti-
cides, herbicides, and lots of others.
Congress is appropriating $750 million annually for soil
conservation, yet this has reduced soil loss very little.
Most of the soil and, even more important in many situa-
tions, the assorted witch's brew of adsorbed toxic chemi-
cals, is still ending up in our Nation's surface waters: low-
ering water quality, damaging fish habitat, and generally
impairing the beneficial uses of others off-site. Since non-
point pollution was acknowledged to be the biggest im-
pediment to achieving water quality standards as of a
couple of years ago in six of EPA's 10 regions, the argu-
ment is compelling for Congress to establish a regulatory
program, and hopefully a meaningful one.
HOW BIG IS THE PROBLEM?
In 1977, the GAO estimated that over half the pollution
entering the Nation's waterways was from nonpoint
sources, and the most important nonpoint sources were
agricultural activities, urban stormwater runoff, silvicul-
ture, and poorly designed and managed septic systems.
In 1984, the Association of State and Interstate Water Pol-
lution Control Administrators found that 1.4 million acres
of surface lake water had been degraded by nonpoint
pollution over the last 10 years.
The tangible costs to society of letting nonpoint pollution
continue for the most part unchecked are very high. The
Nation loses productivity from its land, killing and smother-
ing fish and wetlands, contaminating fish and shellfish so
they're inedible, and spending over $300 million a year
dredging the resultant silt from our rivers, lakes, and har-
bors. The list could go on. And the less tangible, but very
real, costs to the American public are also great.
CURRENT PROPOSALS IN CONGRESS
AND EPA
Congress and EPA are setting a new course for controlling
nonpoint source pollution. A year-long effort by an EPA
task force under the direction of Jack Ravan culminated in
a national nonpoint policy, which hopefully will help to fo-
cus local, State, and Federal agency attention on imple-
menting meaningful control programs in priority water-
bodies. However, it falls short of committing EPA to any
strong action, of assuring that controls are actually used.
In Congress, both the House and the Senate have legis-
lation pending that would encourage States by means of a
grants program to institute runoff pollution controls. Given
the massive Federal deficit, the National Wildlife Federa-
tion believes such a program may not be funded for some
time to come. We continue to press the Congress to adopt
sanctions, or enforcement mechanisms, rather than sim-
ply relying on more Federal money. Otherwise, we may
see very little progress in cleaning up the Nation's number
1 surface water quality problem.
ON-THE-GROUND IMPLEMENTATION
States should be required to identify, develop, and actively
promote the use of appropriate and effective best man-
agement practices for pollution sources. And I emphasize
the words "appropriate" and "effective" because many
so-called BMP's are neither very appropriate nor very ef-
fective in actually protecting off-site surface water quality
and aquatic habitat.
Based on my experience, both research and very practi-
cal, I know that available BMP's—if properly selected, de-
signed, constructed, operated, and maintained, individu-
ally and as coherent systems of BMP's—can effectively
reduce the export of soil, nutrients, herbicides, pesticides,
oxygen-demanding materials, bacteria, heavy metals, and
other toxics off-site and into the Nation's surface waters.
This can be done cost effectively if a genuine commitment
is made to cope with this serious, long-standing problem
and to move on toward protecting our Nation's streams,
reservoirs, and wetlands.
A good bit of money can be spent on so-called BMP's
that don't actually reduce much erosion or soil mobiliza-
17
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
tion at the source or remove sediment, especially the fine
sediment. At Clemson, I've been directing Federal-
agency-supported and industry-supported research and
development work for over a decade dealing with effects
of nonpoint source pollution on surface water quality and
aquatic habitat, erosion from disturbed lands, row-crop
agriculture and heavy construction, and evaluation of indi-
vidual nonpoint source pollution controls and systems of
controls.
The EPA, the water-quality-oriented committees of the
Congress, and the State agencies with water quality re-
sponsibilities simply have to be more active players. They
have to insist that water quality be the—or at least a—real
focus and that the right BMP's be specified more often
and operated properly.
We can't continue to leave it primarily to the agricultural
committees in Congress, USDA, and Soil Conservation
Service county offices, and the land- and agriculture-
oriented State agencies whose focus is properly the agri-
cultural producer, not off-site water quality. We can't leave
it all—or most of it—to these agencies simply "because
they know how to work with the farmers." The rest of us
have to do our part. Many of the nonpoint source prob-
lems involve other than farmers. Too often the approach
that seems most politically palatable is to exempt or
grandfather all the agricultural activities, administratively
declaring them no longer a part of the problem. In my
opinion, this completely begs most of the substantive wa-
ter quality and environmental protection questions facing
us.
And I think my own environmental engineering disci-
pline simply has to recognize the problem and use the
latest in proven cost-effective process design approaches
and procedures for dealing with nonpoint pollution. Con-
gress and the American taxpayers should expect the
same level of reliability, effectiveness, and performance
for nonpoint source control facilities as we all expect of
well-trained modern environmental engineers in dealing
with industrial and municipal wastes that are piped into a
modern and well-operated wastewater treatment plant.
This can be done, and as far as I'm concerned there's
absolutely no reason for it not to be both expected and
accomplished.
18
-------
A LIVESTOCK INDUSTRY PERSPECTIVE ON NONPOINT SOURCE
POLLUTION CONTROL
LESTER COY
Elmore, Minnesota
Our family corporation has a cattle-feeding and farming
operation on 486 hectares (1,200 acres) in Kossuth
County, Iowa. We annually feed 2,500 to 3,000 head of
cattle in confinement and open lot facilities and raise seed
corn and soybeans. I am currently chairman of the Na-
tional Cattlemen's Association's 208 Water Quality Sub-
committee, and chairman of the Lands, Water, and Envi-
ronment Committee of the Iowa Cattlemen's Association. I
also serve on the board of that State organization.
Soil erosion and water runoff have been occurring on
this planet since time began. The most productive agricul-
tural areas of this country were formed by erosion and
sediment from the huge ice caps that melted millions of
years ago. Such erosion produced soils up to more than 1
meter deep in the Midwest. And the loess that covers
much of North America to a depth of several hectometers
is the result of wind erosion over thousands of years.
Man, of course, has aggravated erosion (now called
nonpoint source pollution) although not to the extent that
some groups and government agencies would have us
believe. Certainly, we ought to be concerned about such
erosion and I would like to mention some actions that I
think will help abate erosion resulting from agricultural
activities.
There is a legitimate public interest in water pollution
abatement. The condition of our Nation's waters is impor-
tant to all of us and we need national laws dealing with
water pollution. These laws, however, should be based on
understanding and common sense, not on emotion and
unsupported cliches.
These laws should recognize that there is no single,
nationwide solution to the problem of nonpoint source pol-
lution. There is no panacea that will prevent nonpoint
source pollution in every place under all circumstances.
What works best in one area may not be the best in an-
other area.
The laws must be flexible enough to allow people to
devise effective solutions that fit the unique circumstances
of individual areas. A cooperative program involving local
people familiar with local resources and conditions is the
best way to effectively abate water pollution.
Furthermore, farmers and ranchers are the most envi-
ronmentally conscious group in the United States. As peo-
ple who deal daily with our environment, we do know
something about local topography, climate, and other ele-
ments of Mother Nature. We are acutely aware of environ-
mental problems because, among other things, such
problems bring us enormous economic costs. Regulation
increases these economic costs. Environmental degrada-
tion poses a threat to our livelihood so we have a selfish
interest in preventing and abating pollution.
In addition to the right kind of laws, we need the right
kind of scientific research—not the kind that we have had
so far. Current research consists of stacks and stacks of
research reports identifying types and amounts and loca-
tions of pollutants. What we do not have is research that
tells the effects of these pollutants on man or on activities
vital to mankind, such as food production. It may be good
to know which substances cause cancer when exposed or
fed to mice at rates many times greater than man would
ever face in a normal lifetime, but such experiments do
not answer the question of how many substances man
can tolerate without harm.
We will probably never eradicate all pollutants from our
environment, even if we have the money. Therefore, we
need to identify those pollutants that have the most ad-
verse impacts upon man and determine what can be
done. We also need to distinguish between pollutants orig-
inating from natural causes and those originating from
man's activities.
We need carefully controlled research telling us how
pollutants arrive at a particular place, exactly how far they
move in the soil or water in a given time, and, above all,
what levels of these pollutants man can tolerate with no
side effects.
Research finds conform in a western mountain
stream—but it does not indicate whether this came from
livestock grazing, elk, or other wildlife in the area. Did the
material enter the stream recently or was it something that
was deposited years ago and then stirred up from the river
bottom during the last storm?
Nitrate is found in a midwestern stream. Did it originate
from fertilizer applied nearby or from decaying vegetation
from a city, or maybe from a wilderness area many thou-
sands of miles away? Research often does not attempt to
provide an answer, but this does not prevent some people
from blaming agriculture.
If we are simply guessing at the answers to such ques-
tions, we may not be addressing the real causes of pollu-
tion.
It also appears that research is often used for purposes
other than factual enlightenment. For example, in recent
years, a barrage of articles in the media and a number of
government reports have given the impression the United
States is in imminent danger of losing all of its topsoil.
A 1980 Federal document indicates that sheet and rill
erosion on cropland in the Mississippi River Basin, and in
many other Eastern States carries off 1.8-5.1 metric tons
per hectare (5 to 13.9 tons per acre) each year (U.S. Dep.
Agric. 1980). If that much silt got into the Mississippi, the
river would no longer flow!
That same document says that wind erosion in several
States exceeds more than 1.8 metric tons per hectare (5
tons per acre) per year (U.S. Dep. Agric. 1980). I have
visions of huge amounts of soil piled like snow drifts in
every road ditch and at every wind break.
I suspect such reports are aimed more at generating
political support for government funding than at giving a
true picture of the current situation.
Of course, it will take more than research to curb non-
point source pollution. Even if we know all the answers, it
doesn't do any good if that information does not reach
those who can do something with it. I am referring not just
to the policymakers or government administrators, for it
should be recognized that—especially in regard to non-
point pollution—much can be done through voluntary ef-
forts. As previously stated, a cooperative approach involv-
ing the farmer and rancher can accomplish more to
reduce runoff or erosion from agricultural lands than a
program solely dependent upon mandatory actions.
Bui the success of any effort—mandatory or volun-
tary—depends upon knowledge and understanding. We
not only need research to find out how to control erosion
19
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
but we need to transmit that how to the landowners and
operators who can use the information. Agencies and or-
ganizations need to do a better job of communicating such
knowledge.
I am pleased to say that the National Cattlemen's Asso-
ciation and affiliated State livestock organizations are
making such an effort. We have taken a constructive ap-
proach to the nonpoint source pollution issue. We are tell-
ing our members what cattlemen have done about this
problem in the past, what they are doing now, and what
they can do in the future.
One of our first efforts was the preparation and release
in the 1970's of a slide program entitled "208 Planning
and the Cattle Industry." This program explained the law
and what the States needed to do. More importantly, it
demonstrated what cattlemen could do to minimize non-
point source pollution.
Some 45 organizations affiliated with the National Cat-
tlemen's Association have shown these slides to over a
quarter million cattlemen at various meetings. In addition,
over 75 universities and colleges have used the slides in
curriculum or special program courses.
The updated version of this slide program is being
shown during this conference. We hope you will take the
time to view it, as it demonstrates the positive actions
cattlemen are taking, voluntarily, to address water quality
programs. The slides depict many of the so-called best
management practices used by cattlemen. They reveal
the diversity of the industry nationwide—cattle raised on
the range, in pasture, and in small confinement facilities.
Another example of an educational effort that our asso-
ciation is conducting is a book entitled A Cattlemen's
Guide to Water Pollution Control and 208 Water Quality
Planning. The writing of this book, a few years ago, was
more of a massive job than you might imagine or, frankly,
than we had anticipated.
When we gathered all of the government laws, rules,
and publications on nonpoint source pollution, we had a
stack of books several feet high. Not one, however, gave a
succinct explanation of nonpoint pollution laws or what
cattlemen could do to curb such pollution. Obviously, no
individual cattleman could read all that material.
So we summarized the law and rules, ignoring the jar-
gon that makes most government and academic language
incomprehensible to the average person. Also, we were
not simply concerned with what cattlemen and States are
required to do about nonpoint source pollution. We were
more concerned with what could be done. This book pro-
vides the latest information on livestock management
practices to control nonpoint source runoff, and it includes
a self-evaluation form for cattlemen. It also lists govern-
ment agencies a person can contact for more information.
In 1983, Profit Potential of Environmental Protection
Practices of Cattlemen was published by the Environmen-
tal Management Committee of the National Cattlemen's
Association. This handbook describes ways cattlemen
can prevent water pollution from cattle grazing or feeding
operations. Based on a seminar at which technical ex-
perts in animal wastes and pollution spoke, this publica-
tion has an interesting focus. Instead of regarding animal
wastes as an expensive disposal problem, it shows that
when handled properly, such byproducts can be substi-
tuted for chemical fertilizer, and be profitable to cattlemen.
The soil is the best garbage disposal in the world.
These are examples of what cattle industry organiza-
tions are doing to educate our members on best manage-
ment practices that will curb water pollution and help pre-
serve water quality.
We intend to continue these efforts, because we recog-
nize mandatory government regulations are not the an-
swer. We will continue to cooperate with government
agencies and others to improve technology and processes
that will help achieve the goal that I think all citizens
want—a clean glass of water.
In summary, I make the following points:
1. Agriculture, including animal agriculture, has been
unfairly accused as the major polluter of water.
2. Laws and rules should be based on facts and com-
mon sense.
3. More unbiased scientific research is needed to pro-
vide answers we must have if we are to have an effective
program to minimize nonpoint source pollution.
4. We need to improve communications among gov-
ernment agencies, scientists, farmers, and the general
public.
5. We need to emphasize a cooperative approach to
solving pollution problems, because a lot more can be
accomplished that way than through government dictum.
6. We need to foster the concept of best management
practices.
REFERENCES
National Cattlemen's Association. 1982. A Cattlemen's Guide to
Water Pollution Control and 208 Water Quality Planning.
Washington, DC.
U.S. Department of Agriculture. 1980. Soil and Water Re-
sources Act Appraisal. Pages 108,161.
20
-------
Monitoring and
Assessment Techniques
THE ST. ALBANS BAY WATERSHED RCWP: A CASE STUDY OF
MONITORING AND ASSESSMENT
JOHN C. CLAUSEN
University Water Resources Research Center
School of Natural Resources
University of Vermont
Burlington, Vermont
ABSTRACT
Excessive nutrients from a municipal point source and
agricultural nonpoint sources have impaired the use of St.
Albans Bay of Lake Champlain in Vermont. A comprehen-
sive monitoring and evaluation approach is evaluating the
effects of agricultural Best Management Practices
(BMP's) on the quality of bay and tributary waters as part
of the Rural Clean Water Program (RCWP). Monitoring
techniques include edge-of-field paired watersheds, in-
stream trend stations, bay trend sampling, and land use
tracking. Related short-term studies are investigating bay
circulation patterns, bay sediment phosphorus content
and release, biological indicators, and the role of a wet-
land in treating both point and nonpoint source nutrients.
Each monitoring technique and its associated assess-
ment methods are described through project results. The
comprehensive monitoring approach is designed to iden-
tify overall programmatic effects in complex watersheds.
INTRODUCTION
The St. Albans Bay Watershed is one of the experimental
Rural Clean Water Programs (RCWP) projects designed
to improve water quality through agricultural best man-
agement practices (BMP). St. Albans Bay has been de-
graded by excessive algal and macrophyte growths and
elevated coliform counts (Vt. Agency Environ. Conserv.,
1977). Abundant nutrients in the bay, which are causing
the accelerated eutrophication, come from both point and
nonpoint sources. Recently, Johnson (1985) estimated
that at least 37 percent of the phosphorus and 48 percent
of the nitrogen originated from nonpoint sources. Im-
proper anima! waste management and cropping practices
have been identified by the Soil Conservation Service
(1981) as being primarily responsible for excessive non-
point nutrient loading to the bay.
In 1981, implementation of agricultural BMP's began
with Federal cost-sharing through RCWP to control non-
point sources of nutrients and sediment. Concurrent with
the agricultural nonpoint source control strategy is a com-
prehensive water quality monitoring and evaluation proj-
ect to determine the effects of BMP's on water quality.
Numerous techniques have been used to assess the
effect of land treatment on water quality. Listed in order of
increasing distance from the source, these techniques in-
clude: runoff plots; fields; single, paired, and multiple wa-
tersheds; and larger, mixed land use watersheds (Striffler,
1965; Hewlett et al. 1969; Ponce, 1980; Clausen and
Brooks, 1983). Advantages and disadvantages of these
techniques have been described (Striffler, 1965; Hewlett et
al. 1969; Clausen and Brooks, 1983). One of the greatest
challenges facing water quality data analysts is the inter-
pretation of water quality changes in streams receiving
nutrients from large complex watersheds.
This paper describes the monitoring and assessment
techniques being used in the St. Albans Bay RCWP and
discusses current findings.
STUDY AREA
The 13,500 ha St. Albans Bay watershed is located in
northwestern Vermont, 40 km north of Burlington (Fig. 1).
Agriculture is the dominant land use in the watershed (68
percent); corn and hay are the principal crops. Forests
cover 22 percent of the area, and urban areas and roads
account for the remaining 10 percent. There are 100 dairy
farms in the watershed averaging 134 ha and 95 animal
units.
Watershed soils include loam (51 percent), half of which
is poorly drained, silt and clay (27 percent), rock outcrop
(15 percent), and sand (7 percent) (Soil Conserv. Serv.
1979). These soils formed on glacial till or lacustrine de-
posits.
21
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
ST. ALBANS BAY WATERSHED
FRANKUN COUNTY VERMONT
Figure 1.—Map of the St. Albans Bay watershed showing
sampling locations.
The mean annual precipitation is 845 mm, and occurs
mainly in the summer. The climate is considered to be the
cool, continental type with a mean annual temperature of
7.3°C. Average annual snowfall is 1,560 mm (Soil Con-
serv. Serv. 1979).
Four major tributaries drain the watershed into St. Al-
bans Bay: Jewett Brook, Stevens Brook, Rugg Brook, and
Mill River (Fig. 1). The city of St. Albans' secondary waste-
water treatment plant discharges to Stevens Brook wet-
land at the head of the bay.
METHODS
Sampling Design
To document water quality changes, several levels of wa-
ter quality sampling have been conducted since 1981.
Level 1 involves bay sampling at four stations, 20 times
each year (Fig. 1). At each station, samples are collected
at the 0.5 m depth and 1.0 m from the bottom. Level 2
includes instream sampling at the four tributaries and the
St. Albans' wastewater treatment plant. At each of the five
Level 2 stations, samples are automatically collected at 8-
hour intervals using ISCO refrigerated samplers and com-
bined into two 48-hour and one 72-hour composites each
week. During stormflow periods, each 8-hour sample is
analyzed discretely. Flow is measured continuously at
each station using ISCO bubbler-type stage recorders.
Three standard, weighing-bucket gauges are used to
measure watershed precipitation.
Level 3 involves edge-of-field sampling to evaluate
changes in the quality of runoff associated with best ma-
nure management. A paired watershed design was used
where two field watersheds received best manure man-
agement during a 2-year calibration period, and then one
field received winter-spread manure during the treatment
period. The control field was 0.9 ha and the treatment field
was 1.9 ha. The treatment field received 8,925 kg/ha of
liquid manure spread during winter 1984. Calibration and
treatment regressions were based on paired daily concen-
tration, discharge, and mass export values.
Level 4 sampling is conducted at four other stream loca-
tions in the watershed (Fig. 1) to characterize additional
tributaries and to isolate subwatersheds. Grab samples
are collected an average of once every 20 days on ran-
domly selected dates.
Sample Analysis
All samples are analyzed for turbidity; total and volatile
suspended solids; total and orthophosphorus; and total
Kjeldahl, ammonia, and nitrate + nitrite nitrogen, accord-
ing to standard methods (U.S. Environ. Prot. Agency,
1983). In situ measurements are made at all bay stations
of temperature, dissolved oxygen, and Secchi disk.
Weekly grab samples are analyzed for pH, conductance,
fecal coliform, and fecal streptococcus. St. Albans Bay
samples are also analyzed for chlorophyll a.
Related Studies
In addition to the long-term monitoring there have been
separate investigations of stream biological characteris-
tics (LaBar, 1984), bay circulation (Laible, 1983), and bay
and wetland sediments (Drake, 1984). An extensive land
use monitoring effort is described in detail in a companion
paper (Hopkins and Clausen, 1985).
RESULTS AND DISCUSSION
BMP Implementation Status
The goals of the RCWP were to manage 75 percent of the
6,174 critical hectares in the watershed (lands receiving
animal waste or fertilizer), and to treat a number of critical
sources by using animal waste and fertilizer management,
conservation cropping systems, and stream protection.
Currently, approximately 90 percent of this goal has been
achieved (Table 1). The major BMP is to provide for animal
waste storage during the winter months to prevent daily
manure spreading on snow covered or frozen soils. Under
the Agricultural Conservation Program (ACP), two manure
storage structures have been built and additional areas in
conservation cropping supplement the RCWP.
Table 1.—BMP Implementation status for the St. Albans
Bay watershed RCWP.
Contracted
Farms
Year (No.)
1981
1982
1983
1984
Total
Goal
21
18
11
6
56
64
Critical Areas
Contracted
(ha)
1,577
1,314
908
398
4,197
4,631
Manure
Storage
(No.)
7
21
9
5
42
64
Conservation
Cropping area
(ha)
357
1,200
161
550
2,268
2,590
'This is a sample for a typical footnote in 6 point helvetica by 19 picas.
St. Albans Bay
A horizontal gradient in concentration is evident in St. Al-
bans Bay. The north end of the bay has much higher
concentration of sediment and nutrients as compared to
the south end which opens into Lake Champlain (Fig. 2).
This gradient is related to mixing between the main lake
22
-------
MONITORING AND ASSESSMENT TECHNIQUES
and the bay (Laible, 1983). Chlorophyll a concentrations
follow these nutrient gradients. The inner bay averages
31 /ig/l chlorophyll a and the outer bay averages 9 /*g/l.
The total phosphorus to total nitrogen ratio in the bay
ranges from 6:1 to 33:1, indicating that the limiting nutrient
may at times be either nitrogen (TN:TP <10) or phos-
phorus (TN:TP > 17) (Smith, 1982).
Detection of trends in the bay will have to consider
these gradients, and both the chemical and biological
characteristics of bay waters. Time trends may be con-
founded with hydrological variability. However, the outer
bay station may serve as a control for comparison with the
inner bay station. Trends could then be identified as differ-
ences between regressions, using the inner bay data as
the dependent treatment variable.
Tributary Streams
Mean concentrations of solids, phosphorus, and nitrogen
for the Level 2 tributary stations show both annual variabil-
ity and differences among watersheds (Fig. 2). Annual
precipitation for the 1982-83 water year was near normal
(859 mm) while precipitation during 1983-84 was 30 per-
cent above normal (1,094 mm). Although trends over 2
years of sampling mean little in water quality interpreta-
tions, observed concentrations do identify critical water-
sheds. For example, Jewett Brook, which has 87 percent
agricultural land use, has elevated concentrations of
phosphorus and nitrogen compared to other watersheds
(Fig. 2). Mass exports in Jewett Brook are also quite high;
during 1983-84 total phosphorus export was 6.7 kg ha*1
yr1, over 20 times the average export from agricultural
watersheds in the eastern United States (Omernik, 1976).
The Jewett Brook Watershed has the most BMP's and
therefore the potential for showing the greatest water qual-
ity changes during the project.
Edge-of-Field
., The effects of winter-spread manure on field runoff con-
centrations are summarized in Figure 3. The dark bars
represent the differences between the concentrations pre-
dicted by the calibration equation and the mean concen-
tration observed during treatment. Winter spreading in-
creased the concentrations of total P, ortho-P, total
Kjeldahl-N, and ammonia-N, but total suspended solids
decreased significantly (p = 0.001). After spreading ma-
nure in the winter, increased concentrations of phos-
phorus and njtrogen have been reported previously based
on plot studies (Hensler et al. 1970; Minshall et al. 1970;
Klausner et al. 1976). The reduction in suspended solids
concentrations has also been reported (Young and Holt,
1977; Young and Mutchler, 1976), presumably resulting
from a mulching effect of animal wastes.
Winter manure application decreased surface runoff
from the field (Fig. 4). Runoff may decline because applied
manure increases soil infiltration (Khaleel et al. 1981;
Zwerman et al. 1970). The decrease in runoff together
with reductions in suspended solids resulted in a de-
creased mass export of total suspended solids by one-half
(Fig- 4).
Even though runoff decreased, phosphorus and nitro-
gen increased in runoff after winter manure applications
(Fig. 4). Total phosphorus export increased 11 percent (p
= 0.08), but orthophosphorus export increased by a fac-
tor of 15 (p = 0.03).
Based on the amount of manure applied to the field in
the winter, 15 percent of the phosphorus and 17 percent of
the nitrogen was lost in surface runoff. These losses are
somewhat greater than the 95 percent retention of phos-
phorus and nitrogen of winter-applied manure reported by
Klausner etal. (1976).
TRIBUTARY STATION
BAY STATION
21 22 23 24 25
JEWETT STEVENS RUGG MILL STP
6.24.8
II 12
OUTER INNER
WO
I]
21 22 23 24 25
4.44.8
II 12
t
21 22 23 24
Jl
I
1982-83
1983-84
21 22 23 24 25
JEWETT STEVENS RUGG MILL STP
11 12
OUTER INNER
Figure 2.—Mean Concentrations of solids, phosphorus, and
nitrogen at the tributary and bay trend stations for 2 years.
TOTAL SUSPENDED
SOLIDSIMG/U
0 8 §
CALIBRA1
ION
TREATME
-68%
"1
NT
LOWER UPPER LOWER UPPER
-24%
ITROL
LOWER
TREATMENT
»224%
n
I
OBSERVED-
PREDICTED
TREATMENT
+1480%
,» p=0.01
* * * PS O.O01
+114%
n
n
+576%
I
LOWER UPPER LOWER UPPER
Figure 3.—Mean concentrations in runoff from the Larose
farm paired watersheds.
23
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
40 _ CALIBRATION TREATMENT
8
20
I
n
2 40°
200-
LOWER UPPER LOWER UPPER
-53%
n
D
UPPER
CONTROL
LOWER
TREATMENT
I
OBSERVED
PREDICTED
TREATMENT
p=005
p=0.0l
O.I -r
LOWER UPPER LOWER UPPER
1500%
771
•(•618%
1
EL
LOWER UPPER LOWER UPPER
Figure 4.—Runoff and mass export from the Larose farm
paired watersheds.
CONCLUSIONS
There has been insufficient time to evaluate water quality
trends in the bay or its tributaries. However, the edge-of-
field paired watershed experiment has shown, in a rela-
tively short time, that proper animal waste management
can reduce phosphorus and nitrogen concentrations and
exports to receiving bodies of water.
Monitoring of water quality and agricultural activities will
continue. Several techniques are available for water qual-
ity trend detection for long-term studies: (1) Time plots, (2)
least squares regression, (3) comparisons of annual
means, (4) Q-Q plots, (5) probability distribution functions,
(6) paired watershed regression, (7) spectral analysis, and
(8) time series analysis. Good descriptions of these meth-
ods appear in UNESCO (1978), Hirsch et al. (1982), and
Montgomery and Reckhow (1984). As additional data be-
come available, these trend assessment techniques will
be applied to determine the changes in water quality asso-
ciated with BMP implementation.
ACKNOWLEDGEMENTS: This project is supported by funds
from the USDA Agricultural Stabilization and Conservation
Services and the University of Vermont in a cooperative agree-
ment with the Soil Conservation Service, Franklin County Natu-
ral Resources Conservation District, Vermont Extension Serv-
ice, and the Vermont Agency of Environmental Conservation.
REFERENCES
Clausen, J.C., and K.N. Brooks. 1983. Quality of runoff from
Minnesota peatlands: II. A method for assessing mining im-
pacts. Water Res. Bull. 19(5): 769-72.
Drake, J.C. 1984. Sedimentological studies in St. Albans Bay
and contiguous wetlands. Annu. Rep. Dep. Geology, Univ.
Vermont, Burlington.
Hensler, R.F., et al. 1970. Effect of method of manure handling
on crop yields, nutrient recovery and runoff losses. Trans. Am.
Soc. Agric. Eng. 13(3): 726-31.
Hewlett, J.D., H.W. Lull, and K.G. Reinhart. 1969. In defense of
experimental watersheds. Water Res. Res. 5(1): 306-15.
Hirsch, R.M., J.R. Slack, and R.A. Smith. 1982. Techniques of
trend analysis for monthly water quality data. Water Res. Res.
18(1): 107-21.
Hopkins, R.B., and J.C. Clausen. This vol. Land use monitoring
and assessment for nonpoint source pollution control. In Per-
spectives on Nonpoint Source Pollution. Proc. Natl. Conf. N.
Am. Lake Manage., Kansas City, MO. May 19-22,1985.
Johnson, G.D. 1985. The effects of Stevens Brook Wetland on
nutrient loading to St. Albans Bay. M.S. Thesis. Univ. Ver-
mont, Burlington.
Khaleel, R., K.R. Reddy, and M.R. Overcash. 1981. Changes in
soil physical properties due to organic waste applications: a
review. J. Environ. Qual. 10(2): 133-41.
Klausner, S.D., P.J. Zwerman, and D.F. Ellis. 1976. Nitrogen and
phosphorus losses from winter disposal of dairy manure. J.
Environ. Qual. 5(1): 47-50.
LaBar, G.W. 1984. St. Albans RCWP fish and benthos monitor-
ing study. Final Report. School Nat. Resour, Univ. Vermont,
Burlington.
Laible, J.R 1983. St. Albans Bay current and pollution transport
study. Progress rept. Univ. Vermont, Burlington.
Minshall, N.E., S.A. Witzel, and M.S. Nichols. 1970. Stream
enrichment from farm operations. Am. Soc. Civil Eng. J. San.
Eng. Div. 96(SA2): 513-24.
Montgomery, R.H., and K.H. Reckhow. 1984. Techniques for
detecting trends in lake water quality. Water Res. Bull. 20(1):
43-52.
Omernik, J.M. 1976. The influence of land use on stream nutri-
ent levels. EPA-600/3-76-014. U.S. Environ. Prot. Agency, Cor-
vallis, OR.
Ponce, S.L. 1980. Water quality monitoring programs. WSDG
Tech. Pap. No. 00002. Watershed Systems Development
Group, Forest Serv., U.S. Dept. Agric., Fort Collins, CO.
Smith, V.H. 1982. The nitrogen and phosphorus dependence of
algal biomass in lakes: an empirical and theoretical analysis.
Limnol. Oceanogr. 27(6): 1101-12.
Soil Conservation Service. 1979. Soil Survey of Franklin County,
Vermont, U.S. Dept. Agric. Burlington, VT.
1981. Rural Clean Water Program St. Albans Bay
Project. Plan of Work. U.S. Dept. of Agric. Franklin County, VT.
Striffler, W.D., 1965. The selection of experimental watersheds
and methods in disturbed forest areas. Pages 464-473 in
Symp. Budapest. Internal. Assn. Scientific Hydrol. Publ. No.
66.
UNESCO. 1978. Water quality surveys—a guide for the collec-
tion and interpretation of water quality data. IHD-WHO Work.
Group Qual. Water. UNESCO/World Health Organ., Geneva,
Switzerland.
U.S. Environmental Protection Agency. 1983. Methods for
Chemical Analysis of Water and Wastes. EPA-600/4-79-020.
Off. Res. Dev., Cincinnati, OH.
Vermont Agency of Environmental Conservation. 1977. Nutrient
loading to Shelburne Bay and St. Albans Bay, Lake Cham-
plain, Vermont 1975-1976. Montpelier.
Young, R.A., and R.F. Holt. 1977. Winter-applied manure: effects
on annual runoff, erosion, and nutrient movement. J. Soil Wa-
ter Conser. 32(5): 219-222.
Young, R.A., and C.K. Mutchler. 1976. Pollution potential of ma-
nure spead on frozen ground. J. Environ. Qual. 5(2): 174-9.
Zwerman, P.J., et al. 1970. Rates of water infiltration resulting
from applications of dairy manure. Pages 263-270 in Relation-
ships of Agriculture to Soil and Water Pollution. Proc. 1970
Cornell Univ. Conf. on Agric. Waste Manage.
24
-------
LAND USE MONITORING AND ASSESSMENT FOR NONPOINT
SOURCE POLLUTION CONTROL
RICHMOND B. HOPKINS, JR.
JOHN C. CLAUSEN
University Water Resources Research Center
School of Natural Resources
University of Vermont
Burlington, Vermont
ABSTRACT
Varying soil characteristics, land use patterns, the rela-
tive timing of agricultural practices, and hydrologic events
complicate quantifying relationships between agricultural
land use and surface water quality. In two Vermont water-
sheds where the effects of best management practice
(BMP) implementation on water quality are continuously
monitored, land use and agricultural activities are being
monitored on a field-by-field level. The land use data are
entered in a computerized Geographic Information Sys-
tem (GIS), and the results mapped. Correlation and
stepwise regression techniques related weekly land use
activities for one subwatershed to surface water quality.
Comparisons of water quality to agricultural land use
were based on proximity to surface drainage and whether
activities had occurred on runoff-producing zones. Ma-
nure application on.Soil Hydrologic Group D was signifi-
cantly related to stream total phosphorus concentration (r
= .62) when manure was accumulated between runoff
events. A predictive equation developed explained 55
percent of the variation in total phosphorus concentra-
tion. GIS offers the potential to inventory critical sources
of nonpoint source pollution and identify changes in water
quality from agricultural land use and BMP's.
STUDY AREA
The 1,333 ha Jewett Brook watershed in northwestern
Franklin County, Vermont, was selected for the study (Fig.
1). Land use is predominantly agricultural (87 percent)
with the remainder woodland or residential. The 16 dairy
farms in the watershed average 65 ha, with herds ranging
from 32 to 260 animals. The average herd size is approxi-
mately 125 animals.
Within the Champlain Lowland physiographic unit, the
Jewett Brook watershed has irregular topography with roll-
ing hills. Over two-thirds of this area has soil formed on
lacustrine deposits; other soils formed on glacial till. Ap-
proximately 50 percent of the watershed has poorly
drained silt and clay soils; another 42 percent has poorly
drained loam or sand; and only 8 percent of the area has
well-drained loam or sand (Soil Conserv. Serv. 1979).
The climate of the watershed is influenced by the pres-
ence of Lake Champlain to the west and south and by its
northern latitude (44° 47' 26"). Long-term average winter
temperature is -7°C, and average summer temperature is
20°C. The average last spring freeze is expected May 2
and the first autumn freeze by Oct. 13. Approximately 61
percent of the total yearly precipitation falls in April
through September, with August the wettest month
INTRODUCTION
The relationship between land use and water quality has
been the subject of much research in the last 10 years. It
is generally accepted that as the percent of agricultural
land in a watershed increases, concentrations of sediment
and* nutrients in streams draining these areas also in-
crease (U.S. Environ. Prot. Agency, 1974; Dillon and
Kirchner, 1975; Smolen et al. 1975; Omernik, 1976, 1977;
Hill, 1981). The proximity of agricultural lands to streams
within a watershed may also influence nutrient contribu-
tions in runoff (Kunkle, 1970; Uttormark et al. 1974).
Dunne (1969), and Lake and Morrison (1977) report that
large nutrient losses in runoff may originate from areas of
low infiltration potential or high soil saturation. These ar-
eas have been termed runoff-producing zones.
Greatest stream nutrient concentrations have been
linked to spring stormflow periods when cultivation is ac-
tive and vegetative cover is poor (Dornbush et al. 1974;
Dendy, 1981; McDowell et al. 1981), but this relationship
may be caused solely by increased discharge in the
spring, rather than agriculturally induced.
Agricultural activities (e.g. nutrient applications, cultiva-
tion) should influence stream water quality, with activities
in runoff-producing zones and near streams having a
greater effect than those elsewhere. These relationships
have not been temporally or spatially examined. The pri-
mary purpose of this study is to relate the location and
timing of agricultural activities to stream water quality.
Figure 1.—Vicinity map for project location.
25
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
SoiI Hydrologic Group A - 1X
Soil Hydrologic Group B - 14X
Soi I Hydrologic Group C - 40X
Soil Hydrologic Group 0 - 45/5
KILOMETERS
Corn - 19X
Alfalfa - 115!
Hay I and - 39X
Permanent Pasture - 14X
Farmstead - 2X
Woodland - 12*
Residential - U
Other Agricultural Land - 2X
K i LOMETERS
Figure 2.—Land use for Jewett Brook watershed.
Figure 3.—Soil hydrologic groups for Jewett Brook watershed
26
-------
MONITORING AND ASSESSMENT TECHNIQUES
(10.0 cm). On the average, December receives the great-
est amounts of snowfall (49.2 cm) (Soil Conserv. Serv.
1979).
METHODS
Maps of the watershed were prepared at a scale of
1:10000. Land use and farm and field boundaries were
identified during interviews with each landowner. Soil
types and characteristics were obtained from the Franklin
County Soil Survey (Soil Conserv. Serv. 1979). Streams
and drainage ditch locations were identified using topo-
graphic maps and aerial photographs. Elevation and wa-
tershed boundaries were obtained from USGS 7.5' topo-
graphic maps (U.S. Geolog. Surv. 1972). Data were
entered into a computerized Geographic Information Sys-
tem (GIS) using a 0.404 ha cell grid overlay. Figures 2 and
3, generated by the GIS, show watershed land use and
soil hydrologic classifications, respectively.
Land Use Monitoring
Land use and areas receiving agricultural activities were
recorded onto field logs that had been distributed to each
landowner within the watershed. Agricultural activity data
were recorded from January to December, 1983. Data in-
cluded the date, amount, location, and method of com-
mercial fertilizer and animal waste application, areas that
had been plowed and cultivated, and fields where crops
had been harvested. Information was gathered during
January, June, and December. Data were mapped using
the GIS.
Computerized geographic overlays were performed us-
ing the GIS. Overlays were created with weekly land use
data, soil hydrologic classifications, and the area within 63
meters of the brook and drainage network. Runoff-produc-
ing zones were areas associated with Soil Hydrologic
Group D (those soils having high runoff potential and low
infiltration rates).
Table 1.—Correlations (r) between mean weekly runoff
concentrations (mg/l) and weekly hydrologic variables.
Total Mean
Water Quality Parameters precipitation discharge
(cm) (m*/sec)
Total phosphorus
Orthophosphorus
Total Kjeldahl nitrogen
Ammonia nitrogen
Total suspended solids
Volatile suspended solids
.07
-.06
.06
.02
.321
.24
.37*
.48"
.442
.331
.61"
.682
'Indicates significance at P - 0.05
'Indicates significance at P - 0.01
Figure 4.—Manure applications (In black) In Jewett Brook
watershed during 1983.
Water Analysis
Streamflow quantity and quality were continuously moni-
tored at the watershed outlet. Two 48-hour and one 72-
hour composite water samples were collected each week
for 52 weeks. Samples were analyzed for total suspended
solids, volatile suspended solids, total Kjeldahl nitrogen,
ammonia nitrogen, total phosphorus, and orthophos-
phorus according to Standard Methods (1980; U.S. Envi-
ron. Prot. Agency, 1983). A detailed description of the
comprehensive water quality monitoring program can be
found in Cassell et al. (1983).
RESULTS AND DISCUSSION
Weekly Activities
Weekly mean stream concentrations were positively cor-
related with weekly mean discharge but were not gener-
ally related to total weekly precipitation (Table 1). Sus-
pended solids concentrations were strongly related to
discharge. This positive relationship between discharge
and streamflow concentrations is characteristic of diffuse
sources of nutrients and sediment (Novotny and Chesters,
1981).
Weekly mean total phosphorus concentrations in
streamflow were positively correlated with weekly manure
applications within the watershed (Table 2). In Figure 4,
the darkened areas represent manure applications in the
watershed during 1983. Correlations generally decreased
when considering smaller components of the watershed
as compared to applications throughout the watershed.
Considering manure applications on the greatest runoff-
producing zones (soil hydrologic group D) did not improve
correlations. Manure applications in closer proximity
(63 m) to stream courses were not as well related to
Table 2.—Correlations (r) between mean weekly runoff concentrations (mg/l) and weekly manure applications (mT).
Applied to:
Total phosphorus
Orthophosphorus
Total Kjeldah! nitrogen
Ammonia nitrogen
Total suspended solids
Volatile suspended solids
Total
watershed
AT2
.26
-.04
-.18
.18
.14
Hydrologic
group D
.382
.15
-.11
-.23
.23
.20
Within
63m
.39*
.16
-.15
-.26
.18
.14
Within
63m on D
.20
.10
-.12
-.321
.15
.21
'indicates significance at P - 0.05
'Indicates significance at P - 0.01
27
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
streamflow total P as were overall watershed manure ap-
plications.
Generally, poor correlations between land use changes
and water quality were observed. For example, the corre-
lation between total phosphorus concentrations and the
percent of corn land was only -.19. A possible explanation
may be that only 3 percent of the watershed changed land
use between corn, alfalfa, and hay during this 1-year
study. Also, poor correlations were generally obtained be-
tween areas receiving field management and total phos-
phorus concentration (e.g., cultivation, r = .26). This lack
of correlation resulted partially from the timing of activi-
ties. For example, 81 percent of the cultivation occurred
during a 9-week period in the spring. During the remaining
43-week period, little or no cultivation occurred, whereas
weekly stream concentrations fluctuated greatly.
c: 10.0-
7i>
E 5.0-
I O.S-
o.
Y=0.189X
R=0.60
0.178
10.0 100 1OOO
MANURE APPLIED (mT)
loioo
Figure 5.—Total phosphorus (mg/L) and manure applied
Mean daily discharge rates were examined to estimate the
weeks of stormflow. Weekly land use activity data were
accumulated between stormflow periods and then com-
pared to stream concentrations. This method of compari-
son assumes primary nutrient and sediment movement
during stormflow.
When manure applications were accumulated between
stormflow periods and compared to in-stream concentra-
tions, stronger correlations resulted (Table 3). Total and
orthophosphorus and total Kjeldahl nitrogen were posi-
tively correlated to applied manure using this lagging tech-
nique. Generally, manure applied throughout the water-
shed correlated better with stream concentrations than
manure applied to runoff-producing zones. Proximity did
not appear to greatly influence these relationships. The
relationship between total phosphorus and accumulated
manure applied is shown in Figure 5.
Since both stream discharge and manure applications
were related to stream phosphorus concentrations, multi-
ple regression was used in an attempt to explain more of
the variation in stream concentrations. The best prediction
of total phosphorus concentration (P = 0.01) resulted
from using manure applied on Group D soils and total
suspended solids concentrations in runoff (Log total P =
0.15 Log Manure on D + 0.34 Log total S.S. - 1.09;
multiple r2 = 0.55). This relationship suggests that ma-
nure applications to low infiltration rate soils combined
with suspended solids in runoff are the primary variables
influencing stream phosphorus concentrations. On the av-
erage, 38 percent of the instream total phosphorus con-
centrations were in paniculate form. During storm events,
up to 90 percent of the total phosphorus was particulate.
Surprisingly, discharge did not significantly add to the re-
gression.
The concentrations of weekly total phosphorus in Jewett
Brook were positively related to mean weekly stream dis-
charge and the weekly amounts of manure applied to the
watershed. Considering manure applications adjacent to
the brook did not improve simple correlation relationships.
Accumulated manure applied between stormflow
events improved correlations with stream phosphorus
concentrations. However, the proximity of these applica-
tions did not greatly improve relationships.
Multiple regressions suggested that manure applica-
tions on low infiltration rate soils and suspended solids in
runoff explained variation in stream phosphorus concen-
trations more than other land use and hydrologic varia-
bles.
To better link land use activities to stream water quality,
one might consider mass export rather than just mean
concentrations using the lagging techniques described.
Shorter time intervals than weekly might also improve re-
lationships. Finally, quantify differences between seasons,
land use should be monitored for more than 1 year.
ACKNOWLEDGEMENTS: This research was supported through
funds supplied from the Rural Clean Water Program and by the
Vermont Water Resources Research Center. Cooperative agree-
ments are with U.S. Department of Agriculture, Soil Conserva-
tion Service and the Agricultural Stabilization and Conservation
Service, Franklin County Natural Resources Conservation Dis-
trict, Vermont Extension Services, and the Vermont Agency of
Environmental Conservation, Department of Water Resources.
The cooperation and generosity of the Jewett Brook farm opera-
tors are sincerely appreciated.
Table 3.—Correlations (r) between mean weekly runoff concentrations (mg/l) and accumulated manure between runoff
events (mT).
Total phosphorus
Orthophosphorus
Total Kjeldahl nitrogen
Ammonia nitrogen
Total suspended solids
Volatile suspended solids
Applied to:
Total
watershed
.602
.442
.52'
.35
.16
.19
Hydrologic
group D
.622
.43'
.501
.34
.24
.28
Within
63m
.602
.431
.51 1
.30
.19
.19
Within
63 m on D
.57*
.38
.481
.25
.19
.28
'Indicates significance at P = 0.05
'Indicates significance at P = 0.01
28
-------
REFERENCES
Cassell, E.A., et al. 1983. St. Albans Bay watershed comprehen-
sive water quality monitoring and evaluation. Rep. No. 1,
Background Infor. Vt. Water Resour. Res. Center, Univ. Ver-
mont, Burlington.
Dendy, F.E. 1981. Sediment yields from a Mississippi delta cot-
ton field. J. Environ. Qual. 10(1): 482-6.
Dillon, P.J., and W.B. Kirchner. 1975. The effects of geology and
land use on the export of phosphorus from watersheds. Water
Res. 9:135-48.
Dornbush, J.N., J.R. Andersen, and L.L. Harms. 1974. Quantifi-
cation of Pollutants in Agricultural Runoff. U.S. Environ. Prot.
Agency. Environ. Prot. Technol. Ser. EPA-660/2-74-005.
Dunne, J. 1969. The significance of 'partial-area' contributions
to storm runoff for the study of sources of agricultural pollu-
tants. Unpubl. mss., Agric. Res. Serv. Danville, VT.
Hill, A.R. 1981. Stream phosphorus exports from watersheds
with contrasting land uses in southern Ontario. Water Res.
Bull. 17(4): 627-34.
Kunkle, S.H. 1970. Sources and transport of bacterial indicators
in rural streams. Pages 105-132 in Proc. Symp. Interdiscipli-
nary Aspects of Watershed Management. Irrigation and Drain-
age Div. Am. Soc. Civil Eng.
Lake, J., and J. Morrison. 1977. Environmental impact of land
use on water quality. EPA-905/9-77-007-B. Final rep. Black
Creek Proj. U.S. Environ. Pro. Agency, Washington, D.C.
McDowell, L.L., et al. 1981. Toxaphene and sediment fields from
MONITORING AND ASSESSMENT TECHNIQUES
a Mississippi delta watershed. J. Environ. Qual. 10(11): 121-
25.
Novotony, V, and G. Chesters. 1981. Handbook of Non-Point
Pollution, Sources and Management. Van Nostrand Reinhold
Co. New York.
Omernik, J.M. 1976. The influence of land use on stream nutri-
ent levels. Ecol. Res. Ser. EPA-600/3-76-014. U.S. Environ.
Prot. Agency, Washington, D.C.
. 1977. Non-point source—stream nutrient level rela-
tionships: a nationwide study. Ecol. Res. Ser. EPA-600/3-77-
105. U.S. Environ. Prot. Agency, Washington, D.C.
Smolen, M.D., M. Rashake, and V.O. Shanholtz. 1975. Effect of
agricultural drainage on water quality. Am. Soc. Agric. Eng.
Pap. No. 75-2561.
Soil Conservation Service. 1979. Franklin County soil survey.
Nat. Coop. Soil Surv. U.S. Dept. Agric.
Standard Methods for the Examination of Water and Wastewa-
ter. 1980.15th ed. Am. Pub. Health Assn., Washington, D.C.
U.S. Environmental Protection Agency. 1974. Relationships be-
tween drainage area characteristics and non-point source nu-
trients in streams. Natl. Eutrophication Surv. Work. Pap. No.
25.
1983. Methods for chemical analysis of water and
wastes. EPA-600/4-79-020. Off. Res. Develop. Cincinnati, OH.
U.S. Geological Survey. 1972. St. Albans Bay, Vermont. SW/4
St. Albans 15' Quadrangle. Reston, VA.
Uttormark, P.D., J.D. Chapin, and K.M. Green. 1974. Estimating
nutrient loadings of lakes from non-point sources. EPA-660/3-
77-020. U.S. Environ. Prot. Agency, Washington, D.C.
29
-------
APPROPRIATE DESIGNS FOR DOCUMENTING WATER QUALITY
IMPROVEMENTS FROM AGRICULTURAL NPS CONTROL PROGRAMS
J. SPOONER
R. R MAAS
S. A. DRESSING
M. D. SMOLEN
R J. HUMENIK
National Water Quality Evaluation Project
North Carolina State University
Raleigh, North Carolina
ABSTRACT
Appropriate experimental designs are a function of the
question to be answered. In the case of agricultural NPS
control programs, the question is usually: How does BMP
implementation affect the magnitude of pollutant concen-
trations or loads? This paper discusses the assumptions,
analysis techniques, and advantages and disadvantages
of three basic experimental designs that can be used in
practical terms. Monitoring above and below an imple-
mentation site is generally more useful for documenting
the severity of an NPS than for documenting BMP effec-
tiveness. Time trend designs may be helpful; however,
water quality trends are a result of complex interactions
between land treatment, hydrology, and meteorologic fac-
tors. Accounting for these variables will therefore greatly
increase the probability of documenting water quality im-
provements associated with BMP's. Paired watershed
designs have the greatest potential for documenting im-
provements from BMP implementation because of the
ability to control for meteorologic and hydrologic variabil-
ity.
INTRODUCTION
A vast amount of information exists about best manage-
ment practices (BMP's) for control of agricultural nonpoint
sources (NPS). Most of this information, however, is from
research efforts that considered only field plots or small
watersheds. The investment of public funds to control
nonpoint source pollution from agriculture requires that
there be some assurance that nonpoint source pollution
control programs be effective in protecting water quality.
Hence, monitoring programs have been incorporated into
many of these programs to verify that their application to
the real world is, indeed, effective.
To evaluate the effectiveness of large-scale programs,
such as the Rural Clean Water Program projects (12,000-
40,000 ha), requires a great deal of money. Therefore,
data analysis should be planned and executed carefully
following a clearly specified experimental design. Lack of
an experimental design often results in wasted data col-
lection efforts, and inconclusive results.
In this paper, we present and discuss three alternative
experimental designs that are applicable to most nonpoint
source control projects. The methodologies are applicable
to surface and ground water studies that deal with BMP
effects on pollutant concentrations, loads, or the fre-
quency of standard violations. Most of our examples are
presented in terms of surface water concentration, but
only for convenience. This treatment is not rigorous statis-
tically, but we have attempted to present useful sugges-
tions and lay out some of the advantages, disadvantages,
and assumptions associated with each design.
MONITORING DESIGNS AND ANALYSES
Before and After (Time Trends or Time
Series Analyses) Uncorrected for
Meteorological Variables
Definition, Advantages, and Disadvantages: The be-
fore and after design is generally characterized by moni-
toring one or more sites in a watershed over time to deter-
mine whether a change in water quality conditions has
occurred. Agricultural nonpoint source control programs
generally involve water quality monitoring over a period of
several years below the agricultural nonpoint source to
assess the concentration or loading changes associated
with BMP implementation.
This design is the easiest to conduct with limited funds
and personnel. Little coordination between land treatment
and water quality monitoring personnel is required. In
nearly all cases the entire project area can be monitored.
There are no physical limitations to applying this basic
design to any watershed.
A disadvantage is that sensitivity is low unless meteoro-
logically related variables are measured (stream flow, pre-
cipitation, lake levels, ground water levels). Thus, it is diffi-
cult to attribute water quality changes to land treatment
measures. A long monitoring period is needed to assess
whether significant changes in water quality have oc-
curred. This is due to the extreme hydrological and mete-
orological variability in most systems.
Appropriate Hypothesis, Data Requirements, and
Assumptions: For conceptual clarity, all the hypotheses
will be stated in the alternative rather than the null form.
When meteorologic variables are not measured, the ap-
propriate hypothesis is:
Ha: Mean annual (or seasonal) pollutant concentrations
will decrease over time as BMP's are implemented.
The data needed to test this hypothesis are important.
The sampling regimes should be similar for pre- and post-
BMP implementation periods. Samples should be col-
lected at equally spaced intervals or other predetermined
schedules. It is important that sampling not be taken more
frequently than scheduled. This allows pre- and post-BMP
data to be compared with a minimum chance of sampling
bias.
One assumption associated with this hypothesis is that
every sample can be classified as belonging to either the
pre- or post-BMP implementation period. If statistical tests
are performed that divide the data into only these two
groups, it is assumed that the level of BMP implementa-
tion is similar in each of the post-BMP years. Since this is
30
-------
MONITORING AND ASSESSMENT TECHNIQUES
often not the case, these tests may produce conservative
estimates of effects.
Hypothesis Test, Conclusions, and Interpretations:
The hypothesis can be tested using the Students t-test:
tsampte
(Cpre -
L
Vpre
Vpost p
where
the number of samples taken in each year or
in each session if stratified, assumed constant
i = Pooled variance =
Y
E
y = the total number of years or seasons of
monitoring
re = the number of years or seasons pre-BMP
st = the number of years or seasons post-BMP
= the mean of the pre-BMP concentrations.
= the mean of the post-BMP concentrations.
This t-sample statistic is compared to a t-table with
(Y«n-Y) degrees of freedom. It should be noted that it may
be advantageous to delete the interim time period if it can
not be classified as pre- or post-BMP for this particular
analysis.
An analysis that takes into account the cumulative na-
ture of land treatment is the regression of concentration
versus BMP application level. A significant negative slope
suggests an improvement of water quality associated with
BMP's. This approach does not require deleting data from
intermediate years.
A third analysis that can be useful is generation of a
Quantile-Quantile (Q-Q) plot. This analysis requires sev-
eral steps. First, one generates a cumulative distribution
of concentration for each site. This involves ranking by
magnitude the concentration data and grouping it into per-
centiles. The mean for each percentile is calculated for
both the pre- and post-BMP periods. These pairs are then
plotted and the slope is tested to determine if it is signifi-
cantly less than 1. An example of this plot is given in
Figure 1 . In this example a slope of less than 1 suggests a
downward concentration trend.
Because uncontrolled variables such as flow have such
a pronounced effect, often a downward concentration
trend will not be observed. Even if a decrease in concen-
tration is seen, no cause and effect relationship with BMP
implementation level can be made. In a physical sense,
there are four possible scenarios that may occur.
1 . Mean flows increase; concentrations increase.
2. Mean flows increase; concentrations decrease.
3. Mean flows decrease; concentrations decrease.
4. Mean flows decrease; concentrations increase.
Of these four scenarios, there is generally only one (2)
that provides strong evidence that BMP applications im-
proved water quality. Also, without flow measurements, it
is not possible to determine which of these four situations
has occurred. Hence, without flow measurements, it is
inevitable that a long-term monitoring program will be re-
quired to average out the fluctuations caused by stream
flows, and to determine true effects of land treatment.
Before and After Time Trends Corrected for
Stream Flows
Definition, Advantages, and Disadvantages: This de-
sign involves monitoring both concentration and flows
over time at one or more sites in a watershed. Based upon
100-i
75 -
£ 50-
s 25 -
= PRE-BMP
= POST-BMP
CONCENTRATION
IMPROVEMENT
PRE-BMP CONCENTRATIONS
Figure 1.—An example of a Quantile-Quantile (Q-Q) plot
derived from a plot of cumulative frequency distributions of
concentration data from a pre- and post-BMP period.
previous studies, the variable with the greatest influence
on surface water loads and concentrations is stream flow
volume. (Froehlich, 1976; Johnson et al. 1974; McCool
and Papendick, 1975). Thus, stream flows will be used in
this and all subsequent examples that attempt to correct
for meteorologic variations.
The basic advantages are the same as for the case just
described. In addition, a stronger association with land
treatment can be made. A long monitoring period is still
needed to determine whether significant changes in water
quality have occurred. Disadvantages are reduced, but
unknown or unmeasured factors that occur during the
project may still greatly reduce sensitivity.
Appropriate Hypothesis, Data Requirements and As-
sumptions: The hypothesis tested in this experimental
design is:
Ha: Mean annual (or seasonal) pollutant concentrations
will decrease over time when corrected for stream
flows.
Flow-concentration pairs (concentration and flow
measurments) need not be taken at equally spaced or
predetermined time intervals. In fact, it can be seen from
Figure 2 that the required data can be generated more
efficiently if the monitoring is weighted toward periods of
high flow. A wide range of flows is needed to establish a
flow-concentration relationship, and the potential effects
of BMP's are often greatest at high flows. Since the flow-
concentration relationship often depends greatly upon
whether the sample is taken during the rising or receding
31
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
limb of the hydrograph (Baker, 1985), it may be advisable
to partition the data on this basis.
All the assumptions stated for .the unconnected, before
and after design still hold. In addition, this design assumes
that the BMP's will decrease pollutant concentrations
more than they will reduce stream flows. In general, the
assumption will hold for sediment and sediment-adsorbed
pollutants, but may be in error for pollutants lost primarily
in the dissolved phase of runoff. The pre- and post-BMP
flow-concentration sample pairs need to reflect similar
ranges in flows. If not, only the post-BMP data taken in the
flow ranges present in the pre-BMP data should be used
in the analyses.
Hypothesis Tests, Conclusions and inteirpiretations:
Separate linear regressions of concentrations versus
flows for the pre- and post-BMP periods can be per-
formed. The slopes are compared for equality for the two
periods as shown in Figure 2. From this analysis we can
determine whether concentrations have changed over
time for a given flow rate. With the establishment of a good
flow-concentration relationship, the effects of BMP's can
be distinguished under all four of the scenarios described.
There may be a significant seasonal influence on the con-
centration-flow relationship. This source of variability in
the data can be eliminated by partitioning the data by
seasons. The cost of this partitioning, however, is a loss in
the number of degrees of freedom (effective sample num-
ber), which decreases the sensitivity of the subsequent
statistical tests.
Deffinition, Advantages and Disadvantages: This exper-
imental design involves sampling a flowing system over
time above and below a potential nonpoint source. This
has classically been the design used to monitor the effects
of nonpoint source discharges to flowing systems.
The primary advantage of this approach is that it can
account for upstream inputs to the area of interest. For
agricultural nonpoint source projects, this will often be im-
o
o
portant for watersheds where the upper portions are in
nonagricultural land uses. In addition, some irrigation
management projects receive irrigation water that varies
greatly in quality on an annual or seasonal basis. Perhaps
the most common use of this design, however, is to docu-
ment the location and magnitude of sources. As with the
before and after design there is also the advantage that
little or no coordination is required between the land treat-
ment and water quality monitoring components of the proj-
ect.
If the surface or ground water system originates within
the nonpoint source area, there will be no suitable above
sites. Also, the design provides only limited control for
meteorologic variables, unless stream flow is monitored
as described in the before and after design. In addition, it
requires twice as many sampling sites as the before and
after design to monitor an equivalent amount of the water-
shed area. The procedure may have low sensitivity be-
cause individual nonpoint source inputs are often small
compared to background.
: This design will generally provide informa-
tion for testing two hypotheses: one concerning problem
identification, and another concerning the effects of
BMP's over time.
Ma a. Agricultural pollutant concentrations will be higher
downstream from a suspected agricultural nonpoint
source as compared to upstream.
Ha b. The difference between upstream and downstream
pollutant concentrations will decrease over time as
BMP's are applied.
Testing hypothesis a. requires paired concentration
data above and below the potential nonpoint source over
time during the pre-BMP period. For hypothesis b. the
same paired data are needed for both the pre- and post-
BMP periods.
The most important assumption for this design is that
sampling is timed so that the same parcel of water is being
sampled at the above and below sites. This requires some
understanding of the hydrology system.
Hypotheses Tests, Conclusions, and (Interpretations:
For hypothesis a. to determine whether there is a signifi-
cant concentration increase, a simple one-sided Student's
t-test is used to determine whether the means of the
paired differences between the upstream (Cup) and down-
stream (Cdown concentrations are different from zero.
where D = the average of the paired differences,
n
E
_ J = 1 (Cup —
n
SD =
sd
In many cases, it is desirable to know what percentage
of the pollutant concentration is attributable to the non-
point source. The best estimate of this can be calculated
from:
NPS Percentage
STREAMFLOW
Figure 2.—An example of separate linear regression of con-
centration versus streamflows for a pre- and post-BRflP per-
iod. Note that In this hypothetical example the data show a
significant decrease in post-BMP concentrations when cor-
rected for sJreamtflow, even though the actual concentration
mean is higher tor the post-BRflP period.
To test hypothesis b., paired differences (D|) must first
be calculated for pre- and post-BMP periods (Di = C, own
- Q Up). Then, each of the four analyses described for the
before and after design can be used to test for water qual-
ity improvements associated with BMP implementation.
Briefly, these include: (a) Student's t-test for determining
32
-------
MONITORING AND ASSESSMENT TECHNIQUES
whether pre- and post-BMP mean concentrations are dif-
ferent, (b) Q-Q plots, (c) linear regression of D, versus
BMP implementation level, and (d) linear regressions of D,
versus flow for pre- and post-BMP periods to test for
equality of flow-corrected D,'s.
From testing hypothesis a. we can conclude whether
the suspected agricultural nonpoint source is actually a
significant contributor to an identified water resource im-
pairment. From this, we can estimate the upper limit of
how such improvement can be accomplished using
BMP's.
For hypothesis b. the interpretations are very similar to
those that can be made for the before and after design. In
the cases where not all the water originates within the
project area this experimental design allows trends to be
established with more certainty than the before and after
design, because of the corrections for incoming concen-
trations.
Paired Watersheds Design
(Controlled-Experimental Design or
Treated-Untreated Design)
Definition, Advantages, and Disadvantages: The de-
sign consists of monitoring downstream from two or more
agricultural drainages where at least one drainage has
BMP implementation, and at least one does not. This de-
sign ideally possesses the following characteristics: (a)
simultaneous monitoring below each drainage, (b) moni-
toring at all sites prior to any land treatment (calibration
period) to establish the relative responses of the drain-
ages, and (c) subsequent monitoring, where at least one
drainage area continues to serve as a control through the
land treatment period, i.e., receives significantly less land
treatment than the other drainage areas.
This design controls for meteorologic (and to some ex-
tent hydrologic) variability, minimizing the need for moni-
toring meteorological parameters. In most cases, water
quality improvements related to BMP implementation can
be documented within a much shorter time frame. In addi-
tion, this design provides stronger statistical evidence of
the cause-effect relationship between agricultural non-
point source control efforts and water quality changes.
A disadvantage of this design is that land treatment and
water quality personnel must coordinate closely to match
implementation efforts with monitoring and data analysis
needs. For some projects it may be difficult to find ade-
quately similar drainages. Close physical proximity is es-
sential. Another disadvantage is the fact that control ba-
sins cannot receive as much land treatment, thus
reducing the potential water quality improvement for the
overall project area. This design is not intended to deter-
mine the location or severity of the nonpoint source.
Appropriate Hypothesis, Data Requirements, and
Assumptions:
Ha: An agricultural drainage with BMP's applied will
exhibit a decrease in pollutant concentrations over
time, relative to an untreated agricultural drainage.
Site selection is crucial to this design. A similarity in hy-
drology and land use is desirable. Sampling from the wa-
tersheds should be conducted consistently (either simulta-
neously or separated by a constant time interval).
Because concentration-flow relationships vary with rising
or falling hydrograph limb, it is desirable to partition data
on this basis.
It is assumed that paired watersheds have similar pre-
cipitation patterns, because of their geographic proximity.
The hydrologic response of the paired watersheds should
be consistent, even if actual concentrations are quite dif-
ferent because of differences in slope, soil type, cropping
CALIBRATION
PERIOD
TREATMENT
. PERIOD
CONTROL
BASINS
NO
RHP 5
NO
BHPs
EXPERIMENTAL
BASINS
NO
RHP 5
BHPs
APPLIED
CALIBRATION,
&
§
TREATMENT
CONTROL BASIN CONCENTRATIONS
Figure 3.—An example of data analysis for the paired water-
sheds experimental design. If the predicted watershed value
is significantly less during the treatment period as com-
pared to the calibration period, a significant improvement in
pollutant concentrations is Indicated.
patterns, and other factors. It is assumed that BMP imple-
mentation levels can be measured accurately. Finally, the
precipitation, stream flows, and cropping patterns should
be at least somewhat similar for the calibration and treat-
ment periods.
Hypothesis Tests, Conclusions, and Interpretations:
Linear regressions of the concentrations (or log concen-
trations) for the treatment versus the control watersheds
for the calibration and land treatment periods can be per-
formed (Fig. 3). A Student's t-test is performed to deter-
mine if the predicted treatment watershed values at the
mean control watershed concentration decrease over
time.
A decrease in the predicted treatment watershed values
suggests a positive effect of BMP's on the water quality.
This is stronger evidence of a cause-effect relationship
than that derived from any of the designs previously dis-
cussed because of greater control over the complex mete-
orologic, hydrologic, and temporal factors. Although this
design compares only a treated drainage with an un-
treated drainage, the results can be interpreted to indicate
that the BMP's have improved water quality in the treated
subbasins relative to the condition that would have existed
without treatment. It should be noted that this design doc-
uments water quality improvements only in the treated
subbasins; the accuracy of extrapolating results from the
test basins to other portions of the project areas will re-
main untested. This experimental design may develop
from a project area by chance, as BMP implementation
progresses in subbasins with varying levels of success.
SUMMARY
For documenting water quality improvements resulting
from BMP's within the shortest possible time period we
believe the paired watershed design is clearly superior.
because of its control of meteorologically-related varia-
bles. To document the magnitude of nonpoint sources
prior to implementing BMP's, the above and below design
33
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
has advantages over the other designs. The before and
after design is often the easiest design to follow, and can
yield useful results provided that streamflows or some
other surrogate measure of meteorologic variability is in-
corporated. Without correction for flow variability, it is un-
likely that the before and after design can document BMP
effects at the watershed level within any practical program
time frame. It should be noted that for many of the experi-
mental designs the time period required to observe BMP-
related changes will depend upon how large a change is
actually being made. For example, a 30 percent concen-
tration reduction will take much longer to observe above
the noise (variability) of the system than will a 90 percent
reduction.
At least one of these experimental designs should be
evident in any nonpoint source control project with water
quality monitoring. The most appropriate monitoring strat-
egies may include more than one of these experimental
designs. The choice of the most appropriate design will
depend upon the nature of the water resource impair-
ment, the water quality objectives of the project, the antici-
pated level and timing of land treatment, the topography
of the project area, and the financial resources available
for monitoring.
REFERENCES
Baker, D.B. 1985. Regional water quality impacts of intensive
row crop agriculture: A Lake Erie basin case study. J. Soil
Water Conserv. 40(1): 125-32.
Froehlich, H.A. 1976. Inorganic Pollution from Forests and
Rangelands. Water Resour. Res. Inst. Publ. No. Seminar-
WR021-76. Oregon State Univ., Corvallis.
Johnson, C.W., et al. 1974. Sediment yield from Southwest
Idaho rangeland watersheds. Am. Soc. Civil Eng. No. 74-
2505.
McCool, O.K., and R.I. Papendick. 1975. Variation of suspended
sediment load in the Palouse Region of the Northwest. Am.
Soc. Civil Eng. Pap. No. 75-2510.
34
-------
MONITORING FOR WATER QUALITY OBJECTIVES IN RESPONSE TO
NONPOINT SOURCE POLLUTION
K. W. THOMSON
Water Quality Branch
Environment Canada
Regina, Saskatchewan
G. D. HAFFNER
Water Quality Branch
Environment Canada
Ottawa, Ontario
ABSTRACT
The broad application and continued use of herbicides
'and pesticides have resulted in a major diffuse source
input of toxic material into aquatic ecosystems. The most
common present water quality assessment practices can-
not account for the entry of such compounds into the
environment. Unlike point source input, where levels,
quantity, and consistency of loadings are known, diffuse
source input must be estimated using assessment proce-
dures. Consequently, these assessment practices are not
designed for the development or site-adaptation of water
quality objectives—in Canada, water quality objectives
are used for determining best land use practice and pro-
viding protection to the aquatic ecosystem. These de-
mands on environmental assessments and the subse-
quent development of relevant water quality objectives
can only be achieved by studies that provide insight into
aquatic system behavior. The different environment proc-
esses and fates that potentially regulate a compound's
effect in the aquatic ecosystem emphasize the need for
system behavior information. Examples from different
systems illustrate the need for more comprehensive wa-
ter quality assessment procedures to develop water, qual-
ity objectives relevant to diffuse source inputs.
INTRODUCTION
Management of water quality within an aquatic ecosystem
involves three facets: measurement, evaluation, and re-
medial- action. Measurement pertains to the collection of
physical, biological, and chemical water quality data. Eval-
uation necessitates a set of criteria with which the meas-
ured water quality can be realistically compared (in Can-
ada these criteria take the form of water quality objec-
tives—negotiated limits designed to protect and support
designated water uses). These objectives provide the link
between water quality information and the water uses to
be protected and maintained within a given waterbody.
Remedial actions, if required for use protection, are based
on the measurement and evaluation information.
Approaches to the monitoring and assessment of water
quality, as well as water quality management, will vary
according to the relative significance of nonpoint and point
source pollution. Specific water quality objectives that are
used for evaluation do not vary similarly; however, their
effectiveness depends upon the related data as well as
the resultant management responses to them. Although
water quality objectives have application for both nonpoint
and point source pollution, the development of these ob-
jectives, their monitoring requirements, and the appropri-
ate management strategies may differ significantly. The
remainder of this discussion focuses on monitoring ap-
proaches, with regard to water quality objectives, and the
assessment of nonpoint source pollution.
Point source inputs to an aquatic ecosystem are usually
a consistent load of a given set of materials or chemicals.
Data sets can be generated in specific areas of a river
basin, and areas of noncompliance established with reme-
dial actions confined to specific sources. Diffuse loadings
from land use or atmospheric inputs tend to be more event
oriented without a quantifiable area of effect in the aquatic
environment. Basically, the complex nature of diffuse-
source inputs results in the need for more comprehensive
and extensive measurements and evaluation for develop-
ment of a suitable management strategy.
Nonpoint source pollution in Canada most often results
from agricultural practices, urban runoff, and atmospheric
deposition. Aspects of these concerns are contained in
three highly interrelated departmental priorities recently
identified by Environment Canada: Toxic Chemicals, Long
Range Transport of Airborne Pollutants, and Water Man-
agement (Environ. Can. 1983). To address these priorities,
data must be assembled, an evaluatory mechanism initi-
ated, and response programs implemented. Data collec-
tion necessitates an effective monitoring program; evalua-
tion may correspond to the use of water quality objectives
and the response usually consists of developing and im-
plementing management options.
This discussion critically examines the measurements
and evaluations required to develop water quality objec-
tives specifically for nonpoint source inputs. Selected ex-
amples illustrate how such measurement might be used to
determine the need and type of remedial action required
to protect the aquatic environment.
MONITORING NONPOINT SOURCE
POLLUTION
Water quality management requires a multiplicity of data
to resolve the conflict of economic uses (industrial, agri-
cultural) of water and the health of the aquatic environ-
ment (drinking water, fisheries, recreation). Historically,
monitoring programs have been expected to yield infor-
mation on many different aspects of water quality, and as a
result data bases were established with many distinct and
often incompatible rationales and designs. Generally,
however, these measurement rationales and designs can
be described in one of the following categories of environ-
mental monitoring: (1) crisis response, (2) general moni-
toring, and (3) understanding aquatic processes.
Crisis Monitoring
Crisis monitoring, the oldest form of environmental data
collection, includes observations such as the collapse of
35
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
certain fisheries in the Great Lakes, loss of potable water
supply because of an epidemic such as the typhoid out-
breaks that took place at the turn of the century, or the
number of beach closures occurring over a certain period
of time as happened on the Ottawa River. Although criteria
information indicates a need for environmental manage-
ment, it does not help make decisions to avoid such situa-
tions or identify solutions to ameliorate the problem; there-
fore, it is not relevant to this discussion.
General Monitoring
To determine the state of the aquatic resource requires a
general monitoring program that will yield data describing
the presence, level, and change over time of specific
chemicals entering the aquatic system as a result of
man's activities. Such programs include the collection of
water samples at regular intervals and usually over the
course of a number of years. Data generated from these
collections are used to describe an average water quality
condition of the sites. An example of such a network has
been the general water quality monitoring carried out by
the Water Quality Branch of Environment Canada, which
is based on fixed sampling sites and monthly sampling
frequencies (Whitlow, 1985). Such a network emphasizes
statistics to quantify the accuracy and precision of the
baseline data generated (see Loftis et al. 1983; Sanders
and Ward, 1978). When operated over a period of time,
the program yields data suitable for long-term trend or
intervention analyses.
Often, these data sets are also used to assess compli-
ance with water quality objectives. Usually, these objec-
tives are a simple concentration of a chemical in water,
and the linking of the measurement and evaluation com-
ponents of water quality management becomes little more
than asking the perennial question, do ambient conditions
comply with the objective? Because of our present reli-
ance on fixed monitoring sites, considerable effort has
been made to study the stochastic nature of general moni-
toring (Ward and Loftis, 1983) and determine the probabil-
ity of exceeding a water quality objective or guideline at
any particular point in time.
Compliance monitoring for water quality objectives in
the Prairie Provinces is based on a two-level approach
that provides a short- and long-term objective for each
water quality variable of concern. The short-term objective
is most commonly based on laboratory-derived criteria,
whereas the long-term objective is developed from system
variability (historic mean concentration ± 2 Standard Devi-
ations) to account for seasonal variations. Considering the
episodic nature of diffuse source loadings, the long-term
objective is more relevant for water quality management
concerned with diffuse source inputs. For example, the
use of herbicides and pesticides in the Prairie Provinces
follows crop cycles; application and land runoff provide
event-oriented inputs to the aquatic ecosystem. General
water quality monitoring in the area has demonstrated the
presence and levels of pesticides throughout the area and
indicated some presence of lindane and alpha-BHC in
locations well beyond the areas of use (Gummer, 1978).
Although such a data set indicates the need for water
quality objectives, it does not provide the information to
site-adapt the objectives with respect to potential effects
within the system.
Process Assessments
Designing environmental monitoring or assessment to
provide scientific advice for a specific issue requires a
third type of assessment—monitoring to characterize the
behavior of the system. Specific questions must be ad-
dressed. Is the correct substrate being sampled? Is the
hydrological regime of the system being taken into ac-
count? Are seasonal variations in concentrations and
loadings being considered? These exemplify the need for
a comprehensive multi-media approach to characterizing
a system, if effective water quality objectives are to be
developed and used to provide advice for sound water
quality management. This requires a knowledge of the
natural processes that regulate and often determine envi-
ronmental quality within an aquatic system.
Environmental priorities such as acid rain or toxic sub-
stances make it critical to know both the environmental
exposure and ecological effect of toxic chemicals. Expo-
sure is a function of partitioning a chemical among the
media under consideration (see Chapman et al. 1982);
whereas, the effect is a function of the system's tolerance
to the imposed stress. The need for process assessments
was emphasized by Chapman et al. (1982). They con-
cluded that a full understanding of the behavior of priority
pollutants in the aquatic environment will require collect-
ing considerably more information than chemical concen-
tration in certain compartments.
By virtue of its diffuse nature, understanding of non-
point source pollution relies more on monitoring and as-
sessment than does point source pollution. Direct meas-
urement of diffuse pollution sources is very difficult if not
impossible; thus evaluation (using water quality objec-
tives) depends upon a more careful monitoring of the sys-
tem. General monitoring is often satisfactory for point
source pollution because what and how much has been
contributed to the system is known. However, without the
benefit of accurate information on pollution inputs, more
comprehensive monitoring is needed to evaluate nonpoint
source pollution.
Process assessment requires measuring the system's
variability and examining the physical-chemical and bio-
logical processes that determine environmental quality.
Variability should consider statistical estimates of variance
as well as include the comparison and analysis of the
different sets of physical-chemical conditions. Under-
standing system behavior is an essential component of
environmental management, and criteria, guidelines, or
water quality objectives developed for good management
practice must be adapted to system behavior. Process
assessments provide the third step in developing relevant
water quality objectives and implementing wise environ-
mental management.
The value of process assessments is perhaps best de-
scribed in the Great Lakes phosphorus management pro-
gram. General monitoring provided estimates of total
phosphorus loads within the lakes. From 1972 to the
present, phosphorus loadings declined dramatically be-
cause of point source controls (1 mg/L), legislative con-
trols (detergents), and nonpoint source controls (no till). To
effectively manage phosphorus, and thus control the eu-
trophication of the Great Lakes, it was essential to deter-
mine what forms of phosphorus were most bioavailable
and what sources should be emphasized for control pro-
grams.
Although it showed decreased loadings and concentra-
tion declines in Total Phosphorus, general monitoring
could not provide the essential data to make such deci-
sions. Process monitoring, such as bioassays of phos-
phorus availability and utilization, could distinguish the
importance of the various sources. Consequently, appro-
priate decisions to target phosphorus loads for each of the
lakes were made and agreement was reached on the
most effective way to achieve the target levels.
During the 1960's, insecticides such as DDT and
Dieldrin represented a major diffuse source input into
Lake Michigan. Following the ban on the use and manu-
facture of these compounds in 1970, greater than 90 per-
36
-------
cent declines of DDT levels were measured in bloater
chubs between 1970 and 1980, and concentrations ap-
proached the Great Lakes Water Quality Agreement Ob-
jective of 1.0 jxg/L Dieldrin, however, increased in bloater
chubs over this time period, and concentrations continue
to remain over the water quality objective of 0.3 /*g/L.
The different environmental behavior of these two com-
pounds following regulatory action emphasizes the need
for process information. When developing water quality
objectives it is essential to know if a specific water quality
objective is achievable and how long it might take to meet
this objective. A lack of diffuse source input information
makes it difficult to discern if further controls are required.
What is the process that regulates levels of dieldrin in the
environment, and why is it different from DDT? Process
information is not yet available but is essential to answer
such a question.
For the Great Lakes, water quality objectives supported
by general monitoring have helped determine the need,
type, and priority of remedial effort required. They pro-
vided an indication of the general health and response of
the system. However, to maximize the effectiveness of
water quality objectives, both in terms of their validity and
especially their management potential, process informa-
tion has been needed. Process assessments better re-
solve how to obtain the specific levels represented by the
water quality objectives. They also evaluate the signifi-
cance of nonpoint sources of pollution to encourage more
efficient water quality management.
This point became apparent during the 1970's general
monitoring programs in the Qu'Appelle River Basin of
Saskatchewan which revealed that Province of Saskatch-
ewan water quality objectives (which are not site-specific)
for nutrients were routinely being exceeded. On the basis
of this monitoring and evaluation, it was assumed that
point source pollution was primarily responsible for this
situation. Management adopted the position that control-
ling point source pollution would alleviate the problem.
Tertiary waste treatment for the upstream cities of Regina
and Moose Jaw was installed. Subsequent monitoring re-
vealed little difference in nutrient values and it was not
until detailed process assessments took place that a sig-
nificant source of nutrients was determined to be of non-
point origin. Present water quality objectives, which are
not site-specific, have limited potential for water quality
management because of the overall significance of non-
point contribution of nutrients to the system. Therefore,
process assessment in this case indicates that water qual-
.ity objectives are probably not achievable through point
source controls but require comprehensive nonpoint
source mitigative measures.
CONCLUSIONS
Water quality monitoring for nonpoint source pollution
must be taken into account for developing and maintain-
MONITORING AND ASSESSMENT TECHNIQUES
ing water quality objectives. Evaluating the significance of
this pollution (through water quality objectives) and formu-
lating management responses rely on more dynamic as-
sessments than those provided by general monitoring. In
some situations general monitoring that relies on describ-
ing average condition may be the most cost-effective solu-
tion to supporting (as opposed to developing) water quality
objectives. However, nonpoint source pollution more often
demands a detailed characterization of a system's water
quality. A knowledge of the processes and interrelation-
ships that regulate environmental quality within any
aquatic system is usually required.
With Great Lakes eutrophication, the system forgave
the immediate lack of process information. However,
present day priority issues such as acid rain and the entry
of toxic substances into the environment will not be as
forgiving. In the interests of protecting and sustaining Ca-
nadian water resources, studies emphasizing the knowl-
edge of the system must be carried out. These data are
essential in developing relevant water quality objectives
and designing networks to support them. Furthermore, as
the use of water quality objectives grows, the information
gained from such studies will provide some interpretable
data, both in terms of information on the system and in
assessing the health of these resources.
ACKNOWLEDGEMENTS: The authors wish to thank William D.
Gummer, Chief, Water Quality Branch, Environment Canada,
Western and Northern Region, for his careful review and critical
comments on this manuscript.
REFERENCES
Chapman, P.M., G.P. Romberg, and G.A. Vigors. 1982. Design
of monitoring studies for priority pollutants. J Water Pollut.
Control Fed. 54(2): 292-7.
Environment Canada. 1983. Annual Report 1982-83. Ottawa.
Gummer, W.D. 1978. Pesticide monitoring in the prairies of
Western Canada. Water Qual. Inter. Rep. No. 4. Inland Waters
Director., Environ. Can., Regina.
Loftis, J.C., R.C. Ward, and G.M. Smillie. 1983. Statistical
models for water quality regulation. J. Water Pollut. Control
Fed. 55(8): 1098-1104.
Sanders, T.G., and R.C. Ward. 1-978. Relating stream standards
to regulatory water quality monitoring practices. In Proc. Am.
Water Resour. Assn. Symp. Establishment of Water Quality
Monitoring Programs. San Francisco.
Ward, R.C., and J.C. Loftis. 1983. Incorporating the stochastic
nature of water quality into management. J. Water Pollut. Con-
trol Fed. 55(4): 408-13.
Whitlow, S.H. 1985. Water quality assessment in Canada. Water
Qual. Bull. 10(2): 75-9.
37
-------
USE OF BJOASSAYS TO DETERMINE POTENTIAL TOXICITY
EFFECTS OF ENVIRONMENTAL POLLUTANTS
S. A. PETERSON
W. E. MILLER
J. C. GREENE
C. A. CALLAHAN
Hazardous Waste and Water Branch
U.S. EPA, Corvallis Environmental Research Laboratory
Corvallis, Oregon
ABSTRACT
Nonpoint source (NPS) runoff from mining, landfills,
roads, croplands, grazing lands, and forests can contain
chemicals harmful to aquatic organisms. Full scale bio-
logical surveys to determine their effects are difficult and
costly. Bioassays of environmental samples integrate the
effects of all toxicants contained in a sample. Biological
organisms are being used more frequently to identify toxi-
cant problems and to rank-order their severity. The Cor-
vallis Environmental Research Laboratory (CERL) has
developed a multi-media (aquatic/terrestrial) bioassess-
ment protocol to assist in the identification of toxicity po-
tentials associated with waste disposal. Similar tech-
niques can be used to identify NPS pollutants. The
bioassay response indicators are particularly useful in
identification of field-site problems where complex mix-
tures of pollutants might be present. Use of the
bioassessment protocol reduces the initial need for ex-
tensive chemical analyses, and produces data (toxicity
LCso information) in a form more readily understood by
the public than bulk chemical concentrations. The CERL
protocol has been used successfully to: (1) define and
rank-order the effects of selected heavy metals, herbi-
cides, and insecticides on microbes, earthworms, plant
seeds, algae, daphnia, and fathead minnow larvae; (2)
determine that rank-order of sensitivity differs with major
toxicant groups; (3) detect the presence of bioactive or-
ganic and heavy metal mixtures in field site samples
when concentrations of priority organic pollutants did not
exceed EPA criteria levels; and (4) identify the basic
chemical component of complex waste mixtures which
produce environmental toxicological effects. These types
of information should be useful in determining the poten-
tial effects of NPS pollutants and in designing measures
for their control.
INTRODUCTION
Nonpoint sources (NPS) pollution problems are among
the most pervasive, persistant, and diverse water quality
problems facing the nation. This presents a definite prob-
lem to water quality decisionmakers who traditionally have
addressed individual pollutants or site-specific sources of
pollutants. The individual chemical-by-chemical approach
requires a great deal of patience, time, money, and intel-
lect to determine the pollutants adverse impact. Also, de-
termining the substance producing the impact, the source
of the substances, and the area! extent of the problem is
difficult to address. Even extensive effort on a chemical-
by-chemical basis does not assure an accurate ecotoxico-
logical assessment, since one still has to relate environ-
mental chemical measurements to biological/ecological
impact. The approach most commonly employed is that of
calculating potential toxicity based on chemical concen-
tration of the 129 EPA consent decree chemicals (priority
pollutants) (Keith and Telliard, 1979) with extrapolation to
water quality criteria. The approach has been useful in
providing relative toxicity guidance, i.e., the relative toxic-
ity of various chemicals under laboratory conditions. How-
ever, it has become increasingly apparent that this ap-
proach has severe limitations concerning realistic and
accurate ecotoxicity estimates. Some of the problems as-
sociated with calculation of toxicity potentials based on
priority pollutant chemical concentrations are that:
1. The data bases for most chemicals are not complete
enough to permit the development of reliable criteria;
2. Most of the chemicals for which complete criteria
exist are not necessarily those most commonly found in
complex NPS or waste site discharges;
3. Application of criteria to field situations usually results
in highly conservative and, therefore, overly restrictive es-
timates of toxicity or misinterpretation of toxicity cause
and effect relationships;
4. Criteria for single chemicals were not intended to be
assembled additively and there is little evidence to sup-
port that use; and
5. For contaminated soil and sediment there are no cri-
teria on which to base decisions for judging if a site consti-
tutes a problem.
Biological assessment of environmental toxicity allevi-
ates most concerns associated with the above problems
and provides a direct indication of potential toxicity (Roop
and Hunsaker, 1985). An example was cited by Samoiloff
et al. (1983) when they discovered that the most toxic
sediment samples were those containing none of the EPA
consent decree chemicals. Miller et al. (1985) have dem-
onstrated similar results with the bioassessment of haz-
ardous waste site samples using a multimedia bioassay
procedure. Brown et al. (1984) demonstrated the inability
of chemical analyses to provide a comprehensive evalua-
tion of the toxicity potential of hazardous industrial wastes.
They demonstrated further that a combined testing proto-
col using bioassays and organic chemical analysis was
effective in identifying the toxicity potential of such wastes.
A recommendation from Brown et al. (1984) was that a
battery of bioassays be used to define the toxicity of
wastes. The purpose of this paper is to demonstrate that
such a bioassay test battery and analysis of results can
be used to (1) identify and rank-order toxicity hazard po-
tential of waste site samples; (2) help define and quantify
areal extent of toxicity potentials; (3) help identify what
chemical fractions of a complex waste contribute signifi-
cantly to their overall toxicity; and (4) suggest that similar
procedures might be used to assess the impacts of a
broad spectrum of NPS pollutants. This paper is based, in
part, on recently published and ongoing research con-
ducted or sponsored by the Hazardous Materials Assess-
ment Team at the EPA Corvallis Environmental Research
Laboratory.
38
-------
MONITORING AND ASSESSMENT TECHNIQUES
METHODS
Biological organisms respond to the adverse effects of a
variety of specific pollutants (Fed. Water Pollut. Control
Admin. 1968; U.S. Environ. Prot. Agency, 1976). However,
there has been relatively little comparative toxicology
done on environmental samples using a broad spectrum
of organisms comprising both aquatic and terrestrial com-
partments of the ecosystem. For this purpose, we have
adopted a multimedia bioassessment protocol described
by Porcella (1983). The bioassays in the Porcella protocol
include assessments of water and soil leachate toxicity on
seed germination/root elongation (lettuce, Lactuca sativa
L), earthworms (Eisenia foetida), algae (Selenastrum ca-
pricornutum), daphnia (Daphnia magna), and fathead min-
now larvae (Pimephales promelas). In addition, we have
conducted Microtox (Photobacterium phosphoreum) tests
(Beckman, 1982). Our approach has been to conduct
comparative lexicological studies on pure chemicals and
mixtures of chemicals in the laboratory to increase our
confidence that biological responses to these substances
are predictable and relatable to environmental samples
(Miller et al. 1985). All toxicity responses are expressed as
ECso or LCso concentrations for comparison.
We have focused on substances in chemical extraction
groupings. Metals, base neutral organics, acid organics,
and pesticides were extracted with water (4 ml water to 1 g
soil). Bioassays were performed using these aqueous ex-
tracts. The predicted bioassay response, based on chemi-
cal concentration and criteria for certain chemicals, was
then compared with bioassay responses on environmental
samples dominated by the mixture of chemicals in ques-
tion. This approach has permitted us to test the hypothe-
sis that bioassay of environmental samples will produce
ECso or LCso estimates significantly different from those
predicted by calculation based on chemical concentra-
tions with extrapolation to water quality criteria. Also, we
have examined the relative toxicity potential of various
metals, priority organics, and nonpriority organics in sam-
ples, from the Western Processing Superfund site at Kent,
Washington. This was accomplished by incremental inac-
tivation of metals with EDTA (at an EDTA to metals molar
ratio of 4:1, based on Cu inactivation) and methylene chlo-
ride extraction of priority organic chemicals (Eichelberger
et al. 1983) followed by algal assay examination. Chemical
quality control was assured by surrogate spike recovery
analysis coupled with daily calibration of the GC/MS sys-
tem.
Extent of chemical contamination was determined us-
ing a modified phytotoxicity test described by Thomas and
Cline (1985). Lettuce seeds were used to test the toxicity
potential of soils collected along four 90 m long parallel
transects that were 15 m apart. Soils from 0-15 cm depth
and 15-30 cm depth were used since they encompassed
the root zone in the area. The site was located downwind,
along a suspected concentration gradient perpendicular
to an open ditch known to have transported liquid organic
wastes associated with the manufacture of herbicides, in-
secticides, and neurotoxin gases at Rocky Mountain Arse-
nal, Colorado. Thomas et al. (1984) have described the
statistical sampling design in greater detail. Phytotoxicity
data from the site were analyzed using kriging. Kriging is a
statistical technique developed in the mining industry
(Clark, 1982). Only a limited number of samples are re-
quired to successfully define a contaminated area using
kriging. The technique employs a weighted moving aver-
age that calculates point estimates or block averages over
a specified grid. Output of the kriging analysis for this
study is a contour map displaying areal variation in phyto-
toxicity.
FINDINGS AND DISCUSSION
Comparative Toxicology
Miller et al. .(1985) conducted comparative toxicological
studies on several known single and complex organic and
metal contaminants in the laboratory using the Porcella
(1983) bioassessment protocol plus the Microtox Test (Be-
ckman, 1982). They concluded that:
1. The protocol test organisms responded differentially
to various pollutants and their ECso or LCso results gener-
• ally conformed to the range of values reported in the litera-
ture for individual chemicals and metals;
2. Test organism rank order of sensitivity differed with
major toxicant groups, suggesting that certain bioassays
are better suited than others to assess given chemical
groups;
3. Algae (Selenastrum capricornutum) was the most uni-
formly sensitive test organism across a broad spectrum of
pollutant groupings; and
4. Differences in sensitivity levels of the test organisms,
relative to the toxicant assayed, can be used to identify
those biotic components most susceptible to the presence
of toxicants and to draw an educated conclusion as to the
contaminant type producing the toxic effect.
Based on the conclusions drawn from bioassay re-
sponses to pure chemical substances in the laboratory
and the assumption that bioassays integrate the toxicity
effects of all sample components regardless of their com-
position, Miller et al. (1985) bioassayed soil and soil elutri-
ate samples from seven diverse hazardous waste sites
(Table 1). The samples were dominated by heavy metals,
solvents, phthlates, phenols, pesticides, and herbicides.
Relative, integrated biotic toxicity of the sites and their
rank ordering could be determined by calculating the
arithmetic average toxicity across the different tests in Ta-
ble 1. If one was concerned primarily with potential
aquatic impacts, the algae, Daphnia, and Microtox tests
probably would be the most applicable indicators. The
sensitivity of algae appears to be much greater than the
other bioassays for most of the samples.
Algae responded adversely to all but one of the sam-
ples. In that case, no aquatic test responded adversely.
Toxicity rank ordering, such as that shown in Table 1,
would be helpful in: (1) determining potential environmen-
tal impacts; (2) directing further chemical analyses within
sites; and (3) ranking cleanup across or within various
sites. Bioassay data might be used to monitor toxicity
changes in samples before and after waste cleanup or the
adoption of various NPS management alternatives, thus
helping to determine the degree of treatment success.
Kriging of Bioassay Data
Another means of assimilating bioassay data into a format
useful for problem solving and remedial design relative to
chemical hazard assessment is that of kriging. Phytoas-
say responses for soil samples from Rocky Mountain Ar-
senal were subjected to kriging as described under meth-
ods. Kriging the 0-15 cm phytotoxicity data, with the
resultant toxicity potential contours is shown in Figure 1.
Thomas et al. (1984) compared kriged phytotoxicity bio-
assay estimates (Figure 1) with sample site-specific plant
mortality data (Figure 2). This type of graphic interpolation
could be very useful in making waste site cleanup deci-
sions or in designing NPS watershed or ecoregional con-
taminant source controls. For example, if it was deter-
mined that the 30 percent mortality contour should be
used as the criterion for remedial action for the conditions
shown in Figure 2, the area below the 30 percent solid
contour line would be targeted for acton.
39
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 1.—ECM response for percent in soil (earthworm) or soil elutriate with associated complex chemical contaminants
from selected hazardous waste sites.
Bloassay Response
(Percent of soil or elutriate required)
to produce an
Waste Site
Holder Chemical
. West Virginia
Western
Processing
Kent, WA #17
Big John Houldt
West Virginia
Hollywood
Memphis, TN
Sharon Steel
Fremont, NY
Sapp Battery
Cottondale, FL
Thiokol
Chester, WV
Major Chemical
Group
Pesticides,
herbicides
Heavy metals,
phenols,
solvents,
pesticides
PAH3, unknown
organics
Pesticides
Heavy metals,
tar, PAH
Heavy metals
Diphenylamine
Algae
2.1
0.2
5.4
24.0
0.6
41.0
NE
Daphnia
3.6
5.6
87.0
22.0
30.0
70.0
NE
Microtox
18.0
2.2
28.0
>90.0
99.0
NE
NE
RE1
3.6
37.0
NE4
NE
NE
NE
NE
Earth-
worm
70.0
55.0
<10.0
>25.0
>75.0
NE
35.0
Arithmetic
Average
19.5
20.0
46.1
52.2
40.4
22.2
87.0
'Root elongation test.
2Earthworm 14 day soil contact test (LC$o).
3PAH = polynuclear aromatic hydrocarbons.
4NE = No effect observed at 100% of the soil or soil elutriate. Therefore, NE is factored into the arithmetic average as 100%, i.e., 100% of the soil or elutriate produced no
effect on the test organisms; the greater the percent soil or elutriate required to produce the EGyt. the less toxic is the sample.
90
80
55 70
CD
Z
co
I
CJ
5
x
cr
o
5
O
oc
_
<
60
50
t=\ o 10
1 I 10« 20
20-30
30* 50
50* 75
75»-100
0 15 30
DISTANCE (m) FROM NORTHEAST CORNER
Figure 1 .—Estimated lettuce seed mortality (based on krlglng) for the 0-15 cm soil fraction from the Rocky Mountain Arsenal
(from Thomas et al. 1984).
40
-------
90
<
z
in
80
70
I «
CO
Z
o
t-
5
z
cc
o
5
o
cc
50
40
E 30
u
z
10
MONITORING AND ASSESSMENT TECHNIQUES
9
VZA
ACTUAL MORTALITY
LESS THAN 30%
ACTUAL MORTALITY
GREATER THAN 30%
KRIGING ESTIMATE >30% MORTALITY
O
o.
15 30
DISTANCE (m) FROM NORTHEAST CORNER
45
Figure 2.—A comparison of greater than 30 percent lettuce seed mortality (estimated from krlglng) to observed lettuce seed
mortality for the 0-15 cm soil fraction from the Rocky Mountain Arsenal (from Thomas et al. 1984).
Unfortunately, the hazardous waste site situation is
more complex than the kriged phytotoxicity surface data
(0-15 cm deep) would indicate. Samples from the 15-30
cm depth at the same site produced the results shown in
Rgure 3. Comparison of kriging estimates with plant mor-
tality data at this depth is shown in Figure 4. It is evident
that site cleanup based on the surface sample greater
than 30 percent mortality results would omit significant
areas of contamination. This information makes the reme-
dial action plan more complicated, but it adds significant
realism to the site assessment. A final remedial action
decision that includes consideration of chemical bioavaila-
bility as determined by integrative bioassay endpoints
should greatly enhance the probability of contaminant
cleanup success. Chemical information alohe cannot as-
sure an accurate assessment of toxicity potentials and in
some instances might lead to misinterpretation of lexico-
logical cause and effect relationships.
Chemical Analysis and Bioassay
Hazardous waste assessment and NFS pollution prob-
lems are similar in that each has traditionally been as-
sessed from a chemical perspective. Severity of the prob-
lem has been assessed relative to the concentration of a
given chemical producing a given type and degree of re-
sponse under laboratory conditions. Controlled condition
laboratory response tests have been used extensively to
develop water quality criteria for various chemicals. Prob-
lems associated with the extrapolation of these criteria to
assess field conditions were mentioned in the introduc-
tion. In addition, combinations of pollutants and different
attenuating characteristics of a site are difficult to assess
when calculating toxicity estimates.
Direct bioassay of samples tends to minimize many of
these problems. Bioassays integrate the lexicological ef-
fects of all sample components regardless of their type
41
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
DISTANCE (m) FROM NORTHEAST CORNER
Figure 3.—Estimated lettuce seed mortality (based on kriging) for the 15-30 cm soil fraction from the Rocky Mountain
Arsenal (from Thomas et al. 1984).
and amount. Chemical presence is of limited concern, but
bioavailability of the chemical and its effect on the test
organism are of great concern. Bioassay of the waste
samples provides a direct estimate of the chemical's toxic-
ity potential.
Hazardous waste cleanup decisions have relied heavily
on analysis of the EPA 1965 consent decree chemicals
(the 129 so-called priority pollutants). Concentrations of
these pollutants in excess of water quality criteria values
have been used to justify various cleanups, but in many
instances environmental criteria do not exist. In these
cases the chemical information may be more misleading
than it is helpful since one suspects there may be some
hazard, but there is little information for determining the
degree of hazard based on the chemical analysis.
Herein lies the benefit of the bioassay procedure. Soil
and water bioassays in the Porcella (1983) bioassessment
protocol will provide an indication of toxicity to various
compartments of the system. Also, it will provide a quanti-
tative (ECso or LCso) ranking of the toxicological impact
potentials among those compartments.
We believe that reliance on chemical criteria alone, and
particularly those for priority pollutants, could lead to erro-
neous decisions concerning remedial actions. The gen-
eral chemical analytical protocol for hazardous waste site
samples calls for priority metal and organic identification
DISTANCE |m| FROM NORTHEAST CORNER
Figure 4.—A comparison of greater than 30 percent lettuce
seed mortality (estimated from kriging) to observed lettuce
seed mortality for the 15-30 cm soil fraction from the Rocky
Mountain Arsenal (from Thomas et al. 1984).
42
-------
MONITORING AND ASSESSMENT TECHNIQUES
and quantification. In some instances the next 10 most
prominent GC/MS peaks beyond the priority organic
might be "identified." Data bases for many of these pollu-
tants are too limited to allow one to develop rigorous water
quality criteria. This is especially true for nonpriority or-
ganics. Where toxicity data are not available, it might be
necessary to "estimate" the potential toxicity of chemicals
based on their similarity to other chemicals for which toxic-
ity data does exist. This introduces yet another uncertainty
factor. At present there seems to be no satisfactory
method of estimating toxicity for organic contaminants
short of direct bioassay of environmental samples.
Figure 5 illustrates how difficult it might be to estimate
environmental toxicology or the cause of toxicity based on
chemical analyses of priority pollutants. The figure repre-
sents a typical GC/MS scan of a waste site sediment lea-
chate sample. Results in Table 2, with the exception of the
onsite ponded water, represent sediment leachates from
an offsite reference control (East Ditch), an onsite refer-
ence (005, thought to be uncontaminated) and two offsite
stream sediment samples (017 downstream and 020, up-
stream). Sample 005 contained four identifiable priority
organics, nine identifiable nonpriority organics, and four-
teen unidentifiable nonpriority organic substances.
Concentrations of phthalates, ethylbenzene, nitro-
samines, and phenol priority pollutant fractions for the var-
ious samples collected at Western Processing are shown
in Table 2. The table also shows the nonpriority organic
fractions and the total organics. Among the four identifi-
able priority pollutants, an environmental criterion exists
only for phenol (3.4 mg/L). Assuming that priority pollutant
concentrations are among the most important consider-
ation of hazard potential at a site and that water quality
criteria are paramount in assessing hazard potentials,
sample 017 should be highly toxic due to the presence of
phenol at a concentration of 18.3 mg/L. Chemical concen-
trations and water solubilites of the other priority pollutants
would suggest that the other samples might be nontoxic.
Bioassay of the samples did not support the conclusion
(Table 3). Comparing the mean EC50 or LC50 value for the
different test organisms it can be seen that the toxicity of
sample 017 was quite similar to the East Ditch Control
sample. The upstream reference sample was not toxic.
The onsite ponded water was highly toxic as was sample
number 005 (thought to be uncontaminated). Toxicity of
the samples increased as the nonpriority organic fractions
increased.
To test the apparent relationship between toxicity and
the nonpriority organic component of the Western Proc-
essing Samples we conducted algal assays on 0-1.0 m
integrated soil core samples taken on site at locations 1,
11, and 17 (the latter should not be confused with sedi-
ment sample 17 above). Results of the algal assays are
shown in Figure 6. The results indicated that soil cores
from site 17 were the most toxic and that toxicity in-
creased across the three samples as the concentrations of
soluble metals, soluble priority organics, and total soluble
organics increased. It was not readily apparent from this
which toxic component was dominant in the system.
There was some evidence that toxicity increased with
depth in the soil column (not shown in these data). There-
fore, we elected to use leachate from the 3 m (integrative
depth from 2-3 m) depth at site 17 to further evaluate the
toxic components of the samples. Bioassays were run se-
quentially on untreated sample, EDTA chelated sample
(metals inactivated) and on combined chelation/priority or-
RIC .
Priority pollutants = 3
Non-priority
identifiables = 9
Unknowns
= 14
SS
786
Ethyl
benzene
1839
.0
o
-c
IS
II76
IS
Dibutyl
phtholote 1626
I 1465 \
~..t l-llll
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 2.—Concentration (mg/L) of priority and nonpriority organlcs contained In sediment leachates from Western
Processing (modified from Miller et al. 1985).
Constituent
Phthalates
Ethylbenzene
Nitrosamines
Phenol
Nonpriority
Organics
Total Organics
Control
0.002
0.002 .
< 0.001
< 0.001
0.3
0.3
East Ditch
Water
0.010
< 0.001
< 0.001
< 0.001
15.9
15.9
Pond
005
0.001
0.001
< 0.001
< 0.001
0.8
0.8
017
0.011
0.001
0.010
18.370
11.9
29.9
020
0.002
0.001
< 0.001
< 0.001
0.2
0.2
Table 3.—ECM or LCM response in % sediment1 (earthworm), sediment elutriate, and surface water to chemical
contaminants in Western Processing Samples (modified from Miller et al. 1985).
Test Organism
Algae
Daphnia
Microtox 5 min
15min
30 min
Lettuce RE
Earthworm4
Mean ECso5
East Ditch
Control
45
90
NE
NE
NE
NE
NE
87.0
Pond
Water
0.8
18.5
82.7
21.3
10.2
NE
—
32.4
005
0.4
3.3
41.2
<5.6
<5.6
61.4
>50<100
34.1
017
24.9
NE
55.4
50.1
43.4
49/1 OO3
>100
73.7
020
NE2
NE
NE
NE
NE
NE
—
>100
1 Sediment soil samples.
2 NE = no significant toxicity was observed.
3 49/100 = 49% inhibition in 100% sediment elutriate.
4 LCso values = concentration at which 50% mortality occurs.
5 Mean of algae, daphnia, microtox 30 min, lettuce RE, and earthworm tests.
ganic extracted (methyiene chloride) sample. Results
show that chelation of soluble metals with EDTA de-
creased toxicity 90-fold, but that the chelated elutriate re-
mained highly toxic (Figure 7). Significant additional toxic-
ity reduction was not realized when the sample was
subjected to combined chelation and priority organic ex-
traction. It appears from this analysis that metal toxicity
dominated the Western Processing samples, but that non-
priority organic chemicals alone were sufficient to classify
the soil leachate as highly toxic. The lexicological influ-
ence of priority organics in these samples appears to have
been minimal. Therefore, predicted toxicity of these sam-
ples based on the concentration of priority pollutants
would have severely underestimated sample toxicity
SUMMARY
We have attempted to develop a biological toxicity screen-
ing protocol that has broad-based application potential.
Based on results to date we believe that
1. A modified Porcella bioassessment protocol can be
used to define and rank order the effects of selected
heavy metals, herbicides, and insecticides.
2. Selected segments of the protocol can be used to
assess the influence of complex wastes under field condi-
tions, i.e., there is a relationship between laboratory bioas-
say responses to environmental samples and actual field
conditions.
3. The protocol can be used to assess environmental
toxicity potentials in situations where water quality criteria
are lacking or nonexistent.
4. Direct bioassay of environmental samples produces
toxicity results significantly different from those predic-
tions based on measured chemical concentrations with
extrapolation to water quality criteria.
5. Experience gained from the bioassay of hazardous
waste site samples should have application to many as-
pects of the NFS pollution problem.
6. Algal assay appears to have great universal toxicant/
stimulant assessment potential based on sensitivity to var-
ious toxicants.
ACKNOWLEDGEMENTS: We thank Walt Burns and Glenn
Wilson for their excellent chemical analytical support. Also,
Mary Debacon, Mike Long, Cathy Lee Bartels, Loren Russell,
and Julius Nwosu are acknowledged for their dedication to ex-
cellence in performing the variety of bioassays reported in this
paper.
REFERENCES
Beckman, Inc. 1982. Microtox System Operating Manual. Beck-
man Instruments, Inc. Microbics Operations. Carlsbad, Cali-
fornia.
Brown, K. W., K. C. Donnelly, and J. C. Thomas. 1984. Use of
short-term bioassays to evaluate environmental impact of
land treatment of hazardous industrial waste. EPA-600/S2-84-
135.
Clark, 1.1982 Practical Geostatistics. Appl. Sci. Publ., London.
Robert S. Kerr, Environmental Research Laboratory, Ada, Okla-
homa 74820. Clark, I. 1982. Practical Geostatistics. Applied
Sci. Publishers, London.
Eicnelberger, J. W, E. H. Kerns, P. Olynyk, and W. L. Budde.
1983. Precision and accuracy in the determination of organics
in water by fused silica capillary column gas chromatography/
mass spectrometry and packed column gas chromatography/
mass spectrometry. Anal. Cnem. 55:1471-79.
44
-------
MONITORING AND ASSESSMENT TECHNIQUES
100,000
10,000
1,000
I H20 SOLUBLE METALS
TOTAL SOLUBLE ORGANICS
I SOLUBLE PRIORITY ORGANICS
ALGAL EC50
-= 100,000
S
I
100
10
1.0
0.1
0.01
0.01
1 11
DRILLING SITES
17
Figure 6.—Response of algal assay, ECM (Se/enasfrum capri-
cornutum) to chemical contaminants In 0-1 m depth soil elu-
triates from well drilling site numbers 1,11, and 17 at the
Western Processing waste site.
100,000 —
10,000 —
1,000 —
— 100,000
-= 10,000
— 1,000
01
a
I
H20 SOLUBLE METALS
TOTAL SOLUBLE ORGANICS
SOLUBLE PRIORITY ORGANICS
ALGAL EC
100 -
-= 100
i
o
ui
3
0.1 =-
0.01
0.01
Figure 7.—Algal assay ECM response to Western Processing
soil elutriate from well drilling site 17 at the 3-4 m depth prior
to and after chelatlon of heavy metals and extraction of pri-
ority organic pollutants.
Federal Water Pollution Central Administration. 1968. Water
Quality Criteria. Report of the National Technical Advisory
Committee to the Secretary of the Interior. U.S. FWPCA,
Washington DC.
Keith, L. H., and W. A. Telliard. 1979. Priority pollutants: a per-
spective view. Environ. Sci. Technol. 13:416-23.
Miller, W. E., S. A. Peterson, J. C. Greene, and C. A. Callahan.
1985. Comparative toxicology of hazardous waste site
bioassessment test organisms. Environ. Qual. (accepted for
publication, May 1985).
Porcella, D. B. 1983. Protocol for Bioassessment of Hazardous
Waste Sites. EPA-600/2-83-054 U.S. Environ. Prot. Agency,
Corvallis, OR.
Roop, R. D., and C. T. Hunsaker. 1985. Biomonitoring for toxics
control in NPDES permitting. J. Water Pollut. Control Fed.
57:271-277.
Samoiloff, M. R., et al. 1983. Combined bioassay-chemical frac-
tionation scheme for the determination and ranking of toxic
chemicals in sediments. Environ. Sci. Technol. 17:329-333.
Thomas, J. M., et al. 1984. Characterization of chemical waste
site contamination and its extent using bioassays. Battelle
PNL-5302. PNL, Richland, WA.
Thomas, J. A., and J. F. Cline. 1985. Modification of the
Neubauer technique to assess toxicity of hazardous chemi-
cals in soil. Environ. Tox. Chem. 4:201-207.
U.S. Environmental Protection Agency, 1976. Quality Criteria for
Water. EPA-440/9-76-023. Washington, DC.
45
-------
Legal Aspects of Nonpoint
Source Pollution
AN OVERVIEW OF THE NATIONAL NONPOINT SOURCE POLICY
AMY MARASCO
CLAIRE GESALMAN
VIVIAN DAUB
The Synectics Group
Washington, D.C.
CARL MYERS
JAMES MEEK
U.S. Environmental Protection Agency
Washington, D.C.
ABSTRACT
A task force composed of representatives of Federal,
State, interstate, and local agencies and several private
groups recently completed work on a new national non-
point source policy. This policy will provide a framework
for all nonpoint source programs. It sets out what activi-
ties are to be undertaken by each group. Among other
things, the policy speaks to development of implementa-
tion strategies by the many agencies involved in nonpoint
source management. In fact, the U.S. Environmental Pro-
tection Agency (EPA), the other Federal agencies in-
volved in the task force, and several States have already
developed implementation strategies. This paper
presents key points from the policy and highlights of the
various strategies, and it focuses on the most critical as-
pects for success. It also describes the evaluation frame-
work EPA plans to use in assessing State program imple-
mentation strategies.
RESEARCH POINTED TO THE NEED FOR
A NATIONAL NONPOINT SOURCE
MANAGEMENT APPROACH
The impetus to embark on the process of developing a
Federal, State, and local nonpoint source pollution policy
began several years ago. In response to a Congressional
mandate, the U.S. Environmental Protection Agency
(EPA) conducted research to identify the magnitude and
scope of this pollution problem. The research effort re-
sulted in a Report to Congress, completed in the fall of
1983, which concludes that the pollutant loads from non-
point sources present continuing problems in our efforts to
achieve water quality goals and maintain designated uses
in many parts of the Nation.
Other research efforts indicated similar findings:
• The 1982 State Section 305(b) reports indicated that
virtually all of the States experienced water quality prob-
lems caused by nonpoint sources. One-half of the States
identified this as a major barrier to achieving individual
State water quality goals.
• The Association of State and Interstate Water Pollu-
tion Control Administrators (ASIWPCA) conducted its
Nonpoint Source Pollution Survey in February 1984. Sur-
vey results showed that 78 percent of the States saw their
nonpoint source problems as greater than or equal to
those caused by point sources.
• The 1983 Environmental Management Reports re-
vealed that 6 out of the 10 EPA Regions considered non-
point source pollution to be the principal remaining cause
of water quality problems for their geographic regions.
In addition to the technical findings about the severity of
the nonpoint source pollution problem, the Report to Con-
gress discussed the institutional and management difficul-
ties associated with addressing the problem, in particular,
program coordination was identified as a problem area.
Because of the number of Federal, State, and local agen-
47
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
cies involved—often with overlapping roles, responsibili-
ties, and jurisdictions—the Report called for a coordinated
management strategy.
A TASK FORCE IS ORGANIZED
Because one of its most direct agency missions is related
to water quality, EPA assumed the leadership role and
organized a task force. A group of 50 individuals repre-
senting Federal, State, and local agencies, as well as the
private sector, worked for about 10 months to design a
policy. The task force was successful in developing a pol-
icy that each agency endorsed.
The final policy establishes the framework and direction
used by each participating agency to design its own indi-
vidual strategy. The individual strategies specifically iden-
tify how—given each agency's perspective, mission, and
capabilities—each agency can follow the policy and ad-
dress the nonpoint source pollution problem. Supported
by these strategies, the policy itself becomes a more pro-
found statement.
THE NONPOINT SOURCE POLICY SETS
A DIRECTION
The objective of the policy is "to support and accelerate
the development and implementation of nonpoint source
management programs that ensure water quality protec-
tion while recognizing the competing uses of resources."
Eight actions are listed as fundamental elements for the
overall policy to succeed.
1. To build upon the current compendium of knowledge
and to promote further research efforts. The group recog-
nized that much has been done in this area, and that
existing work should be enhanced, not recreated.
2. To identify the appropriate roles of each agency, un-
derstanding that both the public and the private sector
must be involved.
3,4. The third and fourth actions go hand in hand. The
policy calls for a coordinated effort, an increased level of
resources, and a commitment to the problem from each
agency.
5. To prepare specific agency strategies with the under-
standing that different geographical regions have different
priority nonpoint source problems and are at different
stages in developing programs.
6. To develop and assess Best Management Practices
' (BMP's) based upon site-specific factors. The group noted
that natural background levels of pollution and the techni-
cal feasibility of the approach must be considered along
with the social, political, and economic climate of the area.
7. To ensure the recognition that nonpoint sources are
fundamentally different from point sources and should
therefore be approached differently, and that nonpoint
source programs must be based on site-specific actions
and application of preventive practices.
8. To establish a working partnership among all partici-
pants: Federal, State, local, areawide, and interstate
agencies, as well as the private sector, including nongov-
ernmental agencies.
THE POLICY OUTLINES ROLES AND
RESPONSIBILITIES
The policy asks for coordination and cooperation from all
levels of government and outlines the major responsibili-
ties for each level.
Federal agencies are asked to develop and implement
their nonpoint source strategies. They are asked to inte-
grate the provisions of the policy into their agency deci-
sionmaking process and into their delivery systems for
funding and technical assistance. EPA is asked to serve
as the lead agency in coordinating interagency and State
actions to manage nonpoint source programs.
States are assigned the lead in developing and imple-
menting nonpoint source management strategies on State
and private lands. Though several different State agen-
cies may be needed to address nonpoint source prob-
lems, the policy asks that a lead State agency be desig-
nated to develop and implement State programs. The lead
agency should have water quality as its primary concern.
Local, areawide, and interstate agencies are directed
to use the mechanisms provided in the continuing water
quality management planning process to develop their
nonpoint source strategies. The policy recognizes that lo-
cal agencies are often a first point of contact for the private
sector. This position allows local agencies—with and
through their areawide agencies—the opportunity to pro-
vide a vehicle for public participation.
The private sector is asked for its cooperation and
effort. Government agencies will assist the landowners
and help them coordinate nonpoint source management
efforts with other components of the private sector. Gov-
ernment agencies will also help the private sector develop
and apply resources to implement nonpoint source man-
agement practices.
SUCCESS RELIES HEAVILY ON THE
INDIVIDUAL AGENCY STRATEGIES
Implementation
Skillful development of the specific agency strategies is
vital to implementation. Each strategy is to include a dis-
cussion on (1) problem assessment, (2) program imple-
mentation, (3) incentives and compliance, (4) coordina-
tion, (5) resources, and (6) program evaluation and
oversight.
The policy clearly recognizes that nonpoint source man-
agement actions must be site-specific. However, it does
request a coordinated and a consistent approach across
all levels of government.
Evaluation
EPA, as part of its responsibilities mandated under the
Clean Water Act, reports to Congress on the effectiveness
of water quality programs. Because each agency will peri-
odically review its own program (the framework for over-
sight and evaluation is a part of each individual agency
strategy), EPA will be able to use these evaluations in its
overall assessment of whether national water quality goals
are being adequately addressed. The direction of the na-
tional nonpoint source effort can be controlled and refined
on the basis of well-planned evaluations.
In summary, the policy, like most policies, is a frame-
work. It was carefully and diligently developed by the task
force, and sets the direction for the next few years. It
recognizes that much work has been done, but that non-
point source problems require further attention if water
quality goals are to be achieved. Most importantly, it chal-
lenges each agency to develop and carry out specific
strategies to ensure implementation. Collectively, these
strategies embody the principles of the policy and serve
as the comprehensive plan to minimize nonpoint source
pollution problems across the Nation.
48
-------
LEGAL ASPECTS OF NONPOINT SOURCE POLLUTION
APPENDIX A
NATIONAL NONPOINT SOURCE POLICY
PREAMBLE
The Clean Water Act establishes goals for the Nation's waters.
Considerable progress has been made in achieving these goals.
However, additional progress in restoring and maintaining the
Nation's water quality and water uses will require greater imple-
mentation of nonpoint source (NPS) management programs in
addition to ongoing point source control efforts. NPS manage-
ment programs must build upon past planning and management
efforts and strive for continued progress in achieving water qual-
ity goals and designated beneficial uses.
The objective of this National Nonpoint Source Policy is to
support and accelerate the development and implementation of
NPS management programs that ensure water quality protec-
tion while recognizing the competing uses of resources. The
success of this policy is dependent on the willingness and ability
of both the private and public sectors to manage their activities
to support water quality goals wherever possible.
Meeting the objective of the Policy will require the following
actions:
1. Use of the existing knowledge and program base and sup-
port of increased research efforts to define and assess NPS
problems.
2. Identification of the appropriate roles of Federal, State, lo-
cal, areawide, and interstate agencies and the private sector in
developing and implementing NPS programs.
3. Provision of the structure, available resources, and com-
mitment by which all levels of government and the private sector
can coordinate their efforts to identify priority needs and develop
and implement cost-effective NPS management programs.
4. Support for an increased level of effort and emphasis on
NPS programs by all levels of government and the private sec-
tor, for the purpose of meeting water quality goals.
5. Preparation by each agency of a strategy for program de-
velopment and implementation that incorporates both short- and
long-term objectives; recognizes that different areas of the coun-
try are at different stages of developing their NPS management
programs; and that different geographical areas have different
priority NPS problems.
6. Development and assessment of Best Management Prac-
tices (BMP's) based upon site-specific conditions that reflect
natural background and natural variability of nonpoint sources,
and that include consideration of political, social, economic, and
technical feasibility.
7. Ensuring the recognition that nonpoint sources are differ-
ent from point sources and that NPS programs are based on
site-specific actions and application of preventive practices. Fur-
ther, recognition of the need for flexibility in water quality stand-
ards to address the impacts of time and space components of
NPS as well as naturally occurring events.
8. Development of working partnerships among all Federal,
State, local, areawide, and interstate agencies and the private
sector, including nongovernmental organizations, to best ad-
dress NPS problems. These organizations, working in partner-
ship, will be responsible for identifying needs, developing NPS
programs, gathering and assessing data, and maximizing avail-
able resources.
STATEMENT OF POLICY
Achievement of national clean water goals requires greater im-
plementation of NPS management programs. Emphasis should
be placed on implementing NPS programs in watersheds affect-
ing priority waters. Sources of nonpoint pollution should be eval-
uated to assess potential water quality impacts and needed pro-
gram actions. NPS management is required to protect high
quality surface and ground waters, and to restore and/or im-
prove water quality for designated uses. In many instances, pre-
vention of degradation has proven to be far more cost-effective
than remedial measures.
NPS management programs must be flexible to allow for site-
specific solutions to problems, to accommodate changes in
technical knowledge, to respond to changes in uses of land, and
to optimize net on- and offsite benefits. A mix of both point and
nonpoint source measures should be considered in developing
cost-effective strategies to improve and maintain water quality.
With Federal leadership and coordination, all levels of govern-
ment and the private sector need to cooperate to provide contin-
ued progress with available programs and delivery systems, to
identify unmet needs, and to develop and implement NPS man-
agement programs where needed.
ROLES AND RESPONSIBILITIES
Following is a general summary of responsibilities of the differ-
ent levels of government agencies and the private sector in
managing NPS programs:
All agencies. All agencies, where appropriate, will determine
what institutional barriers to NPS management and implementa-
tion exist and work to remove them. All agencies will work to
coordinate their NPS related data collection and research activi-
ties. In addition, inter- and intra-agency mechanisms will be de-
veloped for coordinating NPS management and implementa-
tion.
Federal agencies. Federal agencies, in preparing their NPS
strategies, will use available and future programs to provide
State and local governments with financial and technical assist-
ance and will conduct research and development. The provi-
sions of this policy will be integrated into the decision processes
of applicable Federal agencies and into their delivery systems
for funding and technical assistance. Where Federal agencies
have the responsibility for direct planning and management of
NPS programs on public lands, they must coordinate NPS man-
agement actions with all levels of government.
As directed by the Clean Water Act, EPA will serve as the lead
agency in coordinating interagency and State actions for man-
aging nonpoint source programs. EPA will promote adoption of
NPS management programs directed at achieving water quality
goals; assist with program development; promote provision of
incentives where needed; provide oversight of its water quality
programs to ensure that they adequately address NPS prob-
lems; and include other agencies' evaluations of the water qual-
ity components of their programs in assessing overall NPS im-
pacts on water quality. EPA will coordinate activities in research,
education, demonstration projects, training, information trans-
fer, technical assistance, and data collection and analysis with
other agencies.
States. States will have the lead in developing and imple-
menting NPS management strategies on State and private
lands, in cooperation with appropriate levels of government and
the private sector. Each strategy should define the State role
and, in consultation with areawide and local agencies, the roles
of areawide and local agencies in managing NPS programs, and
designate a lead agency for managing NPS programs at the
State level (several different State agencies may be needed to
address different types of nonpoint sources). The lead State
agency is responsible for developing and implementing strate-
gies for managing NPS programs and should have water quality
as its primary concern. States with effective NPS management
programs should share their experiences with other States.
Local, areawide, and Interstate agencies. Local, areawide,
and interstate agencies, through the mechanisms provided in
continuing WQM planning processes, will develop NPS strate-
gies in coordination with their respective States and will imple-
ment the programs within their jurisdictions using direct or dele-
gated authorities.
Local agencies, often the first point of contact for the private
sector, are in a unique position to solve NPS problems. The
active involvement of these local agencies, with and through
their areawide agencies in the preparation of strategies, will help
to ensure consistency among strategies and provide a vehicle
for public participation.
Private sector. For activities other than those on Federal and
State lands, successful implementation of the NPS Policy and
agencies' strategies is dependent on the cooperation and effort
of the private sector. It is the policy of the government agencies
to assist landowners and coordinate efforts with involved organi-
zations, associations, and industry. It is the further intention of
these agencies to help develop the potential for application of
managerial and other private resources in the implementation of
NPS management practices as part of each agency's strategy.
Private investment in nonpoint source research and develop-
ment of BMP's is strongly encouraged and will be supported
with agency resources where feasible and available.
49
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
IMPLEMENTATION
To implement this National Policy, Federal, State, local,
areawide and interstate agencies will develop and implement
NPS strategies. Key strategy activities for policy implementation
include: problem assessment (e.g., problem identification, moni-
toring and data maintenance, research and development); pro-
gram implementation (e.g., program planning, development,
and implementation, targeting technical assistance and educa-
tion, BMP emplacement); incentives and compliance (including
enforcement); coordination; resources; program evaluation and
oversight. The strategies will be refined as existing programs
and authorities are reviewed for consistency with existing and
future State NFS management objectives and as institutional
barriers are identified.
Responsibility for NFS implementation will depend on the na-
ture of the NFS problem, the area in question, and the statutory
framework. Implementation activities will emphasize site-spe-
cific solutions but will maintain a consistent NFS management
approach across all levels of government and the private sector.
Where appropriate, all agencies should consider and include as
part of their strategies minimum eligibility requirements to en-
hance implementation of NFS management practices. Coopera-
tive agreements will be developed, as needed, to ensure contin-
ued progress toward meeting national water quality objectives.
A schedule for strategy development and implementation
should be drafted recognizing the nationwide variability in pro-
gram implementation.
EVALUATION
The Clean Water Act requires EPA to oversee the implementa-
tion of water quality programs and to report to Congress on the
effectiveness of these programs. Given that all agencies are
individually responsible for the periodic evaluation of their pro-
grams, EPA will include these evaluations in its assessment of
NFS management efforts in determining if national water quality
goals are being adequately addressed.
GLOSSARY
Agencies: All governmental bodies and entities that—under
their mandates—have a role in addressing and/or mitigating
NPS pollution. Federal, State, local, areawide, and interstate
agencies are included.
Benefits (onsite and offsite): The whole range of direct and
indirect benefits including, but not limited to, water quality, soil
conservation, recreational and other beneficial uses, habitat and
wildlife protection, increased productivity, flood control, and eco-
nomic benefits to landowners or the public at large.
Best Management Practices (BMP's): Methods, measures, or
practices to prevent or reduce water pollution, including, but not
limited to, structural and nonstructural controls and operation
and maintenance procedures. Usually, BMP's are applied as a
system of practices rather than a single practice. BMP's are
selected on the basis of site-specific conditions that reflect natu-
ral background conditions and political, social, economic, and
technical feasibility.
Net Benefits: Since trade-offs (competitive relationships, in the
language of economics) may exist between water quality and
other social benefits, the social objective must be in terms of
optimizing net benefits.
NPS Management Programs: All programs conducted by the
public and/or private sector toward the goal of preventing or
abating nonpoint source pollution. A wide range of activities may
be pursued to this end, including BMP identification, training,
dissemination of educational materials, technical assistance,
monitoring, research and development, and oversight/evalua-
tion. Cost-sharing programs and other incentives can also play
vital roles. Programs may be regulatory or nonregulatory (volun-
tary), or combinations of both.
Nonpoint Source (NPS) Pollution: Diffuse sources of water
pollution that are not regulated as point sources and normally
include agricultural and urban runoff, runoff from construction
activities, etc. In practical terms, nonpoint sources do not dis-
charge at a specific, single location (such as a single pipe).
Nonpoint source pollutants are generally carried over or through
the soil and ground cover via stormflow processes. Unlike point
sources of pollution (such as industrial and municipal effluent
discharge pipes), nonpoint sources are diffuse and can come
from any land area. It must be kept in mind that this definition is
necessarily general; legal and regulatory decisions have some-
times resulted in certain sources being assigned to either the
point or nonpoint source categories because of considerations
other than their manner of discharge (for example, irrigation
return flows are designated as "nonpoint sources" by law, even
though the discharge is through a discrete conveyance).
Partnership: As defined in this policy statement, "partnership"
describes the arrangement between interested parties for solv-
ing the problem of NPS pollution. The key quality of this arrange-
ment is cooperation. The NPS problem inherently requires that
the private sector and all levels of government contribute to its
solution. All entities act as decisionmakers within their respec-
tive roles and areas of responsibility, and the one that can most
appropriately address a particular problem takes the lead. The
specific arrangements that implement a partnership may vary
from informal public agency/private entity cooperation in non-
regulatory programs to memoranda of understanding, contrac-
tual agreements, and cooperative agreements as defined by
OMB under Federal guidance (Federal Register, Vol. 43, No. 61,
August 18,1978).
Strategies: Written documents that specifically outline an Agen-
cy's plan of action to address nonpoint source problems that fall
within its jurisdiction or legislative mandate. Strategy activities
should be defined under six broad topics: problem assessment;
program implementation; incentives and compliance; coordina-
tion; resources; program evaluation and oversight. A consider-
ation of timeliness should be included.
50
-------
INTERGOVERNMENTAL COORDINATION: FEAST OR FAMINE?
ROBERT J. MASSARELLI
Executive Director
South Brevard Water Authority
Melbourne, Florida
ABSTRACT
Lake Washington, on Florida's east coast, is the sole wa-
ter supply for over 100,000 people in south Brevard
County. This lake is one of a chain of lakes on the St.
Johns River. Water quality problems in Lake Washington,
as well as the St. Johns River, have been described as
natural problems aggravated by man. All of the man-in-
duced water quality problems are the result of nonpoint
sources. Historically, diking and draining of the St. Johns
River marsh and continued agricultural drainage has
been the principal nonpoint source of pollution. During
the last 15 to 20 years, urban drainage has been intro-
duced into the lake. An interagency task force was
formed to protect and improve the water quality of Lake
Washington. This task force included the major regula-
tory agencies, water resource managers, and water users
of Lake Washington. The task force's effectiveness was
governed by (1) problem definition, (2) agency statutory
power, (3) agency program priority, and (4) interaction by
policy level individuals.
Intergovernmental coordination has been a popular battle
cry in recent years. Most public work projects and, in par-
ticular, water resources projects will affect more than one
unit of government. The hierarchical nature of American
government with cities, counties or parishes, sub-State,
State, and Federal levels dictates that several levels of
government will be affected. The division of labor by
agency responsibility will require that numerous agencies
become involved. In an attempt to put some order into this
Medusa, intergovernmental coordination is often pre-
sented as a solution.
Numerous examples exist. The planning advisory com-
mittees required by the 201 and 208 programs are a re-
cent past institutional requirement for coordination. In
Florida, the Local Government Comprehensive Planning
Act required an intergovernmental coordination element
in all local plans.
This paper discusses the recent experience of intergov-
ernmental coordination in a nonpoint source control pro-
gram in Brevard County on Florida's east coast. This pa-
per presents several observations on the efforts of the
Lake Washington Water Quality Improvement Task Force
and discusses some reasons for the success or failure of
that Task Force.
BACKGROUND
In May and June of 1970, the Brevard County Board of
County Commissioners constructed a canal approxi-
mately 540 m in length, known as the Sands Canal, con-
necting upland drainage with Lake Washington. This ca-
nal was constructed without the necessary State permits.
Negotiations from 1972 to 1975 between the Florida De-
partment of Environmental Regulation and Brevard
County attempted to resolve this matter.
In 1976, the Department advised the County of its intent
to deny the County an after-the-fact permit application. At
this point, the County requested a formal administrative
hearing. This hearing resulted in a recommendation of
denial from the hearing officer, and the Department issued
a final order on Oct. 15, 1977, denying the after-the-fact
permit and directing the County to submit a plan of resto-
ration for the canal.
Brevard County then appealed the permit denial to the
governor and cabinet, sitting as the Board of Trustees of
the Internal Improvement Trust Fund. In May 1978, the
Trustees entered an order remanding the case to the re-
spondent, directing the Department to formulate accept-
able modifications. On Feb. 22,1983, a consent order was
agreed upon which settled this case.
The consent order required the following major actions.
First, the County is to construct a permanent weir struc-
ture at the end of the Sands Canal. The purpose of the
weir is to ensure the separation of waters between Lake
Washington and the canal during periods when Lake
Washington is below 405 cm msl. To provide navigational
access across the weir, the weir may contain a movable
gate. The gate's lowest elevation is 315 cm msl.
The County also is required to maintain a water quality
monitoring program. Two sampling stations, one within the
Sands Canal landward of the permanent structure previ-
ously described, and the second waterward of the canal
entrance to reflect background conditions. Monitoring is
required not less than once a month.
The third condition of the consent order concerns the
operation of the adjustable weir gate. When the water
quality monitoring of the lake shows no violations of Class
1 water quality standards in the canal and the lake stage is
below 405 cm msl, the gate may be open. Any time lake
stages exceed 405 cm msl the gates may be opened
since the crest elevation of the weir is 405 cm msl.
The final requirement is the establishment of a Lake
Washington Water Quality Improvement Task Force. This
paper will discuss, in detail, the Task Force and its effec-
tiveness.
LAKE WASHINGTON
Lake Washington is located on the St. Johns River in
south Brevard County on Florida's east coast. The city of
Melbourne is located just east of the lake (Fig. 1). The
headwater of the lake begins in a large marsh in Indian
River and Okeechobee County, 32 km to the south. The
overall drainage basin is approximately 275,485 ha.
The lake is relatively shallow with a bottom elevation of
2.3 meters msl. The stage duration curves developed for
the lake stages are less than 4.1 m msl 50 percent of the
time. The bottom is typically of unconsolidated organic
matter. At this time, few submergent species of vegetation
exist in the lake. The shoreline is dominated by wetlands
composed of saw grass (Cladium jamaicense), maiden-
cane (Panicum hemitomon), spikerush (Eleocharis sp.),
nutgrass (Cyperus sp.) and swamp willow (Salix carolinia).
The eastern shore has one small area of urban land use,
including a public boat ramp, marina, a home, and two
water treatment plants.
The water quality of the lake is highly variable, depend-
ing on the time of year, lake stages, and local climatic
conditions. In general, water quality and water quantity
appear to be closely related, as water quality deteriorates
with a decrease in flew. However, lew dissolved oxygen
and high color are associated with high flow conditions.
The water quality of Lake Washington is affected by
51
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
several factors. The drainage basin of Lake Washington is
dominated by large wetland systems, including marshes
and hardwood swamps. These wetlands contribute to the
color and organic loadings to the lake. In addition, during
the last 50 to 60 years, much of the basin has been con-
verted to agricultural lands. This has resulted in the loss of
floodplain wetlands, the channelization of the marsh, agri-
cultural runoff, and finally, significant alteration of the natu-
ral hydrograph. Other factors affecting the water quality of
the lake, particularly during periods of low flow are: (1)
inflow of ground water having higher chlorides and total
dissolved solids concentrations, (2) canals that drain the
uplands east of the lake, and (3) evaporation and evapo-
transpiration during the dry season.
Cultural pollution is from agricultural runoff and urban
drainage. No sewage treatment plants discharge into the
lake or its drainage basin.
Lake Washington is used mostly as a potable water
supply Since 1959, the city of Melbourne, which serves
approximately 109,000 people, has used Lake Washing-
ton as its source of drinking water. The water quality has
been described by the city as the most difficult to treat in
the nation, primarily because of rapid changes in color
and TOC. An algae bloom in the lake during 1984 resulted
in taste and odor complaints for several days. Another
concern is chlorides and total dissolved solids. As lake
levels drop, these parameters increase in concentration.
During a drought in 1980-81, the drinking water standard
for chlorides was exceeded for 90 consecutive days and
the TDS standard was exceeded twice for 140 and 141
consecutive days.
The lake is classified by the Florida Department of Envi-
ronmental Regulation as a Class 1 surface potable water
supply
Flgur* 1.—Location map.
TASK FORCE
One of the requirements of the consent order was the
establishment of a Lake Washington Water Quality Im-
provement Task Force. This Task Force, led by Brevard
County, was to be composed of governmental agencies
and interested parties who are involved in the preserva-
tion and protection of Lake Washington.
The Task Force had five specific purposes. The first was
to evaluate existing and potential sources of pollution in
Lake Washington. This included any canals or ditches
leading into the lake. The second task was to identify ex-
isting and prospective uses of the Lake Washington water
resources.
The Task Force was also to review land use planning
and implementation. In reviewing policies and ordinances,
potential sources of domestic or industrial wastes,
stormwater sources, and the loss of filtrative vegetation
were to be considered.
The fourth item was to develop educational materials on
pollution abatement, stormwater management, and strat-
egies to preserve and restore Lake Washington water
quality. These are to be provided to existing and future
property owners in the Lake Washington watershed. Fi-
nally, the Task Force was to identify sources of funding that
may be used to mitigate water pollution in the lake. The
work of this Task Force was to be completed in 24 months
from the entry of the consent order.
In May 1983, the Brevard County Board of County Com-
missioners organized the Lake Washington Water Quality
Improvement Task Force. It was decided that the Task
Force would be a policymaking board. The membership
consisted of:
1. The District 3 Brevard County commissioner.
2. The District 5 Brevard County commissioner.
3. The district manager of the St. Johns River District
Office of the Florida Department of Environmental Regula-
tion (FDER).
4. The Executive Director of the St. Johns River Water
Management District (SJRWMD).
5. The representative of the Florida Game and Fresh-
water Fish Commission (GFWFC)
6. The city manager of the city of Melbourne.
7. A representative from the Lake Washington Home-
owners' Association.
In the summer of 1983 the South Brevard Water Author-
ity, the agency responsible for public water supply in the
South Brevard area, was created and the Authority's exec-
utive director appointed to the Task Force.
In addition, a Technical Advisory Committee was estab-
lished to perform the technical aspects of the study and to
advise the Task Force. Each of the following agencies had
one technical representative:
• Florida Department of Environmental Regulation
• St. Johns River Water Management District
• Florida Game and Freshwater Fish Commission
• Florida Institute of Technology Staff
The Brevard County Water Resources Department Staff
acted as staff to the Task Force and the Technical Advisory
Committee.
A plan of study was developed to facilitate the work of
the Task Force. The overall program was divided into four
subprograms: historical data, resource management, im-
plementation, and post-implementation. Table 1 outlines
this plan of study. A schedule provided that the work could
be completed in 18 months, allowing for flexibility in meet-
ing the 24-month deadline requirements of the consent
order.
The Task Force met six times. A final report was
adopted at the Task Force's last meeting, February 1985.
52
-------
LEGAL ASPECTS OF NONPOINT SOURCE POLLUTION
Table 1 .—Plan of study.
*Task Force Meeting—Organizational.
History
1. Lake Washington boat tour.
2. Historical and existing information (water quality, water
quantity, land and water uses).
3. Plans (Upper Basin, Brevard Co. Comprehensive Plan,
Melbourne Comprehensive Plan).
•Task Force Meeting—adoption of a comprehensive report on
the background of Lake Washington.
Resource Management
4. Resource evaluation: surface and ground water
hydrology, water chemistry.
5. Resource evaluation: ecology, sociology (meeting with
agricultural committee).
6. Resource management alternatives and funding.
*Task Force Meeting—prioritize management alternatives.
Implementation
7. Land acquisition program.
8. Water and land management regulations.
9. Property rights/compensation.
'Task Force Meeting—selection of implementation program.
Post-Implementation
10. Educational material.
11. Establish continuing planning/coordination program.
12. Final report.
Task Force Meeting—review final report.
EVALUATION
The effectiveness of an intergovernmental activity is hard
to measure. If one agency provides strong leadership, a
program may be implemented even without intergovern-
mental activity. Intergovernmental coordination may slow
down or enhance implementation.
Accordingly, the success or failure of the Lake Washing-
ton Water Quality Improvement Task Force is also very
difficult to evaluate. First, the recommendations of the
Task Force have only recently been completed, in Febru-
ary 1985. Sufficient time has not passed to evaluate the
success or failure of implementation. Secondly, many of
the Task Force's recommendations resemble existing pro-
grams of the participant agencies. What effect the Task
Force will have on them is not yet known.
However, various aspects of the Task Force can be dis-
cussed at this time. The four major factors affecting the
effectiveness of the Task Force were (1) problem defini-
tion, (2) agency statutory power, (3) agency program prior-
ity, and (4) interaction by policy level individuals.
The first step in solving any problem is defining the
problem. The consent order makes some vague reference
to water quality standards. Studies have shown that the
water quality of the St. Johns River and Lake Washington
may be affected by numerous factors. However, direct
cause and effect relationships have not been accurately
defined. To minimize the area studied by the Task Force,
the study area's boundaries were limited to the immediate
area of Lake Washington (Fig. 2). This is just a small por-
tion of the lake's total watershed.
As a result, the Task Force had a poorly defined prob-
lem. The cause and effect relationships needed to pro-
duce defensible solutions were not available. The study
area boundaries needed to keep the study manageable
also eliminated the most significant flow contribution to
the lake—the watershed of the St. Johns River.
Each of the participants in the Task Force was iimited by
statutory authority In a gross generalization, the FDER is
responsible for water quality, the SJRWMD is responsible
I \
1 r—A J
: <* j»«*
Figure 2.—Lake Washington study area.
for water quantity and the GFWFC is responsible for wild-
life. Each agency's responsibilities directly affect the oth-
ers. However, because of their statutory limitations, it is
difficult to estimate official agency interest outside of their
jurisdictional boundaries.
Often, statutory authority is not considered. The con-
sent order was developed between Brevard County and
FDER. The SJRWMD was not a party to it even though
they had to issue the permit for the required weir.
Because of statutory constraints as well as budget con-
siderations, policies, and competing issues, each agency
has its own program priority Brevard County had a high
priority in seeing the completion of the Task Force's work
because of the consent order. The other participants did
not have the same priority. The Task Force's work could be
considered someone else's responsibility In addition, be-
cause of incomplete programs which directly affect Lake
Washington, agencies were reluctant to make specific rec-
ommendations or commitments. At that time, the
SJRWMD was completing the Upper Basin Management
Plan for the St. Johns River, and the Task Force final re-
port was completed before SJRWMD publicly released
their plan.
Finally, the individual level of participation affected the
effectiveness of the Task Force. The GFWFC never ap-
pointed a representative. The FDER Task Force member
never attended; however, his alternate did, occasionally.
The SJRWMD member attended only the initial meeting.
The Task Force was established as a policy level group.
Without the attendance of these individuals, resolution of
policy conflict or commitment of resources could not be
made.
53
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
CONCLUSIONS
The Lake Washington Water Quality Improvement Task
Force was established to provide intergovernmental coor-
dination in developing management policies for Lake
Washington. The principal agencies concerned with the
resources of the lake were involved; however, their effec-
tiveness was limited by problem definition, statute, pro-
gram priority and participation.
To enhance the effectiveness of intergovernmental co-
ordination, the following is recommended:
1. Problem definition. It is important to keep the prob-
lem small enough to minimize necessary variables. How-
ever, in establishing the boundaries of the study do not
exclude variables that may have a significant impact.
If there is not enough information to define the problem,
then wait. Do not attempt to develop a solution with a
poorly defined problem. This will only result in an inade-
quate solution.
2. Statutory power. The powers given to an agency by
the legislature must not limit or hinder intergovernmental
coordination. One alternative is to divide agency responsi-
bility by broad subject areas such as transportation, edu-
cation, and natural resources, rather than by specific re-
sponsibilities such as water quantity, water quality, and
wildlife.
A second alternative is to give each agency the specific
authority to consider cumulative effects or multidiscipli-
nary effects. While the agency's responsibility may remain
specific, the ability to consider other effects will force inter-
governmental coordination.
3. Program priority. Agency-established priorities will
not enhance coordination. Program priority must come
from one centralized authority This can be done by an
office of planning and budgeting within the governor's of-
fice, or a legislatively established budget review process.
A statewide, regional specific plan of State priorities must
be developed and periodically updated.
4. Participation. In general, intergovernmental coordi-
nation at the technical staff level appears to exist and
often works well. It is at the policy level where coordination
is often missing. One way to improve this is to hold regular
symposia to discuss issues of mutual concern.
Intergovernmental coordination is not a cure-all or a
curse. Because of the nature of American government,
multiple agency involvement will occur. Coordination of
these agencies' activities is essential to minimize wasteful
duplication and unnecessary delays. However, coordina-
tion must be carefully managed to avoid prolonged discus-
sion of the problem. Don't assume that since several
agencies are meeting and discussing a problem they are
coordinating their efforts. Strong leadership and individual
commitment will help ensure intergovernmental coordina-
tion.
54
-------
THE BASIC LEGAL ISSUES
JAMES T. B. TRIPP
Counsel
Environmental Defense Fund
New York, New York
ABSTRACT
Nonpoint source water pollution is thought difficult to con-
trol because it derives from diffuse sources over a broad
area. Many forms of nonpoint source pollution derive,
however, from inappropriate use of land or water re-
sources which are subject to regulation through Federal
and State wetland protection laws, such as the Clean
Water Act section 404 Program, local government zoning
powers, and other local, State or Federal laws which con-
trol the siting of major infrastructure facilities which con-
tribute directly or indirectly to nonpoint source pollution.
With respect to urban or highway runoff, vigorous en-
forcement of Clean Air Act programs relating to mobile
and stationary sources may result in significant reduction
of impacts of metals, organics, and acid rain. Non-en-
forceable BMP's are virtually useless at controlling non-
point source pollution. For some existing nonpoint source
pollution, such as agricultural soil, nutrient and pesticide
runoff, we must identify cost-effective economic invest-
ments which control such pollution through alternative
use of waste resources and find the institutional mecha-
nisms to facilitate those investments.
INTRODUCTION
The Federal Clean Water Act, 33 U.S.C. Section 1251 et
seq, divides all causes of surface water degradation into
two parts: point and nonpoint source pollution. In general,
that Act prescribes regulatory programs to control dis-
charges of pollutants and establishes planning programs
to promote control of all other pollution, defined in Section
502(19) to be the manmade or man-induced alteration of
the chemical, physical, biological or radiological integrity
of water. Some courts have held and the U.S. Environ-
mental Protection Agency has taken the position that the
Clean Water Act's permit programs do not regulate dis-
charges into ground water; therefore, under the Clean Wa-
ter Act, pollution of ground water and in turn degradation
of surface waters by contaminated ground water, are
viewed as nonpoint sources of pollution. Subsequent Fed-
eral and related laws, however, including the Resource
Conservation and Recovery Act (RCRA), 42 U.S.C. Sec-
tion 6901, and the Safe Drinking Water Act (SDWA). 42
U.S.C. Section 300f, provide for regulation of major dis-
crete sources of ground water pollution, such as industrial
impoundments, storage tanks, landfills, and underground
injection wells. We can therefore characterize them as
point sources of water pollution as well.
In general, we can view nonpoint source pollution as
any pollution of ground or surface waters associated with
diffuse land use activities that cumulatively result in water
quality degradation. Agricultural, mining, and construc-
tion-related activities, urban or highway runoff, and resi-
dential cr corr.rnercia! septic system and lawn discharges
are typical nonpoint pollution sources. Such pollution
sources are recognized as major causes of degradation of
many surface and ground water systems. However, since
they are viewed as land use problems, the Congress and
most States have not adopted regulatory programs to con-
trol them.
The theme of this paper is that Congressional hesitation
notwithstanding, much nonpoint source pollution can in
fact be controlled or prevented by existing State and Fed-
eral programs, coupled with local government land use
powers; that local land use authority is not, by itself, typi-
cally effective at controlling nonpoint source pollution and
that we need a Federal legal program that gives State and
local governments and Federal agencies a compelling
framework for using existing authorities to control non-
point source pollution.
BASIC CONCEPTS
In designing a meaningful local, State or Federal nonpoint
source pollution control strategy, we should keep in mind
three basic principles.
First, nonpoint source pollution results from and is asso-
ciated with a loss of natural vegetative cover. Natural for-
est or other plant communities do not generate pollution
as defined in the Clean Water Act. Differently stated, re-
tention or re-establishment of natural plant cover prevents
or abates such pollution. The objective of a nonpoint
source pollution control program must therefore be to re-
tain or re-establish natural plant communities as much as
possible.
Second, while most nonpoint source pollution is not di-
rectly controlled or regulated under State and Federal en-
vironmental laws, a good portion of it arises from preced-
ing individual acts which are in fact point sources of
pollution and regulatable as such through existing permit
programs. Increasingly therefore, nonpoint source pollu-
tion is a reflection of ineffective or inadequate implementa-
tion of point source pollution permitting authorities. If we
recognize this fact and intend to do better in the future, it
makes sense to distinguish future from existing nonpoint
source pollution.
In many cases, effective use of point source pollution
control programs to prevent future nonpoint source pollu-
tion may make economic and social sense. Emphasizing
remedial action is always more problematic. In a world of
limited resources, the pros and cons of preventive and
remedial actions must always be assessed. We should
also recognize that much nonpoint source pollution is a
result of Federal and State-funded activities and may
therefore be controlled through budgetary and planning
processes.
Third, the siting of an agricultural, forestry, mining,
transportation, commercial or residential activity within a
surface or ground water watershed is central to its poten-
tial nonpoint source pollution impact on receiving surface
or ground water quality In terms of effect, the siting of
such activities is as important as and often more important
than the operational design.
55
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
It is appropriate to consider the critical portion of a sur-
face or ground water recharge watershed as a basis for
developing siting criteria for activities which may cause
nonpoint source pollution. For a surface water basin, the
most critical portion of the watershed may be its wetlands
within the floodplain of the receiving river, lake, or estuary.
For a ground water system, it is likewise possible to deline-
ate a critical recharge zone in terms of soil conditions,
recharge areas, and ground water residence times.
Imposing controls based on the location of an activity
within a watershed and its nonpoint source pollution po-
tential has a legal basis in many Federal and State envi-
ronmental laws, including the Clean Water Act, the Clean
Air Act, 42 U.S.C. Section 7601, RCRA, the Federal Insec-
ticide, Fungicide, and Rodenticide Act (FIFRA), 7 U.S.C.
Section 136, the Toxic Substance Control Act (TOSCA), 15
U.S.C. Section 2601, the Safe Drinking Water Act (SDWA)
and the Surface Mining Control and Reclamation Act, 30
U.S.C. Section 1201. These Federal laws and their State
counterparts in some measure authorize regulatory pro-
grams that incorporate geographic or siting criteria based
on water quality, among other values, and can be used to
prohibit siting of nonpoint source pollution-generating land
uses in critical watersheds. Federal and State agencies,
however, typically underutilize these legal authorities.
We can see how these general principles operate in a
variety of contexts involving nonpoint source pollution ac-
tivities.
Agricultural runoff with sediments, nutrients, or toxic
chemicals is a prime example of nonpoint source pollu-
tion. The siting of agricultural activity can have a major
impact on receiving water quality. Row crop operations in
low lying floodplain areas, including freshwater wetlands,
can have enormous impacts on receiving water quality
because of their proximity to receiving waters and the loss
of the natural wetland communities that could buffer up-
land runoff. Similarly, agricultural operations in critical
ground water watershed areas can often result in high and
damaging levels of nutrients and pesticides in recharged
ground water.
With some 98 million acres of wetlands remaining
(some 95 percent freshwater wetlands), the 48 cotermi-
nous States have lost more than 50 percent of their wet-
lands, more than 80 percent because of agricultural con-
version, clearing, and drainage. In many riverine
floodplains, including those in the Lower Mississippi Allu-
vial Valley, floodplain vegetation has been cleared to
stream banks, with massive water quality degradation re-
sulting. Agricultural conversion of floodplain wetland com-
munities must therefore be viewed as a major nonpoint
pollution source. Once the conversion has occurred, non-
point source pollution inevitably increases dramatically.
While agricultural conversions of wetland systems tradi-
tionally have not been regulated at all (indeed, Federal
water resource development programs have subsidized
and promoted them), they are now regulatable under Sec-
tion 404 of the Clean Water Act, 33 U.S.C. Section 1344,
and some State wetland programs as well. In Avoyelles
Sportsmen's League v. Marsh, 715 F. 2d 897 (5th Cir.
1983), upholding a district court opinion, 473 F. Supp. 525
(W.D. La. 1979), the Fifth Circuit held that agricultural con-
versions of Section 404 wetlands are not exempted as
normal agricultural activities under Section 404(f)(1)(A);
instead, mechanized clearing operations are "point
sources" of pollution which "redeposit" cleared material.
In addition, they are clearly regulated under Section
404(f)(2) which provides for any discharge of dredged or
fill material into waters of the United States "incidental" to
a change in use where the reach of those waters is im-
paired. The Seventh Circuit has recently rendered a simi-
lar opinion, United States v. Huebner, 752 F. 2d 1235 (7th
Cir. 1985). Thus, future nonpoint source pollution resulting
from agricultural conversion of wetlands can be regulated
and avoided.
In terms of water quality and aquatic ecosystem protec-
tion, the fact is that row crop agriculture should not be
sited in wetland areas. The Clean Water Act Section 404
program provides a legal basis for preventing such non-
point source pollution. Some State wetland laws, for ex-
ample, Florida's Warren S. Henderson Wetlands Protec-
tion Act, Fla. Stat. Section 403.901 et seq.\ provide some
limited protection from water quality degradation although
other State wetland laws generally exempt agriculture. We
need effective enforcement of the Section 404 to limit agri-
cultural conversion of wetlands.2
The role of wetlands in maintaining water quality and
filtering water laden with sediments and nutrients would
suggest that reforestation of wetland riparian areas, as
well as converting high erodible lands to pasture or forest
cover, should be prime remedial action strategies where
existing agricultural activities are a major nonpoint pollu-
tion source. However, the Clean Water Act provides no
legal basis to compel such a result.
The recent experiences with contamination of ground
water by aldicarb in Suffolk County, Long Island, and by
aldicarb and EDB in Florida show that pesticide-related
degradation of ground water quality can be severe when
the agricultural operations that use such pesticides are
located in central recharge areas with soils not effective at
retarding movement of such toxic pollutants. While EPA
may have properly registered these pesticides, their use in
such sensitive recharge areas was clearly inappropriate.
FIFRA provides a legal basis for restricting the use of
registered pesticides geographically, although EPA has
used this authority sparingly at best. Thus, FIFRA could
be used to restrict use of specific pesticides in specific
ground water recharge areas. Preparation of EPA's
Ground-Water Protection Strategy (Aug. 1984) by EPA's
Office of Ground-Water Protection may stimuate such use
of FIFRA. Thus, although pesticide contamination of
ground water is perceived as a nonpoint source of pollu-
tion, such contamination, whenever it occurs, is a conse-
quence of a clearly regulatable act—the use of that pesti-
cide in that area.
Pollution of ground and surface waters associated with
surface mining and other forms of mineral extraction has
traditionally been viewed as nonpoint source. Certainly,
Section 208(b)(2)(G) of the Clean Water Act, 33 U.S.C.
Section 1288(b)(2)(G), viewed mining-related wastewater
runoff in this light. Yet, many aspects of mineral extraction
processes that can result in nonpoint source pollution are
in fact subject to regulation.
Most aspects of surface mining, including development
of mining and reclamation plans, are subject to Federal or,
through delegation, State review and approval under the
Surface Mining Control and Reclamation Act, 33 U.S.C.
Section 1201. Section 101(c) of that Act, 30 U.S.C. Sec-
tion 1201(c), recognizes that surface mining operations
disturb surface waters, cause erosion, and pollute waters.
Section 102(c) of this Act, 30 U.S.C. Section 1201(c), pro-
vides that mining is not to be conducted where reclama-
tion is not feasible. If properly enforced, this Act should
dramatically reduce contamination of surface and ground
water arising from new mining operations that the Act reg-
ulates.
56
-------
LEGAL ASPECTS OF NONPOINT SOURCE POLLUTION
Further, to some degree, the Act provides for reclama-
tion of abandoned stripmined sites which are a major
cause of ground and surface water pollution in many parts
of the country. The Act also provides legal authority to
restrict surface mining operations that could cause irre-
versible pollution so that reclamation would be infeasible.
In other words, the Act could be used to restrict or prohibit
the siting of surface mining in critical surface or ground
water watershed areas highly sensitive to contaminants.
As another example, saltwater intrusion in coastal Loui-
siana, which contains 40 percent of the country's coastal
wetlands, can be viewed as a nonpoint source pollution
problem. Such saltwater intrusion, which contributes to
the accelerating erosion of the Louisiana coastal zone
(now eroding at a rate of some 40 square miles or 32,000
acres of wetland annually—an inexcusable, manmade bi-
ological travesty), results from the construction of canals
in this richly convoluted wetland maze of subtle salinity
and vegegation gradients. These canals have been built
for navigation, drainage, and water supply. Thousands of
miles of canals have also been built to transport oil and
gas exploration and development equipment and to pro-
vide for pipelines to transport extracted oil and gas. Con-
sequences of the construction of this maze of canals in-
clude massive saltwater intrusion, interference with
natural hydrological flows, extensive bank erosion, and
accelerating rates of land loss. Yet, the construction of
these canals has been subject to permit regulation for
more than 10 years under the Corps of Engineers dredge
and fill discharge permitting authority under Section 404
of the Clean Water Act, 33 U.S.C. Section 1344, and in the
last 5 years under the Louisiana Coastal Zone Manage-
ment Program.
Despite this State and Federal authority, permits con-
tinue to be routinely issued for construction of such canals
with some conditions imposed to alter the alignment of
canals and the design of dredged spoils. Neither the
Coastal Zone Section of the State Department of Natural
Resources nor the Corps of Engineers has used its legal
authority to promote or force use of alternatives that in fact
exist. Thus, while the Louisiana coastal zone suffers from
increasing nonpoint source pollution in the form of saltwa-
ter intrusion and loss of sediments through erosion, con-
struction of canals, the primary courses of such pollution,
has been a regulated act for more than a decade.
The Suwannee River, an outstanding Florida water with
its headwaters in the Okefenokee Swamp, a National
Wildlife Refuge, and its mouth north of Cedar Key, Florida,
is one of the few relatively pristine river systems remaining
in Florida, indeed, in the entire Southeast. The Upper Su-
wannee River is characterized by unusual water quality—
low both in nutrients and pH and high in color, a reflection
of the swampy origins of its waters. The major cause of
degradation of the Suwannee River is phosphate stripmin-
ing, mostly in Hamilton County, Florida. This pollution
results from point source discharges into tributaries of the
Upper Suwannee River, subject to NPDES permit require-
ments of Section 402 of the Clean Water Act, 33 U.S.C.
Section 1342, and State water quality permit restrictions.
It also results from loss of critical watershed wetlands that
are stripmined or used as waste disposal sites. The mine-
land wastewater discharges, while regulated, furthermore
cause degradation because they directly destroy tributary
wetlands.
The fact is that most aspects of this phosphate stripmin-
ing, including chemical plant and mineland wastewater
discharges, design and siting of waste disposal sites, and
the siting of mining operations in wetlands, are subject to
State and Federal regulation. The Corps of Engineers has
asserted jurisdiction over the wetlands in Hamilton County
under Section 404 of the Clean Water Act, 33 U.S.C. Sec-
tion 1344, and has released a draft environmental impact
statement intended to assess the impacts of proposed
and alternative mining and waste disposal operations on
the aquatic environment. Thus, existing Federal and State
law together provide express regulatory control over the
siting of mining operations in wetland systems within the
Suwannee River's tributary watersheds. Vigorous restric-
tions on the siting of such operations as well as appropri-
ate controls on point source wastewater discharges could
protect the Suwannee River. Thus, continued degradation
of the Suwannee River resulting from phosphate stripmin-
ing is a consequence of discrete regulatable acts.
Example III: Publicly Funded Infrastructure
Another source of what is typically considered nonpoint
source pollution is the construction and siting of public
infrastructure facilities, such as highways or dams, funded
by Federal, State or county agencies.
Highways generate runoff with organic chemical con-
taminants and nutrients. In turn, with other public infra-
structure investments such as sewers, they typically spur
residential or commercial development that causes more
nonpoint source pollution. The siting of highways in critical
surface watershed areas, including wetlands, and in sen-
sitive ground water recharge zones can greatly increase
the magnitude of their impact on receiving waters.
Aside from direct funding controls, the siting of high-
ways in wetland areas is regulatable under Section 404 of
the Clean Water Act and under many State wetland laws.
In addition, the siting of highways in a recharge zone of
ground water designated as a sole source aquifer under
Section 1424(e) of the Safe Drinking Water Act, 42 U.S.C.
Section 300h-3(e), may be prohibited as a potential cause
of ground water contamination by EPA. Needless to say,
EPA has not used this veto authority aggressively. Further,
because sole source aquifer designation now provides so
little regulatory authority to control the siting of pollution
sources in sensitive recharge areas, Section 1424(e)
should be amended and strengthened. Senate Bill S.124
and House Bill H.R. 1650, the Safe Drinking Water
Amendments of 1985 represent a step in this direction,
although H.R. 1038 and S.24 would be a preferable
amendment.
Much of the pollution associated with highway runoff
stems from the exhaust and tire wear of automobiles,
trucks, and buses. Although the organic chemicals and
toxic metals in such runoff are deemed to be nonpoint
sources of pollution, in fact these motor vehicle pollution
sources are regulated under Subchapter II of the Clean Air
Act, 42 U.S.C. Section 7521. Unfortunately, air pollution
emission standards for trucks and buses are very lax, and
emission standards for automobiles, as well as trucks and
buses, are not stringent enough to prevent significant mo-
tor vehicle-related pollution runoff. Strengthened motor
vehicle source emission standards would, of course, have
innumerable benefits in terms of reducing concentrations
of air pollutants, such as hydrocarbons, nitrogen oxides,
carbon monoxide, and toxic air pollutants, as well as re-
ducing concentrations of toxic contaminants in urban run-
off. Highway runoff, which is presented as an example of
nonpoint source pollution and which cannot be adequately
treated by secondary treatment plants, arises in large part
from a great number of regulated air pollution emission
sources.
We have already mentioned one major cause of saltwa-
ter intrusion in coastal Louisiana—the construction of ca-
nals. Another cause of coastal riverine saltwater intrusion,
such as is the case in the Gulf Coast of Texas, is construct-
ing riverine dams for water supply and other purposes.
57
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Aside from the issue of funding control, construction and,
to a limited degree, the operation of such dams is subject
to some regulatory control under the National Environ-
mental Policy Act, 42 U.S.C. Section 4321 ef seq., the
1958 Fish and Wildlife Coordination Act, 16 U.S.C. Sec-
tion 661, and, in some cases, Section 404 of the Clean
Water Act. The construction of massive regional sewer
systems in Long Island that collect and treat wastewaters
in a ground water-dependent system and discharge them
into ocean water has also resulted in saltwater intrusion
into Great South Bay. Construction of such sewer systems
is subject to the same statutory requirements as dams in
addition to other legal requirements specified in Section
201 of the Clean Water Act, 33 U.S.C. Section 1281.
Example IV: Residential and Commercial
Development
Residential and commercial development typically brings
with it nonpoint sources of pollution—septic system dis-
charges into ground water and lawn-related fertilizers and
particles. The siting of such development is a major factor
in determining the magnitude of its associated nonpoint
source pollution on surface or ground water. As with all
pollution sources, the siting of such development in wet-
lands, low lying floodplains, and other portions of critical
surface or ground water watersheds can cause high levels
of nitrate or organic chemical pollution. Its siting in less
sensitive areas results in a far lesser impact.
Local governments in many parts of the country have
recently been using their zoning authority to limit residen-
tial densities and prevent undue clearing of natural vege-
tation to protect ground and surface waters in critical wa-
tersheds. Long Island townships have used 2- and 5-acre
zoning in part to limit residential pollution of ground water.
Such zoning, when challenged, has been sustained. The
New Jersey Pinelands Commission in its Pinelands Com-
prehensive Management Plan has severely restricted resi-
dential development in some 80 percent of the 1 million-
acre Pine Barrens of southeastern New Jersey, in part to
maintain the remarkably pristine quality of its surface and
ground water, characterized by exceedingly low (below 0.2
parts per million) levels of nitrates. Indeed, that Commis-
sion has adopted the country's most ambitious multi-
county transfer of development rights program in pursuing
its objective to severely restrict development in the most
sensitive areas of the Pinelands. A Virginia Court, Aldre
Properties v. Board of Supervisors, Chancery Nos. 7846.3-
A 19th Judic. Cir. V. Jan. 7,1985, has recently upheld the
rezoning of some 40,000 acres of a critical part of the
watershed of the Occoquan reservoir in Fairfax County
Dade County, Florida, is prohibiting the siting of industrial
facilities with any potential for discharge of broadly-de-
fined hazardous wastes within the zones of influence of its
new water supply well fields west of the most urbanized
portions of the county
Restrictions on residential and commercial potential
nonpoint source pollution in sensitive watersheds is not,
however, limited to exercise of the zoning power. Local or
State governments can and do ban the use of certain
septic tank solvents and other toxic organic compounds in
such watershed areas. In particularly sensitive areas, lo-
cal or State governments could intensively apply such
bans. In addition, under TOSCA, 15 U.S.C. Section 2601,
EPA has the legal authority to prohibit use or disposal of
specific chemicals. EPA could use this authority to limit or
prohibit such use in sensitive ground or surface water
watersheds.
Example V: Atmospheric Pollution—Aeid
Rain
Atmospheric pollution in the form of dry and wet deposi-
tion of oxides of sulfur and nitrogen—acid rain—is gradu-
ally being recognized as a major cause of acidification and
resulting contamination by acid, sulfates, and mobilized
toxic metals of surface waters in geologically sensitive
areas. These sensitive areas are widespread—northern
New England, the Adirondacks, portions of the Hudson
Valley, the Catskills, the Appalachian Region, portions of
Florida, the Upper Midwest, high elevation lakes and
streams in the Rocky Mountain Region, and parts of the
Northwest. Atmospheric deposition is therefore a major
nonpoint source of water pollution.
While emissions of oxides of sulfur and nitrogen, the
precursors of acid deposition from utilities, smelters, and
other industrial sources, are not regulatable under the
Clean Water Act, they are clearly regulatable under the
Clean Air Act. The principal sources of sulfur oxides in the
East and West are all "stationary sources" of air pollution.
Further, acid deposition causes a range of adverse im-
pacts on water quality, forests, crops, manmade materials,
and visibility—all recognized as "welfare effects" as de-
fined in Section 302(h) of the Clean Air Act, 42 U.S.C.
Section 7602(h).
Unfortunately the Administrator has not exercised his
authority or performed his duty to establish a secondary
annual national ambient air quality standard for sulfur dep-
osition in the form of a sulfur deposition rate at a level
designed to avoid sulfur's adverse welfare effects either
based on an existing criteria pollutant, sulfur dioxide, un-
der Section 109(b)(2), 42 U.S.C. Section 7409(b)(2), or by
listing atmospheric sulfur in any chemical form as a new
air pollutant under Section 108(a)(1) of the Clean Air Act
42, U.S.C. Section 7408(3X1), and subsequently estab-
lishing air quality criteria and a secondary standard. De-
spite this failure to act, however, this nonpoint source of
pollution of water quality is controllable under the Clean
Air Act. Needless to say, in the face of continued EPA
disregard of its legal duties, we can hope that the Con-
gress will establish a program to rapidly reduce sulfur ox-
ide emissions from major stationary sources.
CONCLUSION
To a large extent, nonpoint sources of water pollution
result from preceding acts that constitute point sources of
pollution subject to regulation under existing Federal envi-
ronmental laws and their State counterparts. Existing law
therefore can control and limit future nonpoint source pol-
lution and, to a more limited degree, be used to remedy
historic nonpoint source pollution. Vigorous enforcement
of the Clean Water Act Section 404 wetland protection
program would, by way of example, enhance control of
nonpoint source pollution.
Siting of nonpoint source pollution-generating activities
within a ground or surface watershed is also a major factor
in determining the magnitude of impact of that pollution of
receiving waters. To control nonpoint source pollution,
EPA and State environmental protection agencies must
take advantage of those provisions in the Safe Drinking
Water Act, RCRA, TOSCA, FIFRA, the Clean Water Act,
and other laws that authorize use of siting criteria. Since
local governments play a major role in making land use
decisions, they too should take advantage of these provi-
sions.
It is certainly the case that State agencies and EPA
have not taken maximum advantage of these statutory
authorities. We need a general legal framework that facili-
58
-------
LEGAL ASPECTS OF NONPOINT SOURCE POLLUTION
tates using these authorities at all levels of government. that State's first permit program expressly designed to regulate
Strengthening the sole source aquifer programs of the activities in wetlands. It takes away from the State's Depart-
Safe Drinking Water Act would be a step in this, direction. rnent of Environmental Regulation authority to regulate agricul-
Adopting a comparable program for critical surface water tural activities; insofar as they are regulated in connection with
signed to stimulate design and implementation of regional zAdministrative implementation of the Avoyelles decision has
nonpoint source pollution control programs would also been slow. In October 1984, the Assistant Administrator of EPA
provide a broad legal framework for taking advantage of for External Affairs issued interim guidance to all EPA regions
existing regulatory authorities. instructing them that agricultural conversion operations in bot-
••r tomland hardwood wetlands in general are subject to Section
404 regulation. The U.S. Army Corps of Engineers which di-
END NOTES rectlv administers the Section 404 program has issued a Regu-
fc latory Guidance Letter No. 85-4 dated March 29, 1985, which
'The Florida Wetlands Protection Act expands State jurisdiction reflects that agency's begrudging accommodation to the man-
over the State's waters, including wetlands, and establishes dates of the Fifth Circuit.
59
-------
COMPELLING ON-THE-GROUND IMPLEMENTATION OF MEASURES
TO CONTROL NONPOINT SOURCE POLLUTION
HOPE M. BABCOCK
Deputy Counsel
Director, Public Land and Water Program
National Audubon Society
Washington, D.C.
ABSTRACT
Experience to date with implementation of measures to
control nonpoint source pollution has been largely volun-
tary, dependent for success on education and subsidiza-
tion of the costs of erosion control. The record is clear
that these measures have not worked—nonpoint pollu-
tion is getting worse, not better. A dramatic change in
attitudes abut how to control nonpoint pollution must be
made if this serious source of water quality degradation is
to be brought under control. Any program to compel im-
plementation of best management practices must ac-
knowledge the differences between types of sources of
nonpoint pollution and the features that distinguish non-
point pollution from point source pollution (e.g., the'inher-
ent problems associated with measuring the amount of
pollution caused by that source). What is an appropriate
incentive to achieve one source's compliance may not be
appropriate for another. A mix of measures, ranging from
traditional enforcement tools like citizen suits, cross com-
pliance penalties, permits, and fines to financial incen-
tives like taxes, subsidies, and rewards should be exam-
ined for their suitability to different sources of nonpoint
pollution and to the particular conditions of a given water-
shed. The exact mix of measures should be determined
at the State level in an EPA-approved nonpoint program.
In applying these measures, off-farm contributors to the
chain of nonpoint pollution should not be immune—for
example, if excessive nitrogen is a water quality problem
associated with nonpoint source pollution, then fertilizer
manufacturers should be brought within the regulatory
program. The final program must be equitable, effective,
and easy to administer.
Although the catch-all title is "enforcement," implementa-
tion of nonpoint controls can be achieved only by a crea-
tive mixture of traditional enforcement or regulatory tools
and economic incentives. The bottom line to any nonpoint
program must be a discernible improvement in water qual-
ity and any proposed technique to achieve that end must
be measured against that goal.
Nonpoint pollution and nonpoint polluters differ from
point source pollution and point source polluters in several
key aspects, but are the same in others. Understanding
these distinctions and similarities is critical to designing
any program to implement nonpoint source controls.
Unlike the industrial point source program, the benefi-
ciaries of nonpoint source pollution control and the ob-
servable impacts of this form of pollution are generally far
away from the pollution's point of origin. Runoff from a
farm field or mine site often creates a water quality prob-
lem miles away from the pollution source, in some receiv-
ing stream, lake or estuary. This fact can create serious
perception as well as enforcement problems. Farmers ask
why they should be required to undertake the costs of
implementing best management practices to produce
benefits for the distant public-at-large; accordingly, they
expect government assistance to offset the costs of imple-
menting these controls. Whether or not that perception
has any validity when compared to industrial pollution is
irrelevant, because it must be dealt with in any nonpoint
source implementation program.
The distance between the origin of pollution and its im-
pact creates specific enforcement problems as well. Not
all eroded soil ends up in a receiving stream. The distance
factor makes it difficult to apportion liability for nonpoint
pollution. Intervening causes of pollution have too much
opportunity to occur between the points of origin of non-
point pollution and its impact.
How can a specific farmer's share of the pollution and
liability for it be distinguished from naturally occurring non-
point pollution or, for that matter, from the runoff coming
from the farm downstream or upstream of him? What if
farmer A's eroded soil is being trapped by off-farm stream-
bank vegetation, but farmer B, who's losing the same
amount of soil, has no such assistance—should an en-
forcement policy distinguish between these farmers be-
cause only farmer B is causing a discernible water quality
impact even though both farmers are losing soil?
How does one measure the percentage of pollution at-
tributable to a particular activity and assess liability for it
with any degree of precision, let alone equity—a basic
premise of most pollution control programs—if one can't
establish, let alone quantify, the relationship between the
polluting activity and the pollution? What unit of measure-
ment can be used to design an effective effluent reduction
program for nonpoint source pollution under these circum-
stances? The Universal Soil Loss Equation, which mea-
sures on-farm erosion, is of no use when it comes to as-
signing liability for an off-farm adverse water quality
impact. These questions are unique to nonpoint pollution.
The fact that a nonpoint source enforcement agency is
faced with trying to regulate pollution that can't be mea-
sured at its source and can't be attributed in many situa-
tions to a single identifiable cause would seem to elimi-
nate those enforcement techniques that depend on
apportioning responsibility between sources: for example,
effluent fees or monetary penalties.
A second distinguishing characteristic of nonpoint
source polluters is that some of them receive a direct eco-
nomic benefit from the application of control measures,
while others do not. Few, if any, point source polluters
receive any benefit from pollution abatement. Thus, the
farmer, the forest products company, the miner, and the
rancher should all benefit from the retention of soil on their
land; however, the industrial or urban source of nonpoint
pollution may not. This distinction between sources has
relevance for assessing whether incentives, like subsidies
or tax relief, are appropriate for a particular source. It
makes no sense to subsidize a farmer or timber products
company for adopting control technologies that are al-
ready in its best interests to employ. Using a subsidy in
those situations amounts to giving those polluters an un-
warranted double benefit.
A separate question that must be raised when evaluat-
ing incentives as a means of achieving implementation of
control practices is whether the particular circumstances
60
-------
of the situation, such as the extremely depressed farm
economy, warrant abandoning the basic premise of most
pollution control statutes that polluters should not be reim-
bursed for the costs of cleaning up their pollution—"The
polluter always pays" maxim. This is certainly a cardinal
rule of the point source regulatory program. Yet, most
discussions on achieving implementation of nonpoint pol-
lution control measures assume the opposite, namely that
the polluter should be reimbursed for his cleanup costs. I
find these discussions deeply troubling and the relief pro-
posed unwarranted in law or fact. Yet, mine is not a politi-
cally popular position to take and, therefore, one unlikely
to prevail. However, if polluters are to be subsidized for the
cost of implementing control technologies in the nonpoint
as distinguished from the point source program, then that
program must have as an indispensible component the
achievement of some demonstrable reduction in nonpoint
pollution. Otherwise, we will risk replicating the wasteful
experience of the Rural Clean Water Program.
Another distinguishing characteristic of nonpoint pollu-
tion is that many factors affect nonpoint pollution, many of
which are beyond the control of the source. Take, for ex-
ample, agricultural nonpoint pollution. The occurrence,
quantity, and quality of agricultural nonpoint pollution are
influenced by weather, land contour, crop choice, plowing
techniques, and pesticide and fertilizer use as well as by
external factors, like the domestic and foreign demand for
farm products, the cost of fuel and fertilizers, and govern-
ment subsidy programs (Harrington et al. 1985).
Instead of looking at this list and concluding that non-
point pollution is too complex to tackle in any regulatory or
incentives program, the length and diversity of the list
should provide multiple opportunities for abating nonpoint
pollution. A nonpoint source implementation strategy
should look broadly at the chain of contributors to the
pollution and not just the last link in the chain, the final
source. For example, in terms of achieving water policy
goals, imposing controls on the production and cost of
pesticides may be more cost effective than requiring the
farmer to build terraces.
The last distinguishing feature of nonpoint pollution that
deserves some mention here is the view that the farmer
cannot be compelled to do anything, that is, that a regula-
tory program with permits, on-site inspections, and penal-
ties simply will not work on the American farm. At the core
of this perception is the myth that the American farm in-
dustry is composed of moderate-size family farms
($100,000 to $200,000 in annual product sales). The myth
is given poignancy by the very real economic plight of the
family farm.
In actuality, the American family farm is disappearing.
Today, the family farm represents less than 11 percent of
moderate-size farms; a decade ago, the number was 21
percent. The American farm industry is clearly in transi-
tion, moving from a diverse collection of various size
farms to a distinctly bipolar structure composed largely of
very small or parttime enterprises and very large, industri-
alized operations. The disappearance of the family farm,
without question, has created very real stress on the farm
economy as well as stress on many watersheds. Good
conservation, which is a long-term investment, may not
appear relevant during an economic depression when
short-term goals hold greatest appeal.
Implementation of nonpoint source controls clearly did
not cause the disappearance of the moderate-size family
farm. Just as clearly, the conditions of this transition per-
iod should not be allowed to define the content or ap-
proach of any nonpoint source program that Congress
creates. Yet the myth of the family farm persists and is at
the center of the different approaches taken toward the
polluter in the nonpoint and point source programs, even
LEGAL ASPECTS OF NONPOINT SOURCE POLLUTION
today. How else can one explain the reluctance to impose
regulatory controls on farming activities and the too-ready
reliance on voluntary, educational and subsidy programs?
Yet, how different are the large, industrialized farms,
which are clearly the wave of the American farm future,
from industry or forest products companies?
A strictly voluntary approach has not worked and will
not work for the farmer any more than it would work with
the chemical industry. Nonpoint source pollution is in-
creasing, not decreasing, perhaps reflecting the stress of
the family farmer, who will opt for short-term gains and
plow his highly erodible land rather than take it out of
production. Why should the family farmer be any more
willing to cut into profit margins than the steel industry,
particularly in his economically distressed condition?
To allow the image of the family farm to dictate how we
approach the nonpoint pollution problem would be a mis-
take. The attitude toward the farming community must
change dramatically if this most serious source of water
quality degradation is to be controlled. We must acknowl-
edge that farming is no different from manufacturing
chemicals, mining coal, or cutting trees when it comes to
assigning responsibility for nonpoint pollution and bring-
ing it under control.
I start with the premise that an effective nonpoint source
control program must be regulatory in nature and gener-
ally indistinguishable from any other basic pollution con-
trol program. Such a program should provide for permits,
objective standards, on-site inspection by State and Fed-
eral officials, and a full panoply of enforcement measures,
including citizen suits. I think a useful model that might be
pursued in designing such a program is that offered in the
Surface Mining Reclamation and Control Act, a statute
significantly dealing with erosion control and water quality.
The regulatory core of that statute is its performance
standards and design criteria, which dictate with some
precision how mining will be conducted in various parts of
the country. The law requires the industry to implement
specific technological controls—like terraces and sedi-
mentation ponds—to prevent environmental problems
from occurring. Additional features of interest in that law
are its bonding, small operators' assistance, and trust
fund programs.
The design criteria approach of the Surface Mining Act,
which is not very different from the technology-based ap-
proach of the Clean Water Act, has many of the same
advantages for both the regulated industry and the regu-
lating agency. Structures are easier to inspect than efflu-
ents, particularly nonpoint source effluents that can
change under different background circumstances. Af-
fected industries are given a measure of certainty that if
they follow the design criteria they will meet the mandated
performance standard and thus be in compliance with the
law. These criteria and standards are objective and incon-
trovertible, limiting opportunities for subjective value judg-
ments about possible violations. Certainty, equity and
ease of administration are important features of any regu-
latory program and, therefore, should be goals of a non-
point source pollution control program.
Regulation, however, is not the only consideration in
developing this program. Affirmative action must be taken
to eliminate the incentives currently encouraging prac-
tices that lead to nonpoint pollution. Crop subsidies, price
supports, disaster assistance, and other financial help
should not be available in those circumstances that can
cause nonpoint pollution. The tax code also should be
reviewed with an eye toward eliminating tax relief for pollu-
tion-causing activities. Instead, incentives should be built
into the commodity and tax programs for nonpoint source
control. The entire chain of contributors to nonpoint
source pollution should be part of this review. The goal
61
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
should be reduction at the source; in many instances this
will not be the last link in the chain, but rather the first, for
example, the producer of pesticides or fertilizers.
Any nonpoint source control program must be flexible
and must offer a mix of measures, both disincentives and
incentives, to achieve program goals. Different measures,
both regulatory and economic, should be examined for
their suitability to specific sources of nonpoint pollution
and to the particular conditions of a given watershed. The
exact mix of measures should be determined at the State
level in an EPA-approved nonpoint program.
Any nonpoint program should have the capacity to dis-
tinguish between problems and to address those prob-
lems in some priority fashion. I would suggest that the first
order of business should be bringing new activities into
compliance, so that the inventory of problems does not
keep growing. As those activities are brought within the
regulatory fold, then the focus can shift to addressing the
backlog of existing operations, which may be decreasing
on its own for totally unrelated reasons.
Without question what I have proposed here is the most
aggressive approach to solving the problem of nonpoint
source pollution. It reflects my deep conviction that the
problem is of sufficient severity to warrant the imposition
of these types of measures and that the affected sources
can absorb this responsibility like any other part of the
economy, with neither less nor greater dislocation. Equally
clear, traditional attitudes must change before this can be
achieved. The legislation pending in Congress is a first
step toward nonpoint source pollution control and does
not bar States from following the approach proposed here.
Should the States elect not to participate in the proposed
program in a meaningful way, then Congress, at the next
reauthorization of the Clean Water Act, should seriously
consider taking the program to the next generation of con-
trol, as I have proposed in this paper.
REFERENCES
Harrington, Krupnick, and Peskin. 1985. Policies for nonpoint
source water pollution control. J. Soil Water Conserv. 40: 27.
62
-------
CONTROLLING NONPOINT SOURCES OF POLLUTION—THE
FEDERAL LEGAL FRAMEWORK AND THE ALTERNATIVE OF
NONFEDERAL ACTION
RICHARD R. GREENFIELD
U.S. Department of Interior
Washington, D.C.
ABSTRACT
This paper considers certain State and Federal legal as-
pects of nonpoint source pollution control including a
general overview of Federal and representative State
laws on the subject. A major emphasis is the benefits
from reinforcing existing State legal and administrative
institutions to serve as the foundation for a national non-
point source pollution control effort. All too often, when a
national regulatory effort is envisioned, individual State
differences and preferences are ignored in the effort to
articulate a single Federal policy. States can in fact de-
velop and implement meaningful nonpoint source pollu-
tion control programs without traditional Federal controls.
Moreover, even where such pollution involves more than
one State and control efforts must be expanded accord-
ingly, there is the potential to use such proven non-Fed-
eral dispute resolution tools as interstate compacts or
interstate agreements. This paper also considers the ade-
quacy of current Federal nonpoint source control efforts
in the context of existing legislative authorities.
As each State strives to improve overall water quality, non-
point source (NFS) pollution increasingly appears as a
problem which has not been fully addressed in prior as
well as current water quality enforcement efforts. (U.S.
Geol. Survey, 1983; Off. Tech. Assessment, 1984). This
conclusion becomes more apparent as existing Federal
and State NFS statutes and regulations are enforced
more strictly. Looking to the future, NFS pollution will have
to be effectively controlled if the Nation's water quality is to
continue to improve. While nonpoint source pollution is
clearly a problem, the diffuse and intermittent nature of
the discharges involved make definition as well as mea-
surement difficult.
Many of the NFS pollution measurements are rather
subjective in nature. For example, in America's Clean Wa-
ter: The States' Evaluation of Progress 1972-1982 (Associ-
ation of State & Interstate Water Pollution Control Admin-
istrators, 1984), States reported "severe" impairment of
designated water uses as the result of nonpoint pollutants
generated through the following activities (number of
States reporting in parentheses): agricultural (16); urban
(11); mining (15); land disposal (12), and construction (6).
States reporting widespread geographic impairments
caused by nonpoint pollutants from these same activities
were as follows: agricultural (29); urban (8); mining (2);
land disposal (5); and construction (6).
An important and still outstanding public policy question
is whether the regulation of NFS pollution should be ad-
dressed through Federal or State control structure or
some combination of the two. As a starting point in an-
swering this question, this paper will examine the basic
legal framework of Federal regulation of nonpoint source
pollution of the Nation's water resources. Emphasis will be
placed, of necessity, on the Federal Water Pollution Con-
trol Act, as amended by the Clean Water Act Amendments
of 1977.
The basic thesis of this paper is that in the context of
applicable Federal laws, States have the latitude and
should take the initiative to develop and implement mean-
ingful NFS pollution programs. There is no need to wait for
Congress to develop and implement a comprehensive
regulatory program. In fact, it can be argued that a tradi-
tional Federal regulatory program is not in the States' col-
lective best interests. Why? To oversimplify, it is all too
often the case that when a Federal regulatory program is
designed, individual State differences and preferences
are overlooked in the implementation of a National regula-
tory structure.
Where NFS pollution involves more than one State and
control efforts must be expanded accordingly, there is the
clear potential to use such proven non-Federal dispute
resolution tools such as interstate compacts. It is the very
nature of the NFS pollution (especially the diffuse and
intermittent nature of the discharges) as well as potential
legal problems associated with individual State action,
that will often tend to support the interstate compact ap-
proach.
INTRODUCTION
It is often difficult to distinguish between point and non-
point sources of water pollution. Many water pollution
sources are not clearly "point" or "nonpoint", but have
characteristics which suggest placement along a contin-
uum between these two classifications. In addition, the
classification of a pollutant as "point" or "nonpoint" may
vary at different stages in the pollutant life cycle. For ex-
ample, a nonpoint source may be reclassified as a point
source if the pollutant materials in question flow into a
more discernible, confined conveyance such as a ditch or
channel. (See, for example, Natural Resources Defense
Council, Inc. v. Costfe, 568 F. 2d 1369 (1977).)
Unfortunately, Federal law does not provide a precise
definition for NFS pollution. To understand the statutory
scope of the concept, a mental definition must be drawn of
the opposite of the following statutory definition for point
source pollution:
... any discernible, confined and discrete conveyance, in-
cluding but not limited to any pipe, ditch, channel, tunnel,
conduit, well, discrete fissure, container, rolling stock, con-
centrated animal feeding operation, or vessel or other float-
ing craft from which pollutants are or may be discharged. (33
USC §1362 (14))
APPLICABLE FEDERAL LAW RELATING
TO POINT SOURCE POLLUTION
It is somewhat paradoxical that to understand nonpoint
source pollution, one must first examine the statutory defi-
nition of point source pollution. Point source pollution is
concerned primarily with pollutants discharged or other-
wise dispersed from a discrete pipe or conveyance.
'The opinions expressed in this paper are strictly those of the
author and should not necessarily be construed as those of the
Department of Interior.
63
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Among other examples of point source pollution are sew-
age effluent and incinerator residues. Applicable Federal
law generally classifies any activity that emits pollution
from an identifiable point source as point source pollution.
Return flows from irrigated agriculture and unchannelled
and uncollected surface water have been specifically ex-
empted from the point source definition (33 USC §1362
(14)).
Since its original enactment in 1948 (pursuant to 62
Stat. 1155), the key Federal legislation for the control of all
forms of water pollution has been the Federal Water Pollu-
tion Control Act (FWPCA). The Act was substantially
amended in 1972 (pursuant to P.L. 92-500; 86 Stat. 816)
and again in 1977 (pursuant to PL 95-217; 91 Stat. 1567).
The 1977 amendments are known as the Clean Water Act
of 1977.
Under the Federal Water Pollution Control Act, as
amended (Clean Water Act), the actual administration of
water quality standards for point sources has been left to
the States, who are in turn free to impose stricter point
source controls than these promulgated by the EPA. How-
ever, if the State standards are less strict than applicable
Federal standards, the EPA may impose its own. Section
303 of the act requires States to identify water quality
limited segments of streams or other watercourses. "Wa-
ter quality limited" refers to that portion of a stream or
watercourse that receives such a large amount of point
source pollution that discharge standards alone are inade-
quate in and of themselves to preserve minimum water
quality. Where such limitations on water quality apply, the
act requires the establishment of total maximum daily
waste loads for each threatened area. The total maximum
waste load is then allocated among current users of the
area.
Section 402 of the Clean Water Act established the Na-
tional Pollution Discharge Elimination System (NPDES).
An NPDES permit is required in order to discharge point
source pollutants into navigable wafers.
To obtain a Section 402 NPDES permit, certification
must be obtained from the applicable State agency (or the
EPA in the absense of responsible State authority) must
certify that the proposed discharge complies with applica-
ble Federal effluent standards. For the purposes of com-
pliance, such standards include those specified by Sec-
tion 301 of the act. Pollution standards prescribed under
Section 301 have become more stringent in recent years.
Under the current schedule, there is an ongoing shift from
the mandatory use of "best practicable technology" (BPT)
to "best available technology" (BAT). The 1977 amend-
ments to the Clean Water Act established a group of "con-
ventional pollutants" (e.g. suspended solids, conforms,
etc.) for which the "best control technology" must be
used.
It should be remembered that the funding authorization
for the Federal Wafer Pollution Control Act is expired but
the regulatory authority continues. Funding reauthoriza-
tion will again be addressed in the 99th Congress.
The Federal Water Pollution Control Act (33 USC §1251
(b)) expressly recognizes "... the primary responsibilities
of the States to prevent, reduce and eliminate pollution."
The act does not in any way affect State authority to allo-
cate quantities of water within State boundaries. The
FWPCA (33 USC §1251 (a)) obligates the EPA Administra-
tor to:
... encourage cooperative activities by the States for the
prevention, reduction, and elimination of pollution, encour-
age the enactment of improved ... uniform State laws relat-
ing to the prevention, reduction and elimination of pollution;
and encourage compacts between States for the prevention
and control of pollution.
There is every reason why this same authority which
encourages cooperation between States in the adminis-
tration of point source control programs could also be
used as support for State NPS control programs.
WonpoW Source Pollution and Applicable Federal
Lara. Nonpoint sources of water pollution include diffuse
pollution sources that are not regulated as point sources.
It normally includes agricultural and urban runoff, runoff
from construction and from surface mining activities,
among other sources. As the court noted in United States
v. Earth Sciences, Inc., 599 F. 2d 368,373 (10th Cir., 1979):
... [t]he legislative history [of the FWPCA] indicates..Con-
gress was classifying nonpoint source pollution as disparate
runoff caused primarily by rainfall around activities that em-
ploy or cause pollutants.
The U.S. Senate Report on what eventually became the
Clean Water Act (33 USC §1314) was cited by the court in
Earth Sciences as indicative of the significance of NPS
pollution in the overall Federal water pollution control ef-
fort. This Report stated, among other things:
Sediment, often associated with agricultural activities is by
volume our major pollutant, not only by the degrading effect
of the sediment, but because it transports other pollutants.
Fertilizer and pesticide runoff are also major agricultural non-
point sources. Poor forestry practices, including indiscrimi-
nate clear-cutting, may also generate substantial soil erosion
problems.
One of the common problems associated with pollution
control is the dramatic increase in storm runoff when the
earth's surface is made impermeable. Thus, highways,
buildings, and parking lots all contribute substantially to the
accelerated runoff of rainwater into natural water systems.
The greater volume and greater velocity produce high rates
of erosion and siltation. In addition, highway runoff often in-
cludes oil, rubber particles, lead asbestos and other ele-
ments or additives deposited on highways as a result of ve-
hicular traffic.
There is some evidence in the legislative history of the
1972 and 1977 amendments to the Federal Water Pollu-
tion Control Act to suggest that Congress might have reg-
ulated nonpoint sources of pollution as well if they could
have found a way to do so. Instead, Congress was forced
to content itself with such statutory tools for addressing
the NPS problem as the following:
Section 201, which declares that one of the key objec-
tives of the United States Code subchapter (Section 201 et
seq. of the FWPCA) relating to grants to municipalities for
the construction of waste treatment works is control "to
the extent practicable" of nonpoint sources.
Section 208, provides for areawide waste treatment
management planning. The preparation of areawide plans
started in the mid-1970's with the publication of guidelines
by the EPA Administrator (pursuant to 40 Federal Register
55, 321, eventually codified as 40 CFR Part 35). The
guidlines enable the Governor of each participating State
to identify areas within the State as the result of urban-
industrial pollution concentrations or other factors have
substantial water quality problems.
Under Section 208, after the Governor of each State
identifies the areas of the State having substantial water
quality problems, he or she is then mandated to: (a) Desig-
nate the boundaries of each such area; and (b) select a
single planning organization which includes local repre-
sentation, capable of developing and implementing a con-
tinuing areawide waste treatment management planning
process.
Each State is required to act as the chief planning
agency for all portions of its territory not otherwise desig-
64
-------
LEGAL ASPECTS OF NONPOINT SOURCE POLLUTION
nated. An alternate procedure for the designation of wa-
ter-quality impaired areas is available in the absence of
gubernatorial action.
Plans developed under Section 208 process are re-
quired to contain alternatives for waste treatment and be
applicable to all wastes (both point and nonpoint) gener-
ated in the area involved. Under Section 208, areawide
plans must also identify municipal and industrial waste
treatment works necessary to meet the anticipated waste
treatment needs of the designated area over a 20-year
period. During the 1970's, a period of strong Federal sup-
port for the Section 208 construction program, the award
of Federal waste treatment funds was based in large part
on the identification of anticipated waste treatment needs.
Finally, Section 208 plans must include a process to
identify and control nonpoint sources of pollution to the
extent feasible. (Section 208 (b) (I) (F) through (H)). Unfor-
tunately, feasibility is not defined in the statute or the appli-
cable regulation (40 CFR §35.1505 (d)). According to 40
CFR §35.1521 -4(c), Section 208 plans must control non-
point sources of pollution through the use of best manage-
ment practices (BMP's). In the nonpoint context, BMP's
are defined as
... those methods, measures, or practices to prevent or re-
duce water pollution and include but are not limited to struc-
tural and nonstructural controls, and operation and mainte-
nance procedures. BMP's can be applied before, during, and
after pollution-producing activities to reduce or eliminate the
introduction of pollutants into receiving waters. Economic,
institutional, and technical factors shall be considered in de-
veloping BMP's. BMP's shall be developed in a continuing
process of identifying control needs and evaluating and mod-
ifying the BMP's as necessary to achieve water quality goals
(see §35.1521-3 (h)). To the extent practicable, BMP's should
be set forth in a document which can be distributed widely in
the planning area. (40 CFR §35.1521-4 (c))
From the beginning, designated planning agencies
found it easier to address point rather than nonpoint
sources. Why? At the risk of considerable oversimplifica-
tion, the key reason is that point sources are easily defin-
able and control technologies are relatively well-devel-
oped. By contrast, the chief techniques for controlling
NPS pollution often involve some form of land use plan-
ning or other public control of private land use, a topic
over which there is much political controversy. The limited
acceptability of key NPS control strategies coupled with
delays on the part of EPA in preparing necessary guide-
lines, resulted in the targeting of such Section 208 waste
treatment construction monies as were available on point
source control efforts. This assymetric targeting has re-
sulted in the construction of a network of waste treatment
facilities which are less than adequately equipped, in the
view of several observers, to handle the more diffuse NPS
problem.
While the 1977 amendments to Section 208 took certain
steps toward regulating nonpoint sources, the nonregula-
tory flavor of the section was retained. On the control side,
subsection (j) established an agricultural NPS control pro-
gram. Under this subsection, the Secretary of Agriculture,
in conjunction with the EPA Administrator, is empowered
to develop and administer a program under which rural
land owners and operators are eligible for Federal finan-
cial assistance for NPS control. In return, the rural land
owner or operator must provide a contractual commitment
of at least 5 years to use best management practices
(BMPs) to control specified agriculturally-based nonpoint
sources of water pollution. By virtue of the fact that con-
tracts are made directly between the Department of Agri-
culture and the rural land owner or operator (rather than
through the State or local government or areawide plan-
ning agency), Section 208 (j) authorizes what amounts to
direct Federal support for NPS control. It should be em-
phasized, however, that participation under subsection (j)
is voluntary.
The 1977 amendments to Section 208 also made clear
that prior to the determination by the Governor of any
State that an NPS control program was necessary under
Section 208 (b)(4) to meet Statewide water quality stand-
ards and implementation plans required by Section 303,
the approval of the EPA Administrator is necessary. It was
previously unclear whether such approval was required.
Section 208 was also amended in 1977 to require that
any NPS control program developed under Subsection
(b)(4) as part of a Statewide program under Section 303,
adequately consider the impact of nonpoint sources on
the Nation's wetlands. This is done through the require-
ment (pursuant to §208 (b)(4)(B)) that any NPS program
designed at least in part to control the discharge of dredge
or fill materials into navigable waters include provisions to
ensure: (1) coordination with approved State Section 404
programs; (2) that discharge activities are conducted pur-
suant to BMP's; and (3) consultation with relevant parties
such as the State agency with primary jurisdiction over
fish and wildlife resources.
Section 304, relating to information and guidelines, con-
tains a mandate to the EPA Administrator to develop (a)
guidelines for identifying and evaluating the nature and
extent of nonpoint sources of pollution; and (b) processes,
procedures, and methods to control pollution resulting
from such sources as:
• agriculture and silvicultural activities, including runoff
from fields and crop and forest lands
• mining activities, including runoff and siltation from
new, currently operating and abandoned surface or under-
ground mines
• all construction activity, including runoff from facilities
resulting from such construction
• the disposal of pollutants in wells or in subsurface
excavations
• salt water intrusion resulting from the reduction in
fresh water flow for any cause
• changes in the movement, flow or circulation of
ground waters.
However, Section 304 does not provide for the actual regu-
lation of NPS pollution as such.
Section 304 was amended in 1977 to authorize the EPA
Administrator to mandate BMP's to address toxic and haz-
ardous pollutants (the section specifically mentions point
sources but implicitly includes nonpoint sources as well)
which are associated with or ancillary to an industrial man-
ufacturing or waste treatment process. Over the longer
term, the importance of these amendments to Section 304
will largely be a function of how the courts construe the
terms "associated with or ancillary to".
Implementation of Applicable Federal Law by EPA.
Enforcement of the NPS provisions of the Federal Water
Pollution Control Act has not been a high priority of the
agency since the passage of the 1977 amendments. Re-
gardless of whether the effect has been positive or nega-
tive it is clear that the issuance of NPS guidelines has, in
some cases, been delayed and certain NPS regulatory
initiatives have not moved beyond the proposal stage.
Over the last year there has been renewed activity On
December 12, 1984, an EPA-directed task force issued a
National Nonpoint Source Policy. The overall objective of
the task force is to support and accelerate the develop-
ment and implementation of NPS management programs
that ensure water quality protection. The statement of
general policy issued by the task force provides a key
sense as to where EPA now wishes to direct its NPS con-
trol efforts:
65
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Achievement of national clean water goals requires imple-
mentation of NPS management programs. Emphasis should
be placed on implementing NPS programs in watersheds
affecting priority waters. Sources of nonpoint pollution
should be evaluated to assess potential water quality im-
pacts and needed program actions. NPS management is
required to protect high quality surface and ground waters
and to restore and/or improve water quality surface uses. In
many instances, prevention of degradation -has proven to be
far more cost-effective than remedial measures.
NPS management programs must be flexible to allow for
site-specific solutions to problems, to accommodate changes
in technical knowledge, to respond to changes in uses of
land, and to optimize net on- and off-site benefits. A mix of
both point and nonpoint source measures should be consid-
ered in developing cost-effective strategies to improve and
maintain water quality.
With federal leadership and coordination, all levels of gov-
ernment and the private sector need to cooperate to provide
continued progress with available programs and delivery
systems, to identify unmet needs, and to develop and imple-
ment NPS managment programs where needed.
Mompointt Source Control) sA th® State level. The pre-
ceding review of Federal law relating to point and nonpoint
sources of water pollution was not meant to suggest that
individual States have not been active. While it is beyond
the scope of this paper to detail the range and variation of
State NPS initiatives, two examples of State actions illus-
trate State involvement. Given the key role that agricul-
tural activities play in the generation of NPS pollution,
both of these examples come from the agricultural sector.
In the 1970's, the State of Iowa enacted a soil conserva-
tion law (pursuant to Iowa Stat. Ann. Chapter 467 A) under
which rural land owners and operators can, under certain
circumstances, be forced to adopt soil conservation mea-
sures to reduce or eliminate NPS pollutants with the as-
sistance of appropriate public agencies. Similarly, New
York has enacted legislation that requires the develop-
ment of soil conservation plans (Soil Conservation Dis-
tricts Law §4 et seq., McKinn. Consol. Laws).
From a more general, policy-oriented perspective, it
seems likely that any attempt on the part of Congress to
move toward Federal management of nonpoint sources of
water pollution would be inherently ineffective because it
would fail to recognize the very significant regional varia-
tions in the NPS problem. It is difficult to conceive of a
system of Federal regulation that could adequately, effec-
tively and equitably recognize different NPS problems in,
for example, New York and Arizona.
The water policy of the current Administration clearly
follows from the established tradition of congressional def-
erence to State management of State water resources.
The Administration takes the view that the States have
primary authority for the management of their own water
resources except where Congress has indicated other-
wise on a case-by-case basis.
If the States are, as I suggest, going to continue to be
the primary managers of their own water resources as well
as address pollution problems that are not susceptible to
cost-effective National regulation, then what mechanisms
can be put forward in the name of effective State manage-
ment. As the next section of this paper indicates, I believe
that interstate compacts, a mechanism with proven suc-
cess in resolving interstate surface water disputes, can be
effective in controlling nonpoint sources of water pollution
that affect more than one State.
Constraints to OndSvldya! State Action. Despite its tra-
ditional deference to State water laws and failure to defini-
tively supersede State regulation of nonpoint sources,
Congress has not granted States the authority to regulate.
The Commerce Clause of the Federal Constitution would
otherwise prohibit any individual State action to regulate
NPS pollution on the basis of the police power reserved
under the Tenth Amendment must be weighed against the
potential burden on commerce. A long series of U.S. Su-
preme Court decisions (e.g., West v. Kansas Natural Gas
Co., 221 U.S. 229 (1911); Pennsylvania v. West Virginia,
262 U.S. 553 (1923); Pike v. Bruce Church, Inc., 397 U.S.
137 (1970); Douglas v. Seacoasf Products, Inc., 431 U.S.
265 (1977); City of Philadelphia v. New Jersey, 437 U.S.
617 (1978); Hughes v. Oklahoma, 441 U.S. 322 (1979);
and New England Power Co. v. New Hampshire, 102 S. Ct.
1096 (1982)) have invalidated State attempts to burden
interstate commerce in the name of simple economic pro-
tectionism. Where this rationale has been utilized for State
legislation seeking to regulate any form of interstate com-
merce, a per se rule of invalidity has traditionally been
employed (Bowman v. Chicago & Northwestern Railroad
Co., 125 U.S. 465 (1888); H.P. Hood & Sons v. Dumond,
336 U.S. 525 (1949); Bread v. City of Alexandria, La., 341
U.S. 622 (1951); Huron Portland Cement Co. v. City of
Detroit, Michigan, 362 U.S. 440 (1960); and Spomase v.
Nebraska, ex re/., Douglas, 102 S. Ct. 3456(1982)). Where
State legislation has been sufficiently related to the public
health, safety and welfare more flexible burden-on-com-
merce balancing test has been used Southern Pacific v.
Arizona, 325 U.S. 761 (1945) and its progeny (especially
Pike v. Bruce Church, Inc., 397 U.S. 137 (1970)). The bur-
den-of-commerce test contained in Pike v. Bruce Church,
Inc., 397 U.S. at 142, is worth repeating:
[WJhere the [State] regulates evenhandedly to effectuate a
legitimate public interest, and its effects on interstate com-
merce are only incidental, it will be upheld unless the burden
imposed on such commerce is clearly excessive in relation to
the putative local benefits.
On the strength of the Commerce Clause, the Supremacy
Clause and Court decisions (International Shoe Co. v.
Washington, 326 U.S. 310 (1945), Prudential Insurance Co.
v. Benjamin, 323 U.S. 408 (1946), and Western and South-
ern Life Ins. Co. v. State Board of Equalization of California,
101 S. Q. 2070 (1981)), Congress may grant to the States
authority to regulate commercial activities in the name of
NPS control in a manner that would not otherwise be per-
missible. Since Congress has not chosen to do so, basic
constitutional restraints on individual State action may en-
courage States to reexamine the compact alternative to
NPS control.
DNTEIft
It is always possible, of course, that Congress will deter-
mine that a comprehensive (and inherently expensive)
Federal program to regulate and control nonpoint sources
of water pollution is necessary. In such event, State laws
(including compacts) could be superseded. I suggest,
however, that in the present Federal budget climate, any
such action is unlikely at best. Moreover, a National pro-
gram to regulate nonpoint sources of water pollution may
unavoidably overlook individual State differences and
preferences in the effort to articulate a comprehensive
Federal policy. What then are the options for States if one
accepts the proposition that nonpoint sources of water
pollution constitute, in certain areas, a potentially serious
threat to public health? I suggest that there are basically
two options. The first is individual State action as repre-
sented by the efforts of Iowa and New York (among other
66
-------
LEGAL ASPECTS OF NONPOINT SOURCE POLLUTION
States) to address NFS problems of particular concern.
Such individual State action may not unduly burden inter-
state commerce. Second, and perhaps a more functional
non-Federal approach given the often regional manifesta-
tions of NFS problems, is that of interstate compacts.
Interstate Compacts and NPS Management. Inter-
state compacts are cooperative agreements enacted by
the legislatures of signatory States and thereafter con-
sented to by Congress the requirement of congressional
consent follows from the U.S. Constitution (Article I, §10)
which (a) precludes States from entering into any treaty,
alliance or confederation and (b) stipulates the consent of
Congress to be a prerequisite for any State to enter into
any agreement or compact with another State. The basic
theory surrounding the requirement of Congressional con-
sent is the purported need to protect the interests of the
Federal Government and of States not parties to the com-
pact. The late Justice Frankfurter has written of interstate
compacts as the primary mechanisms available to the
States to circumvent the institutional barriers to regional
development. (Frankfurter and Landis, "The Compact
Clause of the Constitution", 34 Yale Law Journal 685
(1925)).
Not every compact requires congressional consent. Fol-
lowing Virginia v. Tennessee, 148 U.S. 503, 518-519
(1893), it appears that consent is required only for those
agreements that increase the political power of signatory
States in contrast to nonsignatory States and thus poten-
tially conflicts with the Supremacy Clause. Because any
interstate compact dealing with nonpoint sources of water
pollution can be viewed as being potentialy in conflict with
the Supremacy Clause, congressional consent is as-
sumed for the purposes of this paper to be necssary.
Application of the compact approach to interstate water
pollution problems is not a totally untried concept. While
an interstate compact has yet to be developed to address
purely nonpoint sources, at least four water pollution com-
pacts have already been enacted:
1. New England Interstate Water Pollution Control Com-
pact. Signatories: Connecticut; Maine; Massachusetts;
New Hampshire; New York; Rhode Island and Vermont.
Approved by Congress pursuant to 61 Stat. 682 (PL. 80-
292 (1947)). Purpose: To establish the New England Inter-
state Water Pollution Control Commission to control and
reduce pollution on interstate waters in the New England
States, including New York.
2. New Hampshire-Vermont Interstate Sewage and
Waste Disposal Facilities Compact. Signatories: New
Hampshire and Vermont. Approved by Congress pursuant
to 90 Stat. 1221 (PL. 94-403 (1976)). Purpose: To provide
authority to local governments and sewage districts in
New Hampshire and Vermont to establish joint sewage
disposal and other waste product treatment facilities as
part of comprehensive pollution abatement efforts.
3. Ohio River Valley Water Sanitation Compact. Signato-
ries: Illinois; Indiana; Kentucky; New York; Ohio; Pennsyl-
vania; Virginia and West Virginia. Approved by Congress
pursuant to PL. 76-739 (54 Stat. 742 (1940)). Purpose:
provided authority for a coordinated State response to wa-
ter and waste treatment problems in the Ohio River Valley.
4. TrhState Sanitation Compact. Signatories: Connecti-
cut, New Jersey, and New York. Approved by Congress
pursuant to Pub. Res. No. 62 (49 Stat. 932 (1935)). Pur-
pose: To establish an Interstate Sanitation Commission
with the overall mandate to improve water quality in the
boundary areas shared by Connecticut, New Jersey, and
New York.
Interstate efforts to address NPS problems of mutual
concern may in time be preempted by a comprehensive
Federal law. Such was in fact the case in the mid-to late-
1960's when several interstate air pollution compacts
were enacted (e.g., Illinois-Indiana Air Pollution Compact;
Mid-Atlantic States Air Pollution Compact; Ohio-Kentucky
Air Pollution Compact; Ohio-West Virginia Air Pollution
Compact; and Kansas-Missouri Air Pollution Compact).
While several of these compacts were pending before
Congress, the Air Quality Act of 1967 was enacted (PL.
90-148; 81 Stat. 485). Similar preemption of interstate
NPS compacts is, of course, possible but it is considerably
less likely for two reasons. First, a general scheme for the
control of water pollution is already a part of Federal law.
Second, Congress has expressly encouraged, pursuant to
33 USC §1251 (a), compacts between States for the pre-
vention and control of water pollution.
67
-------
State Nonpoint
Source Programs
FUNDING NONPOINT CONTROL PROJECTS IN MISSOURI
JOHN HOWLAND
Missouri Department of Natural Resources
Jefferson City, Missouri
On June 15, 1983, the Missouri General Assembly ap-
proved House Joint Resolution No. 21. This measure,
called Constitutional Amendment No. 2, was passed by
public vote in the November 1984 general election. This
amendment increased the State sales tax by 0.1 percent.
Taxation begins July 1, 1985, and will be in effect for five
years. The sales tax will generate approximately $30.5
million annually, to be divided equally between State parks
and historic sites, and soil conservation.
Missouri's Constitutional Amendment No. 2 is a partial
answer to solving the problem with funding nonpoint con-
trols related to soil conservation.
The Soil and Water Districts Commission proposed to
use 77 percent of its $15 million annual share for direct
financial assistance to landowners; 19.7 percent for tech-
nical planning and clerical expenses at the county level;
and 3.3 percent for program administration and State of-
fice personnel. This paper describes how the soil protec-
tion revenues will be used.
MISSOURI SOIL AND WATER
CONSERVATION COST-SHARE
PROGRAM
The Missouri cost share program equals 50.8 percent of
amendment revenues for soils.
Farmers realize the long-term benefits of soil and water
conservation. In the short term, however, the costs out-
weigh the profits. Through the cost-share program, the
public directly assists the farmer and his conservation ef-
forts. The long-term benefit for the public is plentiful food
at reasonable prices.
It has been estimated that $250,000,000 of cost-sharing
funds are needed by the end of the century to protect
Missouri's topsoil. The U.S. Department of Agriculture's
Agricultural Stabilization and Conservation Service
(ASCS), through its agricultural conservation program,
supplies approximately $8 million annually to its cost-
share program. In addition, State funding of at least $8
million per year is needed for an effective operation. The
amendment will fund the state's cost-share program at
approximately this level for 5 years.
Through the cost-share program, the farmer pays ap-
proximately half of the installation costs and the state pays
the rest. Conservation programs eligible under the cost-
share program include terracing, conservation tillage,
strip cropping, and other proven soil and water conserva-
tion techniques. The intent of this portion of amendment
revenues is to make more funds available to the farmer as
an incentive to install soil and water conservation prac-
tices.
SMALL WATERSHED PROTECTION AND
FLOOD PREVENTION PROGRAM
This program equals 13.3 percent of amendment reve-
nues for soils.
Water that does not evaporate or soak into the soil usu-
ally drains off the land into ditches, streams, marshes, or
lakes. The area drained by a stream makes up a water-
shed. Watersheds sometimes can be complex, such as
when land is drained by small streams that flow into a
larger stream. Because several different properties may
be involved, a cooperative watershed program among
neighbors is very important for soil and water conserva-
tion. This also explains why more than one conservation
measure within a watershed is necessary for best results.
More than 100 such watersheds have been designated for
planning in Missouri.
The watershed protection and flood prevention program
does more than conserve soil and water. It also keeps
sediment from entering streams and lakes; this sediment
can reduce the volume of the lake or interfere with fishing.
New revenues will be used to accelerate the watershed
program by funding several completed watershed plans.
Money will be available for cost-sharing assistance to
69
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
landowners for soil and water conservation projects within
selected watersheds. These projects include terraces and
strip cropping to help water soak into the soil instead of
running off, and small dams to hold back runoff water that
otherwise would cause flooding.
MISSOURI SOIL AND WATER
CONSERVATION LOAN INTEREST-SHARE
PROGRAM
This program equals 13.3 percent of amendment reve-
nues for soils.
Many farmers feel they cannot install conservation
practices because of cash-flow problems, and they cannot
borrow money because of high interest rates. The inter-
est-share program provides financial incentives to land-
owners who are conserving soil without the benefit of
other available programs.
Amendment No. 2 establishes a permanent fund to
serve as a financial base for reduced-interest loans. It
provides further incentive to landowners to install soil and
water conservation measures.
WATER QUALITY ASPECTS OF
AMENDMENT NO. 2
While 77 percent of the anticipated $15 million will be for
direct assistance to landowners, only 13.3 percent (about
$2 million per year) will be available for water quality re-
lated land treatments through the small watershed protec-
tion and flood prevention program. This situation makes
targeting extremely important if the State wishes to
achieve measurable water quality improvements.
Therefore, applications will be requested from water-
shed districts and evaluated on the basis of percent land-
owner participation, likelihood of success, potential for
water quality improvement or protection and other factors.
One key drawback may involve lack of interest in areas
that have good potential for environmental improvement.
While Missouri DNR's Water Pollution Control Program
has identified numerous areas where protection or im-
provement is desired, the watershed districts are gov-
erned by a board of supervisors who operate indepen-
dently. Similarly, problem areas may exist because
landowners want to operate independently of government
assistance programs.
Initial watershed protection areas will be identified in the
spring of 1985 and land treatment will begin shortly there-
after. Project monitoring will be conducted prior to, during,
and following land treatment. Because of the difficulties
associated with quantifying runoff-transported pollutants,
monitoring efforts will focus on habitat quality index
changes and alterations in fish community structure. This
study should contribute to the not-well-understood rela-
tions between stream biota and land use activities.
70
-------
STATE OF MARYLAND NONPOINT SOURCE CONTROL
IMPLEMENTATION PROGRAM
KENNETH E. McELROY
MARIE C. HALKA
Maryland Office of Environmental Programs
Baltimore, Maryland
ABSTRACT
The State of Maryland has had a number of nonpoint
source control implementation programs dating back to
the late 60's and early 70's. Beginning in 1982, the re-
search findings of the Chesapeake Bay Program added
momentum. In January 1984, the General Assembly
added a great many new programs and modified some
existing programs. Not all these programs are outgrowths
of the Chesapeake Bay emphasis. Many of them pre-
ceded that event. Each program has a different political
and institutional situation from which it has been derived.
These 12 different programs will indicate the variety of
political situations in which support can be built for new
programs. This paper covers each program, how it came
into being, how it is institutionally implemented, what the
responsibilities at the Federal or State or local level are,
how it is financed, and what the State of Maryland has
done to date in implementing the program.
SEDIMENT CONTROL
The first program I would like to cover is sediment control.
Roy Benner of the State's Water Resources Administra-
tion has written a very lengthy article, "Urban Sediment
and Stormwater Control: the Maryland Experience," pub-
lished in the February 1985 Journal of Soil and Water
Conservation and from which much of this information
comes. The Maryland Attorney General's Office declared
sediment a pollutant on July 31, 1961. That ruling stated
that silt discharged into the waters of the State resulting
from stormwater runoff over land areas exposed from land
clearing or development operations was legally subject to
regulatory control by the State agency. It was largely the
result of this decision, and subsequent analysis of the
extent to which sediment contributed to the State's water
pollution problems, that led Maryland to enact the first
statewide erosion and sediment control legislation on
Earth Day, April 22, 1970. (Nat. Resour. Article, Title 8,
Subtitle 11, Annotated Code of Maryland.)
The major features of the 1970 sediment control legisla-
tion are:
1. No clearing, grading, or transporting of soil can take
place until the developer submits an erosion and sediment
control plan to the local soil conservation district for ap-
proval. The developer must specify that he will carry out
the plan. Only then is he granted a local grading or build-
ing permit.
2. Maryland's 23 counties and 151 municipalities are
required to adopt grading and sediment control ordi-
nances acceptable to the Water Resources Adminstration.
These ordinances include the necessary procedures and
provisions needed to implement and enforce the local sed-
iment control programs.
3. Exemptions from the law include agricultural land
management practices and, in some counties, construc-
tion of single family homes on lots larger than 2 acres.
4. The Water Resources Administration has leadership
for assisting local governments in conservation districts in
carrying out their responsibilities under the law. Moreover,
the Administration must review and approve all land clear-
ing or construction projects conducted on any State or
Federal property.
5. Penalties for not carrying out the provisions of the
law are deemed a criminal misdemeanor. Conviction sub-
jects one to a $5,000 fine, 1 year in prison, or both, for
each violation.
This understanding of the basis of the program is es-
sential to understand the changes made to it since 1970.
The State implemented the program with plan review and
enforcement at the local level, but provided no local fund-
ing. Primary emphasis was placed on the training and
effective use of existing agencies and staff.
After 10 years of experience with the program, three
deficiencies appeared to be at the root of most of the
inadequacies: lack of an administrative commitment to the
program, inadequate field inspection, and an inadequate
enforcement process.
Many local jurisdictions failed to commit themselves to
developing an effective erosion and sediment control pro-
gram for several reasons. Most of them do not have the
financial resources or personnel to administer the pro-
gram effectively. This is particularly true of small municipal
governments which are often run by an administrator or a
small clerical staff. Other local governments may have
had the financial resources to develop an effective pro-
gram but for various reasons did not devote sufficient ef-
fort to their erosion and sediment control program. When
local administrators failed to commit themselves to devel-
oping an effective program, the inspection and enforce-
ment efforts generally proved ineffective as well.
Evaluations of local program effectiveness throughout
the State have consistently indicated that erosion and sed-
imentation caused by mankind's acitivities are not being
effectively controlled, and that the best practical combina-
tion of procedures and people may not always be at the
local level. For this reason, in 1978, the General Assembly
amended the sediment control law to require that appli-
cants for erosion and sediment control plans certify that
any project engineer, superintendent, or foreman in
charge of on-site clearing must have attended a State
training program. This had been done previously on a
voluntary basis only.
The law was also amended in 1984 to add a civil penalty
as an alternative to a criminal sanction. The civil penalty is
a fine that i& double the cost of installing or maintaining
the controls as shown in the approved plan.
The most significant change made in 1984, however,
was to provide that, as of April 1, 1985, the State will
assume all inspection and enforcement of local erosion
and sediment control programs. A local jurisdiction may
request and be granted delegation of enforcement author-
ity by the State. In keeping with this shift in authority, about
20 new inspectors were added to the State staff of 14
inspectors. In March 1985, the Department of Natural Re-
sources granted sediment control inspection and enforce-
ment authority to eight counties and Baltimore City. The
71
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
authority was denied to eight other counties and the
Washington Suburban Sanitary Commission, which oper-
ates outside of Washington, D.C.
Although it is obviously too soon to predict the effective-
ness of the amended sediment control program, we are
hopeful that it will achieve its original goals and we will
have it very much back on track. It is an example of a
delegation to a local government which did not work and,
therefore, was taken back with more control and oversight
at the State level.
STORMWATER CONTROL
Maryland has two stormwater programs. One is a regula-
tory program requiring that a stormwater management
ordinance be adopted at the local level subject to State
criteria (Nat. Resour. Article, Title 8, Subtitle 11 A, Anno-
tated Code of Maryland). The second stormwater program
is an incentive grants program for demonstration projects
to show the effectiveness of urban stormwater practices.
In 1980, it became obvious to the State that 11 of our 23
counties had stormwater management ordinances that
contained many different provisions. We were concerned
about this not only from the developer's perspective of
having to comply with different requirements, but also
from the perspective of determining the most desirable
provisions to be put into regulations. Of particular concern
to us was the issue of whether or not to maintain as nearly
as possible natural runoff characteristics. This could be
accomplished by augmenting infiltration, by controlling
the release of development-related stormflow increases,
or both.
In 1981, regulating stormwater and its downstream im-
pacts was the subject of extensive oversight hearings by a
joint committee of our General Assembly. That commit-
tee's efforts led to the passage, in 1982, of the State
stormwater management law. The State's stormwater
management regulations represent a diversified approach
to controlling the hydrologic consequences of urban de-
velopment rather than simply focusing on controlling peak
flow. Consideration is given to volume reduction, low flow
augmentation, water quality control, and ecological pro-
tection.
Having learned from our sediment control experience,
the State enacted in 1984 a new grant program of $1.7
million to make startup money available to local jurisdic-
tions to implement their local stormwater management
programs. Local stormwater programs were to be in effect
by July 1, 1984. With the threat of a building permit ban,
most counties and about two-thirds of the municipalities
had adopted ordinances and received State approval by
that date.
As of January 1985, grant agreements for local
stormwater program development have been executed
and funds awarded in 13 counties and four municipalities.
The total amount of funds awarded out of the $1.7 million
was $870,000. The State's regulatory requirements for
local stormwater management programs are contained in
the Code of Maryland Regulations 08.05.05.
The second State program having to do with stormwa-
ter management provides State bond funds as an incen-
tive for demonstration projects using best management
practices in existing urban areas. These grants are pro-
vided as 75% State/25% local grants to local govern-
ments to ascertain the cost and effectiveness of methods
of solving stormwater runoff problems created by existing
development. New development is covered by the regula-
tory program previously described.
In 1984, the State authorized $1 million for grants to
local governments for demonstration projects. In addition,
the State authorized $750,000 of General Construction
Loan funds for retrofitting stormwater best management
practices at State facilities. In the first quarter of FY '85,
preliminary proposals were received from 12 local govern-
ments for demonstration stormwater control projects in
existing developed areas. Standards and criteria were
completed for the demonstration grant program and regu-
lations were promulgated, effective April 8, 1985. Eight
potential State projects have also been identified.
It is important to note that this program, in part, ad-
dresses a loophole in the overall abatement of stormwater
pollution. Although the new regulatory program will deal
with new development, thousands of acres of the State
that require best management practices are not subject to
a regulatory program. We are hopeful that the National
Urban Runoff Project reports prepared for the Washington
Metropolitan area and for the Baltimore Metropolitan area
will be of value to us in deciding the types of demonstra-
tion projects to fund. We are also hopeful that this financial
commitment of $1 million at the State level will encourage
local governments in the State to implement similar proj-
ects.
Finally, we have received $875,000 from the Federal
government for nonpoint source abatement projects in the
Chesapeake Bay drainage area in Maryland. Several of
these projects involve retrofitting stormwater facilities on
highways and in existing developed areas. This combina-
tion of a variety of funding sources with regulatory and
incentive programs allows us to more fully address control
of stormwater pollution from existing developed areas.
AGRICULTURAL RUNOFF CONTROL
One of our agricultural nonpoint source control programs
is agricultural cost-sharing. The history of how this pro-
gram came into being is interesting. In 1979, as an option
provided under section 208 of the Clean Water Act, the
State formulated, adopted, and submitted to the U.S. En-
vironmental Protection Agency a Water Quality Manage-
ment Program for the Control of Sediment and Animal
Waste from agricultural lands. This was adopted and ap-
proved by EPA as an applicable statewide nonpoint con-
trol program pursuant to section 208 (b)(4)(A).
We persuaded the agricultural community to support
this program, although not all of the cause and effect rela-
tionships of agricultural runoff affecting water quality and
living resources were well defined. Several decisions were
instrumental in gaining agriculture's support. First, we
asked the agricultural community to write the 208 agricul-
tural control plan. We provided the EPA and State pro-
gram format, and they provided the technical and institu-
tional details pertaining to determining priority problem
areas, best management practices, and implementation.
Second, we made a commitment to work with them to
secure funding for cost-sharing.
We began to succeed with our funding commitment
when, in what we believe to be a unique move, we se-
cured approval from our 1982 General Assembly to reallo-
cate $5 million of State sewerage facilities construction
grant bond funds to agricultural cost-sharing. In 1984, we
secured another $2 million of State bond funds for agricul-
tural cost-sharing. Also, in 1984, we secured approval of
an additional $1.4 million in State general funds to hire 42
new people to work in soil conservation districts to imple-
ment agricultural cost-sharing. With these approvals, we
felt we had kept our 1979 promise to the agricultural com-
munity to get funds to implement the agricultural 208 plan.
The purpose of our agricultural cost-sharing program is
to implement best management practices within priority
watershed areas that contribute the greatest amounts of
pollution. Our goal is to have conservation plans in place
72
-------
STATE NONPOINT SOURCE PROGRAMS
for the farms in these priority watershed areas within 5
years.
For FY '86, we received from the General Assembly an
additional appropriation of $5 million in bond funds for
cost-sharing. So, to date, we have secured approval of a
total of $12 million in State funds for agricultural cost-
sharing. Of this amount, we have obligated all of the initial
$5 million for 2,000 projects, of which 628 are now com-
pleted. We have estimated the total cost of installing best
management practices on all agricultural lands in the
State by the end of the century to be $90 million.
We are also proposing to use some of the Federal Ches-
apeake Bay implementation funds for FY '85 to install
agricultural BMP's in priority watersheds. So it is really a
combination of State bond funds and Federal funds that
we are using for agricultural cost-sharing.
In 1984, the Maryland Department of Agriculture, in co-
operation with our Office of Environmental Programs, de-
veloped and approved a report entitled Statewide Priority
Watersheds for the Potential Release of Agricultural Non-
point Phosphorus and Nitrogen. The report ranked all wa-
tershed segments that drain to the Chesapeake Bay in
order of their relative potential to release phosphorus and
nitrogen as a result of agricultural activities. Factors in-
cluded in the ranking of the watersheds were: (1) the in-
tensity of agricultural land use; (2) intensity of agricultural
cropping; (3) the amount of cropland under conventional
tillage; (4) the fraction of cropland on steep and erodable
or, for nitrogen, highly permeable soil; (5) the potential
intensity of animal waste application to cropland; and (6)
an estimate of the influence of topography upon phos-
phorus movement. In setting priorities, we met with our
Department of Natural Resources to learn where stressed
aquatic areas corresponded with critical agricultural ar-
eas.
The 42 new State-funded positions have been assigned
as technical teams to work in the soil conservation dis-
tricts serving the priority watersheds. These technical
teams in the districts are being supported by the Univer-
sity of Maryland for educational and demonstration activi-
ties.
In summary, at the present time, thousands of farmers
in Maryland are applying for available cost-sharing funds.
We are seeing a harmonious coming together of the Fed-
eral agricultural community, our State Soil Conservation
Committee, soil conservation districts, and water quality
agency staff, to get best management practices on farms.
An additional agricultural nonpoint source program pro-
vides for enforcement in problem areas. Enforcement
actions are taken against landowners when water pollu-
tion standards are clearly being violated and landowners
refuse to install best management practices. In these in-
stances, we work through the appropriate soil conserva-
tion district to try to get BMP's on the land as a voluntary
action. If the district runs into resistance, then the case is
refered to the water quality agency. We exercise our water
quality authority to bring the landowner into compliance.
This approach has been supported by our agricultural
community. They are making the utmost effort to achieve
voluntary compliance. We estimate taking approximately
30 enforcement actions in FY'86 and 80 in each succeed-
ing year with new staff.
AGRICULTURAL DRAINAGE CONTROL
Prior to 1984, the EPA Chesapeake Bay Study had docu-
mented agricultural runoff as one of the major sources of
nutrient and sediment inputs to the Bay. It was also ob-
served that several large agricultural drainage projects
were being planned, financed, and constructed by the
U.S. Soil Conservation Service and local public drainage
associations with little or no opportunity for State regula-
tion. We felt this process was not adequately protecting
the State's natural resources and water quality. Inspection
during project construction, as well as for periodic channel
maintenance, has traditionally been the responsibility of
the U.S. Soil Conservation Service.
For all of these reasons, we prepared legislation which
was enacted in 1984 to require that, prior to constructing
or reconstructing an agricultural drainage project, a local
public drainage association must develop a construction
operation and maintenance plan for approval by our De-
partment of Agriculture, with concurrent review by our De-
partments of Health and Mental Hygiene, and Natural Re-
sources. The statute also requires the three Departments
to jointly establish criteria for plan approval, including
standards for design, construction, operation and mainte-
nance of agricultural drainage projects. To protect against
sedimentation, flooding, nutrient runoff, and habitat loss,
inspection and enforcement of plan compliance is carried
out by the State. The legislation also provides a civil sanc-
tion for violations. Regulations to implement the statute
are undergoing final review now. We are hoping to add
additional staff to implement this program in future years.
SHORELINE EROSION CONTROL
The next program provdes for the abatement of shoreline
erosion around the Chesapeake Bay and its tributaries.
The shoreline erosion control program in the State before
1984 addressed only critical eroding areas and promoted
structural controls such as bulkheads and riprap. Less
critically eroding areas can be stabilized through less ex-
pensive vegetative means, using, wherever possible,
clean spoil from maintenance dredging of channels to re-
duce annual dredging costs. Maryland has 376 miles of
critically eroding areas (more than 2 feet per year of bank
loss) and 965 miles where erosion is less critical. In 1984,
we expanded the program to triple the current level of
abatement in critical areas. We established a two-pronged
nonstructural approach. One prong gave financial assist-
ance to private landowners in the form of 50/50 matching
grants. The second provided for State planning in con-
junction with dredging projects. In addition, the Shore Ero-
sion Control Loan of 1984 authorized $3 million for loans
to property owners to continue structural shore erosion
control.
To implement the program, operating funds of $300,000
were approved with a staff of five. The program is now
operating with projects being actively designed and con-
structed. To facilitate implementation of the program, a
number of workshops were held in the first year with the
State Soil Conservation Committee, Federal soil conser-
vation officials, and various county and regional agencies.
Some of the FY'84 Chesapeake Bay implementation
funds are also being used for nonstructural vegetative
measures to reduce shoreline erosion.
CRITICAL AREAS COMMISSION
The next nonpoint source program involved the creation
of the Chesapeake Bay Critical Area Commission pursu-
ant to legislation enacted by the 1984 General Assembly.
The purpose of creating the Commission was to establish
a State policy of protection, restoration, and enhancement
of the critical shoreline area surrounding the Bay and its
tributaries, to the head of tide. Through a State/local part-
nership, the Commission works to develop and adopt pro-
tection plans for the critical shoreline area. The ultimate
goal is to foster more sensitive development activities to
minimize damage to water quality, natural habitat, and
scenic values.
73
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
The shoreline areas of the Bay system are particularly
fragile environments very susceptible to being adversely
impacted by human activity. Pollutants associated with de-
velopment in these areas may reach waters of the Bay
and its tributaries in greater amounts than those associ-
ated with development in more inland areas. Before the
existence of the Commission, some local governments in
the State had established protection programs. However,
as of 1984, there was no uniform protection program
along the shoreline area. The Commission is now fully
operational. Regional public hearings have solicited pub-
lic comments on criteria for managing activities within the
critical area. The Commission operates with approxi-
mately $500,000 per year of State general funds. We have
high hopes that this nonpoint source program will be ex-
tremely effective over the long run in seeing that land in
the critical area around the Chesapeake Bay will be used
and managed to minimize water pollution.. It has suc-
ceeded in raising the consciousness of many of our
State's citizens to the important role their land plays in the
overall ecological cycle.
RETENTION OF EXISTING FORESTLAND
The purpose of the retention of the existing forestland
program is to maintain existing forest buffer areas around
the Bay and its tributaries to intercept surface runoff and
to infiltrate it to the forest soil profile before reaching the
water. The program consists of several stages: (1) defining
and mapping the critical land areas currently forested ad-
jacent to the Bay and its tributaries; (2) providing technical
assistance to landowners including the preparation of for-
est management plans; and (3) cooperating with local soil
conservation districts in developing forested buffers as
best management practices for agricultural land.
Approximately one-third of the land in Maryland's por-
tion of the Chesapeake Bay basin is currently forested. In
most cases, this land is subject to conversion to other less
protective land uses. Program implementation involves
foresters working with landowners in targeted areas
around the Bay and its tributaries.
No new legislation was required to implement this pro-
gram. Approximately $100,000 of State operating funds is
being used for four forester positions. In the first half of FY
'85, the new foresters developed five forest management
plans covering 365 acres. They are also using student
volunteers to compile the names and addresses of people
owning forestland within the critical areas.
CONSERVATION EASEMENTS
The State program of acquiring conservation easements
encourages private landowners to preserve and protect
undeveloped or low density areas along the shoreline of
the Chesapeake Bay and tributaries by executing ease-
ments pursuant to the existing Maryland Environmental
Trust -Easement Program. Easements offer landowners
the opportunity to make an individual contribution to pro-
tecting the Bay. Because they are permanent, the total
number of easements increases the amount of long-term
protection.
The Maryland Environmental Trust program was started
in 1974 to substantially increase the acreage placed under
easements through cooperative efforts of the Chesapeake
Bay Foundation. It was modified in 1984 to assist ease-
ment owners in identifying and putting into use conserva-
tion practices appropriate for their properties. A staff of
three and operating funds of $60,000 per year were appro-
priated for this program. Easements so far in 1985 amount
to about 2,000 acres, covering 3 miles of shoreline.
DREDGE AND FILL PROJECTS
Another nonpoint source initiative approved in 1984 was
an expansion of the State water quality certification pro-
gram pursuant to sections 401 and 404 of the Clean Water
Act. The Office of Environmental Programs is expected to
review approximately 2,000 construction projects each
year for which water quality certificates are required by
section 401. The Corps of Engineers may not issue a
section 404 dredge or fill permit unless a State water qual-
ity certificate is issued. Certification is a process through
which the State may ensure that certain conditions are
attached to 404 permits. The increased staff will be able to
review 250 to 300 permits per year and conduct 350 to
425 site visits per year related to these permits. This is an
example of using an existing Federal program and the
interest in the Chesapeake Bay to acquire the political
support and resources to perform the job more effectively.
NONTIDAL WETLANDS
A cooperative program is designed to protect non-tidal
wetlands with responsibilities shared by the State and
county governments. Maryland's non-tidal wetlands are
transitional environments existing as isolated entities or
between open waters and dry land. These wetlands pos-
sess many of the same values as tidal wetlands. They
have complex and extensive root systems that stabilize
stream banks, reduce the velocity of sediment laden wa-
ter, and trap sediments and pollutants contained in these
waters. They also provide wildlife habitat and food, partic-
ularly to waterfowl and fur-bearing animals. However, cur-
rent State law directly protects only tidal wetlands. Since
1973, Maryland has lost 14,150 acres of non-tidal wet-
lands. By comparison, only 250 acres of vegetated tidal
wetlands were filled with dredge material from 1971 to
1983.
The initiative relating to non-tidal wetlands did not in-
volve new legislation. Rather, it created funding of approxi-
mately $150,000 to: (1) encourage and assist local govern-
ments with the design and implementation of locally
administered non-tidal wetlands management programs;
(2) initiate a non-tidal wetlands resource assessment and
monitoring system that will provide for a quantitative anal-
ysis of wetlands types; and (3) establish criteria for soil
and water conservation plans to help maintain the integ-
rity of non-tidal wetlands systems.
The new State staff has prepared a handbook regarding
non-tidal wetlands protection and is preparing maps in
cooperation with the U.S. Fish and Wildlife Service. The
staff is expanding training programs and utilization of edu-
cational materials for the protection of non-tidal wetlands.
Staff members organized and recently conducted a Ches-
apeake Wetlands Conference.
MINING (NONCOAL)
Another nonpoint source program regulates surface min-
ing in the State. In 1975, the Maryland Surface Mining Act
(Nat. Resour. Article, Section 7-6A-01) was passed. This
law requires mitigation of the effects of land disturbance,
elimination of hazards to public safety, and prevention of
the waste of mineral resources. The law and regulations
allow only licensed operators to obtain surface mining per-
mits. To obtain a permit for a specific site, a detailed min-
ing and reclamation plan is required, indicating the steps
to be taken to minimize adverse environmental effects and
to restore the landscape. The law also requires that a
performance bond be deposited by the permittee. This
bond is released only after satisfactory fulfillment of all
permit conditions and completion of reclamation. In gen-
74
-------
era), industry compliance with this program has been
good.
A related program, funded by the Surface Mined Land
Reclamation Fund (Nat. Resour. Article, Section 7-6A-04),
provides for reclamation of existing abandoned mines and
pits. The fund receives money from surface mine permit
fees, forfeited bonds, and fines. In June 1981, we com-
pleted an inventory of abandoned mines. Priority sites are
now being reclaimed using the current accumulated fund
of approximately $800,000.
Failing Septic Systems
Maryland also controls on-site waste disposal systems.
State regulations specify that domestic sewage or sewage
effluent may not be disposed of in any manner that will
cause pollution of the ground surface, ground water, bath-
ing area, lake, pond, watercourse, or tidewater, or create a
nuisance (Comar 10.17.02). A permit must be obtained
from local health departments to on-site disposal systems.
STATE NONPOINT SOURCE PROGRAMS
In addition, a second regulation provides that subdivision
development may not be allowed where infiltration of indi-
vidual sewage system wastes might result in ground water
contamination (Comar 10.17.03). Violation of either regu-
lation brings a $100 fine each day on which a violation
occurs.
Presently, the State is considering adopting new regula-
tions that would greatly facilitate the use of innovative on-
site disposal systems. A demonstration project, using 201
construction grant funds, is testing a clustered mound sys-
tem on Maryland's Eastern Shore. This innovative system
is designed to serve more than one dwelling unit in a part
of the State in which conventional septic systems fre-
quently fail.
COALMINING
The State created a Land Reclamation Committee (Nat.
Resour. Article, Annotated Code of Maryland) some years
ago to regulate strip mining for coal in the western portion
of the State.
75
-------
THE WISCONSIN NONPOINT SOURCE PROGRAM
JOHN G. KONRAD
Wisconsin Department of Natural Resources
Madison, Wisconsin
Nonpoint sources are land areas where activities, includ-
ing land management, result in the transport of pollutants
or contaminants, generally by runoff water, to lakes,
streams, or ground water. Pollutants from point sources
usually are discharged directly to waterbodies in fairly
constant concentrations and amounts, whereas pollutants
from nonpoint sources may follow transport paths which
partially deposit them before they reach receiving waters.
The concentrations and volumes vary greatly by season
and year; therefore, nonpoint sources are usually more
difficult to identify, and produce chronic degradation of
water quality. Nonpoint source pollution problems also
vary greatly between geographic regions of the United
States and between individual States.
Water quality problems associated with organic and nu-
trient loads as well as sediment exist in areas of the United
States where livestock-based agriculture is prevalent. In
the Upper Midwest, nonpoint source pollutants from both
croplands and livestock operations have degraded many
surface water resources.
Since major portions of Wisconsin are in this critical
area, officials recognized years ago that fishable and
swimmable water quality will not be reached in many lakes
and streams unless an aggressive program for controlling
urban and rural nonpoint sources is pursued. The degra-
dation of smallmouth bass and trout fisheries, accelerated
eutrophication of inland lakes, and impaired water quality
of the nearshore waters of Lake Michigan are examples of
the water quality problems that require the control of both
point and nonpoint sources. Water resources such as
these are vital to Wisconsin's economy because of their
important recreational use.
The Wisconsin legislature recognized this need and re-
sponded in 1978 by creating and funding the Wisconsin
Nonpoint Source Water Pollution Abatement Program.
The program was tailored to the nonpoint source needs in
urban and rural areas of Wisconsin by incorporating as-
pects of various existing programs as well as devising new
approaches.
Overall responsibility for the Wisconsin nonpoint source
control program is assigned to the Wisconsin Department
of Natural Resources, which administers both resource
management and environmental protection (including wa-
ter quality) programs. Cities, villages, and counties are
assigned the responsibility for local implementation in
project areas. In rural areas, this framework is designed to
maximize local agency contact with individual landowners
and is based as much as possible on existing agencies
and institutions. In urban areas, this framework is de-
signed to maximize city and village involvement.
PROGRAM PURPOSE AND OBJECTIVES
The basic purpose of the program is to systematically con-
trol nonpoint source pollution so surface water and ground
water quality goals can be met within a reasonable time-
frame. The program is designed to deal with the varying
nature of nonpoint sources throughout the State. This in-
cludes sediments from croplands, construction sites,
streambanks and grazed woodlots, and nutrient loads
from barnyard runoff, cropland erosion, manure spread on
croplands, and runoff from city lawns and streets.
The three major program objectives are: (1) to identify
the most effective approach for achieving specific water
quality objectives, and to provide adequate financial and
technical assistance to landowners and operators to assist
in installing of approved nonpoint source control prac-
tices; (2) to coordinate nonpoint source pollution control
with other elements of the State's water quality program;
and (3) to focus limited technical and financial resources
in critical geographic areas.
The third objective warrants specific attention. Unlike
many erosion control programs, the Wisconsin program
(1) concentrates on entire hydrologic units rather than on
random or political boundaries; (2) deals with all urban
and rural categories of nonpoint sources rather than se-
lected categories; and (3) relies on systematic processes
to identify, rank, and select critical watersheds and por-
tions of watersheds to receive comprehensive attention.
Single source management programs will achieve
many onsite land management objectives and may
achieve some pollution control. However, these programs
often are of limited value in solving pollution problems
arising in larger hydrologic units because of their scat-
tered installation. The Wisconsin program concentrates
available funds for technical and educational support into
selected hydrologic units where maximum comprehensive
improvements in water quality can be achieved.
This hydrologic unit approach, called the Priority Water-
shed Approach, allows all categories of urban and rural
nonpoint sources within specific critical areas of a water-
shed to be identified and controlled through the installa-
tion of management practices. Specific areas within a wa-
tershed that contribute pollutants to lakes and streams are
collectively called Priority Management Areas.
In addition to identifying problems and sources, the pri-
ority watershed approach has proven an effective frame-
work for project implementation. Through Priority Water-
shed Projects, this approach concentrates available
educational, financial and technical resources in those
critical watersheds where maximum water quality benefits
will result from investing money and personnel. To date,
there are 26 Priority Watershed Projects in varying stages
from project development to final implementation. Each
project requires 1 year for identifying critical problem ar-
eas, 3 years for landowner signup, and 5 more years for
installing control practices.
PRIORITY WATERSHED PROJECT
OBJECTIVES AND CRITICAL SOURCE
IDENTIFICATION
Selection of a Priority Watershed Project is followed by an
8- to 9-year planning and implementation process. An im-
plementation plan is prepared based on a detailed inven-
tory and assessment of critical source areas in the water-
shed and the project's water quality objectives. Generally,
about 1 year is required to complete the assessment and
prepare the plan. The Priority Watershed Plan consoli-
dates water quality and land use information so the spe-
cific causes and critical areas contributing to the water
quality problem can be identified and the most practical
means of controlling the pollution can be developed. The
plan guides the Priority Watershed Project and details pro-
cedures and responsibilities to help local staff work more
76
-------
STATE NONPOINT SOURCE PROGRAMS
effectively. It can also be important educationally by show-
ing the cause and effect relationship between land man-
agement and water quality.
Central to each Priority Watershed Project are the water
quality objectives identified for its lakes and streams. The
determination of critical pollutants, significant sources, the
level of desired nonpoint source pollutant load reduction,
and the measurement of accomplishments are all based
on these specific water quality objectives. In addition, the
severity of water quality problems and the attainability of
water quality objectives are primary factors in selecting
projects.
Pollutant impacts on water resources must be under-
stood to determine water quality objectives. The objec-
tives must be based on potential use. However, with objec-
tives related to nonpoint sources, the type of impairment
rather than the numerical criteria commonly used for in-
stream standards is more important. Impairments such as
degraded fish habitat caused by sedimentation of the bot-
tom substrate, which commonly occurs in many of Wis-
consin's trout streams, do not relate well to numerical
standards. Identifying water quality problems and objec-
tives in Wisconsin depends, to a large degree, on biologi-
cal and physical techniques that relate to the type of im-
pairment and use. Reliance on chemical parameters
alone could easily result in many impaired uses being
overlooked. Biological indicators often integrate fluctua-
tions in chemical parameters and retain an overall mea-
sure of water quality impacts for a long period of time.
Identifying water quality objectives in this manner requires
the efforts of aquatic biologists and fish managers.
In Wisconsin, some of the water quality objectives iden-
tified for Priority Watershed Projects are: (1) protection of
the nearshore waters of Lake Michigan, (2) rehabilitation
of a warmwater fishery, (3) rehabilitation of a coldwater
fishery such as the upgrading of a trout stream through
habitat improvement, (4) protection of a desired warmwa-
ter fishery, (5) protection of a desired coldwater fishery, (6)
rehabilitation of an inland lake, and (7) protection of an
inland lake.
With the variety of dairy and cash crop farming and
urban land uses in Wisconsin, water quality problems are
seldom caused by a single type of nonpoint source. Thus,
a categorical approach, one that deals just with one cate-
gory of sources, such as eroding croplands, will not be
effective in controlling nonpoint source pollution. Con-
versely, involving all landowners is inefficient and not cost
effective because not all land management activities con-
tribute significantly to the water quality problems.
A comprehensive assessment of all nonpoint sources is
conducted prior to implementing a Priority Watershed
Project. Barnyards, fields where manure is spread, erod-
ing streambanks, eroding croplands, construction sites,
and existing urban areas are all inventoried. These inven-
tories enable more efficient use of time and money during
implementation. For example, 25 to 50 barnyards can be
inventoried in the time required to design and install barn-
yard runoff controls on one or two barnyards. Thus, sub-
stantial time and money are saved by not designing and
installing practices for barnyards that might have been
considered significant using less detailed or more subjec-
tive inventories.
DEVELOPMENT OF IMPLEMENTATION
STRATEGIES
An equally important and potentially overlooked program
aspect is design of the project and the detailed strategies
for implementation. Currently, implementation strategies
include detailed landowner contact lists based on the
results of watershed inventories. These lists are accompa-
nied by a preliminary assessment of the severity and ex-
tent of nonpoint sources for each operation on the lists.
Project implementation strategies also identify and sched-
ule educational activities, outline fiscal management pro-
cedures, discuss preliminary project budgets, and esti-
mate staff needs.
STATE BUDGET SUPPORT
Wisconsin provides financial support in three major cate-
gories: (1) cost-share funds for landowners and municipal-
ities to install management practices; (2) aids for local
governments to fund additional technical assistance, edu-
cation and information, and financial and project manage-
ment; and (3) administrative and planning funds for State
administration and preparation of Priority Watershed
Plans.
Individual management practices are cost shared at 50
to 70 percent of the installation cost. Higher cost-share
rates are used for practices where the capital costs for
installation are high and the offsite water quality benefits
exceed the landowner's onsite benefits. Since 1978, the
State has appropriated over $23 million to implement the
nonpoint source program. Over 80 percent of these funds
have been used to help landowners install control prac-
tices.
MANAGEMENT PRACTICE PACKAGE
APPROACH
Since effective management practices must operate as
systems, the Wisconsin cost-share agreements must con-
tain all management practices necessary to control non-
point sources on each participating farm or municipality.
The landowner or land manager may not limit participation
to the practices most directly useful. This approach is simi-
lar to the Experimental Rural Clean Water Program, but is
quite different from that of the traditional Agricultural Con-
servation Program. Many installed practices and non-
structural controls would not be applied without the sys-
tems package requirement.
ACCOMPLISHMENT TRACKING
Wisconsin's program also includes progress or accom-
plishment tracking. Accomplishment indicators have been
used to some degree in all projects and are being used to
a greater degree in new projects. The accomplishment
indicators used: (1) relate directly to the water quality ob-
jectives and the pollutants causing the problems, (2) relate
to the type and significance of the sources to be con-
trolled, so that pollutant load reductions can be calculated,
(3) provide feedback to the implementing governmental
unit so progress can be determined on a frequent basis,
and (4) provide sufficient detail on the location and level of
control to guide and interpret monitoring results.
SUMMARY
Although participation by landowners and operators is vol-
untary in this State funded program, substantial pollutant
load reductions have been achieved in Priority Watershed
Projects. However, no voluntary program will achieve the
desired levels of control in all situations. In those cases,
regulatory mechanisms must be considered.
The elements of the Wisconsin program are designed to
effectively and efficiently achieve water quality objectives
impaired by nonpoint source ponutants. These program
77
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
elements, along with the experiences gained during the lures. However, the principles used as the foundation for
past 6 years, have resulted in a program structure that is the Wisconsin nonpoint source control program can be
well defined and adaptable to changing needs. Different applied to developing effective programs to control a vari-
areas have different needs and existing institutional struc- ety of nonpoint source problems in any State.
78
-------
Institutional/Financial
Aspects of Nonpoint
Source Controls
BRIDGING THE GAP BETWEEN WATER QUALITY AND NONPOINT
SOURCE ACTIVITIES: A CONTINUUM OF INSTITUTIONAL
ARRANGEMENTS
BART HAGUE
U.S. Environmental Protection Agency
Boston, Massachusetts
ABSTRACT
Successful nonpoint source control implementation re-
quires devising institutional/legal arrangements to draw
formally the various interests and agencies responsible
for Best Management Practices (BMP's) into the water
quality management process; yet, at the same time, for-
malize the role of the State Water Quality Management
(WQM) agencies. EPA and the New England States have
established an array of mechanisms by which State WQM
agencies formally involve the State forester, timber indus-
try, construction industry, and agricultural interests in ac-
tivities ranging from technical assistance on BMP certifi-
cation and plan review to limited inspection.
Corresponding mechanisms for backup enforcement by
the State WQM agency, Attorney General, and EPA vary.
This presentation outlines and evaluates the experience
with these mechanisms over the past 5 years, suggesting
improvements, refinements, or new mechanisms for the
future. The evaluation covers a continuum of measures
from the voluntary to backup enforcement, and from pri-
vate to public responsibilities.
8NTRODUCTION
Nonpoint sources (NFS) by definition are diffuse, wide-
spread, and subtle. Their control touches the daily lives of
countless individuals, groups, and enterprises. Best man-
agement practices (BMP's) to control nonpoint pollution
must become integrated in diverse activities through a mix
of informal and formal, or voluntary and contractual rela-
tionships—sometimes labeled nonregulatory and regula-
tory.
Nonpoint source programs include both formal arid in-
formal steps: formal standards setting, informal education
and technical assistance, followed by formal inspection
and enforcement. First, Federal/State water quality man-
agement agencies must formally adopt water quality
standards and criteria under the water pollution control
laws prohibiting discharge of pollutants into the waters of
the Nation.
Then, informal, voluntary education, technical assist-
ance, inspection, and self-policing programs may help
people adapt their activities to prevent or reduce NFS
pollution. Informal education and technical assistance ef-
forts must reach the farmer in the field, the logger in the
woods and the builder on the back lot. Often, the most
effective way of reaching them is through associates who
share their interests, professional knowledge, or commu-
nity values.
The final step, formal inspection and regulation, must
be waiting in the wings as backup. Here, the State Water
Resources investigator and the compliance officer be-
come involved. If violations persist, the attorney general
may prosecute. Finally, the responsible Federal or State
agency must evaluate the effectiveness of the informal
arrangements in carrying out the formal mandates of the
laws, standards and prescribed BMP's.
A Continuum of Formal and Informal
Arrangements for Water Quality BMP's
For successful nonpoint source control, the institutional
arrangements must draw those affected by the controls
into the formal water quality management process. The
formal environmental agency objectives, standards, and
BMP's must be incorporated into diverse economic activi-
ties. Usually, water quality objectives and BMP's can be
best integrated into these activities through informal ar-
rangements involving fellow workers or professionals,
friends, or neighbors in whom the operator places per-
sonal confidence and trust. At the same time, the formal
mandate for the public interest must be met. Social scien-
tists have developed a body of theory on the role of infor-
mal and formal groups in the adoption of new practices
(Homans, 1950; Spicer, 1952; Wilkening, 1950).
79
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
This presentation outlines and evaluates experience
with informal, voluntary arrangements for providing edu-
cation and technical assistance and, in some cases, in-
spection and compliance to carry out the mandate of the
formal Federal/State standards. It suggests improve-
ments, refinements, and conditions for success of various
approaches.
Though programs are labeled regulatory and nonregu-
latory, in practice nonpoint source programs entail a series
of formal and informal steps using a range of approaches:
formal for setting standards; informal for education, tech-
nical assistance, and initial implementation; and formal for
backup enforcement if informal efforts fail.
Informal arrangements are most appropriate for activi-
ties that are dispersed, intimately associated with family
operations, voluntary in contrast to contractual, and rural
in contrast to urban. Generally, activities rank in order from
the informal/voluntary to the formal/contractual in the fol-
lowing order: agriculture, forestry, on-site waste disposal,
individual home construction, oil and hazardous materials
handling, and large-scale construction.
The major types of institutional arrangements bringing
formal programs closer to the people are:
1. VoOumtoiry associations. Enlist voluntary associa-
tions of the industry or activity that the formal agency is
trying to reach, e.g., trade associations and lake or water-
shed associations.
2. PVoftessJoirtsD affiliation. Enlist fellow professionals.
They can be presumed to be more knowledgeable and
understanding of one's activities and problems, even if
they have formal regulatory responsibilities.
3. State programs with local option. Involve locali-
ties, regional agencies or district. Local governments are
perceived to be more responsive to local needs and activi-
ties than State or Federal agencies.
The cases to be evaluated cover a continuum of measures
from the informal/voluntary to the formal/contractual,
demonstrating both private and public responsibilities.
The cases range, in order of increasing formality, from the
Vermont Timber Harvesters and Truckers Association self-
pblicing program, New Hampshire regulations on earth-
disturbing construction and forestry activities, and Massa-
chusetts Minimum Forest Cutting Practices Regulations,
to Maine and Vermont Statewide Environmental Laws with
Local Option.
Vermont Timber Truckers and Producers
Association
To help implement BMP's for forest practices recom-
mended in the Vermont Water Quality Management Plan,
5 years ago the Vermont Timber Truckers and Producers
Association set up a Committee to provide education and
technical assistance and to investigate complaints. The
Association, made up of over 200 loggers, truckers, mill
owners and operators, landowners, and professional for-
esters, reflects the logger and his values, reaching out to
him through his own peers. (Vt. Timber Truck. Prod. Assn.,
1984).
Initially, the Vermont Timber Truckers and Producers
Association, the Vermont Agency for Environmental Con-
servation, Cooperative Extension Service, and Soil Con-
servation Service jointly prepared a pocket handbook,
Guide for Controlling Soil Erosion and Water Pollution on
Logging Jobs in Vermont, and conducted workshops with
loggers throughout the State (Vt. Agency Environ. Con-
serv., 1979). The booklet and workshops were funded by a
grant from the U.S. Environmental Protection Agency.
To follow up, the Vermont Agency of Environmental
Conservation refers complaints about logging jobs pollut-
ing streams and lakes to the Vermont Timber Truckers and
Producers Association. A local Association committee
member visits the site with the logger to investigate the
complaint. If there is a problem, the committee member
encourages the logger to apply the appropriate BMP's. If
the logger does not voluntarily comply, the case is referred
to a State Water Resource investigator for formal investi-
gation, technical assistance and possible legal action. A
violator risks having his job shut down and may be subject
to fines of up to $25,000 a day under the laws governing
turbidity and discharge of pollutants.
The process for registering complaints, followup, and
reporting results is formally spelled out for the public re-
cord. The steps are clearly outlined on a Department of
Water Resources form: location, nature and source of
complaint; investigation and followup action by an Associ-
ation committee member; results of reinspection; and
action taken in case of noncompliance. Figure 1 is a copy
of the Complaint Record Memo.
Two years ago, the Association, Agency of Environmen-
tal Conservation and other parties held a workshop to
review the progress of the program, to emphasize the
continuing mission, and to motivate those involved.
The program's success can be measured by the volun-
tary adoption of BMP's as a routine part of logging opera-
tions and by the decline in complaints. Settling basins are
now installed in the course of clear-cutting. A major paper
company requires filter strips and water bars as part of the
job, holding back $1.00 a cord in payment to the logger
until BMP's have proved successful. The volume of com-
plaints has fallen nearly 75 percent since the beginning of
the program 5 years ago. Only one problem has been
referred to the Attorney General. This decline in com-
plaints occurred during a period of increased logging, in-
creased clear-cutting, and heightened concern over water
quality. The State's Water Resources investigators find
that involvement of fellow loggers in education and en-
forcement encouraged adoption of BMP's. They are satis-
fied that adequate BMP's have been chosen and imple-
mented.
New Hampshire Statewide Erosion and
Sediment Control Program
The New Hampshire Water Supply and Pollution Control
Commission amended its dredge and fill regulations un-
der the Water Pollution Control Statutes (RSA 149: 8-a) on
April 18, 1981, to require permits for timber harvesting
and construction activities that significantly alter the ter-
rain or affect water quality (N.H. Water Supply Pollut. Con-
trol Comm., 1982). Anyone undertaking earth-disturbing
activities must obtain a permit from the Commission for
commercial logging or for residential or commercial con-
struction affecting over 100,000 square feet in or adjacent
to surface waters.
Under the forestry permit, an operator acknowledges
familiarity with and agrees to apply BMP's such as those
outlined in New Hampshire's pocket handbook, Timber
Harvesting Practices for Controlling Erosion (N.H. Water
Supply Pollut. Control Comm., 1979). State forest rangers
advise operators on these practices. If voluntary efforts
fail and complaints are registered, the Water Supply and
Pollution Control Commission investigates and issues
cease and desist orders. The Commission devotes the
equivalent of one full-time person to inspection and en-
forcement. As many as one or two cease and desist orders
are issued per week.
Before the program began operating (1980-83), com-
plaints averaged five a week, but now have fallen to two a
week. Of these, approximately 60 percent are resolved at
or near initial contact.
80
-------
INSTITUTIONAL/FINANCIAL ASPECTS OF NONPOINT SOURCE CONTROLS
Department of Water Resources
COMPLAINT RECORD MEMO FOR LOGGING JOBS
To Be Completed by Water Resources Investigator.
A.M.
Report No. Date Rec , 19 Time P.M.
Complaint by Address
Phone
Received by Title
A.M.
Screened by Date , 19 Time P.M.
Nature of Complaint:
D Siltation of D Stream d Pond D Other
D Tree Tops in Stream
D Skidding through Active Stream
D Other
Explain: :
To be Completed by Vermont Timber Truckers and Producers Association Committee /Member
Person Investigated—
Name Phone
Address
Date Contacted , 19 Committee Member
Exact Location of Log Job
Cause of Complaint
Recommended Action to Eliminate Problem:
D Install water bags or dips Q Relocate landing
D Mulch or seed landings or skid paths D Remove treetops or brush from stream
D Bridge stream D
Reinspected by Date
Water Quality Problem Eliminated D Yes No D
Signature of Committee Member
NON COMPLIANCE ONLY
Turned over to Water Resources Investigator Name
Date , 19 :..
Action Taken
Figure 1 .—Complaint form used in Vermont was developed by a committee from the Vermont Timber Truckers and Pro-
ducers Association. This self-policing program represents high cooperation among private industry, State, and Federal
personnel.
81
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Unlike Vermont, New Hampshire has not formalized or
publicized procedures for referring violations from the
State forester to the Water Supply and Pollution Commis-
sion for investigation and enforcement. Fewer than 20 per-
cent of the complaints were made after encouragement by
the foresters; approximately 80 percent came directly
from the public. Some foresters have been more involved
that others, as would be expected. In addition, the Coop-
erative Extension Foresters have limited their role to edu-
cational activities, avoiding involvement with the regula-
tory activities of the Commission. Without formal
procedures, the State and Extension foresters may not
feel comfortable taking action or reporting violations to the
Commission. Split authority makes "buck passing" a con-
cern.
Under a construction permit, an applicant submits a
plan for erosion/sediment control measures, runoff treat-
ment, and flood management. Drawing on the Durham,
New Hampshire, National Urban Runoff Project (U.S. En-
viron. Prot. Agency, 1984), the Commission has worked
with builders and designers to develop design criteria for
swales and vegetative surfaces to absorb runoff. Commis-
sion staff site visits may lead to redesign or subsequently
to enforcement. All applications are ultimately approved,
with conditions. The Commission staff feels that appli-
cants prefer to negotiate on reasonable conditions rather
than incur costs in delays in approval. Only a couple of
permit violations have been reported. A very small per-
centage (5 to 10 percent) of construction activities are
undertaken without a permit, according to staff.
Workload and staff limitations preclude much attention
to followup visits and compliance monitoring. An average
of two applications per day are received with as many as
five per day during the peak construction season. Aside
from a special coastal officer, the Commission can devote
the equivalent of only one staff-year to the program
throughout the State, including the rapidly developing
south central Interstate 93 corridor. Efforts are limited to
reviewing applications, with limited site evaluation. Site
visits average three to five per week. Few followup visits
for compliance monitoring take place. Coastal Zone Man-
agement grants provide an additional officer to serve the
seven coastal towns. This officer is able to work more
closely with the towns and applicants. The officer can visit
each site and conduct followup inspections. On the basis
of the success demonstrated in the coastal towns, two
officers would be added to followup compliance through-
out the State if funds were available.
No cases have been referred to the attorney general
during the past few years because of the State's enforce-
ment priorities.
Massachusetts Minimum Forest Cutting
Practices Regulations
Massachusetts has recently undergone a metamorphosis
from a rarely enforced formal law on the books toward
regulations perceived as more practical, more informal,
and in most parties' mutual interest. On January 1,1984,
the Massachusetts Division of Forests and Parks adopted
new Minimum Forest Cutting Practices Regulations (Ma.
State Forest. Comm., 1984) requiring operators to file a
cutting plan. The plan includes BMP's to protect water
quality, such as filter strips and road and skid trail stan-
dards. The State foresters and wardens review and ap-
prove the plans and follow up on compliance. Local con-
servation commissions can notify the State foresters of
concerns and violations.
This type of regulatory program only recently gained
acceptance as a realistic way to implement a long-dor-
mant law, on the books since 1943, requiring forest plans
and cutting permits. Several mutual interests converged
to support the change. The timber industry sought mini-
mum standards applicable to all operators to assure equi-
table competition in the face of alleged fly-by-night opera-
tors who would bid high, disregard cutting standards, and
leave landowners dissatisfied with harvesting timber. Sev-
eral towns had recently adopted their own individual regu-
lations, setting a trend toward crazy quilt regulation. For
the first time, new State Wetlands Protection Regulations
specified minimum cutting practices, but exempted an op-
erator from the more lengthy and complex wetlands regu-
latory process if he had a State-approved forest cutting
plan (Ma. Dep. of Environ. Qual. Eng., 1983). Loggers felt
more comfortable in dealing with State foresters than with
local conservation commissioners.
Despite past opposition to this regulatory scheme, the
various parties now express satisfaction. Though site vis-
its are mandatory only for wetlands or steep slopes, the
foresters or wardens have actually been visiting most
sites, educating loggers on BMP's. Landowners and log-
gers are just learning of the new regulations so considera-
ble cutting is taking place without plans. The importance
of publicity that actually reaches landowners and loggers
cannot be overstated.
State Programs with Local Option
Maine and Vermont have statewide minimum standards
for new development. Maine gives localities the option to
administer their own programs, while Vermont delegates
administration to nine districts, retaining a greater degree
of State control.
Maine's stoteraidte environmental Daws, notably the
Shoreland Zoning Act and the Site Location Act, provide
statewide minimum standards and a framework within
which localities can play as formal or informal a role as
they choose (Maine State Plann. Off., 1984). Effectiveness
depends on the degree of local initiative, the dedication of
resources, and, above all, the will to exercise persuasion,
approval/denial, and enforcement. Communities have the
opportunity to adapt laws to local conditions, but, by the
same token, they can remain passive participants in a
local network of intergroup and personal relationships that
condone lax practices and violations.
The role of local code enforcement officials is being
formalized so that the responsible individual acquires a
sense of professionalism and an official role beyond the
network of local, often familial, relationships.
For over 10 years, State trained and certified evaluators
have determined the suitability of sites for septic systems.
Certification has formalized their role and set public ex-
pectations that they will follow the law. In 1984, the Maine
Legislature considered requiring local code enforcement
officers to become certified through training. Although the
requirement did not pass in its entirety, certification is now
a prerequisite to presenting cases in court. This eliminates
the extra expense of hiring special legal counsel, giving
towns a financial incentive to train their code enforcement
officers.
Vermont's stotewfide land use and development Daw,
Acft 250 sets up the most systematic formal statewide
framework for regulating land use activities (Vt. Environ.
Board, 1982). A State Environmental Board sets policy
and hears appeals. Nine District Environmental Boards
review and pass on permit applications, including all for-
estry, construction, and earth-disturbing activities above
2,500 feet elevation. Although the District Boards are ap-
pointed by the Governor, they try to involve localities and
bring education, technical assistance, and regulation
closer to the people. Districts vary in their handling of
environmental issues—a problem associated with some
informal approaches.
82
-------
INSTITUTIONAL/FINANCIAL ASPECTS OF NONPOINT SOURCE CONTROLS
Evaluation: Conditions for Strengths &
Weaknesses
Reviewing strengths and weaknesses of these ap-
proaches can help tailor voluntary approaches for differ-
ent situations.
Voluntary Associations
Voluntary associations offer the following advantages:
1. Because of close relationships and trust, often,
peers can best arouse concern about water quality and
suggest controls.
2. Fellow workers may be able to tailor effective yet
acceptable controls.
Disadvantages are as follows:
1. Fellow workers may find it difficult to criticize or take
exception to their peers' operations or practices.
2. Fellow workers may be unduly influenced by the in-
terests of the operator or by personal relationships. Peer
pressure cuts two ways.
3. A recalcitrant operator may not respect or accept his
peers' advice. He may seek the authority of an official
agency
4. The public may not know how to refer or follow up on
complaints.
Professional Affiliation
Professional affiliation offers the following advantages:
1. Fellow professionals respect one another.
2. Operators look to professionals in their field of activ-
ity for information and advice.
3. Professionals can often prescribe the most effective
and acceptable BMP's tailored to the situation.
Disadvantages are as follows:
1. Fellow professionals may be more concerned about
the economic interests of the operator than about water
quality.
2. Professionals may be set in conventional ways of
doing their business, closing out consideration of some
BMP's.
3. Professionals in one sector may be reluctant to refer
failures to those in another, especially to regulators.
4. The public may not understand that the formal
agency has a responsibility when the professional does
not secure compliance.
Statewide Programs with Local Option
Statewide programs with local option have the following
advantages:
1. State standards assure a minimum program
throughout the State.
2. Mandatory provisions provide an incentive for locali-
ties to enact laws and develop programs.
3. Localities can tailor the programs to local situations,
including special concerns and needs.
4. Localities can adopt higher standards than the state-
wide minimum.
5. Localities can informally and formally keep in closer
touch with activities, problems, and violations than can
distant, limited State agency staff.
Disadvantages are as follows:
1. Statewide minimum standards can reduce local
standards to the lowest common denominator.
2. Local officials can remain bound to local interests
rather than broader public environmental interests.
3. Localities may lack the resources or expertise.
4. States and localities may pass the buck, each feeling
the other should act or take the heat.
CONCLUSION: CONDITIONS FOR
TAILORING EFFECTIVE FORMAL AND
INFORMAL INSTITUTIONAL
ARRANGEMENTS
With all informal/voluntary institutional arrangements, the
greatest problem is involvement in and loyalty to the sys-
tem, rather than to environmental quality. The formal envi-
ronmental agency must clearly define ultimate responsibil-
ity under its mandate. The public must have a clear
understanding pf the reciprocal responsibilities of the for-
mal public agency and the informal arrangements. If the
informal arrangement fails, the public must know its rights
and procedures for referral and followup action by the
public agency.
Informal/voluntary arrangements appear to work most
effectively when:
1. The voluntary association or group depends on envi-
ronmental quality for its continued livelihood or cares in-
tensely about the environment in its value system;
2. The voluntary association has a stake in maintaining
minimum standards that eliminate unfair competition and
insure equity. Fly-by-night operators using short cuts lower
potential competitors' costs.
3. Professional loyalties and standards transcend indi-
vidual or local interests. For example, loggers, timberland
owners and consulting foresters and engineers perceive a
professional bond with State foresters. They accept their
advice, usually voluntarily. Even though the public forester
is a regulator, he is also a fellow professional.
4. Professionals in the operator's field have specialized
knowledge of BMP's tailored to his activity. The operator
perceives that they have this expertise.
5. Local officials have status so that fellow citizens ex-
pect them to transcend the local web of personal relation-
ships and loyalties. The community has come to expect
the site evaluator, for example, to follow the law.
6. The State or Federal environmental agency formally
states the law, standards, criteria and procedures within
which the voluntary association or professional is to oper-
ate. Roles and responsibilities are clearly defined.
7. Staff of the public agency and of the informal associ-
ation, profession, or locality cooperate in a relationship of
mutual trust and concern for the environment.
8. The landowner and operator are fully aware of their
responsibilities to submit plan applications and carry out
BMP's. Education programs are tailored to reach all land-
owners and operators.
9. The role of associations, professionals, or local offi-
cials in environmental programs is clear not only to the
public agency and to the responsible group, but also to
their respective constituencies and the public. Support is
essential to their public interest role. It lets the public know
what is expected of the group. It makes them accountable.
Further, it lets the public know what specific remedies are
available should voluntary action fail.
10. Referral procedures are agreed upon, specified
and widely publicized. In several cases, the public was not
aware that the environmental agency would take action if
the professional voluntary association failed to act.
11. Backup enforcement by the environmental agency
is certain and prompt. Demonstrated investigation and en-
forcement action encourages voluntary BMP's. If enforce-
ment standards are unclear and enforcement inconsis-
tent, the voluntary program loses credibility. Violations
persist, requiring more agency staff time.
12. The formal agencies and parties to a voluntary pro-
gram meet periodically to evaluate progress, refine the
program, and reaffirm their responsibilities. Continuing
publicity is essential.
83
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
REFERENCES
Homans, G.C. 1950. The Human Group. Harcourt, Brace and
Co., New York.
Maine State Planning Off. 1984. Maine Planning and Land Use
Laws (excerpts from Titles 1,12, 30 and 38 of Maine Revised
Statutes Annotated). Augusta.
Massachusetts Division of Forest and Parks. 1984. Report of the
State Forestry Committee on Minimum Forest Cutting Prac-
tices Regulations. Boston.
Massachusetts Department of Environmental Quality Engineer-
ing. 1983. Wetlands Protection. 310 CMR (Common. Mass.
Reg.) 10.00. Boston.
New Hampshire Water Supply and Pollution Control Commis-
sion. 1982. Rules and Regulations Implementing RSA Chap-
ter 149:8-a Relative to the Prevention of Pollution from Dredg-
ing, Filling, Mining, Transporting Forest Products, or Other
Constructions. Concord.
1979. Timber Harvesting Practices for Controlling
Erosion. Concord.
Spicer, E.H. 1952. Human Problems in Technological Change.
Russell Sage Foundation, New York.
U.S. Environmental Protection Agency. 1984. Nationwide Urban
Runoff Program, Durham, New Hampshire, Urban Runoff Pro-
gram. Summary rep. Washigton, DC.
Vermont Agency of Environmental Conservation. 1979. Guides
for Controlling Soil Erosion and Water Pollution on Logging
Jobs in Vermont. Montpelier.
Vermont Environmental Board. 1982. Act 250: Vermont's Land
Use and Development Law (Annotated Title 10, Chapter 151
of Vermont Statutes). Rev., Montpelier.
Vermont Timber Truckers and Producers Association. 1984.
1984 Members. Wolcott, Vt.
Wilkening, E.A. 1950. A socio-psychological approach to the
study of the acceptance of innovations in farming. Rural So-
ciol. XV(4).
84
-------
THE UTAH AGRICULTURE RESOURCE DEVELOPMENT
LOAN PROGRAM
JAMES A. PARASKEVA
Utah Department of Agriculture
Salt Lake City, Utah
ABSTRACT
In 1983, the Utah Legislature provided $2.4 million for soil
and water conservation practices to the Utah Agricultural
Research Development Loan (ARDL) program. Loans are
made at a 3 percent interest rate with a one-time 4 per-
cent administrative fee and a maximum 12-year loan
length. The Utah Soil Conservation Commission adminis-
ters the program and local Soil Conservation Districts are
responsible for plan approval. After 2 years of operation,
over $11 million has been loaned to farmers for conserva-
tion work. The ARDL program is divided into three cate-
gories: (1) the regular ARDL program for soil and water
conservation practices; (2) the watershed program for
conservation and water quality practices in special tar-
geted areas; and (3) the emergency conservation pro-
gram. This program has been successful in implementing
conservation practices and improving water quality in
Utah. The program is a revolving fund loan and provides
operators with an incentive to install practices that benefit
the public at a low cost to the taxpayers. Utah is currently
the only State in the Nation operating a program of this
kind.
In 1976, the Utah legislature provided $250,000 and be-
gan the Rangeland Development Fund. Over the next sev-
eral years this fund continued to provide low interest loans
to applicants for making range improvements.
This fund was expanded in 1983 to $2.48 million to
include cropland conservation measures. This program is
under the direction of the Utah Soil Conservation Commis-
sion and staff support is provided by the Utah Department
of Agriculture. The Soil Conservation Service in Utah
agreed to provide technical assistance to begin conserva-
tion measures under the direction of the local soil conser-
vation districts.
The program was expanded because of Federal budget
cuts and a growing need for conservation in the State. The
Utah Soil Conservation Commission lobbied the legisla-
ture for a 20-year plan that would result in an $80 million
revolving loan program. Approximately half of the initial
request was met by the legislature and they have demon-
strated a continuing commitment by providing an addi-
tional $1.9 million in fiscal year 1985 and $2 million for FY
86. In addition to these appropriations, the legislature
chose the loan program as a vehicle to assist farmers
damaged by flooding, allocating an additional $3.6 million
for emergency measures.
The loans are available to all farmers and ranchers in
the State for use on private and State lands. Loans are
made at a 3 percent interest rate and carry a one-time 4
percent administrative fee. The maximum life of a loan is
12 years and conservation practices must be maintained
at operator's expense for the full life of the loan.
Early in the program it was recognized that local soil
conservation districts represent a valuable and underuti-
lized resource. These district supervisors are most aware
of the resource needs for their respective areas. Rather
than add to State staff for program administration, the
Commission turned to these local districts.
The districts pooled their resources through the Utah
Association of Conservation Districts and created a frame-
work to assist in the administration of the program. The
State is divided into six zones, each comprised of six or
seven districts. Loan funds are allocated to the zones by
the Commission based on resource needs as demon-
strated by loan applications received and annual plans
and reports. The zones then allocate funds to the local
districts. The districts are responsible for receiving and
processing applications, as well as approving plans and
monitoring projects. A local supervisor monitors each proj-
ect (Fig. 1).
To assist the zones and districts, the Utah Association
employed three regional coordinators. These coordinators
provide staff support for the loan program and other dis-
trict educational and resource activities. The State did not
increase its staff. The 4 percent administrative fee is dis-
tributed as follows: 1 percent to the State for program
administration; 1 percent to the district in which the loan
originates; and 2 percent to the Utah Association for the
regional coordinators.
The program's early success was due to two critical
factors. First, the program was decentralized and re-
source needs determined at the local level. This gives the
program grass roots support and uses the potential of
local districts as resource managers. The second critical
factor was the support of the Soil Conservation Service
(SCS). The State SCS and local officers totally supported
the program and agreed to provide technical assistance.
SCS participated in the development of program guide-
lines and is an ongoing partner.
Conservation practices eligible for funding under the
program are essentially the same as those eligible under
the Agriculture Stabilization and Conservation Service,
Agriculture Conservation Program (ACP). These practices
were adopted by the Commission with only slight modifi-
cations. It was felt that the broadest set of practices should
be made available for selection as local districts deter-
mine which activities are necessary and appropriate for
their areas.
In addition to the regular program, the Commission rec-
ognized that special needs may exist across the State. To
meet these needs, the Commission established the prior-
ity watershed program and energy conservation program
and made special funding set-asides. Later, the emer-
gency program was added to meet the needs of farmers
and ranchers damaged by flooding.
The ARDL watershed subprogram was set up to meet
special conservation needs in priority areas. Projects un-
vna owr. of KaicuLTun — BOIL OOHSOVKTIOM OCMOSBU
67* of Tot*l
5% of Tout
Figure 1.—ARDL fund allocation process.
85
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
der this program are designed to control water pollution,
erosion, or flooding. The strategy behind these projects is
to improve the entire watershed and to develop a coordi-
nated approach to watershed improvements. The guide-
lines require the Commission to designate priority water-
sheds and focus attention on these areas.
The Upper Weber and Upper Provo River drainages
provide most of the drinking water needs for residents of
the Wasatch Front (Utah's most populated area). These
rivers are also the main channels for heavy spring runoff
experienced along the Front. As the headwaters and
source of Utah's most important water resources, the high
watersheds have been recognized by the Utah Soil Con-
servation Commission, Utah Department of Health, and
Soil Conservation Service as being the most critical areas
for improvement in the State.
The Commission has established watershed funds for
use exclusively in these designated areas. To meet the
technical demands resulting from these projects, funds
have also been set aside from the watershed grant fund.
The purpose of these funds is to provide program coordi-
nation and assistance for project implementation within
the priority areas.
This designation and special funding is intended to pro-
vide many benefits within the priority area. Targeting will
offer the opportunity for extensive and coordinated use of
conservation measures. The priority areas have major
conservation, water quality, and flood control needs that
cannot be adequately addressed through the regular pro-
gram.
Early in 1984 the Utah Soil Conservation Commission
appointed a subcommittee known as the Priority Water-
shed Committee to look at these problems and to help
develop solutions. The Committee consists of representa-
tives from Wasatch, Summit, Morgan, and Kamas Valley
soil conservation districts.
The main goal of the Committee is to begin projects that
will meet the needs of the watershed and to facilitate other
projects. The Committee has met with several other agen-
cies and discussed developing joint projects. Wasatch,
Morgan, and Summit Counties have been involved in
streambank improvements and other watershed treat-
ments. In addition, the Weber Basin Water Conservancy
District has begun projects in this area. The Priority Water-
shed Committee is coordinating the efforts of many agen-
cies involved in watershed protection, providing funding
as well as guidance for priority projects. The Committee
has also assisted in providing technical assistance for
many watershed improvement projects.
The priority watershed program has gained the support
of the Soil Conservation Service, Utah Division of Wildlife
Resources, local counties, and State and local water qual-
ity agencies. The following projects have been approved
for funding:
Streambank improvements $ 75,500
Animal waste control $ 58,000
Range improvements $103,500
Irrigation water management for
water quality $218,900
$455,900
These projects have resulted in multiple benefits to the
watershed area. Water quality has been improved through
the control of animal wastes and sediments; streambank
erosion has been controlled by placement of riprap; and
streamside vegetation and runoff waters have been re-
duced through increased infiltration of water into improved
rangelands.
Perhaps most important, the Committee has provided a
mechanism for coordinated action to avoid duplication
and ensure that projects do not have a detrimental effect
on the environment or downstream users. This Committee
is beginning to become a force in the watershed area for
dealing with critical needs in a coordinated manner.
Through its contacts, the Committee will provide technical
assistance for projects, assist in obtaining permits, and
set priorities for program implementation.
The Commission set aside 5 percent of the total pro-
gram funding for energy conservation projects. This pro-
gram is administered by a special subcommittee with rep-
resentatives from Utah State University Extension
Service, Utah Energy Office, and Utah Power and Light.
Projects approved to date include conservation tillage, hy-
droelectric generation, and irrigation water management
programs.
In 1983, Utah experienced the devastation of a 100-year
flood. Although much of the reported damage occurred
within developed communities, farmers and ranchers
across the State suffered large losses. Most land in Utah
adjacent to streams is currently in agricultural use. Utah
Lake and the Great Salt Lake are swallowing large por-
tions of pasture and cropland. Again in 1984, flood dam-
aged many acres of quality agricultural lands. Thousands
of acres of crop and pastureland have literally been
washed away and many more acres have been covered
with water, gravel and sediment. Diversion structures, ca-
nals, irrigation systems, fences, and farm roads were de-
stroyed during these periods of high runoff.
The Utah Department of Agriculture documented over
$71 million in physical damages, crop, and livestock loss
during 1983. During 1984 the Department recorded nearly
$13 million in agricultural damage.
The Utah legislature provided $3.6 million in 1983 and
1984 in low interest loans to farmers and ranchers for
flood damage and prevention. These loans were used to
restore irrigation structures, diversions, level land, clear
debris, restore land fertility, rebuild fences and roads, sta-
bilize streambanks, and install measures to reduce the
risk of future flooding.
These emergency loans were channeled through the
Utah Department of Agriculture to the Utah Soil Conserva-
tion Commission. Local soil conservation districts as-
sessed and reported damage to the Department and
made requests for emergency money based on these as-
sessments.
The sum of $1,972,500 was loaned to repair irrigation
diversion systems, canals, laterals, fences, debris re-
moval, clearing, and releveling. Of that amount, $700,200
was dedicated to streambank protection and stabilization
and for measures to prevent or reduce the risk of damage
from future flooding.
SUMMARY AND CONCLUSIONS
The Utah ARDL program is still evolving. The State and
the Commission have only 2 years of experience with the
expanded program. Yet, early signs are positive. To date,
over $11 million has been put into conservation projects
across the State. These projects have protected soil and
water resources, improved water quality, and reduced the
risk of damage caused by flooding.
Perhaps the greatest achievement of this program is the
revival of the local soil conservation districts. These dis-
tricts, in their role as natural resource managers and water
quality management agencies, have great potential for
protecting and improving water quality.
The local districts have the support of area landowners,
are locally elected, and understand the problems of their
areas. Through the loan program the districts have a
meaningful function. They have been given a reason to
evaluate the resources in their areas and to set priorities
for implementation. Several districts have become in-
86
-------
INSTITUTIONAL/FINANCIAL ASPECTS OF NONPOINT SOURCE CONTROLS
volved with county planning agencies and are working
cooperatively on resource issues. The districts are gaining
an understanding of how they might affect the resource
base and many are undertaking broad programs to bene-
fit the land, water, and the people living there.
The Soil Conservation Commission is supporting the
development of the districts, seeing them as the alterna-
tive to Federal funding. Staff support has been provided to
the local areas and the loan program emphasizes local
control. Other State grant programs are being applied to
district programs and additional resources are being
sought from the legislature to support this program.
The response of landowners to the program has also
been positive. Because the program is a loan, some of the
reluctance to accept grants has been removed. Farmers
feel more responsible for the project, heightening their
sense of achievement. While there has been an over-
whelming response to the program and applications ex-
ceed available funds, some practices are still undersub-
scribed. Some of the soil conservation and water quality
practices with a low economic return, such as terraces or
animal waste control systems, do not receive much atten-
tion. These practices often require the additional incentive
of an ACP cost share used in conjunction with a loan. The
State set-asides are used to balance out the funds used
for any particular type of project.
The State has also tried to minimize the paperwork re-
quired for processing loans. State regulations are less
cumbersome than those for the Federal ACP; however,
some landowners are still reluctant to fill out the required
forms and many balk at the financial statements.
Overall, the program has succeeded in getting conser-
vation on the ground. There are administrative problems
in processing loans, and set-asides have not proven ex-
tremely successful in attracting desired projects. Cur-
rently, the Commission is exploring alternatives such as
varying the interest rate for different practices to encour-
age some desired applications. Many other changes are
due as the program matures, but the groundwork has
been laid for a successful, long-term program that will
enhance the natural resource base of the State of Utah.
ARDL APPLICATION PROCESS
I. First District Board Meeting
A. Applicant completes application form.
B. Soil Conservation District (SCO) Board reviews
application. Checks for completeness, preliminary indication of
credit made, and application screened to determine if request
complies with ARDL program.
C. Application will be approved or disapproved for
planning. The applicant will be notified in writing by the SCO
Board as to decision and given a financial statement form to fill
out and send to the Soil Conservation Commission (SCC).
D. Technical assistance is assigned by the SCO Board to
develop conservation plan for the approved applications.
E. SCO Board assigns a supervisor to track application
progress and planning.
II. Interim
A. Individual applicant sends financial statement and
supporting data as required on the financial data request form to
the Soil Conservation Commission (SCC) within 15 days.
B. SCO Board sends copy of completed application form to
the Zone Coordinator (ZC) and ZC in turn forwards application to
the SCC.
C. Technical assistance agency develops plan with the
individual.
D. SCC investigates applicant's credit and repayment
ability. Upon finding negative information, the SCC will notify the
SCS Field Office and SCO Supervisor and the applicant.
III. Second District Board Meeting
A. Completed conservation plan is presented by the
applicant to the full SCO Board for final approval and funding
(provided funds are available). SCO Board will notify applicant in
writing if final plan receives approval for funding, pending final
determination by the SCC. Work cannot begin on projects until
loan contracts are signed.
B. SCO Board sends copy of final plan to the ZC and ZC
forwards plan to the SCC.
IV. Post Project Approval
A. Security agreement and repayment schedule is
developed between the State and the individual. (Applicant will
be responsible for a portion of loan initiation fees beyond the 4
percent administrative fee.)
B .The SCC will notify the SCO Board, SCS Field Office
and ZC when final contracts are completed and project is ready
to begin.
V. Practice Installation and Certification
A. Technical Assistance (TA) agency will design and
monitor practice installation.
B. SCO Board representative monitors implementation of
project and follows up on loan activities as necessary.
C. TA agency will certify to the State that the practice is or
is not installed according to standards and specifications.
87
-------
DEVELOPING NONPOINT SOURCE CONTROL STRATEGIES FOR
BIG STONE LAKE: TWO APPROACHES
GAYLEN F. REETZ
Minnesota Pollution Control Agency
Roseville, Minnesota
TIMOTHY BJORK
South Dakota Department of Water
and Natural Resources
Pierre, South Dakota
PATRICK J.MULLOY
Minnesota Pollution Control Agency
Roseville, Minnesota
DAVID R. GERMAN
South Dakota Department of Water
and Natural Resources
Pierre, South Dakota
STEVEN A. HEISKARY
Minnesota Pollution Control Agency
Roseville, Minnesota
DONALD ROBERTS
U.S. Environmental Protection Agency
Region 5
Chicago, Illinois
ABSTRACT
Big Stone Lake, a hypereutrophic lake located on the
Minnesota-South Dakota border, suffers from algae
blooms, excessive weed growth, and sedimentation. The
South Dakota Department of Water and Natural Re-
sources, with support from the Minnesota Pollution Con-
trol Agency, completed a Diagnostic-Feasibility Study that
identified nonpoint source pollution from agricultural land
use practices in the lake's 2,938 km2 watershed as the
major source of pollution to the lake. Institutional barriers
often present a greater task for nonpoint source projects
than the technical factors involved in addressing nonpoint
source problems. The Big Stone Lake Project provides an
interesting case study because its initiation involved and
required the cooperation of two regional EPA offices, two
States, five counties, and a multitude of State and local
agencies. The large size of Big Stone Lake's watershed
has also required innovative approaches to identifying
and prioritizing nonpoint source pollution strategies. A
computer model will be used to target nonpoint source
control projects within subwatersheds.
INTRODUCTION
"In recent years there have been complaints of increasing
growths of rooted aquatic plants (weeds) and nonrooted,
generally small, scum-forming plants (blue-green algae) in
the lower or southern end of Big Stone Lake, especially in
the vicinity of Ortonville, Minnesota. ..." This excerpt is
from a report requested by the governors of South Dakota
and Minnesota after a meeting of their representatives at
Milbank, South Dakota, in 1967 (S. Dak.-Minn. Comm.
1967). As can be seen from this nearly 20-year-old report,
concern for eutrophication of Big Stone Lake by South
Dakota and Minnesota is not new. What is new is the
coordinated effort by both States to solve many of the
problems contributing to the lake's degradation.
The information presented here is meant to provide an
understanding of the management philosophies of the two
States involved in the project and to show how institutional
differences have been meshed to develop this joint resto-
ration effort.
Basin Description
Big Stone Lake is located on the border of South Dakota
and Minnesota (Fig. 1). Big Stone is a large, hypereu-
trophic, warm water lake with a surface area of 5,002 ha
(12,360 acres), a shoreline length of 96.4 km (59.9 m), and
an average depth of 2.4 m (8 ft). Big Stone Lake was
formed about 8,000 years ago by an alluvial fan deposited
by the Whetstone River in the glacial valley of the River
Warren (Bray, 1977). In 1939, the lake became a reservoir
when a concrete dam replaced the natural outlet following
the diversion of the Whetstone River into the lake for flood
control. This diversion increased the watershed of Big
Stone Lake from 178,588 ha to 295,367 ha (729,841
acres) and significantly increased problems of nutrient en-
richment and sedimentation. Of the 295,367 ha, two-thirds
lie in South Dakota and one-third in Minnesota (Fig. 2).
Water quality of Big Stone Lake is best described as
hypereutrophic. Growth of blue-green algae dominated by
Aphanizomenon is the primary factor limiting recreational
use of the lake from early July to October. Algal density is
usually the principal factor limiting water transparency,
which typically ranges from over 4 m during the spring
zooplankton pulse to less than .5 m in August. Water
transparency is occasionally limited by resuspension of
sediment in the shallow areas adjacent to major tributary
inlets. These and many other shallow areas are covered
by extensive aquatic macrophyte growth during the sum-
mer (S. Dak. Dep. Water Nat. Resour. 1983). Water quality
degradation over the past 20 years has led to a significant
decline in sport fishing and water-based recreational use
of the lake, which has been an important regional resort
and vacation area for the past 100 years.
The major sources of pollution to Big Stone Lake arise
from agricultural land use in the watershed. Erosion from
cropland and runoff from animal feeding operations are
major sources of nutrient and sediment loadings to Big
Stone Lake. Rapid runoff characteristics and streambank
erosion in some subwatersheds also contribute to lake
pollution loadings. Water quality monitoring on tributary
streams has shown unacceptable loads of both nutrients
and sediment. While nonpoint source pollution from inten-
sive agricultural land use is the major source of pollutants
to Big Stone Lake, other sources such as the municipal
sewage facilities at Browns Valley, Minnesota, and Sisse-
ton, South Dakota, contribute to water quality degradation
(S. Dak. Dep. Water Nat. Resour. 1983).
88
-------
INSTITUTIONAL/FINANCIAL ASPECTS OF NONPOINT SOURCE CONTROLS
UTH DAKOTA I I <
ii
u.s M ai J &/,!
BIG STONE LAKE LOCATION MAP
Figure 1.—Big Stone Lake location map.
STATE NONPOINT STRATEGIES AND
PROGRAMS
Although the public has been interested in restoring Big
Stone Lake for at least 20 years, efforts have been limited
by differences in program priorities and organizational phi-
losophies on either side of the lake.
The South Dakota Strategy and Program
In 1976, designated by the Governor as the statewide
management agency responsible for the "formulation of
implementable water quality management plans," the
South Dakota Department of Water and Natural Re-
sources (SDDWNR) began developing a methodology for
preparing a comprehensive 208 management plan. Seri-
ous consideration was given to a variety of methods. Fi-
nally, the SDDWNR decided not to prepare an all-encom-
passing State plan, but rather to target areas for intensive
efforts; and, as individual plans were prepared, more ar-
eas would be added, eventually encompassing all prob-
lem areas of the State. As a rural State with agriculture as
the primary industry, agricultural nonpoint source prob-
lems were expected to be prominent.
Having selected this management approach,
SDDWNR, then the Department of Environmental Protec-
tion, solicited potential candidates for water quality study
areas from planning districts, soil conservation districts,
lake associations, and various other public and private
BIG STONE LAKE WATERSHED MAP
Figure 2.—Big Stone Lake watershed map.
89
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
groups. Each group was asked to submit recommenda-
tions to the appropriate planning districts, which then sub-
mitted their top three choices to SDDWNR. Selections
were based mainly on available data, public support, and
perceived problems. All of the original selections were ru-
ral watersheds with lake or stream problems resulting
from nonpoint sources of pollution.
The SDDWNR collected the preliminary data necessary
to prepare individual plans. Soil conservation districts
were contracted to assist with water sample collection,
compilation of land use data, and dissemination of public
information. The SDDWNR evaluated the data and pre-
pared reports and plan recommendations. South Dakota
used the 208 program to fund promotion of best manage-
ment practices (BMP's) on selected critical areas.
Soil conservation district employees were responsible
for BMP promotion. Although not State employees, soil
conservation district staff activities in areas of water qual-
ity and nonpoint source pollution were directed by
SDDWNR staff. As is evident, the 208 planning process
for South Dakota was not only managed, but many ele-
ments were actually conducted by the SDDWNR from
project initiation through the preparation of final evalua-
tions and reports. Assistance was provided by other agen-
cies.
This somewhat independent management philosophy
has since carried over into all lake projects in the State.
The application of this philosophy to the Big Stone Lake
Restoration Project occurred naturally, considering past
project management. Although the Big Stone Lake project
did not evolve through the 208 process, it had the same
attributes as other State projects: local support, serious
water quality problems, and extensive baseline data. The
difference with this watershed is that, instead of going
through the 208 planning process as a targeted water
quality study area, Big Stone Lake and its associated wa-
tershed went from preliminary baseline data into a Phase I
Study.
In preparing the Phase I grant application, SDDWNR
used existing staff, secured matching funds, stationed a
fulltime employee in the watershed, and purchased the
required monitoring equipment. The fulltime coordinator
collected, compiled, and evaluated all the data required
for a Phase I report, and prepared major sections of the
report, the remainder of which were prepared by
SDDWNR headquarters staff. Almost all of the agencies
previously mentioned, as well as the local agricultural
agencies and the South Dakota Department of Game,
Fish and Parks collected data.
Once the Phase I report and Phase II application were
completed and submitted to U.S. EPA, preparations were
made for implementing the Phase II grant award. As with
the Phase I, the project coordinator assumed responsibil-
ity for finalizing the matching fund commitments, model-
ing feedlots for implementation, and selecting a model
with which to identify necessary BMP's. After the grant
was awarded, SDDWNR continued to actively participate
in the project through the coordinator, with direct assist-
ance from headquarters staff.
The Minnesota Strategy and Approach
The Minnesota Pollution Control Agency (MPCA) is the
Minnesota water quality management agency. Minneso-
ta's nonpoint water pollution effort began in 1976 with the
development of the Minnesota Water Quality Manage-
ment Plan (208 Plan) led by the MPCA. Its purpose was to
identify significant water quality problems caused by non-
point sources of pollution and to set forth effective pro-
grams to address those problems. Unlike South Dakota,
the Minnesota 208 plan was not a blueprint for action in
individual watersheds; rather, it summarized existing man-
agement policies and programs as well as recommended
future policies and actions. The plan recognized that a
continuing nonpoint program would involve three func-
tions: (1) continued study of nonpoint source issues, (2)
preimplementation activities that would lead to putting rec-
ommended programs into operation, and (3) actual imple-
mentation of management programs.
In 1983, the MPCA initiated a study to identify major
barriers to implementation of integrated water quality and
land management in Minnesota. The four barriers identi-
fied were: (1) a poor understanding by the public of the
existence and economic significance of water quality
problems resulting from land management, (2) poor un-
derstanding of available solutions to nonpoint pollution, (3)
government fragmentation of water quality and land man-
agement, and (4) the limited funds available to solve the
problems (Richfield, 1983).
Minnesota then delineated three strategies to address
these problems. First, MPCA completed an information
strategy to develop public awareness of the economic and
recreational impact of nonpoint pollution. Second, Minne-
sota initiated meetings with other State and Federal agen-
cies to encourage inclusion of water quality management
in their existing programs, to encourage their assumption
of new water quality activities, and to provide technical
support. Third, MPCA helped organize and apply for U.S.
EPA Clean Lakes funds for two watershed projects where
nonpoint problems adversely affect water uses, to demon-
strate successful approaches to nonpoint source manage-
ment. These projects are intended to demonstrate techni-
cal solutions to nonpoint control, the viability of an
integrated land and water management approach, the im-
portance of cooperation in overriding fragmented re-
source management, and actual implementation costs,
thus providing an accurate assessment of the control ef-
fectiveness of project funds. Big Stone Lake is one of
Minnesota's nonpoint demonstration projects.
Minnesota's involvement in the Big Stone Lake project
is based on a program approach developed through the
U.S. EPA Clean Lakes Program. MPCA provides funding
and technical support, while contracting with a local unit of
government to lead the effort locally. This approach allows
local project control and decisionmaking to best meet the
local needs and conditions while providing technical over-
sight.
In the case of Big Stone Lake, the Upper Minnesota
River Watershed District is the grantee. The watershed
district, a local unit of government whose purpose is de-
veloping and coordinating water management programs,
is a five-member board of managers with taxing authority
keyed to hydrologic boundaries. The unique form of local
government is a natural local leader for this project al-
though initially State sponsorship was sought. The water-
shed district was experienced, having sponsored a U.S.
Army Corps of Engineers project to modify the Big Stone
Lake outlet, by which more of the Whetstone River will
bypass Big Stone Lake, reducing nutrient and sediment
loading from the Whetstone River.
In addition to the technical review, the Big Stone Project
has benefited from other ongoing nonpoint program activi-
ties. The MPCA instituted a feedlot permit program in the
early 1970's, designed to eliminate and prevent pollution
hazards from livestock and poultry operations. The Minne-
sota Feedlot Computer Model, developed by the USDA
Agricultural Research Service in cooperation with the
MPCA, determines the pollution hazard, and prioritizes
cost-share funds for cleanup of feedlot problems. This pro-
gram, in cooperation with local soil and water conserva-
tion district activities, has solved most of the feedlot prob-
lems contributing to the Big Stone Lake from the
90
-------
INSTITUTIONAL/FINANCIAL ASPECTS OF NONPOINT SOURCE CONTROLS
Minnesota watershed. Three serious existing problems
are now receiving attention from the MPCA enforcement
staff.
Identifying the sources of nonpoint source pollutants
and tracing their path through a watershed is a complex
and time-consuming process. To more efficiently identify
and trace nonpoint pollution, the MPCA funded and joined
with several conservation agencies to develop two com-
puter water quality models (AGNPS I and II). The Agency
used one of the subwatersheds from Big Stone Lake to
verify and test these models. The Upper Minnesota River
Watershed District will use this information and these
models to prioritize problems and assist in designing the
implementation program at Big Stone.
The MPCA actively solicited project support from State
and Federal agencies already engaged in nonpoint con-
trol, and is coordinating the considerable support re-
ceived. Important to that effort was a meeting organized
by MPCA staff, attended by local representatives of the
Watershed District, SDDWNR, the Minnesota Soil and
Water Conservation Board, University of Minnesota Agri-
cultural Extension Service, Minnesota Water Resources
Board, the Soil Conservation Service, and the Agricultural
Stabilization and Conservation Service. The meeting re-
sulted in additional support and interest for this project.
The Minnesota Soil and Water Conservation Board
(SWCB) will target several subwatersheds to receive an
intensive communications program over a 2-year period.
The program will communicate to farm operators the eco-
nomic and social implications of soil erosion, nutrient loss,
and degraded water quality. The SWCB is also expected
to directly provide additional implementation funds
through two State programs for erosion control and water
management.
THE COORDINATED TWO-STATE
APPROACH
Big Stone Lake and its contributing watershed lie in two
U.S. Environmental Protection Agency regions, two
States, five counties, one watershed district, and a multi-
tude of other local governments and governmental agen-
cies. This project represents an extreme case of frag-
mented political boundaries and consequent limits to
water quality protection. The same organizational com-
plexity which once threatened this project is now recog-
nized as a project asset, flexible in overcoming obstacles
to water quality improvement.
Initially, both States were concerned about the other's
management philosophy, although both States recognized
that any improvement in water quality would require in-
volvement by both South Dakota and Minnesota. Although
both States expressed an interest in the restoration of Big
Stone Lake, they had to overcome several barriers and
differences in approach at the regional, State, and local
level. These differences centered on several areas:
1. Project evaluation criteria
2. Pollution control standards
3. Approaches to pollution problems
4. Project management approaches
5. Clean Lake project prioritization criteria
6. The strengths and weaknesses of the two agencies.
A smaller project, involving a more easily defined and
straightforward solution, would have eliminated several of
these barriers and differences. The enormity and nature of
the lake's problems also complicated joint efforts, making
it difficult to complete the Phase I report within the budget-
ary constraints.
The MPCA and Region V EPA had more experience
with engineering approaches to lake problems and at-
tempted to apply these criteria to a nonpoint source con-
trol project focusing on best management practices.
SDDWNR, on the other hand, emphasized direct imple-,
mentation during the planning process and felt that the
emphasis on planning could delay the project's implemen-
tation. Staff changes at both agencies and regions during
the Phase I project and during the interim period between
the completion of the Phase I and implementation of the
Phase II project also complicated the development of a
cooperative working relationship.
The solution involved developing greater flexibility on
the part of Region V EPA and the MPCA to allow consider-
ation of a nonpoint source control project developed on a
limited budget. This also required SDDWNR and Region
VIII to agree to accept some of the stricter standards and
procedures implemented by Region V Both EPA regions
had to agree to allow some activities, considered as "plan-
ning" in more traditional Clean Lakes projects, to receive
funding under the Phase II effort. Further, staff from both
State agencies had to sell the need for a different ap-
proach to the rest of their agencies and to other State
programs.
While both SDDWNR and the MPCA could have even-
tually resolved these differences and overcome the barri-
ers to a cooperative effort, the time required would have
jeopardized the project's momentum and reduced
chances for FY 1984 Clean Lakes funding.
Both EPA regions played a crucial role in speeding up
the negotiation process, cutting through red tape, finding
solutions to the problems that emerged, and helping the
State agency staffs sell the project to the rest of their
agencies. Because many of the differences emanated
from different approaches' by the two EPA regions, deci-
sionmaking at the regional level was necessary for a com-
promise solution. In other instances, where the differ-
ences arose from differing State approaches, EPA
intervention helped avoid lengthy rulemaking processes
and overcome bureaucratic barriers that could have
slowed the negotiation process. In areas where the two
regions and States continued to differ, the EPA regions
helped the States work out solutions that converted these
differences to variations in emphasis, rather than barriers
to cooperation.
The resulting merger of the two different approaches
has led to a stronger project. The resulting cross-fertiliza-
tion has allowed each State to learn from the other's ap-
proach, management style, and legislation. The fact that
some differences remain has allowed greater flexibility on
the local project level. For example, for some measures
that both States felt important, one EPA region had a
greater likelihood of approving and funding than the other.
In other cases, one or the other of the two States might be
better equipped to implement a certain required measure.
By allowing differences to remain, the local project bene-
fits from the strengths of each State agency and both
regions.
This project has provided the following lessons:
1. Geographic and political fragmentation should not
bar project initiation. Addressing these problems is as im-
portant to improving water quality as are the technical
issues.
2. Many of the differences between States and other
governmental units, while barriers at first, can work to the
advantage of a project because the different groups bring
different sets of experiences, skills, and tools to the proj-
ect.
3. In this project, misunderstood communications be-
tween States and between the States and local units of
government impeded the project. When open, effective
communications were established, cooperation overcame
philosophical and political causes for disagreement.
4. For the two States to agree, they needed to develop
91
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
procedures to work out how and where these differences
would be resolved.
5. Projects involving more than one State and region
require a high degree of flexibility on the part of the parties
involved.
6. NFS projects require more planning and coordina-
tion than more traditional Clean Lakes projects; the par-
ties involved either have to accept a less rigidly defined
project or allow for a greater planning effort.
7. Active involvement by EPA can facilitate and expe-
dite negotiations between States in their attempts to ad-
dress interstate pollution problems.
8. Nonpoint source projects need strong local coopera-
tion. Although the MPCA typically does not get involved in
a project until this is developed, the SDDWNR actively
helped develop the local cooperation during the Phase I
study by involving them in the process. This played a key
role in the project's success.
CONCLUSION
Institutional barriers often present a greater task for non-
point source control projects than do the technical factors
involved. The Big Stone Lake Project provides a case
study because its initiation involved and required coopera-
tion of two regional U.S. EPA offices, two States, five
counties, and a multitude of local units of governments
and government agencies. This same organizational com-
plexity that once threatened this project is now recognized
as a project asset allowing the programs the flexibility
necessary to overcome obstacles to water quality im-
provement.
REFERENCES
Bray, E.G. 1977. Billions of Years in Minnesota, the Geological
Story of the State. Sci. Museum Minnesota. St. Paul.
Richfield, D.T. 1983. Integration of water quality management in
Minnesota. Unpubl. rep. Minn. Pollut. Control Agency, Rose-
ville.
South Dakota Department of Water and Natural Resources.
1983. Big Stone Lake Phase I Diagnostic Feasibility Study:
final report. Unpubl. Pierre.
South Dakota-Minnesota Governors Action Committee. 1967.
Big Stone Lake Problem study. Unpubl. rep.
92
-------
NONPOINT SOURCE POLLUTION OF RESERVOIRS:
WHAT THE TENNESSEE VALLEY AUTHORITY IS DOING ABOUT IT
LARRY R. CLARK
Tennessee Valley Authority
Chattanooga, Tennessee
ABSTRACT
The Tennessee Valley Authority has constructed a multi-
purpose reservoir system that is recognized throughout
the world as a model for water resources management on
a watershed basis. As early as the 1930's TVA recognized
the importance of controlling soil erosion to prevent the
premature filling of reservoirs and began working with
Valley farmers in implementing soil conservation prac-
tices. In the 1980's indications of declining water quality
in TVA reservoirs prompted a renewed emphasis on re-
ducing nonpoint source pollution and relies heavily on
Valley States and other Federal agencies to assist in im-
plementing corrective measures in cooperation with pri-
vate landowners. TVA uses a variety of techniques to re-
duce nonpoint sources. These are discussed.
Nonpoint source pollution is adversely affecting water
quality in the Tennessee Valley. In a recent survey of water
quality in the region over half of the 10 most critical water
quality problems resulted from nonpoint sources (Clark et
al. 1980); three additional problems have been identified
since that survey (Tenn. Valley Author. 1984a). The types
of water quality impacts that can be attributed to nonpoint
sources in the Tennessee River watershed include silta-
tion and filling of reservoirs, bacteria contamination, accel-
erated eutrophication of reservoirs, low dissolved oxygen,
and elevated levels of metals.
Although many of the nonpoint-source-related water
quality problems in the Tennessee Valley have a very inter-
esting history, the primary objectives of this paper are to
examine the Tennessee Valley Authority's (TVA) role in
protecting its reservoirs from nonpoint source pollution
and describe TVA's efforts to resolve existing problems
and improve the overall nonpoint source management
throughout the region. This discussion is based only on
activities in TVA's water resources programs.
TVA was created by Congress in 1933 as a corporate
agency of the Federal government. Not part of any Fed-
eral cabinet department, it is an independent agency that
operates with a certain degree of the autonomy and flexi-
bility of a private corporation. TVA planned, built, and now
manages a unified water control system of 40 dams and
reservoirs that regulate the entire length of this Nation's
fifth largest river plus key stretches of its principal tributar-
ies. TVA water resources activities are supported by an-
nual appropriations from Congress (Tenn. Valley Author.
1985).
TVA follows a stewardship philosophy in management,
resulting in the maximum beneficial uses today and in the
future. Also, TVA promotes the economic growth and de-
velopment of the region while ensuring the enhancement
of the Valley's natural resources. Water pollution resulting
from nonpoint sources can affect not only water use in the
TVA region and TVA's ability to manage the reservoir sys-
tem, but can also hinder or preclude regional develop-
ment.
Although it is not a regulatory agency for controlling
pollution, TVA does not depend on the Valley State regula-
tory agencies to carry the entire burden of improving wa-
ter quality in Valley reservoirs. Several nonregulatory TVA
activities help the Valley States keep TVA reservoirs clean
and suitable for beneficial uses.
STEWARDSHIP
TVA is a steward for the water resources of the Tennessee
Valley and four specific water resources activities that help
the agency fulfill that role: (1) controlling nonpoint source
pollution emanating from properties under TVA's custody
or control, (2) reservoir water quality management plan-
ning, (3) septic tank suitability analysis for reservoir shore-
lines, and (4) reservoir release improvements.
Controlling Nonpoint Source Pollution from
TVA Properties
TVA has fee-owned lands and flowage easement rights
along its reservoirs. Fee-owned lands located above the
normal maximum pool are managed under short-term re-
newable license or long-term land use agreements for
multipurpose uses that include agriculture, recreation,
wildlife, and silviculture.
Since 1981 TVA has been recognized by the State of
Tennessee and the U.S. Environmental Protection Agency
(EPA) as the management agency for controlling nonpoint
source pollution emanating from properties under TVA
custody or control. This recognition is pursuant to Section
208(c) of the Clean Water Act of 1977 and its implement-
ing regulations, 40 CFR 35.1521-3. A memorandum of
understanding between TVA and the State of Alabama for
similar recognition in that State is being finalized and
agreements are being pursued with the five other Valley
States.
As a recognized management agency TVA has devel-
oped provisions to be included in deeds, easements,
leases, and licenses requiring the use of best manage-
ment practices (BMP's) for controlling erosion and sedi-
mentation resulting from land disturbing activities. Special
procedures now used in issuing agricultural licenses en-
sure that TVA lands are suitable for row crops and that
State-approved BMP's are followed to protect water qual-
ity and the long-term agricultural capability of the land. In
addition, TVA has developed BMP's for timber harvesting
activities on TVA lands.
Reservoir Water Quality Management Plans
TVA reservoirs, like large settling basins, are particularly
vulnerable to nonpoint source pollution. The beneficial ef-
fects of reservoirs on water quality are well documented
as are the consequences of uncontrolled nonpoint source
pollution (Churchill, 1957; Clark et al. 1980). Improving
and protecting water quality in the TVA reservoir system is
the major reason behind TVA's involvement in nonpoint
source pollution control. The cornerstone of TVA's efforts
is the Reservoir Water Quality Management Plan.
Through its reservoir water quality management plan-
ning process TVA has an active role in defining water
quality problem areas, identifying corrective actions, and
implementing appropriate management actions. These
plans help States carry out their regulatory programs and
93
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
guide TVA itself in operating and managing the reservoir
system.
The reservoir management planning process includes
the following five phases:
1. Identifying water quality problems and management
issues.
2. Developing a data base appropriate to the problems
and needs identified.
3. Identifying cause and effect relationships and using
those relationships to predict changes in water quality that
would result from applying alternative pollution control
strategies and further development.
4. Developing a management plan that synthesizes the
information into recommendations for correcting existing
use impairments and preventing future water quality prob-
lems.
5. Implementing the management plan recommenda-
tions.
By the end of 1985 TVA will have completed manage-
ment plans for five reservoirs and be well into the imple-
mentation phase (Tenn. Valley Author, 1984b). Three
other reservoirs will have management plans in one of the
other four phases.
Septic Tank Soil Suitability Analysis
Soil conditions along the reservoir shorelines of many TVA
reservoirs are unsuitable for conventional septic tank soil
absorption systems. Because of this, many conventional
systems are failing and may be contributing bacteria and
nutrients to TVA reservoirs. In 1985 TVA is attempting to
document the extent of water quality degradation resulting
from falling septic tank systems along reservoir proper-
ties. To combat this nonpoint source TVA is providing guid-
ance to local and State health departments and land de-
velopers on the capability of shoreline properties to
handle onsite sewage disposal systems. This guidance is
a conceptual plan that identifies shoreline properties suit-
able for conventional or alternative onsite systems and
also properties not suitable for any type of onsite system.
In the latter case these properties must be sewered or
remain undeveloped. This analysis is performed using soil
survey information digitized on TVA's Geographic Infor-
mation System. The soil suitability analyses are per-
formed by an experienced soil scientist and environmen-
tal engineer familiar with the soil requirements for
conventional and alternative onsite systems. Conceptual
plans have been completed for two TVA reservoirs (Sa-
gona, 1985); another is scheduled to be completed in
1985.
Reservoir Release Improvements
Nonpoint sources contribute to the natural dissolved oxy-
gen depletion processes occurring in TVA's deep, ther-
mally stratified reservoirs. The primary result of this dis-
solved oxygen depletion is almost 300 miles of stream
below TVA dams that are low in dissolved oxygen. One
way of dealing with this condition is to increase the dis-
solved oxygen at the point of release, the dam. In 1981
TVA began a 3-year experimental program designed to
study and test alternative methods of enhancing dissolved
oxygen levels in reservoir releases. This program has
been very successful (Tenn. Valley Author. 1984c). The
implementation phase will probably continue for at least
another 3 years.
To complement the reservoir release improvement pro-
gram, in 1985 TVA initiated a basin rehabilitation project
for the South Fork Holston River. One of the purposes of
this project is to reduce point and nonpoint source contri-
butions in the watershed above two TVA reservoirs experi-
encing dissolved oxygen depletion. The results of this
project will help TVA determine the degree of improve-
ment that could be expected from improved reservoir
quality.
RESOURCE ENHANCEMENT
An adequate supply of water capable of supporting a vari-
ety of beneficial uses is essential to economic growth and
future development that may be in the public interest. The
resolution of nonpoint source-related water quality prob-
lems will aid TVA's efforts to promote natural resource-
based economic development.
Three activities that support TVA's resource enhance-
ment role include: (1) identification of nonpoint source
problem areas, (2) serving in a facilitator role to resolve
nonpoint source pollution problems, and (3) conducting
demonstrations of solutions to nonpoint source problems.
Identification of Nonpoint Sources
Three years after its creation in 1936 TVA conducted a
survey of water pollution in the Tennessee River (Scott,
1941). Since that time water quality monitoring and as-
sessments have continued to be a key component of
TVA's water resources programs. Although the emphasis
of the monitoring program has shifted from time to time,
the primary objective remains: to identify problem areas
and evaluate the effectiveness of corrective actions. Tradi-
tional TVA ambient monitoring programs have been only
partially effective in identifying nonpoint source-related
problems; therefore, TVA has recently turned to intensive
surveys of suspected problem areas with rainfall event
sampling for targeted water quality parameters (Milligan et
al. 1984; Carriker and Mullins, 1983).
The diffuse nature of nonpoint source pollution coupled
with its seasonal and hydrologic variation make source
identification technically difficult and expensive. TVA uses
aerial photography and stereoscopic interpretation tech-
niques to reduce costs and improve the extent of cover-
age and accuracy of nonpoint source identification. These
techniques are not new. However, their extensive use in
identifying nonpoint sources is new.
TVA uses color infrared photography and personnel
trained in the characterization of nonpoint source pollution
from aerial photographs to identify animal waste runoff
and failing septic tanks. In cooperation with the Soil Con-
servation Service (SCS) district conservationist, aerial
photography and county soil survey information is also
used to estimate soil erosion rates from individual farm
fields.
Results of all TVA monitoring and data analysis are
made available to the State regulatory agencies. TVA data
complements the State's monitoring programs and helps
to prioritize problem areas. When nonpoint source prob-
lems are identified, the Valley States initiate appropriate
regulatory or voluntary cleanup actions and often TVA co-
operates in the problem resolution process.
Catalyst for Solving Water Problems
When a nonpoint source water quality problem is identi-
fied, TVA works cooperatively with State and other Federal
agencies to solve the problem. TVA uses the data col-
lected during the problem identification phase to focus
public attention on priority problems and issues. TVA en-
courages public involvement in controlling nonpoint
sources. One approach that has been effective in correct-
ing some of the more complex water quality problems in
the Tennessee Valley has been the formation of an inter-
agency task force to plan and direct cleanup activities.
Federal agencies such as SCS, Agricultural Stabilization
and Conservation Service (ASCS), U.S. Geological Sur-
94
-------
vey, and EPA, along with the State regulatory agencies
have worked cooperatively with TVA on nonpoint source
pollution problems.
Demonstration of Solutions to Nonpoint
Source Problems
Often the correction of a nonpoint source problem cannot
proceed because cost-effective corrective techniques are
not available. In the case of high-priority problems TVA
develops and implements projects to demonstrate effec-
tive and economical solutions. TVA demonstrations also
serve as an education tool to encourage participation in
water quality improvement efforts. On one such project
that involved reclamation of abandoned mineral mine
lands, TVA developed a minimal land reclamation tech-
nique that controls offsite erosion at a low cost of $2,4707
hectare ($1,000/acre) (Muncy, 1981). This demonstration
encouraged the State's legislature fo provide funding to
the county governments to complete the project. The end
result was the control of erosion from over 242 hectares
(600 acres) of abandoned mine lands, erosion that was
adversely affecting downstream water supplies and con-
tributing to the siltation of TVA reservoirs.
In another project TVA is working with SCS, ASCS, and
farmers to control animal waste runoff in a major tributary
watershed. While helping farmers install animal waste
systems TVA is developing information on the amount of
cost-share necessary to stimulate landowner interest and
identify the animal waste treatment components with the
most water quality benefits. Through this demonstration
animal waste treatment system designs have been im-
proved and educational material on operation and mainte-
nance of animal waste systems has been developed.
CONCLUSION
TVA's role as a steward for the water resources of the
Tennessee Valley and its mission of resource enhance-
INSTITUTIONAUFINANCIAL ASPECTS OF NONPOINT SOURCE CONTROLS
ment dictates an active involvement in helping control
nonpoint source pollution. The lack of direct regulatory
responsibility for pollution control should not discourage
water resources agencies in working cooperatively with
others to resolve nonpoint source pollution problems. The
fact that TVA is not burdened with regulatory responsibili-
ties provides more opportunities and flexibility in dealing
with nonpoint sources.
REFERENCES
Carriker, N.E., and J. Mullins. 1983. Big Nance Creek water
quality investigation, Lawrence County, Ala. Tenn. Valley Au-
thor.
Churchill, M.A. 1957. Effects of storage impoundments on water
quality. Am. Soc. Civil Eng. Pages 419-64 in Trans. 123.
Clark, L.R., et al. 1980. Is the water getting cleaner. Tenn. Valley
Author, Chattanooga, Tenn.
Milligan, J.D., I.E. Wallace, and R.P. Betson. 1984. The relation-
ships of urban runoff to land use and groundwater resources.
TVA/ONRED/AWQ-84/1. Tenn. Valley Author.
Muncy, J.A. 1981. The Tennessee Valley Authority's cooperative
noncoal minerals abandoned mine land reclamation demon-
stration project: Avery Mitchell, and Yancy Counties, N.C.
Tenn. Valley Author.
Sagona, F.J. 1985. Conceptual onsite wastewater management
plan for residential developments along Cedar Creek Reser-
voir. TVA/ONRED/AWR-85/4. Tenn. Valley Author.
Scott, G.R. 1941. Studies of the pollution of the Tennessee River
system. Tenn. Valley Author.
Tennessee Valley Authority. 1984a. Regional Water Manage-
ment Program. FY 1985 Program Plan.
. I984b. Water Quality Management Plan for Chero-
kee Reservoir. TVA/ONRED/INRQ-84/1.
_. 1984c. Improving reservoir releases. TVA/ONRED/
ASWR-84-27.
. 1985. TVA and the Office of Natural Resources and
Economic Development. D. Rucker, ed. Impact. 8(1): 2, 6.
95
-------
COMPREHENSIVE PROTECTION FOR TWO MULTIPURPOSE
RESERVOIRS IN CENTRAL NORTH CAROLINA—EPA'S NATIONAL
NONPOINT SOURCE POLICY CAN WORK
EDWARD A. HOLLAND
Triangle J Council of Governments
Research Triangle Park, North Carolina
ALAN W. KLIMEK
North Carolina Department of Natural
Resources and Community Development
Raleigh, North Carolina
ABSTRACT
Federal, State, and local agencies are carrying out an
aggressive watershed protection program to prevent de-
radation of two new multipurpose reservoirs in the Ra-
leigh-Durham-Chapel Hill area of North Carolina. The
U.S. Army Corps of Engineers impounded the B. Everett
Jordan Lake and Falls of the Neuse Reservoirs in 1981
and 1983 for flood control, recreation, and water supply.
With a drainage area of almost 6,500 km2, the reservoirs
have a combined surface area of 10,000 ha, and repre-
sent a potential raw water source of 200 million gallons
per day. An interagency strategy was developed in re-
sponse to growing public demands for water supply pro-
tection amid accelerating urban development and evi-
dence of excessive nutrients in the reservoirs. The
strategy is preventive in focus—none of the intended
uses has yet been impaired. Phosphorus removal will be
required from all new wastewater discharges and from
selected existing facilities in the watersheds. The North
Carolina General Assembly will consider a statewide ban
on the sale of phosphate-containing laundry detergents
during its 1985 session. Cities and counties have enacted
land use controls and a $2 million a year State-funded
cost-share program is helping farmers finance much-
needed agricultural BMP's in critical portions of the wa-
tersheds. Initial success of the overall strategy appears to
support the principles of EPA's proposed National Non-
point Source Policy.
The B. Everett Jordan and Falls of the Neuse Reservoirs
lie in North Carolina's piedmont physiographic province
(Fig. 1). The U.S. Army Corps of Engineers impounded
the lakes in 1981 and 1983 for flood control, recreation,
and water supply. With a combined drainage area of 2,500
square miles, they represent a potential raw water source
of 200 mgd for the Research Triangle area of Raleigh,
Durham, and Chapel Hill, North Carolina.
Local officials and the general public recognizing the
reservoirs' value to the region have demanded increased
protection during an unprecedented period of economic
growth and development. Although a great deal of public
and editorial attention has focused on possible water qual-
ity effects of urbanization, none of the intended uses of
either lake have yet been impaired. The Falls/Jordan wa-
tershed efforts described here represent an important
public commitment to preventive—rather than correc-
tive—action.
Efforts begun in 1983 have resulted in several accom-
plishments:
• Phosphorus removal is now required at all new
wastewater plants in the 2,500 square-mile watershed and
at selected existing facilities.
• The North Carolina House of Representatives ap-
proved a ban on the sale of phosphate detergents (to be
considered by the State Senate in 1986).
• Cities and counties have enacted aggressive and
controversial land use controls for new development re-
stricting sewer extension policies, impervious surface cov-
erage, gross density, industrial siting, underground chemi-
cal and petroleum storage, and vegetated stream buffer
requirements.
• The North Carolina General Assembly created a
State-funded cost-share program for agricultural conser-
vation practices, and provided a two million dollar biennial
appropriation for use by 15 counties in the State's desig-
nated Nutrient Sensitive Watersheds.
THE WATERSHEDS
Table 1 highlights several features of the Falls and Jordan
watersheds. Both lakes are shallow, with mean depths of
12 and 16 feet, respectively. Wastewater treatment plant
effluent equals or exceeds the volume of natural stream-
flow entering the lakes during low flow periods. Both wa-
tersheds are large and heavily populated, containing
about 10 percent of North Carolina's total population (Div.
Environ. Manage. 1983).
Figure 2 depicts gross land use and phosphorus load-
ing. Approximately 63 percent of the land is forested, and
28 percent is in agricultural use (tobacco, corn, poultry,
dairy, and hog production). The relatively small proportion
(9 percent) of urbanized land is replacing forested and
agricultural areas at an increasing rate. The largest frac-
tion of phosphorus input (55 percent) comes from munici-
pal wastewater plants, none of which removed phos-
phorus before the current initiative (Div. Environ. Manage.
1983).
Falls and Jordan Lakes are two of the most highly en-
riched water bodies in North Carolina, but their quality
tends to be typical of mainstem piedmont reservoirs in the
southeastern United States. Low Secchi depths are due to
high algal biomass and inorganic sediment; pH and dis-
solved oxygen data reflect the high productivity, photosyn-
thesis, and thermal stratification of hot summer condi-
tions. Phosphorus and chlorophyll a concentrations,
which clearly exceed "acceptable" levels for northern
Figure 1.—Location map of Falls and Jordan Lake water-
sheds, North Carolina.
96
-------
INSTirUTIONAUFINANCIAL ASPECTS OF NONPOINT SOURCE CONTROLS
Table 1.—Selected hydrologic and morphometrlc features
of the Falls and Jordan Lake watersheds, North Carolina
(Dlv. Environ. Manage. 1983).
Falls .Jordan
Surface Area (ac)
Volume (ac-ft)
Mean Depth (ft)
Streamflow (cfs)
Mean annual
7Q10
WWTP Flow (cfs)
Watershed (sq mi)
Population
12,500
154,000
12.3
800
17
16
770
150,000
14,300
235,000
16.4
1700
76
143
1690
480,000
table 2.—Generalized summertime water quality data, Falls
and Jordan Lakes, North Carolina (Correale, 1985).
Surface Bottom
Conductivity
Secchi (ft)
PH
D.O. (% sat.)
Total P 0*9/1-)
Chi a G*g/L)
100
2.5
9.3
130
80
100
250
—
6.5
0
350
—
temperate lakes, have not resulted in algal mat formation,
and do not represent nuisance conditions in Falls and
Jordan Reservoirs. Table 2 displays generalized water
quality data representing surface and bottom conditions
typical of hot summer periods.
In addition to nutrients, watershed protection strategies
have focused on sediment loads and the possible pres-
ence of toxic materials. The North Carolina Environmental
Management Commission classified both reservoirs as
public water supply sources but will not authorize the pota-
ble use of Jordan Lake until more data are gathered about
trace metals and synthetic organic chemicals present in
the watershed (Environ. Manage. Comm. 1983). Local offi-
cials and the general public have consistently demanded
assurances that the potential 200 mgd water supply will be
safe for consumption. To date, no synthetic organic chemi-
cals have been measurable in either Falls or Jordan Lake
(D'rv. Environ. Manage. 1985a).
PUBLIC CONCERN
A high and sustained public concern expressed by local
governing bodies, newspaper editorials, and radio/TV fea-
tures was an important factor behind the Falls/Jordan wa-
tershed protection effort. The reservoirs' recent impound-
ment occurred during a period of unprecedented growth
in the Research Triangle area. A proliferation of new subdi-
visions, office parks, and shopping centers had height-
ened the public's awareness of potential water quality ef-
fects on their new reservoirs.
Chronology
The period from impoundment to active protection encom-
passed several activities in the following order:
Construction/impoundment. Jordan and Falls Reser-
voirs were filled in 1981 and 1983, respectively.
Call for Action. A resounding call for action, as de-
scribed above, received a quick and positive election year
response from cabinet level state officials.
Steering Committee. The Secretary of North Caroli-
na's Department of Natural Resources and Community
Development (NRCD) created a Steering Committee of
mayors and county board chairmen from each of the 16
political jurisdictions in the Falls and Jordan watersheds.
Nutrient Sensitive Designation. The North Carolina
Environmental Management Commission classified the
Falls and Jordan watersheds "Nutrient Sensitive," provid-
ing an explicit regulatory mechanism for point source
phosphorus control.
Point and Nonpoint Tradeoffs. The Secretary of
NRCD proposed a basic tradeoff: "If you (local govern-
ments) take certain actions to reduce nonpoint runoff in
your jurisdictions, then we (State government) might not
have to require phosphorus removal at your treatment
plants..."
State-Local Action Plan. State and local officials
agreed to a semi-formal "action agenda" setting basic
goals and responsibilities for the participants.
Implementation.
LAND USE
2460 sq mi
PHOSPHORUS LOADING
1,800,000 Ibs/yr
Municipal
WWTPs
55%
Figure 2.—Gross land use and phosphorus loading, Falls and Jordan Lake watersheds (combined data), North Caro-
lina (Dlv. Environ. Manage. 1983).
97
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
THE STATE-LOCAL ACTION PLAN
As noted in Table 3, basic targets of the Action Plan were
agricultural and urban runoff, point source phosphorus,
and hazardous materials leaks and spills.
One of the most significant accomplishments to date
has been the passage of a $2 million biennial state cost-
share program which provides up to 75 percent funding
for agricultural best management practices (BMP's). (The
Orange Water and Sewer Authority, serving Chapel Hill,
also offers up to 50 percent of the remaining costs,
thereby reducing the private share to 12.5 percent of total
BMP cost in certain portions of the Jordan watershed.)
Another area of substantial progress has been the adop-
tion of aggressive land use controls by nearby cities and
counties.
LOCAL LAND USE GUIDELINES-
MANAGING THE TYPE AND LOCATION
OF NEW DEVELOPMENT
The Triangle J Council of Governments developed a three-
tiered set of recommendations for the type and location of
new development in the watersheds based on the princi-
ple of providing greater protection to areas closest to the
lakes (Triangle J, 1984). The three tiers correspond gener-
ally to distance from the reservoirs:
• Wafer Quality Critical Areas—Land within one mile of
the shoreline.
• Limited Industry Areas—Land beyond the Critical Ar-
eas, but within public water supply portions of the water-
shed.
• Basinwide Guidelines—All land throughout the Falls
and Jordan watersheds.
Water Quality Critical Areas. The strictest and most
controversial recommendations applied to the Critical Ar-
eas within 1 mile of each lake. The primary goal was to
minimize urban runoff and the risk of chemical spills by
maintaining the patterns of low intensity rural residential
development that already existed. Accordingly, the guide-
lines called for a 6 percent limit on impervious coverage;
no new industrial development whatsoever; and, no mu-
nicipal sewer extensions into the Water Quality Critical
Areas.
Limited Industry Areas. Beyond the Critical Areas, but
within water supply portions of the watersheds, the guide-
lines were less restrictive, and focused on special safe-
guards for industries that use, produce, store, or transport
specified amounts of certain hazardous materials. Before
receiving a local development permit in a Limited Industry
Area, the applicant would have to provide detailed infor-
Table 3.—Major elements of the State-local action plan for
Falls and Jordan Lakes, North Carolina (Grimsley, 1983).
Agricultural runoff
• State funding for agricultural cost-share program
Urban runoff from new development
• Stricter zoning and land use controls by local
governments
• Stormwater management requirements for water quality
control
• Local erosion & sediment programs for new construction
Point source phosphorus removal
• Phosphate detergent ban
• Phosphorus removal to 1 mg/Lg at selected treatment
plants
Hazardous materials
• Local inventories of use, storage, production
• Contingency response plans for leaks and spills
• Additional toxics research and monitoring by state
agencies
mation on materials present on site, special plans for con-
taining and cleaning up any spills, and compliance with
siting and monitoring standards for chemical storage
tanks.
Basinwide Guidelines. Beyond the Water Quality Criti-
cal and Limited Industry Areas, certain recommendations
applied to new development throughout the 2,500 square
mile watershed. These included controlling 1/2 inch of run-
off from all impervious surfaces (preferably through natu-
ral infiltration), maintaining 50-foot vegetated buffers
along all streams, and adopting 12 and 30 percent imper-
vious limits for sewered and unsewered areas, respec-
tively.
Nearby cities and counties have made substantial pro-
gress incorporating these often unpopular guidelines into
local zoning ordinances and subdivision regulations. De-
tails of local programs in the Falls and Jordan watersheds
are reported elsewhere (Triangle J, 1985).
OTHER ACCOMPLISHMENTS
In addition to aggressive local development controls and
the agricultural cost-share program, other accomplish-
ments are notable:
Phosphate Detergent Ban. The North Carolina House
of Representatives passed legislation banning the sale of
household detergents containing more than 0.5 percent
phosphorus in the Falls and Jordan watersheds. The de-
tergent ban has been widely supported by citizens and
local governments, but is vigorously opposed by industry
groups led by the Soap and Detergent Association. The
legislation will be considered by North Carolina's Senate
in 1986.
Expanded Toxics Program. Concern about the possi-
ble presence of toxic chemicals in the water of Falls and
Jordan Lakes highlighted a statewide need for additional
chemical and biological monitoring of North Carolina's
waters. In response, the General Assembly appropriated
funds to expand the State's water quality monitoring net-
work and analytical capability for toxic substances.
Increased Public Awareness. An important result of
the Falls/Jordan initiative has been the greater aware-
ness, support, and commitment to a sophisticated menu
of water quality issues by the general public and elected
officials of the Research Triangle area.
FACTORS FOR SUCCESS
In terms of substantial State and local efforts focused on a
complex problem and an action-oriented commitment by a
wide range of agencies and interest groups, the Falls/
Jordan watershed project has been more successful than
other initiatives in North Carolina and elsewhere. Several
factors contributed to these accomplishments.
Common Perception of A Problem. The overall reser-
voir strategy has been preventive. To date, none of the
intended uses of either lake have been impaired by water
quality problems. Nevertheless, watershed efforts drew
strength from a sustained and widespread sense of public
urgency, due in part to the general awareness that Raleigh
would soon depend solely on Falls Lake for its water sup-
ply, and that the region's spectacular economic growth
included some unwanted side effects: unsightly commer-
cial development, traffic congestion, and water pollution.
Much of the urgency to "do something" was expressed in
the deliberations of local policy board and in editorials of
local newspapers.
Effective Political Leadership. A quick and incisive
response by Governor James B. Hunt, Jr. and Natural
Resources Secretary Joseph W. Grimsley created an ad
hoc steering committee of mayors and county board chair-
98
-------
INSTITUTIONAL/FINANCIAL ASPECTS OF NONPOINT SOURCE CONTROLS
men from 16 jurisdictions in the watersheds, and mobi-
lized the resources of state and local government into a
working partnership. The clear commitment of key State
and local leaders provided the administrative momentum
for overcoming traditional bureaucratic barriers.
Expertise In Place. Technical work and policy recom-
mendations for the Falls/Jordan strategy were drafted by
existing State, Federal, and local staff well versed in the
array of land use-water quality issues. Most of the techni-
cal information on nutrient loading, sediment sources, and
hydrology had been developed previously by the North
Carolina Division of Environmental Management, the
USDA Soil Conservation Service, county Soil and Water
Conservation Districts, and the Triangle J Council of Gov-
ernments. Given the top level political commitment for
action, it remained only to organize relevant technical in-
formation into a coherent policy framework and implemen-
tation program.
The 208 Experience. Many key agencies and individ-
uals at both the State and areawide levels had developed
their water quality management expertise and familiarity
with nonpoint pollution issues through EPA's 208 process.
In some ways, Falls and Jordan became the "main event"
for which earlier 208 exercises were the warmup.
CONCLUSIONS
A comprehensive program for protecting the 2,500 square
mile watershed of two multipurpose reservoirs is being
accomplished by State and local governments in central
North Carolina. The preventive strategy includes phos-
phorus removal at selected treatment plants; a phosphate
detergent ban; State-funded cost-share program for agri-
cultural BMP's; local development restrictions on impervi-
ous coverage, density, industrial siting, hazardous materi-
als storage, and utility extension policies. State and local
political leaders effectively mobilized existing expertise
and public concern about the effects of rapid economic
growth on the region's two new reservoirs.
REFERENCES
Correale, Christina. 1985. Personal commun., U.S. Army Corps
of Engineers, Wilmington, NC.
Division of Environmental Management. 1983. Water Quality
discussions of Falls of the Neuse and B. Everett Jordan
Lakes. Rep. No. 83-06. NC Dep. Nat. Resour. Commun. De-
velop. Raleigh.
1985a. Toxic Substances in Surface Waters of the B.
Everett Jordan Lake Watershed. Rep. No. 85-02. NC Dep.
Nat. Resour. Commun. Develop., Raleigh.
1985b. Toxic Substances in Surface Waters of the
Falls of the Neuse Lake Watershed. Rep. No. 85-08. NC Dep.
Nat. Resour. Commun. Develop., Raleigh.
Environmental Management Commission. 1983. Report of pro-
ceedings concerning proposed reclassification of B. Everett
Jordan Lake and watershed and Falls of the Neuse Lake and
watershed—public hearing, August 9, 1983. NC Dep. Nat.
Resour. Commun. Develop., Raleigh.
Grimsley, Joseph W. 1983. State-local action agenda: Falls and
Jordan watersheds, Remarks to the Falls/Jordan Steering
Committee, Oct. 7,1983, Research Triangle Park, NC.
Powell, D. 1983, 1984. (Editorial Cartoons) Raleigh News & Ob-
server, June 5,1983 and January 6,1984, Raleigh, NC.
Triangle J Council of Governments. 1984. Clean water and new
development—guidelines for Falls and Jordan Lakes. Re-
search Triangle Park, NC, April 25.
1985. Falls/Jordan Watershed protection—report of
local progress in Region J. Research Triangle Park, NC, Feb-
ruary 27.
99
-------
Ground Water Quality
GROUND WATER CONTAMINATION BY ALDICARB PESTICIDE IN
EASTERN SUFFOLK COUNTY, LONG ISLAND
JULIAN SOREN
U.S. Geological Survey
Syosset, New York
W. G. STELZ
U.S. Environmental Protection Agency
New York, New York
ABSTRACT
Highly toxic aldicarb pesticide, initially considered incap-
able of contaminating ground water, was used on about
9,700 ha of potato fields in eastern Suffolk County on
Long Island, N.Y., during 1975-79. In 1979, aldicarb was
found in substantial concentrations in ground water sam-
ples. Subsequent extensive analyses of the area's
ground water showed widespread contamination, with
concentrations ranging from traces to as much as 515 /xg/
L. The New York State Department of Health has set a
limit for aldicarb in drinking water at 7 /ig/L, and, in early
1980, the U.S. Environmental Protection Agency revoked
its approval for the use of aldicarb on Long Island. In
1980, a total area of about 198 km2—nearly four times the
area to which aldicarb had been applied—contained
small to excessive aldicarb concentrations in ground wa-
ter; a 54 km2 area, approximately half of the area to which
aldicarb had been applied, contained excessive concen-
trations. By 1984, aldicarb's manufacturer had installed
more than 2,000 activated-charcoal filters on drinking-
water supply-well systems with excessive aldicarb con-
centrations to reduce concentration to less than 7 ^g/L,
and agreed to continue the filter installations as needed.
Despite aldicarb's reputed toxicity, no case of aldicarb
poisoning from drinking water on Long Island has been
documented. The aldicarb may remain within the system
for decades. Its duration in the soil and ground water is
under investigation. Estimates of its "half-life" range from
2 years to as much as 10 years. Current estimates of
aldicarb's deterioration on Long Island suggest that con-
centration may decline to less than 7 /*g/L between 1990
and 2030.
INTRODUCTION
Aldicarb is a highly toxic oxime-carbamate pesticide de-
veloped for agricultural use. Aldicarb was used on about
9,700 ha of potato fields on the north and south forks of
eastern Long Island (Fig. 1) during 1975-79 and seemed
highly effective in controlling the Colorado potato beetle
and golden nematode. Although the pesticide was thought
incapable of contaminating ground water, several ground
water analyses in 1979 revealed contamination of the up-
per glacial aquifer, which is the principal source of drinking
water supply in this rural and suburban area. Subsequent
extensive ground water analyses through 1980 by Suffolk
County's Department of Health Services (SCDHS)
showed the contamination to be widespread.
The contamination led the U.S. Environmental Protec-
tion Agency to revoke aldicarb's approval for use on Long
20KILOMETEKS
Figure 1 .—Area of aldicarb contamination in eastern Suffolk
County and location of study area.
101
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
LONG
ISLAND
•ws*
r L A N r i c
South
OC
| | Area uncontaminated byaldicarb
Aldicarb-contaminated area
£76 Aldicarb concentrations to>7wg/L were detected.
1 but data are insufficient to delineate area, upper
number is number of wells sampled: number below
is number of wells in which aldicarb concentrations
exceeded 7ug/L
Base from U.S. Geological Survey
State base map, 1:500.000.1974
Aldicarb data adapted from Baierand Bobbins i1"P2aand 1982b)
Figure 2.—Areal distribution of aldlcarb-contaminated ground water in 1980.
Island in February 1980 at the manufacturer's request,
and the New York State Department of Health set an in-
terim maximum of 7 ppb, equivalent to 7 ng/L, for aldicarb
in drinking water. Much of the area's ground water was
found to exceed this limit; concentrations in the 1979-80
analyses ranged to as much as 515 /tg/L. The manufac-
turer, in an agreement with SCDHS, provided for installa-
tion of activated-charcoal filters through June 1983 on
contaminated drinking water supply-well systems to bring
the supply's concentrations to below 7 /tg/L (Baier and
Moran, 1981). Through an extension of the period for filter
installations, more than 2,000 well systems had been
equipped with filters through 1984 (Moran, 1984).
In 1980, areas of about 145 km2 and 53 km2 on the
north and south forks, respectively (Fig. 2), contained
small to excessive aldicarb concentrations. The combined
contaminated areas totaled nearly three times the area to
which aldicarb had been applied. The total area of both
forks in which aldicarb concentrations exceeded 7^g/L
was about 54 km2, approximately half of the area to which
aldicarb had been applied.
The acute symptoms of aldicarb overdosage in humans
include headache, giddiness, nervousness, blurred vi-
sion, weakness, nausea, cramps, and chest discomfort;
signs of overdosage include sweating, myosis, tears, sali-
vation, excessive respiratory secretions, vomiting, im-
paired breathing, muscle twitching, convulsions, and
coma (Union Carbide Corp., 1975). Through 1984, how-
ever, no signs or symptoms of aldicarb poisoning that
could be traced to drinking water in Suffolk County have
been documented.
In response to public concern about the presence of
aldicarb in the ground water, the U.S. Geological Survey
in 1980 entered into a cooperative program with SCDHS
and the Suffolk County Water Authority to study the
ground water flow patterns and the concentration and
depth of infiltration of aldicarb over time in the aquifer to
estimate its rate of removal from the aquifer. Results of the
study will be published in Soren and Stelz (1985). Much
previous work had been done by SCDHS in locating aldi-
carb-contaminated areas, and the manufacturer had pro-
vided SCDHS with results of more than 8,000 analyses of
ground water from wells in the potato farm locales. The
Geological Survey selected a small representative area in
the Jamesport vicinity of Long Island's north fork (Fig. 1)
to conduct its study.
This paper summarizes the data and interpretations
given in the paper by Soren and Stelz (1985).
Location and Extent of Area
Most of the aldicarb contamination on the north fork is in
the towns of Riverhead and Southold; significant contami-
nation also occurred on the south fork, mainly in the town
of Southampton and to lesser extents in southeastern and
southwestern Brookhaven and East Hampton, respec-
tively (Fig. 2). The SCDHS's investigation of aldicarb con-
tamination in 1982 showed similar distribution (Baier and
Bobbins, 1982a,b).
The Jamesport area, the location selected for the Geo-
logical Survey's study, is a strip 2.4 km wide across the
north fork that contains about 364 ha of farmed land (Fig.
1), approximately 243 ha of which are used for growing
potatoes. This site was selected because it had not been
studied in detail by the SCDHS.
Hydrogeology
The hydrogeology of the Jamesport vicinity is typical of
most of eastern Long Island. Therefore, findings at Ja-
mesport are probably applicable to the rest of eastern
Long Island.
The upper glacial aquifer consists mainly of glacial sand
and gravel deposits. It is the principal source of ground
102
-------
GROUND WATER QUALITY
water supply in eastern Long Island. The aquifer surface
is mostly near land surface and is therefore readily con-
taminated by surface contaminants. The sand and gravel
deposits have high lateral and vertical hydraulic conduc-
tivity, which gives high mobility to dissolved substances.
Methods of Investigation
The Geological Survey installed test and observation
wells during 1980-82 at 13 sites from north to south
across the Jamesport area to obtain samples for labora-
tory analyses and to aid in preparing a water table map to
determine directions of ground water flow. Several other
wells in the area were also used for sampling. Holes for
the test and observation wells were 100 mm in diameter,
drilled to depths of about 49 m, with 50-mm inside-diame-
ter casings and 0.6-m screens. The holes were backfilled
with cuttings from them. Screen settings were developed
and pumped for sampling. After a screen setting was
pumped, the casing was pulled up about 6 m and pumped
and sampled again. Two observation wells were installed
at each of the 13 test sites; after laboratory analyses,
screen settings at each of the sites ranged from 2 to 4 m
and from 11 to 15 m below the water table. Well locations
are shown in Figure 3. A total of 132 water samples were
obtained from 101 well-screen settings that ranged in
depth from about 2-45 m below the water table; these
were analyzed for aldicarb content.
The complex procedure for aldicarb analyses was de-
veloped and described by the manufacturer (Union Car-
bide Corp., 1975). Other analysts who measured the pesti-
cide in this study claimed they used the manufacturer's
basic method with slight modifications in a few proce-
dures. To check the reliability of analytical data, split-sam-
ple analyses were done on more than 20 percent of the
samples collected. Many of the split-sample analyses
Figure 3.—Locations of selected wells in the Jamesport
area.
showed small to large discrepancies that made interpreta-
tion of aldicarb contamination uncertain at several sites.
ALDICARB IN THE UPPER GLACIAL
AQUIFER
Degradation Products of Aldicarb
Aldicarb was applied to plant furrows when the potato
seeds were sown, and to the soil shortly after the plants
began developing. The precise amount of aldicarb applied
to the soil is unknown. Recommended total applications
per crop on Long Island ranged from 4.5 to 8 kg/ha. Re-
cords of aldicarb sales on Long Island indicate that the
average application was about 5.5 kg/ha. The pesticide
readily degrades by oxidation and hydrolysis to a sulfoxide
and a sulfone. Aldicarb sulfoxide's toxicity is comparable
to that of aldicarb; the sulfone is considerably less toxic
(Union Carbide Corp., 1975). Aldicarb and its sulfoxide
and sulfone are looked for in analyses, and the constitu-
ents are reported collectively as aldicarb. Degradation
products of sulfoxide and sulfone are not considered to be
toxic. Recent studies (Hansen and Spiegel, 1982) state
that aldicarb itself has not been found in ground water;
rather, the water contained nearly equal parts of the sul-
foxide and sulfone.
Results of Chemical Analyses
Aldicarb concentrations in the ground water samples from
the Jamesport area ranged from undetected to 180 ^g/L;
results are given in Table 1. At 14 of the test well screens'
settings, and at 6 of the older wells, samples were split for
analyses by several laboratories. Results of the split-sam-
ple analyses are given in Table 2. Percentage differences
among the laboratories ranged from -49 to +50. The
standard laboratories used in the determination of the per-
centage differences were those of the Union Carbide
Corp., and the SCDHS. The SCDHS procedure was
adopted by Union Carbide Corp., because it expedited the
analyses and provided similar results (Romine, 1984). Un-
ion Carbide Corp. analyses, however, generally showed
larger concentrations than the other laboratories' analy-
ses of the same samples.
The preceding discussion of aldicarb analyses by differ-
ent laboratories suggests that analyzing for very low con-
centrations (several /*g/L) of a complex constituent that
undergoes multiple degradations in the soil and ground
water presents difficulties in interpretation of the degree of
contamination. New York State's maximum allowable aldi-
carb concentration of 7 /*g/L allows for a conservative un-
certainty factor of 100 times the allowance (Baier and
Moran, 1981).
DISTRIBUTION AND MOVEMENT OF
ALDICARB WITHIN THE
GROUND WATER SYSTEM
Movement of Aldicarb through the Soil and
Unsaturated Zone
Before aldicarb's use on Long Island, its manufacturer
believed that it would not infiltrate much deeper than
about 1 m into the soil, its half-life was less than 7 days,
and it would degrade to undetectable levels within 21 days
after application (Union Carbide Corp., 1975). Thus, aldi-
carb was not expected to leach into the ground water.
However, its discovery in large concentrations in the upper
glacial aquifer in 1979 required reexamination of its leach-
ing and degradation characteristics.
103
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 1.—Aldlcarta analyses of water from wells In the Jamesport vicinity, town of Riverhead,
Suffolk County, N.Y., 1980-82. (Well locations are shown In Fig. 3.)
Well
number
S3705
4620
8421
10256
10364
22855
31281
40407
47989
48944
51581
51581
51581
51581
51582
51582
51582
51582
51587
51587
51587
51589
51589
51589
71566
71566
71566a
71566b
71566C
71566d
S71567
71567a
71567b
71567C
71568
71568
71568a
71568b
71568C
71569
71569
71569a
71570
71570a
71570b
71571
71571
715713
71 571 b
71572
71572
715728
71572b
71572C
71572d
71572B
71573
71573
71574
71574
715748
71574b
71575
71575
71577
71577
715778
71577b
71577C
71578
71578
Depth of
well below
land
surface
(m)
35
35
33
32
16
29
13
43
26
5
13
do.
do.
do.
25
do.
do.
do.
24
do.
do.
12
do.
do.
8
do.
14
20
27
39
5
4
6
7
5
do.
11
37
46
10
do.
13
16
16
19
13
do.
19
22
17
do.
23
30
36
42
48
23
do.
33
do.
39
48
20
do.
18
do.
24
33
42
14
do.
Aldlcarb analysis
Use of
well
IRR
do.
do.
do.
do.
do.
DOM
FP
IRR
DOM
OBS
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
OBS
do.
TEST
do.
do.
do.
OBS
TEST
do.
do.
OBS
do.
TEST
do.
do.
OBS
do.
TEST
OBS
TEST
do.
OBS
do.
TEST
do.
OBS
do.
TEST
do.
TEST
do.
do.
OBS
do.
do.
do.
TEST
do.
OBS
do.
do.
do.
TEST
do.
do.
OBS
do.
Date
sampled
8-16-82
do.
do.
8-24-82
7-29-82
7-7-82
8-31-82
8-20-82
do.
8-31-82
3-17-80
6-29-81
2-1-82
8-3-82
3-17-80
6-29-81
2-17-81
8-3-82
3-17-80
7-7-81
8-3-82
3-17-80
7-7-81
8-3-82
1-27-82
8-3-82
1-27-82
do.
do.
do.
8-3-82
1-26-82
2-18-82
1-26-82
1-7-82
8-4-82
1-7-82
do.
1-6-82
1-27-82
8-4-82
1-27-82
8-4-82
2-18-82
1-22-82
1-19-82
8-10-82
1-19-82
do.
2-8-82
8-10-82
2-28-82
do.
2-28-82
do.
do.
7-1-82
8-10-82
2-16-82
8-10-82
2-16-82
do.
6-30-82
8-11-82
11-9-82
8-12-82
11-9-81
do.
do.
2-2-82
8-5-82
Concentration
G»g/L)
5.2
9.2
5
12
18
14
<1
<1
<1
140
48
33
ND4
<1
<1
ND
13
ND
63
33
3.6
<1
<2
<2
180
35
35
27
3
< 1
< 1
150
1
< 1
3.2
3
6
65
29
1
3
1
1
< 1
1.3
< 1
1.7
< 1
< 1
ND
do.
46
< 1
5
1.4
Analyst
A
A
A
A
A
A
B
A
A
A
C
D
E
C
D
E
C
D
C
D
E
A
B
E
E
E
A
E
E
B
B
A
E
E
E
B
A
B
A
E
B
B
A
E
E
E
A
E
E
E
E
E
A
A
B
A
E
E
A
A
B
A
B
D
D
B
A
Approximate
depth of
well below
water table
(m)
16
18
16
16
12
16
12
41
12
4
5
5
5
5
8
8
8
8
8
8
8
8
8
8
5
5
11
17
24
37
3
2
4
5
4
4
10
36
45
4
4
7
8
9
11
5
5
11
13
5
5
11
18
24
30
37
11
11
16
16
22
31
2
2
5
5
11
20
29
7
7
104
-------
GROUND WATER QUALITY
Table 1.—Aldlcarb analyses of water from wells In the Jamesport vicinity, town of Rlverhead,
Suffolk County, N.Y., 1980-82. (Continued)
Well
number
S71578a
71578b
71578c
71579
71579
715793
71579b
71579C
71580
715803
71580b
71581
71581
72840
72840
72841
72842
72843
72843
72844
72844
72845
72845
72846
72846
72847
72847
72848
72848
73272
Depth of
well below
land
surface
(m)
20
27
36
8
do.
14
40
44
8
7
10
11
do.
14
do.
12
10
12
do.
18
do.
23
do.
20
do.
14
do.
23
do.
25
Aldlcarb analysis
Use of
well
TEST
do.
do.
OBS
do.
TEST
do.
do.
OBS
TEST
do.
OBS
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
do.
DOM
Date
sampled
2-2-82
do.
do.
12-30-81
8-4-82
12-30-81
do.
do.
8-6-82
2-17-82
do.
7-2-82
8-5-82
6-21-82
8-3-82
6-22-82
8-30-82
6-23-82
8-4-82
7-2-82
8-10-82
6-28-82
8-12-82
6-30-82
8-5-82
6-23-82
8-4-82
7-7-82
8-11-82
8-30-82
Concentration
(M9/L)
< 1
< 1
< 2
< 1
< 2
< 1
< 1
28
37
18
19
ND
< 1
4.7
< 1
2
37
55
25
2
ND
do.
do.
1.1
140
43
Analyst
E
E
E
B
A
E
E
E
A
B
E
A
A
A
A
A
B
A
A
A
A
A
A
A
A
A
A
A
A
B
Approximate
depth of
well below
water table
(m)
13
20
29
5
5
11
36
41
4
4
7
8
8
11
11
10
9
6
6
10
10
10
10
13
13
10
10
9
9
7
Use of well: DOM. domestic; FP, fire protection; IRR, Irrigation; OBS, observation; TEST, water sample obtained and well screen raised to other depth (TEST well Is
at the site of the numbered well).
Analyst: A, U.S. Geological Survey; B, split-sample analyses made, see table 2; C, U.S. Environmental Protection Agency laboratory, from oral communication,
S.V. Gary, SCDHS, 1982; D, H2M Corp., from oral communication, Richard Market, SCDHS, 1982; E, H2M Corp., MelMlle, N.Y.
Table 2.—Results of split-sample analyses for aldlcarb by different laboratories, November 1981 through August 1982.
(Well locations are shown In fig. 3.)
Concentration (/ig/L)
Well No.
S31281
51581
51582
51587
51589
71566a
71567C
71568
71569
71569a
71570
71571
71574
71577
715773
71578
71579
71580
72842
73272
Date of
sample
8-3-82
8-3-82
8-3-82
8-3-82
8-3-82
1-27-82
1-26-82
1-7-82
1-27-82
1-27-82
1-22-82
1-19-82
2-16-82
11-9-81
11-9-81
2-2-82
12-30-81
2-17-82
8-30-82
8-30-82
Depth of
well below
land
surface
(m)
13
13
25
24
12
14
7
5
10
13
19
13
33
18
24
14
8
7
10
25
Analyst
USGS
2.1
30
< 1
55
76
ND
18
87
40
ND
do.
41
2.2
35
< 1
ND
do.
33
69
< 1
ucc
2.
—
—
—
—
13
35.
137.
65.
ND
do.
61.
ND
54.
1.
3.
ND
54.
74.
5.
H2M
3.
—
—
—
—
<1.
34.
—
—
<1.
—
—
1.
45.
<1.
<1.
<1.
—
44.
4.
SCDHS
—
49.
<1.
73.
77.
—
—
—
—
—
—
—
—
—
—
—
—
—
—
a
Approximate
depth below
water table
(m)
12
5
8
8
8
11
5
4
4
7
11
5
16
5
11
7
5
4
9
7
•A concentration of 9 ftg/L was analyzed in July 1982 in a sample provided by the well owner.
Analyst: USGS, U.S. Geological Survey, Ooraville, Ga.. UCC, Union Carbide Corp., Research Triangle Park, N.C., H2M, Holzmacher, McLendon & Murrell Corp.,
Melville, N.Y, SCDHS, Suffolk County Department of Health Services, Hauppauge, N.Y
105
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Aldicarb is highly mobile in the sandy deposits of Long
Island's upper glacial aquifer, and its degradation is re-
tarded by low ground water temperatures, lack of organic
matter in the soil, and acidic conditions in the environ-
ment. At temperatures typical for eastern Long Island,
where the mean air temperature for the growing season is
about 18°C (Mordoff, 1949) and the water temperature in
the upper glacial aquifer ranges from 6 to 15°C (Soren,
1977), the solubility of aldicarb and its sulfone are each
less than 1 percent, and that of the sulfoxide is about 30
percent (Union Carbide Corp., 1975), nearly the same as
sugar and common table salt.
The ground water of Long Island is acidic in most
places. Water collected in 1973-75 from 137 wells tapping
the upper glacial aquifer from the town of Brookhaven
eastward had a pH range of 5.1-7.8; only six of the analy-
ses showed pH values of 7 or higher (Soren, 1977). Seven
of the samples in that report were in the Jamesport vicinity
and had a pH range of 5.2-5.9.
The leaching potential of aldicarb's degradation prod-
ucts to the ground water on Long Island is enhanced by
the high permeability of the unsaturated zone and by the
general lack of organic material. The high hydraulic con-
ductivity of the deposits provides considerable vertical
and lateral mobility of dissolved constituents, and the acid-
ity of the ground water environment (commonly lower than
pH 6) decreases aldicarb's rate of degradation. Heavy rain
or too much irrigation shortly after application would ac-
celerate the downward movement of aldicarb and its deg-
radation products, especially the more soluble sulfoxide,
to the water table. More than one-third of the mean annual
precipitation on eastern Long Island (460-510 mm) falls
during the growing season, mostly in the spring (Mordoff,
1949); therefore, the leaching potential of the aldicarb is
probably greatest at the times of its application in the
growing season.
Water moving from the land surface to the water table
travels most quickly when soil moisture requirements in
the unsaturated zone above the water table are satisfied
and significant precipitation or too heavy irrigation occurs.
The rates of infiltration to the water table beneath re-
charge basins on Long Island under these conditions have
ranged from 1 m/h in winter to as high as 2.1 m/h in sum-
mer (Seaburn and Aronson, 1974). Thus, rate of flow to
the water table below land surface in the farmed areas of
Long Island can be rapid. Because aldicarb is soluble and
not strongly adsorbed by soil and rock particles (Jones,
1983), it probably moves at the same rate as the water in
which it is dissolved. In the farmed parts of the Jamesport
area, depths to the water table range from less than 3 m to
about 18m. At an infiltration rate of 0.9 m/h, aldicarb
could have reached the water table within a day or several
days after application, if soil moisture requirements were
satisfied and if heavy precipitation or too heavy irrigation
followed application.
Not all aldicarb in the unsaturated zone is transmitted to
the water table immediately; some of it lags in transit
through the unsaturated zone, but eventually it is carried
to the water table with recharge. Sand samples were col-
lected from the unsaturated zone to a depth of about 26 m
at the Cornell University Research Farm (9.7 km west of
the Jamesport area) in November 1980 (more than a year
after the last aldicarb application) and analyzed for aldi-
carb (Trautmann and Hughes, 1983). These samples were
reported to contain aldicarb concentrations ranging from
less than 5 ppb to about 10 ppb. In that study, the detec-
tion limit for aldicarb in soil was stated to be 5 ppb, and the
concentrations in the soil samples were equivalent to as
much as 300 pg/L in water. Sand samples collected by
Hughes and Porter (1984) from the unsaturated zone in
1983 in the Wading River area (17.5km west of the
Jamesport area) at depths from about 9 m-32 m, showed
aldicarb concentrations ranging from 2 to 25 ppb and in-
creasing with depth. Most concentrations were 10 ppb or
less, however.
The depth distribution of aldicarb in ground water in the
Jamesport vicinity is depicted in a hydrogeologic section
in Rgure 4. The maximum depth of penetration below the
water table in 1982 was generally about 12 m (Fig. 4). The
analyses in Table 1 show that 67 percent of the sample
collected from depths of 2-12 m below the water table had
aldicarb concentrations greater than 7 ^g/L. From depths
of 12-45 m below the water table, water from 16 percent
of the wells at 25 sites had aldicarb concentrations greater
than 7 pg/L, and all samples exceeding 7 pg/L were from
irrigation wells that produce large drawdowns in the water
table—commonly as much as 6 m. All samples indicating
aldicarb at depths greater than 12m below the water table
were from irrigation wells or test wells close by. Concentra-
tions and depths of aldicarb penetration below the water
table are summarized in Table 3.
The highest aldicarb concentrations in the Jamesport
area in 1981-82 were at depths of less than 3-9 m below
the water table. The weighted average of the reported
concentrations in Table 1 in the top 12m of the ground
water body is 26 /ig/L. Of the 37 samples that exceeded
the weighted average, only one was from less than 3 m
below the water table, and only three were from below
9 m. The traces to small quantities of aldicarb shown in
and below the clay beds in Figure 4 are attributed to con-
tamination during drilling operations, discrepant labora-
tory analyses, or both.
The depth of aldicarb infiltration in the Jamesport vicin-
ity, as described, seems typical of all potato farming areas
in eastern Suffolk County. Baier and Robbins (1982a) re-
port similar depths of infiltration in most other affected
areas of the county.
inug/L at OapBi indicated: m*i*»r» in paranthaaa* aiatplM-aarnpla
rwau'U. Whai* (nor* than ona ana>v«B <• aliown ml a dopth. m>*t
Figure 4.—Distribution and concentrations of aldicarb with
depth along a north-south section In the Jamesport area In
1982.
106
-------
GROUND WATER QUALITY
Table 3.—Total aldlcarb concentrations In relation to depth below the water table In the Jamesport area, Suffolk County,
N.Y., 1980-82.
(Data summarized from Table 1.)
Depth below
water table
On)
<3to6
>6to9
>9to12
> 12 to 15
>15to18
> 18 to 21
>21 to 24
> 24 to 27
>27to30
> 30 (to 45)
Number of
analyses
40
33
28
4
11
2
3
—
3
8
Aldlcarb concentration (/ig/L)
Range
NO to 180
NDtO 140
ND to 55
NO to 6
NDto 14
ND
NDtOl
—
1 to 5
NDtol
Mean
42
24
6
1.5
5
0
1
No samples obtained
3.5
0
Median
35
3
1
ND
3
0
1
—
1
0
Aldlcarb concentration: NO, not detected, considered to be 0. (Concentrations less than 1 pg/L are considered as not detected; concentrations less than 2 /»g/L
are considered as 1 for practical purposes in this table.)
The north-south section in Figure 4 indicates the great-
est depths of penetration to be near irrigation wells near
the ground water divide that are screened deeper than
12m below the water table. The depth of penetration is
generally less near the south shore, probably as a result of
the diminishing downward gradients away from the
ground water divide and the upward flow paths in
nearshore discharge areas.
Residence Time of Aldicarb in the Upper
Glacial Aquifer
The duration of aldicarb contamination in the upper glacial
aquifer of eastern Long Island is imprecisely known. Aldi-
carb's rate of degradation (commonly called half-life) var-
ies with environmental conditions (unlike the half-life of
radioactive decay, which is constant). For example, aldi-
carb was determined not to be significantly harmful to
ground water when used properly in Florida. Although its
use there was banned in January 1983 when it was dis-
covered at some agricultural sites, the ban was lifted in
January 1984, because environmental conditions were
determined to be conducive to rapid degradation (Environ.
Sci. Techno!., 1983). The higher temperature and pH of
the ground water in Florida, and probably the presence of
organic matter in the fossiliferous calcareous rocks, con-
tribute to aldicarb's rapid degradation. By contrast, the
rocks that form the aquifers on Long Island are mainly
silicate, with little or no organic matter below a thin surface
soil zone, and the ground water temperature and pH are
significantly lower.
Aldicarb's half-life on Long Island has been estimated in
terms of years and decades rather than days and months,
as in Florida. The pesticide's manufacturer estimates that
the half-life is 2-3 years (Jones, 1984), which would cause
aldicarb concentrations to decrease to below 7 jtg/L by
about 1990. A study by Cornell University (Trautmann et
al. 1983a) estimates that aldicarb's half-life (averaging
best and worse cases with half-lives of 3 years to infinity) is
10 years; that report also states that a 10-year half-life
would decrease aldicarb concentrations to below 7 /ig/L
by 2010. (We compute the year to be about 2030 at the
10-year half-life estimate.) The Cornell University investi-
gators, however, are reevaluating aldicarb's half-life on
Long Island (Trautmann et al. I983b).
Most of the aldicarb in the upper glacial aquifer of east-
ern Long Island will degrade to harmless compounds be-
fore it reaches saline ground water near the shorelines,
because the rate of ground water movement is very slow.
Several decades to as much as several centuries would
be required for water to move from the ground water di-
vides of the north and south forks to the shores (Fig. 5).
The retarding effect of ground water pumpage, mainly
from many irrigation wells, together with the slow rate of
ground water movement toward the shore, will prevent
most of the aldicarb from leaving the aquifer before it
degrades to concentrations below 7 /*g/L. As long as the
aldicarb remains in the ground water system, it will move
downward at rates ranging from 1.5 to 1.8 m per year near
the ground water divide and at lesser rates away from the
divide; it will also move laterally with the area! ground
water flow pattern. Until all aldicarb concentrations de-
cline to below 7 pg/L, concentrations at individual wells
probably will fluctuate widely according to patterns of
ground water movement from natural conditions and
ground water pumpage.
Figure 5.—Average water table configuration in the
Jamesport area, 1975-82, and selected paths of ground
water flow near the water table showing times of travel from
the ground water divide to discharge points.
107
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
REFERENCES
Baier, J.H., and D. Moran. 1981. Status report on aldicarb con-
tamination of groundwater as of September 1981. Suffolk
County Dep. Health Serv., Hauppauge, N.Y.
Baier, J.H., and S.F. Robbins. I982a. Report on the occurrence
and movement of agricultural chemicals in groundwater—
north fork of Suffolk County. Suffolk County Dep. Health Serv.,
Hauppauge, N.Y
1982b. Report on the occurrence and movement of
agricultural chemicals in groundwater—south fork of Suffolk
County. Suffolk County Dep. Health Serv., Hauppauge, N.Y
Environmental Science and Technology. 1983.17(11): 513A.
Hansen, J.L., and M.H. Spiegel. 1982. Personal communication.
Union Carbide Corp., South Charleston, W.Va.
Hughes, H.B.F., and K.S. Porter. 1984. Interim results tracking
aldicarb residues in Long Island ground water. Cornell Univer-
sity Center for Environmental Research. Ithaca, NY
Jones, R. 1983. Personal communication. Union Carbide Corp.
December 14,1983.
Jones, R. 1984. Personal communication. Union Carbide Corp.
January 26,1984.
Moran, D. 1984. Suffolk County Department of Health Services.
Mordoff, R.A. 1949. The climate of New York State. Bull. No. 764
(reprinted 1953). Cornell Univ. Ext., Ithaca, NY.
Romine, R.R. Personal communication, January 4,1984. Union
Carbide Corp.
Seaburn, G.E., and D.A. Aronson. 1974. Influence of recharge
basins on the hydrology of Nassau' and Suffolk Counties,
Long Island, New York. U.S. Geolog. Surv. Water-Supply Pap.
2031.
Soren, J. 1977. Ground water quality near the water table in
Suffolk County, Long Island, New York. Long Island Water
Resour. Bull. 8. Suffolk County Dep. Environ. Control, Haup-
pauge, NY.
Soren, J., and W.G. Stelz. 1985. Aldicarb-pesticide contamina-
tion of ground water in eastern Suffolk County, Long Island,
New York. U.S. Geolog. Surv. Water Resour. Invest. Rep. 84-
4251.
Trautmann, N.M., and H.B. Hughes. 1983. Personal communi-
cation. Cornell Univ. Center Environ. Res.
Trautmann, N.M., K.S. Porter, and H.B. Hughes. 1983a. Protec-
tion and restoration of ground water in Southold, N.Y. Cornell
Univ. Center Environ. Res., Ithaca, NY.
I983b. Southold demonstration site, New York State
fertilizer and pesticide demonstration project. Cornell Univ.
Center Environ. Res., Ithaca, NY.
Union Carbide Corporation. 1975. Technical information, TEMIK
aldicarb pesticide. Salinas, Cat.
108
-------
NONPOINT SOURCE CONTAMINATION OF GROUND WATER IN
KARST-CARBONATE AQUIFERS IN IOWA
GEORGE R. HALLBERG
ROBERT D. LIBRA
BERNARD E. HOYER
Iowa Geological Survey
Iowa City, Iowa
ABSTRACT
There are two components of ground water recharge in
karst-carbonate aquifer systems: (1) conventional infiltra-
tion, and (2) the direct entry of surface water through
sinkholes. Thus, in karst areas the better known prob-
lems of nonpoint source pollution, associated with the
runoff of sediment and chemicals from agricultural land
into surface waters, directly merge with the ground water
system and the poorly-understood problems of the infiltra-
tion of agricultural chemicals. Three years of detailed wa-
ter-quality monitoring and water, mass-balance studies
show: (1) during major surface-runoff events high con-
centration of suspended sediment, pesticides, and bacte-
ria enter the ground water and moye as a "slug" through
the carbonate aquifer, creating brief, but acute water
quality problems; but, (2) over a water year the infiltration
component delivers to ground water the greatest mass
and highest concentrations of NO3, and the greatest
mass of the pesticides detected. Many of the more widely
used herbicides are detected commonly in ground water,
and are now present year-round. The amount of NO3-N
discharged in ground water and surface water per year,
from a 267 km3 study basin, has equaled about 30-50
percent of the fertilizer-N applied, an economic as well as
an environmental concern.
INTRODUCTION
Since 1980 the Iowa Geological Survey, in conjunction
with other State, Federal, and local agencies, and univer-
sity researchers, has been conducting studies of ground
water quality problems in northeastern Iowa. During the
past decade, reports of problems associated with nitrate
and bacterial ground water contamination have increased
in this area. Although these reports are, in most cases,
unquantified, they have been made by experienced well
drillers, well owners, and dairy operators, and records
from public water supplies throughout northeastern Iowa
support them. The underlying rocks of northeastern Iowa
are predominantly composed of limestone and dolomite,
or carbonate rocks, which serve as aquifers and are the
primary source of ground water drinking water.
These carbonate aquifers are mantled by Quaternary
glacial and fluvial deposits ranging from zero to over
150 m in thickness. In over half of the region (17,600 km2),
these deposits are less than 15m thick. In some areas,
where the Quaternary deposits are less than 8 m thick, a
karst topography—with sinkholes, sinking streams, and
blind valleys—is well expressed (Hallberg and Hoyer,
1982). Such karst features swallow surface runoff water—
and the contaminants contained in the water—and allow it
to enter the carbonate-aquifer ground water system. This
run-in water may move rapidly through conduits in the
fractured carbonate rocks.
In over 6,000 analyses of ground water from private
wells in the region, nitrate contamination was associated
with the areal geologic setting. Both karst areas and non-
karst regions, where the aquifers were covered by less
than 15 m of glacial deposits, showed significant nitrate
contamination (Hallberg and Hoyer, 1982). In areas where
the aquifers were more deeply buried (> 15 m to 150 m),
essentially no nitrate (<1 mg/L NO3-N) occurred in the
aquifer (Hallberg and Hoyer, 1982; Hallberg et al. 1983a,
b).
BIG SPRING BASIN STUDY
Detailed studies were undertaken to further define the
nature of this ground water contamination. Some of the
most enlightening data come from the Big Spring basin
study. The Big Spring ground water basin includes
267 km2 in Clayton County, Iowa, which drains to the Tur-
key River. It was chosen for study because of prior studies
in the area, local concern with water quality problems, and
because State-owned structures built at a trout hatchery
afforded the rare opportunity to gage ground water dis-
charge at Big Spring (a large, carbonate ground water
spring).
An extensive data base was developed to define the
hydrogeology, soils, and land use of the basin. A few items
of particular interest are described here (see Hallberg et
al. 1983a, 1984a). The ground water basin has been de-
fined by studying the potentiometric surface in the Galena
aquifer, by dye-trace studies, and through assessment of
gaining and losing stream reaches. These data, combined
with spring and stream gaging, show that Big Spring ac-
counts for 85-90 percent of the ground water discharged
from the basin. Land use mapping and inventories show
that the basin is wholly agricultural (with no significant
industries, ag-chemical facilities, etc.); 91 percent of the
area is used for corn, pasture, or hay, planted in rotation,
in part to support the common dairy and livestock opera-
tions. Except during the PIK program in 1983, approxi-
mately 50-60 percent of the area has been planted in corn
since 1979. With the assistance of the USDA Soil Conser-
vation Service and Iowa State University's Cooperative
Extension Service, ag-chemical use also has been moni-
tored.
Since 1981 the discharge from the Big Spring basin has
been gaged. Water quality sampling also has been con-
ducted from a network of surface water sites, tile lines,
wells, the Big Spring, and other small springs. Sampling
intensity at Big Spring has varied for different constitu-
ents, in relation to research goals and hydrologic condi-
tions. For NO3 and pesticides, sampling has been con-
ducted weekly, and during some events hourly, or even
more frequently since the 1982 water year. All water analy-
ses have been performed using standard methods by the
University Hygienic Laboratory, which has an EPA-ap-
proved QA/QC plan.
Anatomy of a Runoff Event
The discharge at Big Spring has varied from 0.9 cms (cu-
_bic meters/sec; 30 cfs) to peaks of over 7 cms (250 cfs).
109
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
During snowmelt or rainfall-runoff events, the ground wa-
ter discharge has increased from 1 to 7 cms in 30 to 40
hours, as the runoff-run-in water moves rapidly through
conduits in the ground water system.
Figure 1 shows a discharge hydrograph for a large
event during the summer of 1983. Approximately 230 mm
of rain fell intermittently during a 3-day period, producing
significant infiltration recharge and generating a sequence
of three runoff events which, in turn, caused three abrupt
rises of discharge at Big Spring (Fig. 1). The third event
produced the highest discharge, which reached 7.1 cms
(250 cfs) on July 1. Prior to these rains, ground water was
undergoing slow, base flow recession, and the discharge
was 1.4 cms (50 cfs). During these events, water chemis-
try was intensively monitored at Big Spring and other net-
work sites, including surface waters that discharge into
the sinkholes.
Figure 1 also shows water quality data superimposed
on the hydrograph. These data illustrate the effects of the
surface run-in wafer on ground water quality. During the
abrupt discharge rises, the ground water chemistry
changes in response to the mixing of different sources of
recharge wafer: run-in versus infiltration. Run-in provides
constituents typical of surface runoff, such as sediment
and chemicals of low mobility; infiltration provides constit-
uents more typical of ground wafer, including various mo-
bile, dissolved ions such as NO3. A few examples illustrate
this.
Suspended sediment increases from negligible values
to concentrations over 4,000 mg/L (a load of over 87,000
kg/hr) as the run-in component discharges at the spring.
In marked contrast is the change in specific conductance
(SpC), which integrates the effects of all dissolved ions.
During stable base flow (before and after the discharge
events), when the ground wafer quality must be controlled
by infiltration processes, the SpC is between 700 and 730
^mhos/cm2. During the major discharge peak, as sedi-
ment from the run-in increases, the SpC drops sharply
from 670 to 450 /^mhos/cm (Fig. 18), reflecting the dilution
of infiltration-derived ground water by the surface run-in
water, which has very low concentrations of dissolved con-
stituents.
Similar changes can be seen in the concentrations of
pesticides and NO3, whose maximum concentrations are
related to the run-in and infiltration components, respec-
tively. Afrazine is the most widely used herbicide in the
basin, and it is the dominant pesticide found in the ground
wafer. During base flow, prior to the discharge events (Fig.
1C, June 24-26), atrazine concentrations were about
0.2 ftg/L. During the discharge events, three peak periods
of afrazine and total pesticide concentrations occurred,
corresponding to the three discharge events, and the
three periods of run-in recharge. The first two pesticide
concentration peaks are out of phase with the discharge
peaks because of time lags between discharge rises and
the arrival of wafer quality changes at these lower dis-
charges (Hallberg et al. 1984a); the third and highest pes-
ticide peak roughly coincides with the peak of suspended
sediment. During the influx of run-in recharge, several
other pesticides also occur in the ground water. The vari-
ous pesticides and their maximum concentrations (^g/L)
were: atrazine, 5.1; cyanazine, 1.2; alachlor, 0.6; meto-
lachlor, 0.6; and fonofos, 0.1 nQ/L. In Big Spring basin
ground wafer afrazine is the only pesticide that persists
year-round, though alachlor and cyanazine have ap-
peared intermittently during ground water base-flow peri-
ods.
Concentrations of NO3 show a complex record (Fig.
1D), which is almost directly out of phase with the pesti-
cides, with minima occurring when run-in components
reach maxima. During peak runoff, the ground surface
» » a 'ao
r
SL,'
Figure 1.—Ground water dlschairgo from @ig Spring basin
(gray line; scale on right) shown with (block linos) A. sus-
pended sediment; El. speclfllc conductance; C. atnsizlne and
total pesticide; and D. nitrate concentrations, during sum-
mar 1883, dischairg® event (dram Hallberg at al. 1884o).
wafers running into sinkholes generally contained less
than 1 mg/L NOrN- During discharge recession (Fig. 1,
from July 3), when infiltration recharge dominates, the
NO3 concentrations rise dramatically to values of 16.5 mg/
L NO3-N (75 mg/L NO3) for a short time and then recess
back to base flow concentrations of 10-11 mg/L NCVN.
Other water quality parameters show the same effects
and vary in relation to their recharge delivery mechanism.
Parameters generally associated with surface water runoff
(e.g., K, PO4, Fe, organic-N, ammonium-N) are found in
their highest concentrations, or only occur in conjunction
110
-------
GROUND WATER QUALITY
with the high sediment and pesticide concentrations and
run-in water. Also, concentrations of bacteria in wells peak
during these periods. Other water quality parameters, typi-
cally associated with slower infiltration recharge and dif-
fuse flow (e.g., Cl, SO4) coincide with SpC and NO3.
Detailed monitoring during such rainfall-run-in events
at Big Spring shows that the high surface water concen-
trations of parameters such as suspended sediment and
pesticides enter the ground water, and move as slugs
through the system. They discharge from the ground wa-
ter in essentially the same concentrations in which they
entered, at least during large hydrologic events.
Quantitative Assessment of Chemical
Recharge
The detailed analysis of such events not only demon-
strates what happens to ground water quality during a
discharge event in the unique karst setting, but also af-
fords a calibration for quantitative estimates of the relative
delivery of recharge water and contaminants by run-in
versus infiltration. A ready analogy can be made between
the Big Spring ground water hydrograph and a humid
climate, surface stream hydrograph: the large peaks in
the ground water discharge are related to surface water
run-in and rapid flow through conduits, while in a stream
this is storm or flood flow. The more gradual, falling, or
recession limb, and the prolonged, stable portions of the
hydrograph are related to typical ground water compo-
nents, infiltration recharge, and bank storage effects in a
stream.
Based on this analogy, various hydrograph separation
techniques, typically used in surface water hydrology
(e.g., Singh and Stall, 1971), can be adapted to these
ground water hydrographs. Just as these methods are
used for quantitative assessment of flood flow and base
flow in a stream, we can adapt them to provide a quantita-
tive estimate of the run-in conduit flow component of re-
charge and the infiltration-diffuse flow component (see
Hallberg et al. 1983a, 1984a). Several analytical and
chemical methods (e.g., Freeze and Cherry, 1979) have
been used to check on the consistency and range of de-
rived values. The various methods produce different
results that provide insights into hydrologic behavior. The
methods have produced quantitative results that are quite
consistent (± 10 percent) and compare quite well with the
actual chemical monitoring data. This allows some refined
conclusions.
Synthesis of 3 water years' data shows that the infiltra-
tion component contributes about 90 percent of the water,
about 95 percent of the NO3-N, and 50-85 percent of the
pesticides. The run-in component delivers about 10 per-
cent of the water, only about 5 percent of the nitrate, and
from 15-50 percent of the pesticides. While the run-in
component delivers to the ground water contaminants of
concern for public health on the local level, the infiltration
component is responsible for regional aquifer contamina-
tion. In summary, these studies show that the infiltration
component delivers to ground water the largest mass and
the highest concentrations of nitrates (and other mobile
ions), and the largest mass of mobile pesticides, but gen-
erally in low concentrations. The run-in component deliv-
ers to ground water high concentrations and large loads of
pesticides and other relatively insoluble or highly ad-
sorbed chemicals, peak turbidity and sediment loads, and
other peak loads of organic and pathogenic organisms,
but only for short periods of time.
In the past, most water quality problems in the karst
areas have been attributed to sinkholes and run-in. These
analyses, however, show that simple infiltration is the ma-
jor mechanism of contaminant delivery, even in a karst
area. Infiltration is the recharge mechanism common to all
aquifers, and it gives these data much broader implica-
tions. These implications clearly affect the type of agricul-
tural management decisions needed to mitigate these
problems.
Annual pesticide losses exhibit more variability because
the pesticide concentrations often vary by one or two or-
ders of magnitude between infiltration, base flow (0.01-
1.0 /ig/L), and the run-in-conduit flow water (1.0-10.0
/tg/L). On a water year basis, the differences in percent-
ages of pesticide contributions varied from wet to dry
years in relation to the number of significant runoff events.
DELIVERY OF CHEMICALS THROUGH
INFILTRATION
Other studies support the conclusion that infiltration is the
primary delivery mechanism of ag-related contaminants—
including pesticides—to ground water. In Floyd and Mitch-
ell counties (Libra et al. 1984), another area of regional
carbonate aquifers, the geologic region that shows the
highest concentrations of NO3 and pesticides in the
ground water has no open sinkholes. It is an intensive row-
cropped area (corn and soybeans) where the aquifer oc-
curs at 1-4 m in depth; the land is relatively flat and
marked by high rates of infiltration. The ground water-
drinking water quality in this area—even in wells open to
significant depths below the surface—is very similar to
that found in tile lines from other row-cropped areas.
Statistical review of water quality data from all across
Iowa shows that shallow wells (less than 15 m deep in
particular) across the State are exhibiting elevated con-
centrations of nitrate. In some alluvial aquifer settings,
denitrification is removing the nitrates, but pesticides are
still infiltrating to (and persisting in) the ground water.
In the spring of 1984 over 40 public water supplies in
Iowa exceeded the drinking water standard for NO3.
These supplies tapped a variety of sources, including sur-
face waters, but even here the high NO3 levels in these
streams is related to ground water base flow contribution
and infiltration recharge. Residues of surface-applied
chemicals may be stored in the soil until infiltrating water
carries them downward into the ground water. Because
infiltration is the principal component of recharge, the tim-
ing of nitrate fluctuations in water supplies is related to
seasonal recharge periods, and not to the timing of sea-
sonal agricultural practices. This is why nitrate concentra-
tions in wells and streams increase during spring re-
charge, often many weeks before N-fertilizers are applied.
For example, in the Big Spring basin, the concentration
and the monthly load of nitrate is highly correlated with the
total ground water discharge (Fig. 2).
Pesticides in Ground Water
Among the unexpected findings of our recent studies in
Iowa is the fact that many of the herbicides most com-
monly used in Iowa occur commonly in ground water, and
persist there year round. Table 1 shows the pesticides
found in ground water and the maximum concentrations
detected to date. (These values are derived from areas of
routine use. In isolated instances, from spills or siphoning
accidents other pesticides or higher concentrations have
been noted.) All the herbicides shown have been detected
in winter samples. Only one insecticide has been detected
in ground water. Fonofos was detected in low concentra-
tions in a few samples in a karst area, and these samples
were related to the run-in water component. To date, how-
ever, the analyses have tested only for parent compounds,
and it is not known if metabolites, or breakdown products,
of these or other pesticides are occurring.
111
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
70-, ®
60-
5O-
-,40H
1 30-
20-
10
®
6 5
"12©
10
8 - WY 1982
© - WY 1983
1.0
2.0 3.0 4.0 5.0 6.O
Total monthly Q; m3 x 10s
7.0
8.0
Figure 2.—Total monthly ground water discharge versus to-
tal monthly NOrN discharge at Big Spring (from Hall berg et
al. 1984a).
In our studies in northeastern Iowa, 70-80 percent of
the wells and springs sampled in karst, shallow bedrock,
and alluvial settings (in the latter two categories only infil-
tration recharge occurs) have shown detectable concen-
trations of some pesticides over the course of a water
year. The maximum concentrations noted on Table 1 are
associated with run-in recharge in karst areas (Big
Spring), or shallow bedrock, high infiltration conditions
(Floyd and Mitchell Counties). The more typical concentra-
tions of these pesticides tend to be much lower, generally
between 0.1 and 1.0 uglL.
Our findings show that, in Iowa, if NO3-N concentrations
greater than 2-3 mg/L occur in ground water samples,
probably the ground water will contain detectable pesti-
cides. However, the concentration of nitrate is not a very
good predictor of the absolute concentration of the pesti-
cides. Also, in some alluvial settings, denitrification is ap-
parently removing the NO3, but pesticides still persist
(Hallberg et al. 1984a).
MAGNITUDE OF CHEMICAL LOSSES
The monitoring of water discharge, water chemistry land
management, and chemical use in the Big Spring basin
allows some mass-balance calculations to be made. Table
2 summarizes 3 complete water years of monitoring. The
amount of NO3-N discharged with ground water and sur-
face water from the basin totaled about 820,000 kg-N to
1,300,000 kg-N, for near normal (1982 and 1984) and wet
years (1983) respectively. This equals approximately 50
and 80 kg-N/ha for the long-term corn acreage in the ba-
sin. These losses are equivalent to about 33-55 percent of
the average amount of fertilizer-IM applied (Table 3). This is
not to imply that all of this NO3-N is derived directly from
the fertilizer-N. However, the large losses occur in re-
sponse to the large amounts applied.
These are minimum figures for the amount of N lost,
because only the NO3-N losses have been computed.
Other forms of IM are also discharged with the water, and
losses by denitrification cannot be estimated. Comparison
with other regional and local studies suggest that these
losses are probably typical for Iowa and much of the Mid-
west under current management practices (e.g., Baker
and Johnson, 1977,1981; Gast et al. 1978; Kanwar et al.
1983; Hallberg et al. 1984a). Beyond the environmental
impact, the magnitude of the N-lorses is of economic con-
Table 1.—Maximum measured concentrations of pesticides
detected In groundwater In Big Spring basin and Floyd-
Mitchell county area, and annual (WY) flow-weighted
mean atrazine concentration at Big Spring. (WY 82-83
from Hallberg et al. 1983a,b; 1984a; Libra
etal. 1984).
Big Spring Basin
Herbicides
atrazine
alachlor
cyanazine
metolachlor
metribuzin
Insecticides
fonofos
Flow-wtd mean
atrazine, /ig/L,
at Big Spring
WY-82
2.5
0.2
0.7
—
—
—
0.181
WY-83
5.1
0.6
1.2
0.6
—
0.1
0.28
WY-84
n
10.0
4.0
1.7
4.5
—
0.3
0.45
Floyd/
Mitchell
Counties
WY-83
1.6
16.6
0.5
0.1
4.4
—
'Not present during all of WY-82; 0.31 pg"- for time present.
(Pesticide identifications based on electron-capture gas chromatography with dual
columns of different polarities, with periodic confirmation from additional GC/MS
analyses. Internal standards and calibrations used for the pesticides commonly used
in Iowa, including atrazine, alachlor, carbofuran, cyanazine, fonofos, metolachlor,
metribuzin, pendimethalin, phorate, terbufos, trifluralin, and chlorinated hydrocar-
bons. Those in italics are used in the Big Spring basin.)
Table 2.—Summary of 3 water years' discharge,
precipitation, and NO3-N discharge data from the Big
Spring basin.
Water-Year
1982' 1983' 1984
x106m3
58.5 63.4 52.5
42.9 85.9 45.1
101.3 149.2 97.6
Water yield
Total ground water discharge
Stream flow discharge
Total discharge
Precipitation and discharge
Precipitation
Water yield
A. Less change in gw storage
B. As % precipitation
NOj-N discharged
In ground water
In surface water
Total
Flow-wtd mean NO3-N concentration
In ground water
mm
864 1,130 833
332 559 353
33
----
478
343
49 42
613
687
466
351
821 1,300 817
mg/l
8.7 10.2 9.6
'(From Hallberg et al. I984a).
Table 3.—Summary of NO3-N loss from the Big Spring basin
In relation to land use for 3 water years; shown as kg-N/ha
equivalent.
19821 1983' 1984
kg-N/ha of total basin
kg-N/ha of long-term row crop (land in
corn rotation)
% of applied chemical-N
31 49 31
52 83 52
33% 53%2 33%
'Data from Hallberg et al. 1983a; 1984a.
2PIK year; % based on 1982 applied chemical-N.
112
-------
GROUND WATER QUALITY
cern as well. When such substantial amounts of N are not
being used for crop production obviously there is room to
improve efficiency and economic gain.
In contrast to the nitrogen losses, the total pesticide
losses are quite small. For example, the total amount of
atrazine discharged in ground water at Big Spring, during
the 3 water years, ranged from 7 kg (1982) to 18 kg
(1984). The loss of atrazine (and total pesticides) in ground
water equaled less than 0.1 percent of the amount ap-
plied. Pesticide concentrations in surface water were not
monitored in sufficient detail (because of costs) to warrant
calculating mass losses with surface water. However, pes-
ticide concentrations in surface waters commonly were 10
to 100 times greater than in ground water, and total pesti-
cide losses are estimated at between 1 and 5 percent of
the amount applied.
TEMPORAL CHANGES IN GROUND
WATER QUALITY IN NORTHEASTERN
IOWA
As noted, a variety of detailed information has been com-
piled for the northeastern Iowa area. Various data indicate
that the natural, or background, concentrations of nitrate
in the aquifers in northeastern Iowa were very low, gener-
ally less than 1 mg/L NOs-N. At Big Spring the nitrate
concentration in the ground water averaged about 3 mg/L
NOa-N during the 1950's up through about 1968. In 1981-
82, when the Iowa Geological Survey began detailed mon-
itoring, the NO3 concentration averaged 9 mg/L NOs-N,
and 10 mg/L NO3-N in 1983. Data from over 50 wells sam-
pled in surrounding counties during 1975 and 1983 show
the same rate of increase (Fig. 3A).
Data compiled by the Crop and Livestock Reporting
Service were used to evaluate changes in land use and
chemical use that took place in the basin during this time.
From the late 1960's to around 1980 the livestock popula-
tion increased about 30 percent, the corn acreage about
40 percent, and the N-fertilizer application rate about 80
percent. These data can be converted to the amount of N
applied in the basin from theyar[ous sources.
~ Using'standard assumptions, the amount of manure-N
is a direct function of the livestock population. Thus, ma-
nure-N increased about 30 percent. The corn acreage in-
crease and the N-fertilizer rate increase are additive fac-
tors; and thus, the amount of fertilizer-N applied in the
basin increased about 250 percent over the same period
that the NO3 concentration in the ground water increased
by about 230 percent. The maximum amount of N har-
vested with the corn can also be estimated from standard
formulas; this estimate is a maximum because the data
cannot be corrected for zero-N treatments. The increase
in ground water NO3 occurs after the amount of fertilizer-N
applied begins to greatly exceed the N removed in the
corn. All these data are graphically shown in Figure 3B.
The increase in NO3 concentration in ground water di-
rectly parallels the increase in the amount of fertilizer-N
applied in the basin. This same direct, linear response is
shown by numerous experiment farm studies (e.g., Gast
etal. 1978; Baker and Laflen, 1983; Hallberg et at. 1984a),
and this is the response that should be predicted in areas
with shallow aquifers, such as in the Big Spring basin. As
noted, there are no other significant nonagronomic activi-
ties in this basin to complicate the interpretation.
Currently NOa-N delivery into ground water appears to
be in relative balance with land treatment and the water
flux through the system (Fig. 2). Annual flow-weighted,
mean NOrN concentrations reflect this, rising and falling
with the ground water discharge, particularly the infiltra-
tion component (Table 2).
50-1
30-
20-
Big Spring Groundwoter
SAAn
1951 I960
1965
1 1 '
1970
Yeors
3,000-i
2,500-
2,000-
" 1,500-
£
1,000-
B.
Fertilizer-N
Maximum estimate of
amount of N harvested
• I • • • ' I • • • • I • • ' ' I '
I960 1963 1970 1975
Years
Figure 3.—A. Change In NO3 concentration versus time at
Big Spring, mean NO3 for a suite of 50 wells (solid square
and triangles), and two individual wells In northeastern Iowa
(dashed lines) (from Hallberg et al. 19846). B. Estimated
tons of fertilizer- and manure-nitrogen applied In the Big
Spring basin, and average nitrate concentration In ground
water at Big Spring (from Hallberg et al. 1983a). Estimated
maximum amount of N harvested with corn shown In gray
(from Hallberg et al. 1984a).
Pesticides in the ground water environment may pose a
problem in the future. Will pesticide concentrations, in
ground water rise over time, as nitrate has? Some evi-
dence suggests they are rising, although from the short
period of data collection we cannot draw conclusions.
Atrazine is the most widely used herbicide in the Big
Spring basin, accounting for over 40 percent of the herbi-
cides currently applied. It comprised a greater proportion
in the past. Since May 1982, it has been present year-
round in the ground water in the basin; it was not detected
in samples earlier in water year 1982. Unlike NO3, which
has varied with the ground water flux during the past 3
water years (Fig. 2), mean (and maximum) atrazine con-
centrations and loads in ground water have steadily in-
creased since 1981 (Table 1). These, and other data cur-
rently under review, suggest that pesticides in ground
water are increasing, perhaps more in terms of persist-
ence than concentration. Future monitoring must address
this question.
113
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
IMPLICATIONS
There are many broad-scale implications for nonpoint
source concerns from the studies in northeastern Iowa.
First, even if only the unique aspects of karst aquifers are
considered (e.g., the surface water run-in of contaminants
to ground water), there are broad regional implications.
Extensive agricultural areas overlie karst terrains from
Iowa to Missouri and Texas, to Florida and Puerto Rico, to
New York and Pennsylvania, to Indiana, Kentucky, and
Ohio. Second, even with the unique problems of these
areas, infiltration is the dominant mode of recharge for
water and chemicals into ground water, and infiltration is
the common denominator for all ground water systems. A
review of surface water, tile-effluent, and other agronomic
research suggests that similar chemical losses are occur-
ring throughout the corn belt, although in many areas
these shallow ground waters are not used as drinking wa-
ter.
These ground water quality problems related to agricul-
tural chemical use can only be resolved through a more
holistic approach to agricultural management. We must
couple our standard concerns for soil conservation and
surface water quality with the need to protect ground wa-
ter. Many standard approaches used to treat soil erosion
can increase the infiltration of chemicals, and new combi-
nations of many current practices may be needed. Better
chemical and nutrient management must play a part. The
magnitude of nitrogen losses, in particular, suggests that
better N-management may increase economic gains,
even if the costs of the impact on the environment are
ignored. Solving these problems will require a concerted
effort by all segments of the agricultural community to
gain the experience and data necessary to effect a satis-
factory balance between efficient agricultural production
and the protection of water supplies.
REFERENCES
Baker, J.L., and H.P. Johnson. 1977. Impact of subsurface drain-
age on water quality. Proc. 3rd Natl. Drainage Symp., Am.
Soc. Agric. Eng., St. Joseph, Mo.
Baker, J.L., and H.P. Johnson. 1981. Nitrate-nitrogen in tile
drainage as affected by fertilization: J. Environ. Qua). 10:
519-22.
Baker, J.L., and J.M. Laflen. 1983. Water quality consequences
of conservation tillage. J. Soil Water Conserv. 38:186-93.
Freeze, R.A., and J.A. Cherry. 1979. Groundwater. Prentice Hall,
Inc., Englewood Cliffs, NJ.
Gast, R.G., W.W. Nelson, and G.W. Randall. 1978. Nitrate accu-
mulation in soils and loss in tile drainage following nitrogen
application to continuous corn. J. Environ. Qua). 7: 258-62.
Hallberg, G.R., and B.E. Hoyer. 1982. Sinkholes, hydrogeology,
and ground-water quality in northeast Iowa. Iowa Geolog.
Surv. Open-file Rep. 82-3:1-120.
Hallberg, G.R., B.E. Hoyer, E.A. Bettis III, and R.D. Libra. 1983a.
Hydrogeology, water quality, and land management in the Big
Spring basin, Clayton County, Iowa. Iowa Geolog. Surv. Open-
file Rep. 83-3:1-191.
Hallberg, G.R., et al. I983b. Additional regional groundwater
quality data from the karst-carbonate aquifers of northeast
Iowa. Iowa Geolog. Surv. Open-file Rep. 83-1:1-16.
Hallberg, G.R., R.D. Libra, E.A. Bettis III, and B.E. Hoyer. I984a.
Hydrogeologic and water quality investigations in the Big
Spring basin, Clayton County, Iowa; 1983 water year. Iowa
Geolog. Surv. Open-file Rep. 84-4:1-231.
Hallberg, G.R., et al. 1984b. Temporal changes in nitrates in
groundwater in northeastern Iowa. Iowa Geolog. Surv. Open-
file Rep. 84-1: 1-10.
Kanwar, R.S., H.P. Johnson, and J.L. Baker. 1983. Comparison
of simulated and measured nitrate losses in tile effluent.
Trans. Am. Soc. Agric. Eng. 26:1451-57.
Libra, R.D., G.R. Hallberg, G.G. Ressmeyer, and B.E. Hoyer.
1984. 1. Groundwater quality and hydrogeology of Devonian
carbonate aquifers in Floyd and Mitchell counties, Iowa. Iowa
Geolog. Surv. Open-file Rep. 84-2:1-106.
Singh, K.P., and J.B. Stall. 1971. Derivation of base flow reces-
sion curves and parameters. Water Resourc. Res. 7(2): 292-
303.
114
-------
AN INTERDISCIPLINARY APPROACH TO SHALLOW GROUND
WATER CONTAMINATION IN NORTH-CENTRAL MONTANA
JANE M. HOLZER
JEFF D. FARKELL
BRIAN J. HARRISON
GLENN A. HOCKETT
Triangle Conservation District
Conrad, Montana
ABSTRACT
Shallow ground water quality over portions of the north-
ern Great Plains has degraded considerably over the past
20 to 40 years. Shallow wells in the affected areas have
shown a range in IDS from 20,000 to 50,000 mg/L. Al-
though the prevalent geologic and climatic conditions are
significant, the degradation is attributed to, for the most
part, the cultural farming practices of the region. The use
of a strict, alternate crop-fallow management practice
results in the inefficient use of annual precipitation and
allows for the downward migration of subsurface salts
into shallow ground water systems. In addition, under the
right geologic conditions, the salts may resurface and
destroy productive cropland in the form of saline seep. It
has been estimated that 113,300 ha (280,000 acres) of
cropland have been lost to this process in Montana. In
north-central Montana, a group of local landowners, in
cooperation with State and Federal officials, have formed
a technical team to address the problem of saline seep
and the degradation of shallow ground water quality. This
team has developed an intensive management plan that
reclaims salinized cropland and prevents further degra-
dation of ground water quality. This has significantly de-
creased the size of saline seeps in as little as 3 to 5 years.
In addition, the knowledge gained has led to potential use
of vegetative management as a reclamation tool for other
water quality problems such as acid mine drainage.
INTRODUCTION
A major resource problem in Montana is the growing sa-
line seep acreage. While the salinized areas first ap-
peared in the early 1950s, real concern and research on
the problem began in the last 10-15 years. Saline seeps
are recently developed low-volume saline springs in nonir-
rigated areas that are intermittently to continuously wet.
The saline soils have salt crusts and result in reduced or
nonproductive crop growth. It is estimated that over
113,300 ha (280,000 ac) of cropland are out of production.
Individual seeps range in size from .4 to 220 ha. The
growth rate is 10 percent per year, so that over the next 20
years the potential exists for 762,753 ha (1,884,000 ac) to
be affected, emphasizing the magnitude of the problem if
it is left unaddressed.
The original concern was loss of crop production and
decreased land value, but a more severe consequence
and concern now is the degradation of local surface water
and shallow ground water quality. The reclamation of the
land has been documented and the technique will be dis-
cussed in detail, but once the water quality is degraded,
reversal to potable limits may not occur within our lifetime.
Therefore, the prevention of saline seeps may be as im-
portant as the reclamation of existing ones. Water quality
in wells and ponds close to salinized areas has shown a
10-20 fold increase in total dissolved solids (IDS) over the
last 15 years. The water sources become unusable for
either human or livestock consumption, thus requiring
new "fresh" sources that may be both scarce and expen-
sive. Loss of a suitable water supply has restricted live-
stock production and can require domestic water supplies
to be trucked or piped. In many parts of the north-central
Triangle Area, the practice of drilling a new water well
every decade has become culturally ingrained. Any one of
these measures results in a significant cost to the individ-
ual consumer. The regionwide deterioration of shallow
ground water presents a real threat to the economic well-
being of rural Montana.
The Triangle Conservation District was formed by land-
owners in a 10-county area in 1979 to address the prob-
lems of saline seep and contaminated shallow ground wa-
ter. The District has an interdisciplinary staff with
backgrounds in agronomy, soil science, hydrogeology, na-
tive vegetation, and cropping systems. The staff works on
a farm-by-farm basis to achieve saline seep prevention
and reclamation, using state-of-the-art recharge area
identification and intensive cropping and reclamation
techniques. The District has a strong working relationship
with the agricultural experiment station, the Montana Bu-
reau of Mines and Geology, conservation districts, cooper-
ative extension service, U.S. Agriculture Research Serv-
ice, U.S. Soil Conservation Service, and U.S. Agriculture
Stabilization and Conservation Service. This paper will
detail the proven technique for the Triangle Area, which
has been refined for other areas in Montana, Alberta, and
Saskatchewan, all areas with varying geological and cli-
matic conditions.
HYDROGEOLOGY OF SALINE SEEP
DEVELOPMENT
The formation and development of saline seep under vari-
able conditions has been thoroughly discussed in pre-
vious works (Dodge et al. 1984; Brown et al. 1982; Miller et
al. 1981; Krall and Brown, 1976, Vander Pluym, 1978;
Halvorsen and Black, 1974; Bahls and Miller, 1973; Black
et al. 1981). Due to the abundance of available literature,
our discussion will review saline seep development, the
hydrogeological conditions and reclamation techniques as
they pertain to the Triangle Area of central Montana (Fig.
1).
Saline seep formation is actually the result of an interac-
tion of climatic and hydrogeologic conditions with local
cultural practices. The climate of north-central Montana is
arid to semiarid, with widely variable precipitation (25-
45 cm), low humidity, warm summers, and cold winters.
Winter low pressure centers sweep southwesterly out of
Canada, bringing strong winds that result in generally lim-
ited snow cover (less than 5-8 cm) and extensive drifting.
The variable precipitation patterns, low humidity, and high
evapotranspiration rates during the growing season have
made moisture one of the principal growth limiting factors
of northern Montana's agricultural industry.
The largest land-use change in the northern Great
Plains since the 1930s has been from native range to a
115
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
SUEETGRASS HILLS
15 30
Figure 1.—Location of the "Golden Wangle" in north-central Montana.
rigid crop-fallow rotational system (Dodge et al. 1984;
Ford and Krall, 1979). This rigid cropping system does not
efficiently use the annual precipitation. In north-central
Montana, native range or perennial vegetation uses water
approximately 14 months out of a 24-month period, while
the crop-fallow system allows water use only 3-6 months
out of this same 2-year cycle. During a fallow year, most
soils are capable of storing 10-25 cm of moisture in a
rooting depth of 120 cm. Thus, any precipitation in excess
of the soil moisture holding capacity is lost to deep perco-
lation.
Normally the water table under fallow ground shows a
rapid rise (up to 3 m) during spring melting. This is fol-
lowed by a gradual decline over the remainder of the
growing season. It has been observed that, under crop-
fallow management, the water table lows do not reach the
previous year's low. This implies a water table buildup
caused by deep percolation of unused annual precipita-
tion. As the water migrates vertically through the soil, it
dissolves soluble salts within the profile. Upon reaching a
less permeable layer (i.e., shale) the ground water begins
to flow laterally down gradient. This saline water can re-
surface upon encountering a change to soil with a heavier
texture, decreased gradient, or physical barrier. With
ground water flow velocity reduced, a water table buildup
occurs. Once the water table is within 0.9 m of the sur-
face, capillary action and evaporation account for the sol-
uble salts being precipitated at or near the surface.
North-central Montana's geologic history has been
characterized by long periods of acquiescence which
have allowed for the deposition of thousands of meters of
mainly marine sediments. Mid-Cretaceous time (83-105
million years) in this area was marked by the widespread
deposition of over 610 m of marine shale (Colorado
Group) in a relatively stable, epicontinental sea. The Colo-
rado epoch was brought to a close by uplift of the present
Rocky Mountains on the western edge of central Mon-
tana. This period of crustal instability produced multiple
transgressions and regressions of the Cretaceous sea and
imparted a cyclical nature to the succeeding Montana
Group (Telegraph Creek, Eagle Sandstone, Claggett
Shale, Judith River, Bearpaw Shale, Fox Hills, and Hell
Creek formations). Each of the transgressive facies (Clag-
gett and Bearpaw) deposited marine shales that vary in
nature only slightly from the Colorado Group. A regional
northeast tilt of less than 3° accounted for the erosion of
Early Tertiary (50-65 million years) sediments and the ex-
posure of the Cretaceous marine shales. Two or more
glacial ice advances during the Pleistocene (15,000-2 mil-
lion years) left behind a thin mantle (1-24 m) of poorly
sorted, unconsolidated deposits. The glacial till greatly al-
tered the preexisting landscape by filling previous lows
and forming a gently rolling terrain.
As previously recognized (Miller et al. 1981), the geol-
ogy of the area influences seep development in the follow-
ing ways:
1. The bedrock units influence the type of and amount
of salt that is incorporated in the overlying glacial drift and
associated ground water.
2. The fresh shale and thin bentonite beds create an
impermeable layer, thereby prohibiting any significant
ground water movement through these units.
3. The weathered shale zone provides a laterally con-
tinuous permeable zone allowing water to migrate down
gradient.
4. The poor drainage associated with glacial terrains
allows water to pond for extended periods of time.
5. The extensive vertical joints and fractures of the gla-
cial till allow water to percolate through the drift.
6. The drift contains an abundant supply of water solu-
ble salts throughout the region.
The shallow ground water quality in areas of high saline
seep density is very poor. In general, the water quality
corresponds to the underlying geology: a TDS concentra-
tion of 20,000-50,000 mg/L in Colorado Shale and Bear-
paw Shale, with a range of 10,000-25,000 mg/L in the
Claggett Shale-Judith River unit (Dodge et al. 1984). The
major constituents present are soluble calcium, magne-
sium, and sodium sulfates with lesser amounts of nitrates,
chlorides, and bicarbonates. In addition, in a study by
Miller et al. 1978, 40 percent of the water samples col-
lected showed increased levels of selenium (greater than
the 0.01 mg/L limit given by the U.S. Public Health Serv-
ice). Table 1 shows an averaged water quality analysis
116
-------
GROUND WATER QUALITY
Table 1.—Averaged analysis of ground water from areas
underlain by Colorado Shale and Judith River-Claggctt
formations (from Miller et al. 1981)
Judith River-
Colorado shale
unit
Claggett Eagle
unit
Mean concentrations (mg R-')
Ca
Mg
Na
K
Fe
SiO2
HCO3
Cl
S04
NO3
Se
Sr
TDS
PH
275
866
1317
17.7
0.3
12
534
141
6041
57
0.308
4.6
9262
7.62
120
71
670
6.1
0.3
12
722
177
1143
6
0.028
2
2928
7.89
Table 2.—Rooting depth and net soil water depletion for
perennial vegetation after 5 years compared to annual
crops, all grown near Fort Benton, Montana
(Brown etal. 1983).
Net soil
Rooting water
depth depletion
Crop
Perennials
Beaver alfalfa
Ladak 65 alfalfa
Kane alfalfa
Intermediate wheatgrass
Crested wheatgrass
Annuals
Safflower
Sunflower
Winter wheat
Barley
Feef
24
22
16
15
13
7
6
6
5
Inches
41
26
21
29
16
10
7
7
6
from an area underlain by Colorado Shale and from an
area underlain by Judith River-Claggett formations. The
increased salt load of ground water in the region has very
serious implications. Not only have domestic wells been
abandoned, but livestock and fishkills have been attrib-
uted to salinized surface water supplies.
Through cultivation, the crop-fallow system has re-
sulted in a significant redistribution and removal of soluble
salts when compared with native range. The upper portion
of the soil profile in the recharge area has been improved
for crop production from this leaching. The greater mobil-
ity of the more soluble cations Mg2+ and Na+ has left
Ca2+ as the dominant ion in the surface layer: thus, the
surface soil structure. Considering only MgSO4 and
NaSO4, greater than 90 T/ha of soluble salts have been
leached from 0-360 cm depth (Ferguson and Bateridge,
1982).
One of the benefits of the crop-fallow system is in-
creased water storage in the root zone, which has helped
stabilize the crop production in the northern Great Plains.
But, side effects over the last 40-50 years have been a
general rise in shallow ground water tables, a serious sa-
linization of ground water, and degradation of many sur-
face waters (Bahls and Miller, 1973). Any precipitation that
infiltrates the soil profile in excess of field capacity will
percolate below the root zone of small grains to accumu-
late and form a perched water table over time. An average
of 2-3 cm of leachate is moved below 180 cm each fallow
year. In a 50-year crop-fallow cycle, between 0 and 75-cm
of water can move below the cereal grain root zone
(Ferguson and Bateridge, 1982).
To quantify the amount of water movement from a single
hectare, approximately 254,120 L/ha would be leached
each fallow year or 5,082,398 Una over the last 40 years.
For each .4 ha of saline seep, there is an average of 4
hectares of recharge area. If 2.5 cm were lost from the 4
hectares, 2,541,199 L/ha would be channeled into the
.4 ha seep area for evaporation from the soil surface
(Halvorson et al. 1974). The water quality of a typical seep
in the Triangle Area is 25,000 mg/L TDS, so over
63,440 kg/ha of salts would be concentrated in the seep
area.
VEGETATIVE RECLAMATION
The practice of summer fallowing to control weeds and
conserve moisture became popular following the drought
years, 1917-20. The accompanying strip cropping to pre-
vent wind erosion became universal in the 1930's with
government incentive payments. The disadvantages of
summer fallow associated with saline seep development
are the inefficient soil water storage capabilities and an-
nual precipitation use. The amount lost can be much more
on above average rainfall years or in light textured soils
with low water-holding capacities. This process has been
going on since the late 1940's with the advent of large-
scale farming equipment that controls virtually all mois-
ture-robbing vegetation in the fallow period.
The saline seep can be divided into two areas defined
as the discharge and recharge. The discharge area is the
wet, salinized soil profile which is actually the symptom,
not the problem. The recharge or upslope area is where
the seep water originates as unused precipitation. Before
the seep itself can be reclaimed, a more intensive crop
water-use program must be initiated on the recharge area.
The drilling of shallow ground water monitoring wells de-
fines the recharge area and determines the soil profile,
depth to bedrock, and amount of stored moisture. From
this information, a site-specific intensive cropping system
is designed, keeping in mind the climate and manage-
ment of the individual land operator.
The recharge area treatment must be as intensive (that
is, plant growth as long and often) as is economically fea-
sible. Alfalfa (Medicago sativa) has proven to be the most
consistent crop recommended because of its many attrib-
utes. Alfalfa benefits, both long and short term, include:
1. Ability to root deeply to use- stored soil moisture be-
low the normal or small grain root zone, and to utilize
moisture from the capillary fringe of the water table.
2. Long growing season, whereby it commences
growth early in the spring and continues to use water into
the fall or about 7 months. The precipitation-use efficiency
is 95-100 percent.
3. Economic yields comparable to cereal grain produc-
tion if managed property, that is, adequate seedbed, fertil-
ity, rhizobia bacteria inoculation, harvesting, and length of
stand.
4. Improved soil tilth, and increased organic matter and
nitrate levels (important for subsequent crops).
Alfalfa will dry out the deep subsoil to provide a dry
reservoir of soil for storage of unused precipitation for
subsequent cropping systems. The alfalfa need not be a
permanent stand, but is usually left in for 5-6 years, de-
pending on the water table levels and the quality of the
stand. The cultivars adaptable to this area vary considera-
bly in rooting depth and soil water depletion. Brown et al.
(1983) determined that, after 5 years, Beaver alfalfa could
root to 7.3 m and used 104 cm of water (in addition to the
117
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
annual precipitation), while Kane rooted only 4.9 m and
used 53 cm of water. In the same studies, other perennial
legumes and grasses were found to vary significantly, but
used less water than alfalfa and more than small grain
production (Table 2).
Intensive cropping systems are implemented on the re-
charge area when: (1) the alfalfa is removed, (2) alfalfa is
not an economically viable alternative, or (3) the soil pro-
file is too shallow or too permeable to support a healthy
stand. The two cropping systems recommended by the
District are annual and flexible. Annual cropping (one crop
each year on successive years) is appropriate under very
limited conditions, therefore it is not widely used. If the soil
profile is shallow, less than 1.5 m, or very permeable, the
water holding capacity is limited. Thus, the efficiency of a
fallow period is limited. The amount of crop produced will
be dependent on the annual precipitation, since little mois-
ture will be stored from the year before. Fallowing in a high
precipitation year with these soil conditions will rapidly
recharge the water table in a saline seep, even if previ-
ously reclaimed. The other situation in which annual crop-
ping is adaptable is where the annual precipitation level is
high, usually greater than 35 cm and the growing season
precipitation is consistent and timely 15 cm or better.
The flexible cropping system can be implemented al-
most universally. The decision to crop, made annually at
seeding time, is based on stored soil moisture plus ex-
pected growing season rainfall. Together, these must ex-
ceed 20 cm to achieve a satisfactory economic yield. The
stored moisture is a measure of the depth of moist soil x
the plant available water per centimeter of moist soil. The
expected rainfall is the average amount received May 1-
July 31 at 70 percent probability, or 7 out of 10 years. With
this information, a yield prediction can be made on individ-
ual fields to assist the land operator in crop rotation deci-
sions that are both economical and conservation minded.
If the moisture conditions are insufficient, then a fallow
year is inserted into the rotation. One guideline is never to
use a fallow rotation for 2 years in a row to avoid loss to
deep percolation.
Intensive cropping systems require a high level of man-
agement by the operators. Careful attention must be paid
to details like soil testing and fertilizer application, select-
ing the most suitable crop and cultivar, adequate weed
control, marketing and so forth. Alternate crops to cereal
grains, specifically oilseeds, may be inserted into the rota-
tion where the climate and market are suitable. The oil-
seed crops break up the weed and disease cycles inher-
ent in small grain monoculture. A problem arises in that
there is not always an adaptable oilseed crop for each
recharge area, since oilseeds are sensitive to the climate
or heat units produced.
The economic feasibility of a flexible cropping system
versus the traditional crop-fallow system can be com-
pared only if accurate records are kept of variable costs
and operations on a long-term basis. It has been shown
that, with good management and marketing skills, the in-
tensive cropping system can outproduce the crop-fallow
system in both yield and net return.
The discharge area treatment may not begin for several
years after the recharge area cropping system has been
intensified. As the ground water table is being lowered,
natural precipitation will leach the'precipitated soluble
salts down through the profile. Soil tests of the top 7-
15 cm are taken periodically to determine the electrical
conductivity (EC). When the soil's EC and water table
have been reduced and an adequate seedbed can be
prepared, salt-tolerant vegetation (primarily grasses) is es-
tablished. The introduced vegetation will further lower the
water table and provide weed and erosion control. It may
be necessary to leave the seep area in perennial vegeta-
Table 3.—Salt-tolerant vegetation species and cultivar
selection based on electrical conductivity (EC)
soil test results In mmhos/cm.1
mmhos/cm
Recommended
Common name variety
Beardless wild rye
Tall wheatgrass
Altai wildrye
Slender wheatgrass
Tall fescue
Western
wheatgrass
Russian wildrye
Barley
Crested wheatgrass
Creeping foxtail
Yellow sweetclover
Alfalfa
Shoshone
Jose
Prairieland
Revenue
Kenmont
Rosana
—
—
Nordan
Garrison
Commercial
Ladak 65
Marginal
level
24
22
20
20
16
14
12
12
10
8
8
6
Upper
limit
28
26
24
24
20
18
16
16
14
12
10
8
'USDA-SCS Job Sheet MT-JS-ES-126.
tion, but, in many cases, the seep can be reclaimed to
original condition and again produce crops.
The saline seep or discharge area may or may not be
seeded to salt-tolerant vegetation, depending on the size
and severity, as well as the recharge area treatment. For
example, if the seep area were less than 2 ha, with an EC
of 12 mmhos/cm, the area would possibly be seeded an-
nually to barley (Hordeum vulgare), especially if the re-
charge area were to be flex-cropped. Barley production at
this EC level would be marginal, but control would be
possible on the broadleaved, salt-tolerant weeds. It is of-
ten not economical to have hay production on small acre-
ages, since many small grain operators no longer have
haying equipment, nor cattle to graze the forage.
On larger salinized areas, where the water table is stabi-
lized or dropping, standard farm equipment will be used to
prepare a weed-free, firm seedbed. Often chemical weed
control is combined with the mechanical treatments. Fox-
tail barley (Hordeum jubatum) is a salt-tolerant weed that
provides much competition to seedling forages and in-
vades established stands.
Soil test results are matched with forages from the salt-
tolerant forage list (Table 3). Beardless wildrye (Elymus
triticoides) is the most tolerant, with excellent grazing and
forage qualities, but seed availability is limited and it is
difficult to establish because of seed dormancy. Tall wheat-
grass (Agropyron elongatum) and altai wildrye (Elymus
angustus) are recommended most often if the EC is 22
mmhos/cm or less. As noted in the table, alfalfa is not
extremely saline tolerant, especially in the seedling stage.
As a frame of reference, soils are considered saline if the
EC is 4 mmhos/cm or greater. Wheat production is se-
verely reduced by salts, while barley is the most tolerant of
all cereal grains. If the EC is greater than 22 mmhos/cm, it
is recommended not to seed, since the cost is too great for
the limited establishment expected. The costs are very
high because seeding rates are usually twice those
needed for nonsaline soils, combined with the high cost of
some seed (e.g., altai wildrye) and the need for a seeding
mixture. If the EC varies across the seep area, a seed mix
will adaot to the condition.
An example of a site-specific saline seep reclamation
plan involved a drainage way in Teton County, leading into
Priest Lake, which had been salinized for a distance of
4.8km (3 mi). In July 1981, the drainageway flowed at
2.5 m3/hr with a TDS ot 78,310 mg.L. The high sulfate
content (58,700 mg/L) had caused severe deterioration of
the concrete road culvert. The recharge area above one of
the seeps was cropped in 1982; alfalfa was seeded in
118
-------
GROUND WATER QUALITY
\^ Jtffttto f n i *"*"
*. SALTY SUBSTRATA •. ^
IMPERMEABLE LAYER
Figure 2.—Typical process of formation of saline seep In north-central Montana.
1983, and a 5.8 mt/ha hay crop was harvested in 1984.
The water level in the seep dropped from 38 cm from the
surface in 1982, to 94cm in 1983 to 198cm in 1984.
Rexible cropping above other seeps has also dropped the
water level, though not as dramatically By July 1984, the
drainageway was no longer flowing. In contrast, another
field was left fallow in 1983, and the associated seep water
level was raised 25 cm to surface level, indicating the deli-
cate balance between recharge and discharge area water
use.
Another reclamation success involved a total of 8 ha of
salinized cropland in Toole County. Observation wells
were installed in the recharge and discharge areas to
monitor the shallow ground water tables. Approximately
36 ha of Ladak 65 alfalfa were seeded in a designated
recharge area. Additional surrounding cropland was flexi-
bly cropped to increase overall water efficiency. In 1984,4
years after plan implementation, the water table level in
the recharge area had dropped from an original level of
160 cm to 452 cm. Likewise, the water table in the dis-
charge area dropped from its 1980 level of 86 cm to a
1984 level of 239 cm. Figure 3 graphically displays the
water table level decreases. The annual precipitation has
leached the salts lower into the soil profile, allowing alfalfa
to spread into formerly salinized areas.
The knowledge gained in vegetation management for
saline seep control is now being adapted for other water
quality problems. The District is participating in a research
project, in conjunction with the Montana Bureau of Mines
and Geology to determine the feasibility of intensive man-
agement as a key to cleaning up acid mine drainage. In
the Stockett-Sand Coulee area of central Montana, ineffi-
cient moisture management on cropped benches has al-
lowed soil moisture to percolate past small grain roots and
into lower-level, abandoned coal mines. Although the
.problem differs in type of pollutant, the process leading to
formation is the same. An attempt is being made to design
an intensive cropping system in which the cost-benefit
ratio favors the local landowner, so as to encourage volun-
tary adoption of the system.
51 ci
20"
104 CI
1,0"
152 ci
60"
203 ci
80"
254 c
100"
305 ci
120"
356 ci
HO"
406 ci
160"
157 ci
180"
| I I I I
TORSHX F«RR (ETHR1D6E. RT.)
DISCHARGE HELL #1
RECHARGE «ELL
6/80 10/80 5/81 9/81 4/82 12/82 5/83 6/83 6/8k
Figure 3.—Effect of alfalfa in lowering ground water tables
In the recharge and discharge areas. (NOTE: The spring salt-
water (4/82), Indicated by the peak in the recharge area wa-
ter table, was fully utilized'by the alfalfa before it could con-
tribute to the discharge area).
SUMMARY
The control of saline seep is crucial for several reasons.
Not only productive cropland is lost, but surface and
ground water quality is degraded. There are also adverse
impacts on livestock production and fisheries and wildlife
populations. The effect on Montana's economy can be
severe. There is much concern over the reversibility of
119
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
degraded water supplies. This question has yet to be an-
swered, and the implications in the future could be devas-
tating.
In most cases, with the implementation of an intensive
vegetation management plan, saline seep can be con-
trolled. More efficient plant water use reduces hydrostatic
pressure on the entire system, allowing the water table to
be lowered in the discharge area. Once the water table is
stabilized below 1.8 m, the salts are isolated below the
rooting zone of small grains by annual precipitation. The
deeper water table allows the natural fluctuation in the
spring flush to occur without the deleterious effect of a
continuously saturated saline profile.
REFERENCES
Bahls, L.L., and M.R. Miller. 1973. Saline seeps in Montana.
Pages 35-44 in Second Annual Report, Montana Environ-
mental Quality Council. Helena, MT.
Black, A.L., PL. Brown, A.D. Halvorson, and F.H. Siddoway.
1981. Dryland cropping strategies for efficient water-use to
control saline seeps in the northern Great Plains, U.S.A.
Agric. Water Manage. J. 4: 295-311.
Brown, PL, A.D. Halvorson, F.H. Siddoway, H.F. Mayland, and
M.R. Miller. 1983. Saline-seep diagnosis, control and reclama-
tion. Conserv. Rep. No. 30. U.S. Dep. Agric., Washington, DC.
Dodge, T., J. Holzer, W. Adams, J. Farkell, and M.R. Miller. 1984.
Triangle Conservation District—a case study in dryland saline
seep control. Pages 56-70 in R. H. French, Salinity in Water-
courses and Reservoirs. Butterworth Pub., Boston.
Ferguson, H., and T. Bateridge. 1982. Salt status of glacial till
soils of north-central Montana as affected by the crop-fallow
system of dryland farming. Soil Sci. Soc. Am. J. 46: 807-10.
Ford, G.L., and J.L Krall. 1979. The history of summer fallow in
Montana. Bull. 704. Montana Agric. Exper. Sta. Bozeman, MT.
Halvorson, A.D., and A.L. Black. 1974. Saline-seep develop-
ment in dryland soils of northeastern Montana. J. Soil Water
Conserv. 29:77-81.
Halvorson, A.D., A.L. Black, and C.A. Ruele. 1974. Saline
seeps—continued bad news for many farmers. Montana
Farmer-Stockman 61(18): 34-38.
Krall, J.L., PL. Brown. 1976. Cultural practices for the control of
saline seep in the Northern Plains, pages 64-77. In: Conser-
vation Tillage, Great Plains Agricultural Council, Pub. 77, Fort
Collins, Colo.
Miller, M.R., PL. Brown, J.J. Donovan, R.N. Bergantino, J.L.
Sonderegger, and FA. Schmidt. 1981. Saline-seep develop-
ment and control in the North American Great Plains: hydro-
geological aspects. Agric. Water Manage. J. 4: 115-41.
Miller, M.R., R.N. Bergantino, W.M. Bermel, FA. Schmidt, and
M.K. Botz. 1978. Regional assessment of the saline-seep
problem and a water quality inventory of the Montana plains.
Open file Rep. 42. Montana Bureau of Mines and Geology.
U.S. Department of Agriculture-Soil Conservation Service.
1981. (Agronomy) Species selection and establishment guide-
lines for saline seeps. Montana Job Sheet JS-ES-126. Boze-
man, MT.
Vander Pluym, H.S.A. 1978. Extent, causes and control of dry-
land saline seepage in the northern Great Plains of North
America. Pages 1.4-1.58 in Dryland Saline-Seep Control,
Proc. Meeting Subcommission on Salt-Affected Soils. 11th
Int. Soil Sci. Soc. Congress. Edmonton, Alberta, Canada.
120
-------
NONPOINT SOURCE IMPACTS ON GROUND WATER QUALITY
IN MAJOR LAND RESOURCE AREAS OF THE SOUTHWEST
S. J. SMITH
J. W. NANEY
Agricultural Research Service
U.S. Department of Agriculture
Durant, Oklahoma
W. A. BERG
Agricultural Research Service
U.S. Department of Agriculture
Woodward, Oklahoma
ABSTRACT
For the past several years, we have assessed the impact
of agricultural management practices on ground water
quality of watersheds in various land resource areas of
Oklahoma and Texas. Typically, the watersheds (1.6-6 ha)
encompass a wide range of soils and managements.
Treatments include different crop, grass, tillage, fertilizer,
pesticide, and grazing practices. Results of water quality
analyses are presented for 34 ground water wells of shal-
low depth (<20 m water table) that are monitored sea-
sonally on the watersheds. Topics considered include the
impacts of soluble nitrogen and phosphorus, irrigation
suitability, sulfate and chloride, salinity, pesticides, and
oil/gas development. Results are related to agricultural
management practices, existing geology, and where ap-
plicable, past and present petroleum production. In some
locations, local geology and intense early date oilfield
activity are reflected in high levels of sulfate and chloride,
respectively. However, with few exceptions, farming and
ranching activities were found to have limited impact on
ground water quality.
INTRODUCTION
The Southwest, representing a vast expanse of farming
and ranching activities, continues to experience increas-
ing rural and urban demands on its water supplies. Conse-
quently, the need has intensified for more detailed infor-
mation on ground water supplies associated with
agricultural management practices. Such information has
a direct bearing on the environmental and economic
wellbeing of the area. For several years part of our re-
search has been to assess the impact of agricultural prac-
tices on ground water quality of watersheds in major land
resource areas of Oklahoma and Texas. Presented here
are results for 34 shallow wells (< 20 m water table depth)
that are monitored periodically on watersheds in the Cross
Timbers (CT), Reddish Prairie (RP), and Rolling Red Plain
(RRP) major land resource areas.
AREAS AND METHODS
The locations of the approximately 30 watersheds (1.6-
6 ha (4-15 acre) size) on which water quality was deter-
mined, and the major land resource areas in which they
occur are indicated in Figure 1. Principal geologic and
management features of the locations, as they relate to
ground water quality studies, are given in Table 1. Within
each major land resource area, the watersheds represent
characteristic settings where ground water quality may be
affected by changes in land use and management prac-
tices.
Typically, the watersheds encompass a wide range of
soils and management treatments. Treatments include dif-
ferent crop, grass, tillage, fertilizer, pesticide, and grazing
practices. Water table depths generally range from 2 to
20 m, with the watershed wells sampled on a seasonal
basis. Additional details about the watersheds and wells
may be found in previous publications (Naney and Smith,
1983; Naney et al. 1984; Smith et al. 1983).
All well samples for chemical analysis were refrigerated
at 0-4°C. Chemical analyses for nitrate-N and ammo-
nium-N were made using standard methods described in
the Federal Water Pollution Control manual (U.S. Dep.
Inter. 1971). Water soluble P was determined by the mo-
lybdenum-blue method described by Murphy and Riley
(1962). Calcium (Ca), magnesium (Mg), and sodium (Na),
used to obtain sodium percentages for irrigation suitability,
were determined by atomic absorption. Sulfate was deter-
mined by turbidimetry, chloride by the specific ion elec-
trode, and specific conductance by the wheatstone
bridge. Samples were sent to the Oklahoma Department
of Agriculture State Laboratory for pesticide analysis.
RESOURCE AREAS
RP REDDISH PRAMES
HP HIGH PLAINS
RRP ROLLING RED PLAINS
BH BLUESTEM HILLS
CT CROSS TIMBERS
CP CHEROKEE PRAIRIES
EP EDWARDS PLATEAU
GS GRANITE SOILS
Figure 1 .—Locations of nonpolnt source water quality moni-
toring wells within major land resource areas In Oklahoma
and Texas.
121
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Publications by the U.S. Environmental Protection Agency
(1973, 1976) and Sawyer and McCarty (1967) were used
as guides for water quality standards.
in wells under fertilized and grazed wheatfields empha-
sized the need for care in well installation and mainte-
nance. A proper well seal is essential to prevent direct
surface flow contamination into the annulus between
borehole and well casing.
m
Both present and potential irrigation well development ex-
ist in the alluvium and sandy lands of the Chickasha, Fort
Cobb, and Woodward watersheds. Therefore, special at-
tention was given to suitability of well water quality for
irrigation purposes. When planning irrigation develop-
ment, several physical and chemical factors such as rain-
fall distribution, irrigation water applied, soil drainage, and
soil chemical properties must be considered in addition to
water quality. Various techniques exist for determining the
suitability of water for irrigation (Hagin and Tucker, 1982;
Richards, 1954); a simple sodium percentage-specific
conductance relationship (Davis, 1955) has been used
here.
The results in Figure 2 were obtained by combining the
Ca, Mg, Na and conductance data from Chickasha,
Ft. Cobb, and Woodward watershed wells with additional
data from the U.S. Geological Survey. No attempt was
made to assess the water quality criteria for development
of irrigation from the El Reno wells because local geology
precluded ground water irrigation.
Twenty-five wells on the Chickasha, Ft. Cobb, and
Woodward watersheds were shown to be of satisfactory
irrigation quality, while four wells yielded water that was
doubtful for irrigation use, and two were classified as un-
Table 1 .—Principal watershed features related to ground water quality studies for NPS assessment.
Soluble N (nitrate and ammonium) and P contents, con-
sidered to reflect any possible fertilizer contaminants in
the wells, are given in Table 2. Ranges of the values indi-
cate few high concentrations associated with nitrate-N
(typically <10mg/L). However, some exceptionally high
soluble P contents (> 5,000 ^g/L) associated with wells
from the Chickasha and El Reno wheatlands are present.
Like soluble P, some high ammonium-N contents (>3 mg/
L) were also observed. For both situations, though, this
was only true for certain instances in 1979 and 1980.
Overall, the data in Table 1 indicate nitrate-N contents
within the acceptable limits of 10 and 100 mg/L for human
and livestock consumption, respectively. Ammonium-N
levels above 2.5 mg/L, harmful to fish, and soluble-P lev-
els above 10 ^g/L, sufficient for eutrophication, were ob-
served in certain instances. In most cases, the high am-
monium-N and soluble-P contents were traced to
improperly installed and maintained well casings that al-
lowed surface runoff from agricultural fields and farm-
steads to flow directly into the wells. Improved well protec-
tion techniques eliminated this problem and reduced
ammonium-N and soluble-P to more acceptable levels.
The ammonium-N concentrations > 3 mg/L at certain
farmsteads and the soluble-P concentrations > 5000
Location
Chickasha
El Reno
Ft. Cobb
Woodward
LRA
RP&CT
RP
RRP & CT
RRP
Wells
12
10
2
10
Drilled
depth
(m)
10-40
10-25
15-30
3-9
Major
Land Use
Native grass, wheat •
Native grass, wheat,
grain sorghum
Peanuts & grain
sorghum
Improved grasses
and alfalfa
Geologic
Age
Quaternary
Quaternary
Permian
Permian
Quaternary
Permian
Quaternary
Permian
Stratigraphy
Terrace deposits
& alluvium
Terrace deposits
El Reno group
Cloud Chief
formation
Whitehorse
El Reno group
Alluvium
Whitehorse group
Terrace deposits
Whitehorse group
LKhology
Sand & gravel
Sand & gravel
Shale & Sandstone
Gypsum, shale and
sandstone
Sand & Silt
Sandstone
Sand & gravel
Shale & sandstone
Table 2.—Range in nutrient concentrations In ground water sampled at four nonpolnt source monitoring areas
from 1979-84 in Oklahoma as affected by land use.
Monitor well
Location
Chickasha
El Reno
Woodward
El Reno
Ft. Cobb
Chickasha
El Reno
Ft. Cobb
Chickasha
LRA
RP&CT
RP
RRP
RP
RRP & CT
RP&CT
RP
RRP & Ct
RP&T
No. of
wells
2
4
I
10
3
1
5
3
1
3
Nitrate N
(mg/L)
0.08-0.90
0.20-1.1
mproved grasses and
0.08-16.0
0.10-8.8
1.87-2.1
0.09-4.1
0.20-2.1
0.0-0.7
0.17-18.4
Soluble P
(M9/L)
6-243
0-641
4-89
8-63
101-103
4-8,183
0-5,299
22-25
0-87
Ammonium N
(mg/L)
0.01-1.99
0.00-0.33
0.0-0.42
0.0-0.41
0.0-0.04
0.0-2.83
0.0-0.42
0.0-0.17
0.0-3.30
122
-------
GROUND WATER QUALITY
suitable. On the whole, water from the wells was satisfac-
tory for irrigation purposes.
Sulfate and Chloride
Contents of these chemical constituents are given in Table
3. Sulfate contents as high as 2,390 mg/L were observed
and, unlike ammonium-N and soluble-P, generally in-
creased from 1979 to 1981 for the Chickasha area water-
sheds. These high sulfate contents (> 250 mg/L is the
recommended domestic standard limit) are associated
with the natural geology and not with agricultural manage-
ment practices. The contents arise from gypsum (CaSO4 •
2H2O) and epsomite (MgSO4 • 7H2O) common to the as-
sociated Permian Red Beds and their weathering prod-
ucts.
The relatively high solution of gypsiferrous materials
within the local geology resulted in conductivities of some
well water samples as high as 1,500 ^mhos/cm. In fact,
variation of electrical conductivity paralleled that of sulfate
content, and may be explained by the relationship be-
tween conductivity and dissolved solids (Reeves and
Miller, 1978). Seasonal changes in sulfate contents of the
wells have been related to fluctuations in rainfall and
ground water levels (Naney and Smith, 1983).
Typically, chloride well contents were low (<50 mg/L)
for all land uses on the watersheds. The high chloride
values in Table 3 were observed initially after well comple-
tion only and are attributed to some chloride salts in the
Permian Red Beds. Therefore, the lower range values are
more representative. Unlike sulfate, the natural geology
appears to contribute little chloride to the ground water.
For the most part, chloride levels are well below the do-
mestic recommended standard limit of 250 mg/L.
Salinity and Livestock
Highly saline waters are well known to cause harmful os-
motic effects, and to exert specific toxicities on livestock.
The chemical constituents include mainly the metals Ca,
Mg, and Na, and the nonmetals sulfate, chloride and bi-
carbonate, with the results expressed as total soluble
salts. A simple estimate of total soluble salts may be ob-
tained by multiplying the well water conductivities (Fig. 2)
by an appropriate factor (Sawyer and McCarty, 1967). Us-
ing the factor 0.65, soluble salt contents were generally far
below 2,000 mg/L by this method. From the standpoint of
salinity and osmotic effects, waters with < 3,000 mg/L sol-
5OO IOOO I9OO ZOOO 29OO 3OOO 35OO
TOTAL CONCENTRATION
AS SPECIFIC CONDUCTANCE (gmhos)
Figure 2.—Chart depicting irrigation suitability of ground
waters in the Chickasha, Ft. Cobb, and Woodward areas,
based on average concentrations; modified from Davis,
1955.
uble salts are satisfactory for livestock under almost any
conditions. Consequently, salinity of the well waters is con-
sidered to pose no particular livestock problems.
Pesticides
Various pesticides at the recommended rates have been
applied to certain watersheds. Therefore, detection of re-
sidual pesticides in the associated ground waters was in-
vestigated. Water samples from two wells each at the Fort
Cobb (peanut and grain sorghum watersheds), El Reno
(wheat and improved practice grassland watersheds) and
Woodward (wheat and virgin grassland watersheds) loca-
tions were analyzed for organochloride, organophos-
phate, and phenoxy pesticide residues. In every case, the
pesticide residue contents were below the limits of analyti-
cal detections (<0.01 ppb). Hence, persistent pesticide
residues are considered to pose no problems in the wells.
Oil and Gas Development
The ground water wells are located in an area of major oil
and gas development. During 1982, a peak development
year, the total number of oil and gas wells drilled was, by
county: Caddo, 188; Canadian, 302; Grady, 162; and
Woodward, 109 (Arndt, 1985). As noted earlier (Naney
and Smith, 1983), high chloride contents (> 2,000 mg/L)
were determined in two observation ground water wells
located down gradient from the Cement oilfield in Grady
County, Oklahoma. This impact on water quality appears
to be related locally to abandoned practices and oil wells
developed during the 1930's. Samples of ground water
wells in the vicinity of current oil wells indicate no water
quality problems (Table 4).
Also during 1982, over a 3-month drilling and installa-
tion of a gas well, surface water quality was monitored
above and below the construction site. The results, given
in Table 4, show no effect on surface water quality. There-
fore, current data indicate technology exists that allows
petroleum drilling and production activities to be con-
ducted without posing particular water quality hazards.
Specific exceptions may occur when available technology
is not used.
REFERENCES
Arndt, R.H. 1985. Statistics in Oklahoma's petroleum industry,
1982-83. Okla. Geo. Notes 45(1): 4-27.
Davis, L.V. 1955. Geology and groundwater resources of Grady
and northern Stephens counties, Oklahoma. Okla. Geolog.
Surv. Bull. 73: 123-6.
Hagin, J.B., and B. Tucker. 1982. Fertilization of Dryland and
Irrigated Soils. Springer-Verlag, Berlin, New York.
Murphy, J., and J.P. Riley. 1962. A modified single solution
method for determination of phosphate in natural waters.
Anal. Chem. Acta 27: 31-6.
Naney, J.W., and S.J. Smith. 1983. Geologic and land use ef-
fects on ground water quality in shallow wells of the Anadarko
Basin. Okla. Geology Notes 43(4): 100-9.
Naney, J.W. et al. 1984. Management and geologic effects on
ground water quality of agricultural lands in the Southern
Plains. Pages 638-48 in M.A. Collins, ed. Water for the 21st
Century: Will it be there? Center for Urban Studies, Southern
Methodist University, Dallas, Texas.
Reeves, C.C., Jr., and W.D. Miller. 1978. Nitrate, chloride and
dissolved solids, Ogallala Aquifer, West Texas. Ground Water
16(3): 167-73.
Richards, L.A. 1954. Diagnosis and Improvement of Saline and
Alkali Soils. Agric. Handbook No. 60. U.S. Govt. Printing Off.,
Washington, DC.
Sawyer, C.N., and T.L. McCarty. 1967. Chemistry for Sanitary
Engineers. 2nd ed. McGraw-Hill Book Co., New York.
Smith, S.J. et al. 1983. Nutrient and sediment discharge from
Southern Plains grasslands. J. Range Manage. 36: 435-9.
123
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
U.S. Department of Interior. 1971. FWPCA Methods for Chemi- 1976. Quality Criteria for Water. U.S. Govt. Printing
cal Analysis of Water Wastes. FCN 16020-7/71. Nat. Environ. Off., Washington, DC.
Res. Center. Anal. Control Lab., Cincinnati, OH. Wood, P.P., and B.L. Stacy. 1965. Geology and Ground Water
U.S. Environmental Protection Agency. 1973. Water Quality Cri- Resources of Woodward County, Oklahoma. U.S. Geolog.
teria 1972. U.S. Govt. Printing Off., Washington, DC. Surv. Bull. 21. Okla. Water Resour. Board, Oklahoma City.
Table 3.—Range in sutfate and chloride concentrations In ground water sampled at four nonpolnt source monitoring areas
from 1979-84 In Oklahoma as affected by land use.
Monitor
Well location
Chickasha
El Reno
Woodward
El Reno
Ft. Cobb
Chickasha
El Reno
Ft. Cobb
Chickasha
LRA
RP&CT
RP
RRP
RP
RRP & CT
RP&CT
RP
RRP & CT
RP&CT
No. of
wells
Sulfate
(mg/L)
2 131-1,212
4 52-870
10 22-1,935
3
1
5
3
i
3
24-593
54-55
12-1,708
33-1 ,551
12-15
10-2,390
Chloride
(mg/L)
6-34
11-1,029
8-427
28-633
9-13
13-99
6-796
3-4
3-45
table 4.—Water quality summary for areas Impacted by petroleum field developments near the Chickasha watersheds.
Monitor
sites
21
Above gas well site2
Below gas well site2
Period
1974-84
1982
1982
Nitrate N
mg/L
0-0.086
0.06
0.06
Ammonium-N
mg/L
0-.21
0.04
0.04
SolP
M9/L
9-16
17
19
Sulfate
mg/L
114-317
21
22
Chloride
mg/L
16-32
16
14
E. cond.
^mhos/cm
630-807
522
546
'Geometric range values tor nearby ground water wells.
2Average of 26 surface water samplings over a 3-month period during gas well construction.
124
-------
MONITORING THE EFFECTS TO THE GROUND WATER SYSTEM
ATTRIBUTABLE TO AGRICULTURAL PRACTICES
.1
CLARK GREGORY KIMBALL
South Dakota Department of Water and Natural Resources
Pierre, South Dakota
ABSTRACT
A 10-year study to evaluate the effect of three conserva-
tion practices—fertilizer management, pesticide manage-
ment, and conservation tillage—on both surface and
ground water is underway in South Dakota. The area of
study is in the lakes region of the glaciated part of the
State. Glacial till and outwash geologic environments are
included. Excessive nutrients have been identified in both
wells and lakes. Field monitoring sites from 8 to 32 ha
have been selected with and without conservation prac-
tices and with and without farming. Monitoring design has
been carefully integrated with the site geology and instru-
mentation is extensive. Surface and ground water are
monitored by sampling for laboratory analysis and by
testing in situ. Nutrient and pesticide concentrations are
determined. Soil moistures are measured and soil sam-
ple extractions analyzed for nutrients to ascertain soil wa-
ter chemistry and flux rates and their contribution to
ground water. All data are filed in a data management
system that can be manipulated as desired. Instrumenta-
tion of all field sites is not yet complete. Initial test results
indicate both definite and relatively rapid response, ef-
fects on ground water quality, in some cases, from both
nutrients and one pesticide. Results Have both limited
value and dependability due to the small number of data
collected to date.
INTRODUCTION
Elevated nutrient levels in the lakes and excess nitrogen
in the ground water are being attributed to nonpoint
source pollution from farming activities, specifically, runoff
and leaching from farm fields.
Oakwood Lakes and Lake Poinsett were designated
208 Water Quality Study Areas in 1976 to study the impair-
ment to the lakes from excessive nutrients. Later the two
areas were combined to form a Water Quality Project to
implement best management practices (BMP's) through
the Agricultural Stabilization and Conservation Services
(ASCS) Agricultural Conservation Program (ACP). Accep-
tance of the area as a Rural Clean Water Program
(RCWP) was received in October 1981. The goal of the
RCWP is to implement BMP's to alleviate water quality
impairment and carry out general chemical and physical
monitoring to document results. Prior to completion of the
RCWP monitoring strategy the project was recommended
for a 10-year Comprehensive Monitoring and Evaluation
Project, granted in October 1982. This Project is to quanti-
tatively document the effectiveness of BMP's.
The Oakwood Lakes-Poinsett Project is located within
the glacial lakes region of east central South Dakota and
encompasses 42,480 ha (106,200 acres), 26,110 ha
(65,275 acres) of which are cropland (corn, soybeans, and
small grain). The project area is partially located in Brook-
ings, Hamlin, and Kingsbury counties approximately
95.5 km (60 miles) north of Sioux Falls. Within the project
area are three lakes and portions of three others. Lake
Poinsett, the largest natural lake in South Dakota, Oak-
wood Lakes, and Lake Albert are wholly within the project
area. Lake St. John, Dry Lake, and Thisted Lake are par-
tially within the project area (Fig. 1). Lake Poinsett and
Oakwood Lakes serve as the focal point for many recrea-
tional activities. The shallow glacio-fluvial aquifer in this
area is the Big Sioux Aquifer.
The water quality of these lakes shows high nutrient
levels in excess of State Water Quality Standards. The
high nutrient levels and subsequent algal blooms, oxygen
depletion with associated fishkills, ammonia approaching
toxic levels, and excessive aquatic macrophytes impair
recreation in the lakes. The high suspended solids and
phosphate levels found in the intermittent streams indi-
cate that a nonpoint source problem exists in this water-
shed. The lakes are phosphorus limited based on the typi-
cal 15:1 nitrogen to phosphorus ratio.
Within the Big Sioux River basin the Big Sioux aquifer is
the major source of potable water, supplying 32 percent of
the State's population. From the State Department of Wa-
ter and Natural Resources, Office of Drinking Water files,
27 percent of 861 private wells in Brookings and Hamlin
Oakwood Lakes - Poinsett Project
Rural Clean Water Program
>II— Sl*t« Highway
——O— PaMd County Road
Unp«*M ROM
Figure 1.—Oakwood Lakes—Poinsett Project Rural Clean
Water Program.
125
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
counties (wherein a portion of the project lies) showed
nitrate levels that exceeded EPA's standard for drinking
water of 10 mg/L nitrate as nitrogen.
The BMP's chosen to counter the surface water prob-
lem are conservation tillage to keep sediment-transported
contaminants on the fields, and fertilizer management to
manage nitrogen and phosphorus application rates and
timing. Pesticide management is also being stressed in
response to typically increased use of pesticides associ-
ated with the implementation of management tillage.
The Big Sioux aquifer is relatively shallow and some of
its overlying soils are thin and quite permeable. This aqui-
fer is potentially contaminable, and, based on existing
data for nitrate in ground water, contamination may al-
ready be underway. The project faces a twofold question:
(1) is the nitrogen excess in ground water attributable to
farming operations; and (2) will the implementation of con-
servation tillage and other approved BMP's affect nitrogen
and pesticide levels in the shallow ground water system?
Moreover, will the effect be positive or negative, that is, will
conservation tillage increase or decrease water-related
contaminant transport to the subsurface? A modeling
study of tillage alternatives stated that conservation tillage
techniques reduced percolation and leaching (Crowder et
al. 1984). However, a recent article on conservation tillage
in Ohio indicated that conservation tillage may have in-
creased pesticides and possibly nitrogen input to ground
water (Agrichem. Age, 1984).
GEOLOGIC SETTING
Non-bedrock aquifers in eastern South Dakota are glacial
outwash deposits from the Pleistocene Glacial Epoch.
Within the project area, the surficial deposits are Wiscon-
sin stage glaciation, specifically, the Gary (late Wisconsin)
and lowan (early Wisconsin) substage till, loess, and out-
wash of thicknesses up to 122-153 m (400-500 ft.). The
Gary outwash is the thickest. Its morphology is that of a
valley train outwash that forms along the margin of a gla-
cier as it melts and recedes. Long, relatively thin, sinuous
deposits of sand and gravel of variable thickness and ar-
ea) extent result.
Three outwash aquifer systems exist in the area and are
referred to as the surficial, intermediate, and basal aqui-
fers. Connections between these aquifers occur in some
instances, but are not reliably known. The Big Sioux aqui-
fer is the surficial aquifer of greatest importance because
of its high quality water, accessibility, and thickness. The
other two aquifers are at a great enough depth that this
project is not directly concerned with them.
The till in the area is of a silty clay texture with some to
little sand and gravel, its high clay content deriving from its
parent material, the Pierre shale. The tills of different sub-
stages are indistinguishable from one another. The major
part of the project area is covered with till, hence, it is of
considerable interest to the overall effort and direction of
the monitoring program.
Topographically, the area is relatively flat; end moraines
are subdued and gently rolling. Areas of ice disintegration
structures, kames, and collapsed or stratified drift exhibit
somewhat more expression but are still low relief.
Streams in the area are influent (lack ground water re-
charge) and thus are intermittent in nature, flowing only in
times of excessive precipitation, or snow melt. The natural
waterbodies are ground water lakes of glacial origin.
MONITORING STRATEGY
Monitoring objectives are to quantitatively describe the
physical and chemical characteristics and the distribution
and movement of water and nutrients within the natural
system of a working farm field. The system is described to
evaluate the effects conservation tillage, fertilizer man-
agement, and pesticide management have had in terms of
nitrogen and pesticide movement through the soil profile
to the ground water, and in terms of nutrient delivery to the
lake.
To accomplish the objectives, the monitoring strategy is
designed to be site specific, with the exception of some
ongoing tributary and lake sampling. The site-specific ap-
proach is desirable for two reasons: (1) the project area is
too large to monitor as an entire unit, and (2) site-specific
monitoring increases the probability of detecting land use-
effected changes in water quality.
The site-specific monitoring will be accomplished with
10 field sites, 8-32 ha (20-80 acres) in size. Nine of the
field sites are working farm fields and one is unfarmed. Of
the nine farmed fields two are control sites, farmed with-
out the BMP's which are being evaluated. Although the
rapid infiltration of a sandy outwash is expected to be the
most probable means of introducing nitrogen to the sub-
surface, both till and outwash sites are monitored.
A master site has been established as an experimental
site to add replication and control to the monitoring effort
of the soil profile. The master site is worked as a farm
field, but various treatments are restricted to plot size ar-
eas 5.5 m x 10.5 m (18 ft x 35 ft), thereby allowing more
control over the input variables: three treatment levels of
fertilizer application, three tillage practices (moldboard
plow, chisel plow, and no till), and strict control of pesticide
application. Only the soil profile is being monitored since
the close mutual proximity of the plots means that ground
water samples would not be separable based on the treat-
ments beneath the plots. Berming and other control mea-
sures preclude runoff at the site which therefore is not
monitored.
INSTRUMENTATION AND MONITORING
SYSTEM DESIGN
To collect the data, various instrumentation must be in
place. Ten to 20 piezometers per site are used to deter-
mine hydraulic head, flow direction, and gradients, and to
collect ground water samples. Piezometers are con-
structed of PVC pipe joined by threaded pipe couplings
with 91-152 cm (3-5 ft) screens. As the need arises, flush
joint pipe and screen are also used. At a majority of the
locations piezometers are nested (one location with two to
four wells screened at different depth intervals).
Tensiometers to determine soil matric potential are in-
stalled at the field sites in banks of five tensiometers at
30 cm (1 ft) intervals from 30-152 cm (1-5 ft). Matric po-
tential determines the gradient and direction of movement
in the soil profile.
Access tubes through which neutron probes can be in-
troduced determine percent soil moisture to depths of
3.7m (12ft). Neutron probe tubes are installed at the
same locations as tensiometers to complement data col-
lected.
H-flumes outfitted with still wells and water level re-
corders monitor surface flow. Flow actuated sampling de-
vices complete the surface water flow instrumentation.
The monitoring system design began early in the proj-
ect as a preliminary drilling program to identify areas geo-
logically favorable for monitoring. Because of abundant till
in the project area the preliminary drilling was to delineate
outwash areas with shallow depth to ground water. Based
on information from large-scale glacial maps, the prelimi-
nary drilling was accomplished with a rotary drill rig. Cross
sections were developed and potential areas for monitor-
ing were identified.
126
-------
GROUND WATER QUALITY
Landowners were contacted within these potential ar-
eas to determine their attitude toward a monitoring site on
their fields. If favorable, permission to drill was obtained.
Second phase drilling determined the geology of each
site, and installed approximately three wells per site to
estimate roughly the ground water flow direction. Drilling
and logging of the holes proceeded with two drill rigs: a
standard flight auger rig, and a hollow flight auger rig,
which took split spoon samples every 1.5 m (5 ft). Wells
were installed with the hollow flight auger rig through the
center of the auger, backfilled with gravel pack, or allowed
to cave as a natural gravel pack, to 30 cm (1 ft) above the
screen. Each well was then sealed with bentonite and
spoils to the surface. As site-specific information became
available, newly constructed cross sections determined
the feasibility of monitoring and possible locations for fu-
ture monitoring wells.
The last phase of site drilling was primariliy to install the
final array of monitoring wells and to fill in information
gaps that may have developed during cross section gen-
eration. Based on cross sections developed from second
phase drilling and estimates of ground water flow direc-
tions from in-place wells, the ground water monitoring sys-
tem was designed. Wells were placed in strata that held
the most potential for chemical change caused by surficial
activities, or near surface water table conditions. Vertical
placement was intended to intercept downward moving
contaminants and describe and predict the vertical distri-
bution of contaminants within the system. Wells were
placed both upgradient and downgradient of the site of
interest; and where consent was obtained, in the field it-
self.
Tensiometer banks and neutron probe access tubes lie
near well nests to correlate results and determine re-
charge through the soil profile to the ground water system.
Two to three banks of tensiometers with neutron probe
tubes exist at each field site. The master site has a bank of
tensiometers and a neutron probe tube at each plot.
DATA COLLECTION
Weekly to monthly water level measurements are taken at
wells from which flow directions can be determined and
horizontal and vertical gradients calculated. Baildown and
slug tests will determine hydraulic conductivity.
Physical system monitoring of the soil profile at the field
sites is conducted throughout the growing season from
spring thaw to fall freeze. Tensiometers are placed in the
ground at spring thaw but must be removed soon after the
ground freezes to avoid damaging the instruments. Neu-
tron probe access tubes remain in place all year. Readings
are taken on a weekly basis during the growing season
and monthly during the nongrowing season.
Runoff monitoring at the field sites is storm event
based. At the time of an event, water level recording de-
vices are flow actuated. Water is directed through an H-
flume equipped with a still well to determine flow volumes.
Two U.S. weather bureau stations, north of the project
at Castlewood and south of the project area at Brookings,
monitor climate. In addition, a recording weather station is
centrally located within the project area at the master site.
Weighing bucket rainfall gauges will be installed at two
field sites to help determine rainfall distribution patterns.
Chemical monitoring of ground water at the field sites is
conducted on a quarterly basis, with an additional monthly
sampling of selected wells at each site. Parameters being
analyzed are nitrate, nitrite, ammonia, total Kjeldahl nitro-
gen, chloride, sulfate, total hardness (occasionally), total
alkalinity (occasionally), pesticide scan (complete list of
pesticides in Table 1), pH, water temperature, electrical
conductivity, and dissolved oxygen.
Table 1.—Pesticides to be analyzed under the category of
pesticide scan.
2,4-D
Atrazine
Lasso (Alachlor)
Dual (Metolachlor)
Sencor (Metribuzin)
Ambien (Chloramben)
Sutan
Furdan
Methoxochlor
Endrin
Counter (Terbufos)
Parathion
Banvel (Dicamba)
Treflan (Trifluralin)
Bladex (Cyanazine)
Tordon (Pichloram)
Eradicane
Ramrod
Paraquat
Toxaphene
Lindane
Thimet (Phorate)
Dyfonate (Fonofos)
Ground water is sampled in such a way as to assure a
representative sample and to preserve the sample integ-
rity before its delivery to the laboratory. After purging the
well, a pneumatic bladder pump or a variable capacity
double check valve (PVC or Teflon) bailer obtains the sam-
ple. Sample containers are new or laboratory cleaned
polyethylene bottles for inorganic parameters and glass
with foil or Teflon lids for the pesticides. Samples are kept
cold until delivered to the laboratory (always within 24
hours).
' Chemical monitoring at the master site is conducted
once a year in the fall by analyzing soil samples via the
extraction of water from a soil slurry for nitrate, phos-
phorus, and pesticide (depending on the pesticide ap-
plied) contents.
Soil profile chemical monitoring at the field sites is con-
ducted once a year in the fall for nitrates, and in years 3,4,
5, 7, and 9 for pesticides. Soil cores are taken across the
field at 0.3 m (1 ft) depth increments to 1.5m (5 ft) and
then aggregated by depth. The depth aggregated soil
samples are analyzed by extracting water from a soil
slurry and then analyzing the extract.
A nutrient budget emerging from the monitoring data for
the field sites requires special monitoring to include that
portion of the nitrogen lost during denitrification. This
monitoring entails collecting soil cores at the field sites,
incubating them in acetylene for a predetermined time,
and then analyzing for acetylene conversion to nitrous
oxide. The conversion takes place because of microbial
action responsible for denitrification. From spring thaw un-
til fall freeze soil cores are collected monthly to twice a
month, based on events affecting soil moisture levels.
Land use histories at the field sites have been tracked
as far back as the farmer can remember, and since the
initial contact, the farmer has been interviewed yearly.
Feasible land use information will be collected or gener-
ated for the whole project area. Some important land.use
parameters include:
1. Land use, irrigated crop, nonirrigated crop;
3. The predominant soil mapping units on a field;
3. BMP's contracted and applied;
4. Planting date and residue on field at planting;
5. Fertilizer and pesticide types applied:
a. quantities applied,
b. method of application,
c. date of application;
6. Tillage methods used:
a. depth of tillage,
b. dates of tillage.
EVALUATION
The RCWP participants began implementing BMP's be-
fore monitoring so a comparison of before and after condi-
tions is not possible. To attribute effects to BMP's the sites
must be examined in terms of water quality trends that
-127
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
may develop because of BMP's. Reids with BMP's will
have to be compared to fields without BMP's and to fields
that have not been farmed at all in a paired watershed
approach.
To handle the voluminous amount of data this project
will generate, a computer data base management system
has been developed. An existing system is currently being
modified to accommodate the project needs. It will store
files ranging from drill logs to water and soil quality data to
land use parameters, with the ability to merge and manip-
ulate these files for acceptance by the computer statistical
analysis package SAS. SAS will be used extensively
through descriptive statistics, regression analysis, analy-
sis of variance, trend analysis, and graphing and plotting
for visual interpretation.
Computer modeling will be an important step in the final
evaluation of BMPs' present and future effects on the
lakes based on field scale studies. Some modeling is al-
ready underway using the AGNPS model (Bosch et al.
1983) in a portion of the project area as part of a thesis
project. Ground water modeling and edge of field surface
water modeling will likely play important roles in the evalu-
ation as well.
CONCLUSIONS
Since October 1982 the major amount of effort has been
spent establishing monitoring sites. Seven out of 10 sites
have been established; three have in place full instrumen-
tation for ground water monitoring. Monitoring has not pro-
ceeded long enough for trend analysis nor has it produced
enough data for statistical comparisons between sites.
Completed monitoring data are showing drinking water
standards for nitrates being exceeded approximately 20
percent of the time in farm fields as opposed to 27 percent
noted for domestic wells. The highest mean nitrate con-
centrations are found in the outwash, followed by the till,
followed by sampling done in deep strata of outwash. Ni-
trate concentrations are higher in the shallow portions of
the aquifer and decrease with depth.
Till sites are not exempt from nitrate contamination as
one might guess. In fact, concentrations as high as 30 mg/
L have been recorded. Till system flow seems to be con-
trolled by fracturing within the till rather than movement
through the silty clay matrix. Outwash sites tend to have
the highest proportion of total nitrogen as nitrate while till
sites have a greater proportion of organic and, in some
cases, ammonia nitrogen. Denitrification studies show
losses of nitrogen to the atmosphere that amount to one-
half to one-third of the total nitrogen applied as fertilizer on
a yearly basis, implying that nitrogen routing field scale
models may underestimate losses from denitrification.
Pesticides (2,4-D) have been detected only once at very
low levels, in one field, and that was the pesticide applied
to that field the year before.
ACKNOWLEDGEMENTS: The Oakwood Lakes-Poinsett Proj-
ect of the Rural Clean Water Program is a cooperative special
study. The following are contributing agencies: Office of Water
Quality, Division of Geological Survey, South Dakota Depart-
ment of Water and Natural Resources, Water Resources Insti-
tute, Station Biochemistry, Cooperative Extension Service, Mi-
crobiology Department; South Dakota State University, U.S.
Environmental Protection Agency; Agricultural Stabilization and
Conservation Service, Economic Research Service, Soil Con-
servation Service, U.S. Department of Agriculture. Special
thanks to James Hyland, Alan Bender, Jeanne Goodman, and
the Office of Water Quality, for their review and comments in the
preparation of this paper.
REFERENCES
Agrichemical Age. 1984. Backtalk. In Aug.-Sept. issue.
Bosch, D.D., C.A. Onstad, and R.A. Young. 1983. A procedure
for prioritizing water quality problem areas. Proc. 1983 Sum-
mer Meet. Am. Soc. Agric. Eng. Montana State Univ., Boze-
man. June 26-29.
Crowder, B.M. et al. 1984. The effects on farm income of con-
straining soil and plant nutrient losses, an application of
CREAMS simulation model. Bull. 850 Pennsylvania State
Univ. Agric. Exp. Sta., University Park.
128
-------
Lake Quality
URBAN NONPOINT SOURCE IMPACTS ON A
SURFACE WATER SUPPLY
WILLIAM W. WALKER, JR.
Environmental Engineer
Concord, Massachusetts
ABSTRACT
Lake drinking water supplies are particularly vulnerable
to nonpoint sources because of cause-effect relation-
ships linking watershed characteristics, nutrient loading,
lake eutrophication, treatment plant disturbances, and
quality of water supplied to consumers (Walker, 1983).
This paper describes results from the first year of an in-
tensive watershed and lake monitoring program under-
taken by St. Paul Water Utility, MM. The objectives of the
program are: (1) to characterize the limnology of the sup-
ply lakes, (2) to quantify major sources of runoff, nutri-
ents, and other pollutants reaching the lakes, and (3) to
identify potential control measures for algal-related taste-
and-odor problems that have developed in recent years.
While diversions from other watersheds account for an
average of 85 percent of the flow through the lakes, runoff
and nutrient loadings from local watersheds undergoing
rapid urban development have become increasingly im-
portant. Site-specific and regional data indicate signifi-
cant effects of urban land uses on runoff and nutrient
export. Linked models relate watershed land uses to lake
water quality conditions and Water Utility impacts, ex-
pressed in terms of the frequency of nuisance-level algal
densities and potential costs of chemicals used for taste-
and-odor control. The models are used to estimate the
impacts of existing and future urban development in the
basin on lake water quality conditions and treatment
costs.
INTRODUCTION
This paper describes interim results from a diagnostic
study of the Vadnais Chain of Lakes, Minnesota, which
serves as the water supply for St. Paul. The study has
been undertaken by the St. Paul Water Utility to identify
causes and corrective measures for taste and odor prob-
lems that have occurred with increasing frequency over
the past several years. Historical data indicate that taste
and odor episodes are generally associated with algal
blooms (particularly, blue-greens) and lake turnover
(Walker, 1985b), as is common for water supplies derived
from eutrophic lakes or reservoirs (Lin, 1977). Following is
a description of monitoring and modeling efforts to quan-
tify the impacts of local watershed urbanization on existing
and future water quality conditions.
ST. PAUL WATER SUPPLY SYSTEM AND
MONITORING NETWORK
The Vadnais Lake Watershed (Fig. 1) is a system of 12
interconnected lakes with a drainage area of 6,227 ha
(15,381 acres). Morphometric, hydrologic, and water qual-
ity characteristics of major lakes in the watershed are
summarized in Table 1. Hydrologic data for the 1978-84
period indicate that the Utility diversions account for an
average of 85 percent of the inflow to the lake chain (66
percent from the Mississippi River (west) and 19 percent
from the Rice Creek Watershed (north)). The remainder of
the inflow is attributed to local watershed runoff (9 per-
cent) and direct precipitation on lake surfaces (6 percent).
The average withdrawal rate at the Utility's Vadnais Lake
intake (the only functional outlet from the watershed) is
187,000 m3/day (49.4 million gallons/day). The hydraulic
residence time of the main lake chain (140 days for Pleas-
ant-Sucker-Vadnais) is essentially determined by the Util-
ity's pumping and withdrawal rates. The Utility throttles
back on diversions from the Mississippi during periods of
high runoff from the local watershed. Lake level fluctua-
tions are relatively minor.
An intensive monitoring network was established in
1984 at various locations in the Vadnais Lake and Rice
Creek Watersheds (Fig. 1). Stations are of three types: six
lake stations (sampled biweekly), 22 tributary or diversion
stations (sampled weekly or biweekly), and four runoff sta-
tions (sampled on a storm event basis using continuous
flow monitoring and automated samplers). Under a coop-
129
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
VADNAIS LAKE WATERSHED.
BS-
Q Hitarshia Unit
» Tributary station
• flunoff Station
A L«*« station
Figure 1.—Vadnais Lake watershed.
erative arrangement, the U.S. Geological Survey operates
the runoff stations and is developing detailed hydrologic
simulation models of major subwatersheds.
Water balance calculations for the May-September
1984 period indicate that 86.7 percent of the total inflow to
the lake chain was subject to direct flow gauging and
quality sampling. The remaining inflows are attributed to
ungauged local watershed runoff (4.1 percent) and direct
precipitation on lake surfaces (9.2 percent). Monitoring
data from 1984 reflect a relatively high runoff period, ow-
ing primarily to an 11.4-cm (4.5-in.) rainstorm that oc-
curred in June. May-September total precipitation (60 cm)
and local watershed yield (7.6 cm) were both above 1978-
84 means for the same months (49 cm and 5.1 cm, re-
spectively). The increased runoff may be attributed to a
combination of climatologic factors and changes in land
use. Results of 1984 and other historical monitoring activi-
ties in the watershed are discussed in an interim report
(Walker, 1985b). The study is continuing to provide per-
MODELING APPROACH
INPUTS
MODELS
OUTWTS
•ATERSHED AREAS
LAIC USES
EXPORT COEFFICIENTS
SEGMENTATION
NORPHONETirf
ATMOSPHERIC LOADS
HISTORICAL COST DATA—' IMPACT -
. KATER3HEO » RUNOFF AND NUTRIENT EXPORT
> MASS-BALANCE »LAKE NUTRIENT CONCENTRATIONS
> OLDROPHYU.-A. TRANSPARENCY
• AL8AL NUISANCE-LEVEL FREQUENCY
ALGAL-DEPENDENT TREATMENT COSTS
Figure 2.—Modeling approach.
spectives on seasonal and year to year variability in water-
shed loadings and lake conditions.
Overall, land use in the watershed is 32 percent open,
21 percent wetlands, 11 percent lake surfaces, 4 percent
agricultural, 12 percent low-density residential (< = Limit/
acre), and 20 percent urban. Most of the high-density ur-
ban development is located in the lower portion of the
watershed draining directly into Vadnais Lake, where the
Utility's epilimnetic intake is located. The current pace of
development is rapid and many open (and, in some cases,
unprotected wetland) areas are being converted to high-
intensity urban uses. Areas to the north and east of Pleas-
ant Lake are slated primarily for low-density residential
development (Ramsey County Soil Water Conservation
Distr. 1985).
MODELING APPROACH
To provide a basis for evaluating eutrophication control
strategies and future land use scenarios, a mathematical
model of the Utility's watershed and lake system is being
developed (Fig. 2). The model consists of four compo-
nents:
1. watershed: estimates runoff and nutrient export from
each subwatershed as a function of land use
2. mass balance: routes water, phosphorus, and nitro-
gen through the stream and lake network to predict lake
nutrient concentrations
Variable
Table 1.—Lake morphometric, hydrologic, and water quality characteristics.
Units
Deep
Charley
• • Lake • •
Pleasant
Sucker
'Hydrologic conditions for May-September 1984.
'Mean concentrations, May-August 1984, Deep Lake and Charley outflows, Others 0-3 m Lake Stations.
Vadnais
Volume
Surface area
Mean depth
Maximum depth
Outflow
Residence time
Total phosphorus
Orthophosphorus
Total nitrogen
Inorganic nitrogen
Organic nitrogen
Reactive silica
Chlorophyll a
Secchi depth
__ . . .
106m3
km2
m
m
1 03 m3/day
days
ppb
ppb
ppb
ppb
ppb
ppm
ppb
m
Hydrologic Vai
0.39
0.28
1.4
5.0
11
35
lity Variables2 •
136
24
2,014
156
1,476
—
—
—
0.22
0.12
1.8
6.9
165
1.3
95
24
1,228
266
962
—
—
—
13.16
2.45
5.4
17.8
179
74
58
<11
1,120
116
1,004
2.0
27
1.5
0.79
0.24
3.3
7.9
181
4.3
58
<12
1,170
165
1,005
2.1
24
1.4
12.56
1.55
8.1
16.5
196
64
50
<10
831
126
705
1.5
19
1.8
130
-------
LAKE QUALITY
3. eutrophication response: predicts mean
chlorophyll a and transparency in each lake segment as
functions of nutrient concentrations
4. utility impact: predicts algal bloom frequency at the
Utility's intake and potential economic impacts, based
upon equivalent costs of treatment chemicals sensitive to
intake algal density or bloom frequency.
This section describes model structure and preliminary
calibration based upon monitoring data from May-Sep-
tember 1984. Results are limited by the fact that the early
spring runoff period was not monitored. Data from subse-
quent years will be used to refine the assessment.
A total of 19 subwatersheds, 12 lakes, and 18 mass-
balance segments are considered in the model, as illus-
trated in Figure 1. Mass-balance segments are of two
types: stream nodes and lake nodes. Stream nodes col-
lect runoff and nutrient loads from subwatersheds above
the lake system and are placed at each of the four runoff
monitoring stations (Charley Creek, Wilkenson Creek,
Lambert Creek, and Vadnais Creek). Nutrients are as-
sumed to be conservative in the stream nodes. Lake
nodes are located in each major lake or lake area. Pleas-
ant and Vadnais Lakes a/e each subdivided into two seg-
ments.
Model computations are performed in three steps. The
first step converts land use areas and export coefficients
for each subwatershed into stream flows and concentra-
tions required for mass-balance calculations. This is a
simple matrix multiplication problem. The second step
routes flow and nutrients through the stream and lake
network to predict average water quality conditions in
each segment using empirical nutrient retention and nutri-
ent/chlorophyll a relationships. The third step converts
predicted mean chlorophyll a concentrations in Vadnais
Lake into algal nuisance-level or bloom frequency (per-
cent of the time chlorophyll a exceeds 30 ppb). This statis-
tic is a reasonable, predictable surrogate for the frequency
of algal-related taste and odor episodes. Potential effects
on water treatment costs are also estimated, as described
here.
WATERSHED EXPORT MODEL
CALIBRATION
Existing land uses in each watershed unit are summarized
in Table 2, based upon maps prepared by the Ramsey
County Soil and Water Conservation District (1985). Cali-
bration of the watershed model involves estimating runoff
and nutrient export coefficients for each land use cate-
gory, based upon 1984 monitoring and other regional data
sets (Oberts, 1983; Payne et al. 1982; Nelson and Brown,
1983). Urban land use is a significant factor contributing to
runoff and nutrient export, as illustrated in Figure 3 for 17
regional watersheds with less than 50 percent agricultural
land use. Runoff and nutrient export coefficients selected
for each land use are summarized in Table 3 and com-
pared with other regional and nationwide estimates.
Export coefficients for agricultural land uses tend to be
highly variable because of differences in the types and
intensities of agriculture and soil characteristics. Agricul-
tural export coefficients in the lower range of those mea-
sured in other watersheds have been selected because
agricultural activities in the local watershed are generally
of low intensity. Model results are very insensitive to agri-
cultural export coefficients because this land use ac-
counts for only 4 percent of the watershed.
Because of lower use intensity and less impervious
area, low-density residential land uses are distinguished
from other urban land uses in the export matrix. Regional
data analyses indicate that phosphorus export is more
strongly correlated with urban land use when low-density
areas are excluded. The export estimates for the urban
land use category show good agreement with other data
sources in Table 3. Estimates for the low-density residen-
tial areas are somewhat subjective and may require fur-
ther investigation.
As applied to the Vadnais Lake watersheds, the esti-
mated export coefficients refer to May-September 1984
conditions. With these coefficients, the total flow, phos-
phorus, and nitrogen export monitored at the four runoff
monitoring stations agree with predicted values to within
10 percent; the predicted water balance on the entire lake
Table 2.—1984 land use breakdown for Vadnais Lake watershed model.
Watershed Land Uses (Acres)
ID
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
NAME
Gilfillan L
Black Lake
Birch Lake
Wilkenson S
Amelia Lake
Wilkenson L
Deep Lake
Charley Cr
Charley Lake
Pleasant W
Pleasant E
Sucker Lake
Gem Lake
Goose Lake
Lambert Crk
Vadnais Crk
Vadnais No
Vadnais So
White Bear
TOTALS
R<1
0
0
127
120
12
0
0
10
0
0
0
120
5
387
733
83
79
0
473
2,147
Areas rounded to nearest acre (1 acre = .40 ha)
ID
R<1
R>1
R— M
Cll
- Watershed Unit Number (Fig. 1)
• Residential < 1 unit/acre
- Residential > = 1 unit/acre
- Residential, Multi-Unit
- Commercial, Industrial, Institutional
R>1
380
125
0
19
60
27
159
177
0
60
526
92
42
22
128
17
7
0
0
1,840
AQR
WET
HI
LAKE
TOTAL
DEPTH
R-M Cll
0 0
0 0
6 39
0 62
0 0
11 0
0 0
8 0
0 13
0 68
0 0
12 19
0 0
27 166
94 111
0 4
0 0
0 0
18 70
176 551
= Agricultural
AGR
0
0
0
19
310
92
0
0
0
0
0
0
20
32
87
44
31
0
0
634
WET
91
61
21
521
183
272
213
264
24
52
440
91
35
17
847
2
92
0
7
3,231
OPEN
132
76
241
774
231
307
357
197
130
141
76
410
67
151
966
105
386
97
117
4,959
HI
0
0
12
15
0
0
0
0
0
0
0
0
0
18
42
0
0
0
0
87
LAKE
86
15
138
0
123
120
63
0
30
150
450
59
20
120
0
0
190
193
0
1,757
TOTAL
689
276
584
1,529
919
829
792
655
197
471
1,492
803
188
939
3,007
255
784
290
684
15,381
DEPTH
5.9
3.3
3.6
—
3.3
1.6
4.6
—
5.9
20.0
17.1
10.8
3.3
6.9
—
3.3
26.6
26.6
—
—
- Wetlands & Lake in Upper Watershed, Excluding LAKE
= Major Interstate Highways
= Lake Segment Surface Area at Lower End of Unit
- Total Watershed Unit Area
- Mean Depth of Lake Segment (Feet)
131
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Sr
a s
H.
fe _
hi
••
pa
40 60 80 100
I
Is
CO
a.
© «
Q »
20 40 60 BO iOO
I
CO
I
M ©
20 40 60 80
PERCENT URBAN
100
Legend
Vadnais Lake Watersheds are circled (period - May-September 1984; total
precipitation - 60 cm.
Data for other Twin Cities watersheds are from Oberts (1983), Payne et al.
(1982). Nelson and Brown (1983) (annual values, 1980 or 1982; total
precipitation - 51 -68 cm).
Watersheds with >50 percent agricultural land use are excluded.
Sandburg site ("sa") has 70 percent impervious area.
PDQ site ("pd") has construction activity.
Figure 3.—Runoff and nutrient export versus urban land use.
chain is accurate to within 1 percent. Additional monitor-
ing data and model refinements are required to translate
the export coefficients to annual estimates and to other
hydrologic years. Water balance calculations indicate that
local watershed runoff for the 1984 water year was 2.0
times the May-September runoff. Annual nutrient export
would probably be less than twice the May-September
export because stream nutrient concentrations generally
tend to be lower during the October-March period.
Future refinements to the watershed export model will
consider variations in soil type as well as land use. This
will require a detailed inventory of soil types and impervi-
ous areas in each watershed unit. To provide a basis for
estimating year to year variability, runoff and nutrient ex-
port coefficients should be tied to measured climatologic
factors (precipitation, potential evapotranspiration, and so
forth), and long-term stream gauging stations in the re-
gion. This will provide a higher resolution tool for evaluat-
ing site-specific development impacts and control strate-
gies.
MASS-BALANCE AND
EUTROPHICATION-RESPONSE MODEL
CALIBRATION
Mass-balance and eutrophication-response calculations
are performed using BATHTUB, a generalized computer
program designed for application of empirical eutrophica-
tion models to segmented lake or reservoir systems
(Walker, 1985a). Nutrient retention is predicted using em-
pirical models calibrated to large lake and reservoir data
sets. Phosphorus retention is estimated using a second-
order decay formulation (Walker, 1984a), based on availa-
ble phosphorus loading (a weighted sum of ortho and
non-orthophosphorus, with a greater weight on the ortho
component). Nitrogen retention is estimated using an em-
pirical formulation developed by Bachman (1980).
Normally, empirical estimates of nutrient retention terms
are accurate to within a factor of 2-3, and some adjust-
ment of the effective retention terms in each model seg-
ment is required to calibrate simulated nutrient profiles
against observed values. In this application, nitrogen and
phosphorus sedimentation rates have been reduced by 50
percent in Wilkenson and Deep Lakes to improve agree-
ment between observed and predicted nutrient concentra-
tions in the outflow from Deep Lake. These lakes are rela-
tively shallow and may have more efficient nutrient
recycling during the summer than the other lake seg-
ments. This adjustment has a minor impact on simulation
of nutrient profiles in the main lake chain (Pleasant-
Sucker-Vadnais) because the outflow from Deep Lake ac-
counts for only 5.1 percent of the flow, 4.7 percent of the
phosphorus, and 6.0 percent of the nitrogen discharged to
the main lake chain. The retention models have been
used without recalibration in other lake segments.
When a nutrient-balance model is used to predict mixed
layer concentrations, the term "retention" refers to loss of
nutrients from the mixed layer attributed to direct sedi-
mentation, adsorption, and algal uptake and settling. Sig-
nificant nutrient accumulation occurs in the hypolimnia of
stratified lake segments (Pleasant and Vadnais) during the
summer, owing to nutrient release from anoxic sediments
and seston. Models of this type are not designed to ac-
count for "internal loading" that occurs when the lakes
turn over and nutrients are recycled into the mixed layer in
the fall. During 1984, this process began during early Sep-
tember in Pleasant Lake and during early October in Vad-
nais Lake. In calibrating the model network, a May-Au-
gust averaging period has been used for the observed
data to exclude the fall turnover period. The lakes are iron
poor, and phosphorus recycled at fall turnover appears to
remain in the water column for extended periods. Despite
substantial increases in Vadnais Lake's mixed layer
orthophosphorus concentrations at fall turnover (from
< 10 to 180 ppb), a major algal bloom did not occur, appar-
ently because of unfavorable light, temperature, or mixing
regimes. Problems relating to the fall turnover period must
be addressed independently of the modeling effort, unless
the model structure is refined to account for effects of lake
turnover.
Empirical models predict mean, mixed layer
chlorophyll a and transparency as functions of total phos-
phorus and total nitrogen concentrations. Nitrogen is in-
cluded as a chlorophyll a predictor because time series
data indicate that the lakes approach a nitrogen-limited
state during certain periods, although phosphorus is the
most important growth-limiting nutrient. The empirical
132
-------
LAKE QUALITY
SECCHI (M)
012
CHL-A (PPB) TOTAL N (PPB) TOTAL P (PPB)
0 20 40 60 0 1000 8000 0 SO 100 190
Deep Leke
Charley Lake
Pleasant West
Pleasant East
Sucker Lake
Vadnals North
Vadnnis South
r—* 1
t-M
h-K
I XcH
Legend
Vertical bars indicate 95 percent confidence range for observed mean value.
0 = model prediction using measured runoff and nutrient loadings.
X = model prediction using runoff and nutrient loadings estimated from land
use.
Figure 4.—Observed and predicted trophic state indicators.
Table 3.—Export coefficients selected for Vadnals Lake Watershed model.
Land use
Rice Creek
Watershed
This District
study1 (1979)
Ayers
etal.
(1980)
Reckhow
etal.
(1980)
Jones &
Lee
(1982)
Low-density residential
Urban
Resid. 1 < acre
Multiple units
Commercial/lndustrial/lnstitutional
Agricultural
Mixed
Row crops
Pasture
Undeveloped
1 Total Phosphorus (kg/km2) •
50 45
120
162
388
149
50
110
24
60-110-2702
50-90-140
90-220-550
30-80-270
10-20-30
100
50
10
Low-density residential
Urban
Agricultural
Mixed
Row crops
Pasture
Undeveloped
Low-density residential
Urban
Agricultural
Undeveloped
loiai nitrogen (Kg/Km*) •
200
600
400
100
9
16
6
7
400-600-1200
900-1400-2500
400-900-2300
200-400-900
90-180-230
19
7.6
500/2503
500/200
300/100
1 Export coefficients for May-September 1984.
Total precipitation = 60 cm versus annual mean of 74 cm; others average annual values.
2Percentiles: 25-50-75 based upon nationwide data summary
3Nitrogen export tor Eastern/Western United States.
133
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
lake segments. Predictions are shown for two cases: (1)
using measured nows and nutrient concentrations at tne
four runoff-gauging sites, and (2) using values predicted
from land use. Runoff and export from ungauged water-
sheds are predicted from land use in both cases. Effects
of errors in the export model are minor, based on the
agreement between the two simulations. The vertical bars
reflect the approximate 95-percent error bounds for the
observed mean concentrations, calculated from the num-
ber of sampling dates and interdate variance at each sta-
tion. Further calibration of the model does not seem ap-
propriate, based upon the fact that predicted
concentrations are generally within the observed error
bounds.
Table 4 summarizes the water, phosphorus, and nitro-
gen balances for the entire watershed and for Vadnais
Lake alone, based on the calibrated watershed and mass-
balance models. Of particular interest is the local water-
shed loading component, which accounts for 15 percent
of the flow, 46 percent of the phosphorus loading, and 23
percent of the nitrogen loading to the entire watershed.
UTILITY IMPACT MODEL CALIBRATION
As a final modeling step, mean chlorophyll a concentra-
tion at the Utility's Vadnais intake is converted into two
measures of impact directly relevant to the Utility's opera-
tions: algal nuisance-level frequency and equivalent treat-
ment costs. These relationships are described here.
Algal nuisance-level or bloom frequency is defined as
the percent of the growing season that chlorophyll a con-
centrations exceed 30 ppb, a reasonable criterion for nui-
sance-level algal densities, based on 1984 Vadnais intake
time series data (Fig. 5) and general literature criteria
(Walmsley, 1984; Walker, 1984b). Bloom frequency is com-
puted from mean chlorophyll a using a frequency distribu-
tion model described by Walker (1984b). This statistic is a
reasonable surrogate for the frequency of algal-related
taste and odor episodes: It is limited by the fact that it does
not distinguish between diatom and blue-green blooms;
the latter are more directly implicated in summer taste and
odor problems, although Figure 5 suggests that diatoms
may be a factor in the spring.
Table 4.—Mass balances based upon gauged watershed loadings, May-September 1984.
Drainage Entire Lake Chain
Watershed
Area
km2
Flow
10«m3
Avail. P
kg
Total N
Ungauged Local Watersheds
Amelia Local
Wilkenson Local
Deep Local
Charley Local
Pleasant W Local
Pleasant E Local
Sucker Local
vadnais N Local
vadnais S Local
Gauged Local Watersheds
Charley Creek
Wilkenson South
Lambert Creek
vadnais Creek
SPWU Diversions
Fridley
Centerville Wells
Summary
Precipitation
External Inflow
Total Inflow
Outflow
Evaporation
Retention
3.22
2.87
2.95
0.67
1.30
4.22
3.01
2.41
0.39
2.65
12.47
19.51
1.03
5.57
56.71
62.28
62.28
0.22
0.20
0.22
0.05
0.12
0.34
0.27
0.20
0.03
0.38
0.88
2.09
0.10
25.00
0.27
3.34
30.37
33.71
30.08
3.62
0.7
0.6
0.6
0.1
0.4
1.0
0.8
0.6
0.1
1.1
2.6
6.2
0.3
74.2
0.8
9.9
90.1
100.0
89.2
10.8
123
75
76
19
74
158
161
97
7
36
378
1420
94
2970
10
152
5698
5849
1418
4431
2.1
1.3
1.3
0.3
1.3
2.7
2.8
1.7
0.1
0.6
6.5
24.3
1.6
50.8
0.2
2.6
97.4
100.0
24.2
75.8
747
432
360
93
292
635
644
440
39
933
1598
5568
242
34954
72
4682
47048
51730
27118
24612
1.4
0.8
0.7
0.2
0.6
1.2
1.2
0.8
0.1
1.8
3.1
10.8
0.5
67.6
0.1
9.1
90.9
100.0
52.4
47.6
Watershed
Sucker Outflow
Vadnais N Local
Vadnais S Local
Lambert Creek
Vadnais Creek
Precipitation
Total inflow
Outflow
Evaporation
Retention
Drainage
Area
km2
37.40
2.41
0.39
19.51
1.03
1.55
62.28
62.28
—
—
27.76
0.20
0.03
2.09
0.10
0.92
31.09
30.08
1.00
—
89.3
0.6
0.1
6.6
0.3
3.0
100.0
96.8
3.1
—
Vadnais Lake only
• • • • Avail. P
kg %
1511
97
7
1420
94
42
3171
1418
—
1752
47.7
3.0
0.2
44.8
3.0
1.3
100.0
44.7
—
55.3
T
kg
28258
440
39
5568
242
1294
35840
27118
—
8834
•%•«•! kl
78.8
1.2
0.1
15.5
0.7
3.6
100.0
75.7
—
24.3
Avail. P - Available Phosphorus Load (Walker, 1984a)
- .33 x Total P + 1.93 x Ortho P, for Lake Inflows
- Tola) R for Lake Outflows
Ortho-P/Total P - .57 for Local Watershed Loads
134
-------
f> 'V -r.-
LAKE QUALITY
Legend
Three-day moving average of dairy measurements at the SPWU Vadnals Lake
intake are shown. Dashed lines Indicate approximate nuisance-level criteria.
Figure 5.—Time series of algal counts and threshold odor
number at the SPWU Vadnals Lake Intake, 1984.
Another impact statistic is designed to provide approxi-
mate perspectives on potential economic impacts, ex-
pressed in terms of chemical treatment costs. Table 5 lists
total 1984 Utility costs for chemicals that are directly or
indirectly related to intake algal density or nuisance-level
frequency. Three chemicals (potassium permanganate,
sodium chlorite, and powdered carbon) are used explicitly
to control taste and odor problems and account for 75
percent of the total costs. Copper sulfate is applied weekly
during the growth season in an attempt to control algal
populations in Pleasant, Sucker, and Vadnais Lakes. Chlo-
rine is used as a disinfectant, and higher dosages are
required during periods of higher algal densities because
of increased chlorine demand attributed to organic materi-
als. Another relatively minor cost factor, anhydrous ammo-
nia, generates chloramines for disinfection; this treatment
has replaced direct chlorination to control trihalomethane
production, which is sensitive to source eutrophication
(Dorin, 1980; Bernhardt, 1980; Walker, 1983).
If the treatment plant operations were "optimized" to
apply these chemicals in exact proportion to their needs
based upon intake water quality, then most of the costs
(particularly for oxidants and carbon) would be nearly pro-
portional to algal bloom frequency. Chemicals would be
used for taste and odor control only when dictated by
intake quality. In practice, however, because of the risks
and uncertainty involved. The plant is operated in a con-
servative fashion; certain control chemicals are fed re-
gardless of intake water quality, but dosages are in-
creased during and following taste and odor episodes..
Thus, the actual cost sensitivity is less than that pre-
dicted by assuming that chemical costs are proportional to
algal bloom frequency. As further studies and experience
improve understanding of the cause-effect relationships
linking watershed conditions, lake dynamics, intake water
quality, treatment plant operations, and taste and odor epi-
sodes, the feasibility of optimizing treatment operations
and the sensitivity of chemical dosages to intake water
quality may increase. The intent of the economic model is
to provide an approximate estimate of cost sensitivity.
Other cost factors not considered include lake and water-
shed monitoring, labor for copper sulfate applications, and
energy Algal-dependent costs would increase by more
than an order of magnitude if major changes in the treat-
ment process train (addition of ozone or granular activated
carbon filtration) were required to solve this problem.
Based on the 1984 chemical costs ($462,266) and nui-
sance-level frequency at Vadnais South station (14.2 per-
cent), potential chemical costs associated with different
nuisance-level frequencies are estimated from:
C = 462 (F*/14.2) (1)
where,
C = annual chemical cost for taste and odor control
($1,000)
F* = algal nuisance level frequency (%)
This relationship is linked with the watershed, lake, and
chlorophyll a frequency distribution models to predict cost
sensitivity to watershed development and to variations in
diversion water quality. Predicted economic impacts
should be interpreted cautiously. Regardless of dosages
or cost, the chemical additions do not always effectively
control taste and odor problems. The resulting impacts
are real (obtained from consumer feedback) but difficult to
express in terms of dollars or to otherwise quantify.
URBAN NONPOINT SOURCE IMPACTS
The models described can provide perspectives on the
long-term effects of urban watershed development on eu-
trophication and related water quality conditions in Vad-
nais Lake. To define the potential range of urban develop-
ment impacts, three land use scenarios have been
simulated:
1. pristine: all existing urban, residential, and agricul-
tural areas converted to open land
2. existing: 1984 land uses
3. developed: all currently undeveloped and agricul-
tural areas (excluding wetlands) converted to urban.
Table 5.—St. Paul Water Utility algal-dependent chemical costs for 1984.
Chemical
Potassium permanganate
Sodium chlorite
Powdered carbon
Copper sulfate
Chlorine
Anhydrous ammonia
Total
Annual Cost
$217,661
102,102
27,630
36,984
61,506
16,383
$462.266
Use
oxidation of taste and odor compounds
oxidation of taste and odor compounds
adsorption of taste and odor compounds
lake applications for algal control
disinfection
disinfection/chloramines
135
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Results are summarized in Table 6 and Figure 6. Be-
cause of the nonlinear relationship between mean chloro-
phyll a and nuisance-level frequency, the latter is more
sensitive to watershed development. Potential algal-de-
pendent chemical costs are estimated at $159,000,
$462,000, and $839,000 per year for the three scenarios,
respectively For a fixed total water demand, pumping re-
quirements from the Mississippi River vary with watershed
land use. Table 6 shows that potential pumping cost sav-
ings attributed to changes in local runoff volume are gen-
erally less than 10 percent of the potential impacts on
chemical costs. Despite limitations in the cost estimates,
their order of magnitude appears to be significant. The
potential costs provide yardsticks for evaluating alterna-
tive control measures (best management practices, intake
techniques, etc.) from a cost-effectiveness standpoint.
The nutrient ratio, (N-i50)/P, decreases from "19.8 to
15.6 as watershed development increases. This ratio is an
approximate indicator of limiting nutrient (N-limited < 8,
Transition 8-16, P-limited > 16) (Walker, 1984a). Aside
from increasing the total nutrient supply and mean chloro-
phyll a concentration, increased urban runoff may drive
the lake system toward a nitrogen-limited state and further
promote the growth of nitrogen-fixing blue-greens, which
are of greater concern from a taste and odor perspective
than are diatoms or green algae.
The fact that a detectable bloom frequency (4.9 percent)
remains for the pristine case suggests that the effective
nutrient loading from Utility diversions (Mississippi River)
may have to be reduced to eliminate taste and odor epi-
8
S?
|"
Is
8 3
IS
13
S 10 19 20 29
MEAN CHLOROPHYLL-A (PPB)
30
Figure 6.—mean chlorophyll a and bloom frequency for
various watershed development scenarios.
sodes via nutrient control. As part of the diagnostic study,
diversion treatment schemes (nutrient removal or inactiva-
tion) and intake management techniques are being inves-
tigated for application to the system, along with watershed
management practices.
Another set of simulations is designed to estimate the
marginal impacts of urban development in each water-
shed unit on lake conditions and potential chemical costs.
These simulations involve converting 20 ha (50 acres) of
open land in each subwatershed into urban uses. Table 7
lists simulated increases in Vadnais South mean chloro-
phyll a, bloom frequency, and potential chemical costs at-
tributed to development in each watershed unit. While the
Table 6.—Simulated Impacts of urban watershed development on Vadnais Lake water quality and potential chemical costs.
Watershed Development
Factor
Mean intake Total P
Mean intake Total N
Mean intake (N-150)/P
Mean intake chl. a
Nuisance-level freq.
Equiv. chemical cost
Pumping cost increase
Net annual cost
Units
ppb
ppb
ppb
% > 30
$1,000/yr
$1,000/yr
$1,000/yr
Pristine
32.9
803
19.8
13.0
4.9
159
15
174
Existing
46.7
920
16.5
18.7
14.2
462
0
462
Developed
58.6
1064
15.6
24.3
25.8
839
-26
813
Table 7.— Marginal Impacts of a 50-acre urban development in each watershed unit.
Mean Chl. a Bloom Freq. Gross Cost
Watershed Unit ppb days $/yr
01 Gillfillan
02 Black
03 Birch
04 Wilkenson South
05 Amelia
06 Wilkenson
07 Deep
08 Charley Creek
09 Charley
10 Pleasant West
1 1 Pleasant East
12 Sucker
13 Gem
14 Goose
15 Lambert Creek
16 Vadnais Creek
17 Vadnais North
18 Vadnais South
19 White Bear
.005*
.011
.007
.016
.007
.016
.022
.030
.030
.033
.035
.092
.022
.032
.110
.142
.110
.142
.110
0.01*
0.03
0.02
0.05
0.02
0.05
0.06
0.09
0.09
0.10
0.10
0.26
0.06
0.09
0.32
0.41
0.32
0.41
0.32
312
687
437
999
437
999
1374
1874
1874
2061
2186
5747
1374
1999
6871
8870
6871
8870
6871
Unit Cost
$/acre-yr
6
14
9
20
9
20
27
37
37
41
44
115
27
40
137
177
137
177
137
•Simulated Increases in mean chlorophyll a and bloom frequency at Vadnais Intake
resulting from 50-acre (20 ha) urban development in each watershed unit.
Total watershed area = 15.381 Acres - 6,227 Ha.
Bloom frequency calculated for a 150-day growing season.
136
-------
LAKE QUALITY
simulated increases for an individual development are
small and would not be statistically detectable in a moni-
toring program, the cumulative effects of many develop-
ments are of major concern.
The results highlight spatial variations in Vadnais Lake
sensitivity to development in specific subwatersheds. Sen-
sitivity ranges over an order of magnitude. As expected,
watershed units closest to Vadnais Lake show the greatest
sensitivity. Units in the upper extremities of the watershed
show lower sensitivity because development impacts are
buffered by nutrient retention in upstream lake segments.
Expressed per unit of developed area, potential increases
in treatment costs attributed to urban development range
from $15-$437/ha/yr ($6-$177/acre/yr). Corresponding
cost savings attributed to reduced pumping costs for di-
versions from the Mississippi River are on the order of $57
ha/yr ($2/acre/yr).
CONCLUSIONS
These simulations provide approximate perspectives on
long-term impacts of urban watershed development on
the St. Paul water supply. The estimates do not reflect
potential short-term impacts of construction sites, which
have considerably greater runoff and nutrient export po-
tential, as compared with stabilized urban areas. To some
extent, dilution afforded by the Utility's diversions from the
Mississippi River tends to buffer the lakes from impacts of
local watershed development. A water supply without
such a significant diversion volume would be expected to
show a much higher land use sensitivity.
The assessment of urban impacts is obviously sensitive
to the selection of export coefficients for the various land
use categories. Refining the assessment will incorporate
error analysis concepts (Reckhow and Chapra, 1983). For-
tunately, in this case, good site-specific and regional data
bases exist for estimating export coefficients.
Additional data collected under the ongoing monitoring
program will refine the model structure and impact analy-
sis by considering soil types and year to year variations in
diversion water quality and in local runoff quantity and
quality The refined model will evaluate alternative mea-
sures for controlling eutrophication and taste and color
problems in the St. Paul water supply.
REFERENCES
Ayers, M.A., G.A. Payne, and M.R. Have. 1980. Effects of urban-
ization on the water quality of lakes in Eagen, Minnesota. U.S.
Geolog. Surv., Water Resources Division, St: Paul, NTIS PB-
81-154932, August 1980.
Bachman, R.W. 1980. Prediction of total nitrogen in lakes and
reservoirs. In Restoration of Lakes and Inland Waters. EPA-
440/5-81-010. U.S. Environ. Prot. Agency. Washington, DC.
Bernhardt, H. 1980. General impacts of eutrophication on pota-
ble water preparation. In Restoration of Lakes and Inland Wa-
ters. EPA-440/5451-010. U.S. Environ. Prot. Agency. Washing-
ton, DC.
Dorin, G. 1980. Organochlorinated compounds in drinking water
as a result of eutrophication. in Restoration of Lakes and In-
land Waters. EPA-440/5-81-010. U.S. Environ. Prot. Agency.
Washington, DC.
Jones, R.A., and G.F. Lee. 1982. Recent Advances in Assessing
Impact of Phosphorus Loads on Eutrophication-Related Water
Quality. Water Research, Vol. 16: 503-15.
Lin, S.D. 1977: Tastes and Odors in Water Supplies—A Review.
Circ. 127. III. State Water Survey. Urbana.
Metropolitan Council of the Twin Cities Area. 1981. A Study of
the Water Quahty of 60 Lakes in the Seven County Metropoli-
tan Area. Publ. No. 10-81-047. Data App. 10-81-047a.
. 1982. A 1981 Study of the Water Quality of 30 Lakes
in the Seven County Metropolitan Area. Publ. No. 10-82-
005047.
Nelson, L. and R.G. Brown, 1983. Streamflow and Water-Quality
Data for Wetland Inflows and Outflows in the Twin Cities Met-
ropolitan Area, Minnesota, 1981-1982. Open-File Rep. 83-
543. U.S. Geolog. Surv. St. Paul.
Oberts, G. 1983. Surface Water Management: Simplified Model-
ing for Watersheds. Publ. No. 10-83-130. Metropolitan Council
Twin Cities Area. St. Paul.
Osgood, R.A. 1984. A 1983 Study of the Water Quality of 28
Metropolitan Area Lakes. Publ. No. 10-84-037. Metropolitan
Council Twin Cities Area. St. Paul.
Payne, G.A., M.A. Ayers, and R.G. Brown. 1982. Quality of Run-
off from Small Watersheds in the Twin Cities Metropolitan
•Area, Minnesota—Hydrologic Data for 1980. Open File Rep.
82-504. U.S. Geolog. Surv. St. Paul.
Ramsey County Soil and Water Conservation District. 1985.
Land Use and Hydrologic Maps of the Vadnais Lake Water-
shed. Prep, for Vadnais Lakes Area Water Manage. Organ.
Reckhow, K.H., and S.C. Chapra. 1983. Engineering Ap-
proaches for Lake Management, Volume 1: Data Analysis and
Empirical Modeling. Butterworth Publ., Boston.
Reckhow, K.H., M.N. Beaulac, and J.T. Simpson. 1980. Model-
ing Phosphorus Loading and Lake Response Under Uncer-
tainty: A Manual and Compilation of Export Coefficients. Prep.
for Clean Lakes Section. U.S. Environ. Prot. Agency.
Rice Creek Watershed District. 1979. Wetland Preservation
Guide, Revised Oct. 3.
Walker, W.W. 1983. Significance of eutrophication in water-sup-
ply reservoirs. J. Am. Water Works Assn., 75(1): 38-42.
1984a. Empirical Methods for Predicting Eutrophica-
tion in Impoundments, Report 3: Model Refinements. Prep.
for Off. Chief Eng. U.S. Army. Tech. Rep. E-81-9. U.S. Army
Environ. Waterways Exp. Sta. Vicksburg, Miss.
_. 1984b. Statistical bases for mean chlorophyll a crite-
ria. Proc. Int. Symp. Lake Watershed Manage. 4th annu.
meet. N.Am. Lake Manage. Soc. Oct. 16-19, McAfee, NJ.
, 1985a. Empirical Methods for Predicting Eutrophica-
tion in Impoundments, Report 4: Applications Manual. Draft
prep, for Off. Chief Eng. U.S. Army. Tech. Rep. E-81-9. U.S.
Army Environ. Waterways Exp. Sta. Vicksburg, Miss.
_. 1985b. Compilation and Analysis of 1984 Monitoring
Data from the Vadnais Lakes Diagnostic Study. Prep, 'for
Board Water Comm. City of St. Paul, Minn.
Walmsley, R.D. 1984. A chlorophyll a trophic status classifica-
tion system for South African impoundments. J. Environ.
Qua). 13(1): 97-104.
~137
-------
NONPOINT SOURCE POLLUTION CONTROL FUNDING FOR LAKE
RESTORATION: A CASE STUDY AT CARLISLE LAKE
BARRY C. MOORE
WILLIAM H. FUNK
RICHARD BAINBRIDGE
Environmental Engineering Department
Washington State University
Pullman, Washington
ABSTRACT
This paper describes the results of a year-long monitoring
program at Carlisle Lake, a small hypereutrophic lake
located in southwestern Washington State. The objec-
tives of the study were to assess the current limnology of
the lake, to determine and quantify the nutrient sources to
the lake, and to make recommendations for possible res-
toration of the lake and for nutrient control in the water-
shed. The monitoring has indicated that nonpoint sources
associated with dairy and cattle operations contribute
most of the extermal nutrient load to the lake. Sponsored
by the Washington State Department of Ecology and con-
ducted by the Washington State University Environmen-
tal Engineering Department and the USDA Soil Conser-
vation Service, the study of Carlisle Lake illustrates how
an interdisciplinary approach may bring together the ex-
pertise to deal with nonpoint source pollution problems.
Some of the control measures that may be employed in
reducing the loading to Carlisle Lake and some nontradi-
tional funding sources and alternatives for these control
measures are discussed.
Nonpoint source pollution has been recognized as a major
problem in this country. However, the diffuse nature of the
pollution has often made abatement and control difficult,
not only from a technical standpoint but also from adminis-
trative and institutional viewpoints. In many cases, non-
point source pollution crosses political boundaries and
agency jurisdictions. Inclusion of the responsible parties
in the decisionmaking process can be essential to the
success of a control plan, and the expertise of various
grouos can be used to strengthen the technical and finan-
cial resources brought to bear on the problem.
This has been especially true in the case of Carlisle
Lake in western Washington State. By realizing that non-
point source pollution is a major contributor to the eutroph-
ication of Carlisle Lake, reduction of the nonpoint source
pollution has become an integral part of a restoration proj-
ect for the lake, and has involved personnel and funds
from the Washington State University Environmental Engi-
neering Section, the U.S. Department of Agriculture Soil
Conservation Service, and the Washington State Depart-
ment of Ecology. Currently, a Phase I feasibility study of
the lake has been completed by the Washington State
University Environmental Engineering Section and the
Phase II implementation project is being planned. This
paper describes some of the findings of the Phase I study
and plans for the lake restoration efforts and funding, with
emphasis on the interagency cooperative approach taken
to solve the technical and financial problems of nonpoint
source pollution control.
STUDY AREA AND METHODS
Carlisle Lake is a small (10 ha), shallow (average depth
.67 m), hypereutrophic lake located in the town of Ona-
laska, Lewis County, Washington (Fig. 1). The lake was
constructed in about 1910 as a mill pond for what was, at
the time, reputed to be the world's largest inland sawmill.
The lake was used for log storage and movement until
1942, when mill operations ceased. For many years, the
lake was a popular recreational site, providing swimming,
fishing, boating, and other activities for local residents. In
recent years, water quality has deteriorated significantly.
Heavy phytoplankton blooms occur in the summer. These
are caused by nonpoint source nutrient enrichment and
result in highly turbid water. Large stands of emergent
macrophytes have become established and are rapidly
filling the lake, compounding the problem. Fish stocked in
early spring do not survive past early June because of
dissolved oxygen depletion. Recreational use of the lake
has almost ceased altogether because of these problems.
The study conducted by Washington State University
included the identification and quantification of the
sources of pollution contributing to eutrophication in Car-
lisle Lake, and recommendations for possible restoration
options. (The project summary and data cited in this re-
port are to be found in Moore et al. 1985). Examination of
the Carlisle Lake watershed showed that dairy and cattle
FEET
500
Explanation:
Line of equal
water depth
Interval 2 Feet
Figure 1.—Map of Carlisle Lake.
138
-------
LAKE QUALITY
operations were the major sources of pollution along the
.inlet stream.
Monitoring stations were established on the inlet and
outlet streams, and samples were taken every 2 weeks,
from October 1983 to September 1984, to cover the entire
1983-84 water year. Sampling parameters included
stream flow, nitrogen and phosphorus fractions, tempera-
ture, dissolved oxygen, conductivity, pH, alkalinity, sili-
cates, and suspended solids. Piezometers were installed
around the lake for ground water analysis.
RESULTS AND DISCUSSIONS
The hydraulic budget (Table 1) and the nutrient (phos-
phorus) budget (Table 2) for the lake were constructed
from the data gathered on the inlets and outlet. Nonpoint
source pollution associated with the dairy and cattle oper-
ations were found to contribute about 75 percent of the
total annual phosphorus loading to the lake. The remain-
ing 25 percent of the total phosphorus load resulted from
internal nutrient recycling from sediment diffusion and
from macrophyte release.
Specific activities and sources that result in nonpoint
source pollution include wastewater spray irrigation, was-
tewater storage lagoon overflow, nutrients leached from
wastes deposited in fields, and direct deposition of animal
wastes in streams. Although the study was not designed
to quantify separately the specific pollution sources, some
inferences can be made from the available data.
Figures 2 and 3 show the stream discharge volumes
and phosphorus concentrations in the two major inlets,
Gheer Creek and the Ditch Inlet. Table 3 shows the sea-
sonal averages for stream discharge and phosphorus
Table 1 .—Summary hydraulic budget.
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Year
Inlets
total
507400
2777275
1435030
2488103
859544
1083250
1055821
995091
306378
64169
81450
84047
11737558
Outlet
-455290
-1233955
-596550
-674725
-1138555
-2129901
-1222997
-751091
-209236
-128312
-123801
-26194
-8690607
Ground
water
-52110
-1543320
838480
-1813378
27901 1
1046651
167176
-244000
-97142
64143
42351
-57853
-3046951
Note: Ground water estimate includes net rainfall, sheet flow, and ground water.
Losses are negative: gains are positive. Flows are in m3.
Table 2.—Summary phosphorus budget.
Gheer Ground
Ck Ditch Outlet water Sediment
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Year
4.829
70.398
46.472
160.064
22.748
16.799
17.84
18.121
17.472
3.136
6.818
4.996
399.693
32.235
43.405
76.534
179.51
16.269
46.176
15.679
35.862
8.567
1.512
4.072
8.202
468.023
-74.364
-270.078
-76.189
-89.681
-96.854
-189.012
-68.702
-50.429
-21.032
-22.890
-19.266
-4.331
-982.828
-2.606
-77.166
-41.924
-90.669
13.951
52.333
8.359
-39.040
-15.543
3.207
2.118
9.256
-196.237
39.906
233.441
-4.893
-159.224
33.886
73.704
26.824
35.486
10.536
15.035
6.258
0.389
311.349
(.' . 1000]
Figure 2.—Gheer Creek discharge and phosphorus concen-
trations.
Figure 3.-
tlons.
-Ditch Inlet discharge and phosphorus concentra-
Note: Losses to the system are negative, gains are positive. Sediment and ground
water values are net estimates in kg.
loadings for the 1983-84 water year. The discharge and
concentrations figures show that an inverse relationship
existed between these stream properties throughout most
of the study period; that is, as discharge increased, phos-
phorus concentration decreased, and vice versa. This pat-
tern may have resulted from a relatively constant source
of nutrients, so that the phosphorus tended to be diluted
during high flows and highly concentrated during low flow
periods.
Some exceptions to the general flow and concentration
pattern occurred in January 1984, in both inlet streams
and, in April/May 1984, in the Ditch Inlet. During these
periods, the phosphorus concentration rose as the dis-
charge increased, indicating either a flow-related
Table 3.—Practices eligible for cost-sharing under
the Agricultural conservation Program
(Soil Conserv. Serv. 1984)
Prevention of soil loss from water and wind erosion.
Permanent vegetative cover establishment
Diversions
Cropland protective cover
Streambank stabilization
Solution to water qualtiy problems
Stream protection
Sod waterways
Animal waste control facilities
Water management systems for pollution control
Conservation of soil and water through forestry
Forest tree plantations
Forest tree stand improvement
139
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
source(s) or an activity that was occurring during these
times and not during others. Substantial nutrient loading
occurred during these periods (Table 2).
The January peak flows were the highest of the water
year and occurred at the end of a 3-month period of high
rainfall and stream flows. As stated, both inlet streams
experienced increases in nutrient concentration and load-
ing associated with the January peak discharges. It is
possible that the high rainfall caused increased leaching
of nutrients, but such increased leaching usually occurs at
the onset of the high flow periods. This was not the case in
the Carlisle Lake watershed. The more likely source of the
increased nutrients was overflow from dairy wastewater
lagoons. The lagoons are used to hold water to be recy-
cled from milk parlor and feed lot cleaning. The lagoons
may have been full by January because of high levels of
rainfall, and the continued rainfall in January—especially
some high precipitation storm events—combined with the
washwater flows probably caused the lagoons to overflow
during this period.
The April/May high-loading period for the Ditch Inlet
has, we believe, a different but related cause. Manure
spreading of lagoon water is employed by the dairy Jo
fertilize some of the pastures, but it is used primarily to
control the water levels in the recycling lagoons. Manure
spreading was performed on fields draining into the Ditch
Inlet. This practice probably accounted for the phosphorus
increase in the Ditch Inlet during the spring.
From the evidence discussed and from other data col-
lected during the study, we can characterize the nonpoint
pollution sources that result in nutrient enrichment to the
inlet streams and to Carlisle Lake. Nutrients leached from
manure in the pastures adjacent to the streams and from
manure deposited directly in the streams seem to provide
a relatively constant source of loading. Manure spreading
and lagoon overflow are intermittent sources that contrib-
ute spike loadings, especially during high rainfall and
stream flow periods.
Controlling the pollution sources to Carlisle Lake will be
the goal of the Phase II project. Dredging will be required
to remove the sediments and macrophytes to deepen the
lake and reduce the internal nutrient loading. In the water-
shed, pollution-control measures will be aimed at reducing
or eliminating the nutrients from animal wastes that reach
the inlets. The amount of nutrient reduction needed in the
watershed was estimated using a phosphorus model
which has been found to yield reasonably accurate predic-
tions of phosphorus concentrations in western lakes (Ma-
hamah and Bhagat, 1982). A Vollenweider-type analysis
was used to predict trophic state in the lake following res-
CARLISLE LAKE TROPHIC STATUS AS A FUNCTION OF STREAM PHOSPHORUS LOADING AND DISCHARGE
Siren Dttclurge (
Figure 4.—Carlisle Lake trophic status predicted by the
Walker model.
toration with various phosphorus loadings and stream dis-
charges (Vollenweider, 1975). The model results are pre-
sented in Figure 4. Permissible loading is defined as that
loading which will result in an average annual phosphorus
concentration of 10 /ig/L or less. Dangerous loadings will
result in average annual phosphorus concentrations of 20
/ig/L or greater. The model predicts that a nutrient-loading
reduction of approximately 85-90 percent would be re-
quired to maintain acceptable water quality in the lake
after the restoration is completed.
The watershed control measures include creating addi-
tional lagoons to increase water storage, so that overflow
can be eliminated. The additional storage capacity should
give the dairy more flexibility in its management and re-
duce the need to use wastewater spray irrigation to control
the lagoon water levels. The wastewater spray irrigation
should be allowed only at distances of 30 m or greater
from the streams to allow for nutrient attenuation by the
vegetation and soils. In addition, the wastewater spray
irrigation will be moved from fields with soils of poorly
drained Lacamas clay to fields with the well-drained Cine-
bar sandy soils.
A vegetated buffer zone will be established around the
streams, which will be fenced to prohibit direct access by
animals. Many of the pastures have high water tables, and
overland sheet flow occurs during most of the year. The
sheet flow picks up nutrients from the fields and allows for
little nutrient attenuation by the vegetation or soils. Drain
tiles will be installed to improve drainage in these fields, to
prevent sheet flow, and provide for nutrient attenuation.
The technical aspects of the proposed pollution control
plan were worked out with the cooperation and help of
personnel from Washington State University, the Soil Con-
servation Service, and the Department of Ecology, work-
ing with local farmers, concerned citizens, and public
service groups from the Carlisle Lake area. It is clear that
close cooperation among these groups will be necessary
to secure the resources needed to put the plan into effect.
Funding for the many lake restoration projects in Wash-
ington is provided by the State from Referendum 39
funds. Referendum 39 established a lake restoration pro-
gram that is administered by the State Department of
Ecology. The program funds qualified restoration projects,
with 75 percent of funds provided by the State and 25
percent by local sources. In the case of Carlisle Lake, the
Lewis County Soil Conservation District has taken the
lead for raising local contributions. The costs of the Phase
II implementation project at Carlisle Lake are estimated to
be about $300,000. This requires about $75,000 in local
matching funds. Because Onalaska has a small popula-
tion base, local matching at this level poses a problem.
However, the Department of Ecology allows for inclusion
of in-kind services as part of the local match. Approxi-
mately 50 percent of the local match is still required to be
in cash outlays, but a lower percentage is allowed in cer-
tain cases.
The in-kind services provision of Referendum 39 may
be a distinct boon for nonpoint source pollution control
associated with lake restoration projects, because of the
nature of nonpoint sources. Nonpoint source pollution is
often associated with small operations that have limited
cash resources. The sources are frequently located in ru-
ral areas with low population densities and a low tax reve-
nue base. However, these operators will often own or have
access to heavy equipment that can be used for services
required in a restoration project, services which would
otherwise have to be purchased as part of the project. For
example, backhoes, a common farm implement, can be
used to dig lagoons, clear channels, or dig drain tile
ditches. The farmer must provide the labor, machinery,
and fuel, but the work can be credited to the project for
140
-------
part of the local match at the market price for comparable
services.
Even with credit for labor for the in-kind services, local
expense requirements can still be substantial. Some of
the control measures for nonpoint source pollution control
may actually improve production for dairies or farms. This
may possibly disqualify them from direct Referendum 39
funds under provisions that prohibit public funding for pri-
vate improvements. There is some ambiguity in the provi-
sions, but the cost to an individual farm, dairy, or range
operation can be large if even moderate pollution control
measures are required. These problems have suggested
the use of the Agricultural Conservation Program, admin-
istered by the Soil Conservation Service.
The Agricultural Conservation Program provides some
cost-sharing funding to individual farms for implementa-
tion of practices that result in soil conservation, pollution
abatement, or forestry practices. The practices eligible for
cost-sharing must meet specifications of the Soil Conser-
vation Service and the State Department of Natural Re-
sources. As one of the stated aims of the program is to
encourage public benefits, especially the goal of "prevent-
ing or abating pollution and other environmental degrada-
tion," linking the ACP with a public lake restoration project
should provide impetus for favorable deposition of the ap-
plication. Table 3 contains a list of the types of practices
eligible for cost-sharing under the ACP. Many of these
LAKE QUALITY
eligible activities are practices for eliminating or abating
nonpoint source pollution.
In conclusion, the Carlisle Lake Phase II restoration im-
plementation project is in the planning stages. Coopera-
tion among personnel from the involved agencies has led
to the unification of some diverse funding sources into a
cohesive program for abatement of a specific nonpoint
source pollution problem. Similar cooperative approaches
in other areas may be required to secure the funds and
technical resources to control nonpoint source pollution.
REFERENCES
Mahamah, D.S., and S.K. Bhagat. 1982. Performance of some
empirical models. J. Environ. Eng. Div. Am. Soc. Civil Eng.
108(EE4): 722-9.
Moore, B.C., et al. 1985. Assessment of the sources of enrich-
ment of Carlisle Lake and possible restoration alternatives.
Rep. to Lewis County Conserv. District. Washington State Wa-
ter Res. Center, Pullman.
Soil Conservation Service. 1984. Lewis County agricultural con-
servation program. Agric. Stabil. Conserv. Serv. Program
Bull. U.S. Dep. Agric., Chehalis, WA.
Vollenweider, R.A. 1975. Input-output models with special refer-
ences to the phosphorus loading concept in limnology. Sch-
weiz A. Hydrol. 37(9): 53-84.
141
-------
WHY SCOFIELD RESERVOIR IS EUTROPHIC: EFFECTS OF
NONPOINT SOURCE POLLUTANTS ON A WATER SUPPLY
RESERVOIR IN UTAH
DOYLE STEPHENS
U.S. Geological Survey
Salt Lake City, Utah
ABSTRACT
Studies since 1979 have classified Scofield Reservoir as
mesoeutrophic or eutrophic. The principal pollutants are
nutrients, trace metals, and sediments associated with.
nonpoint sources, such as construction of roads and
mine portals, domestic waste disposal, animal grazing,
and natural deposits of rock containing phosphate.
Blooms of blue-green algae, which have resulted in
fishkills, have corresponded to years of decreased inflow.
Biota populations during wet years were quite diverse,
with only minimal numbers of undesirable blue-green al-
gae. During 1981, however, the minimum usable water
storage decreased to 31 percent of the 25-year average
and blue-green algae blooms resulted in serious fishkills.
Concentrations of mercury in the water in Scofield Reser-
voir have caused concern for water users, but none of the
concentrations has exceeded revised State standards.
The mercury originates from coal particles within the
drainage basin, and most of it is bound as silicate in the
reservoir sediments and is not readily soluble. Several
management practices have been implemented to de-
crease nonpoint source pollution. Among these are an
improved waste disposal station for recreational vehicles,
a containerized waste system for fish cleaning, and
streambank stabilization.
OBJECTIVES OF THE INVESTIGATION
Extensive research on Scofield Reservoir in central Utah
has been conducted during the last few years by the Utah
Division of Environmental Health and the U.S. Geological
Survey. The objectives of the State studies were to quan-
tify nonpoint sources of nutrients entering the reservoir,
determine the trophic conditions, and identify possible
restoration methods. The Geological Survey, in coopera-
tion with the Bureau of Land Management, was interested
primarily in determining the effects on the reservoir of coal
mining within the drainage basin. Eutrophic conditions,
large sediment loads, and trace metal contamination were
problems anticipated prior to the start of the studies. Be-
cause of differences in objectives and the period of study,
there was little duplication of effort, and complementary
data bases were compiled.
DESCRIPTION OF THE AREA
Scofield Reservoir, at an elevation of about 2,320 m in
central Utah (Fig. 1), was formed in 1926 and enlarged in
1945 by dams on the Price River. Usable capacity of the
reservoir is 81 hm3 and maximum depth is 14m. The
drainage basin for the reservoir is 400 km2 and is primarily
mountainous, with elevations ranging from 2,312 to
3,183 m. The drainage basin contains large coal deposits
and the area has been mined since the late 1800's. The
major source of inflow to the reservoir is Fish Creek, which
contributed 72 percent of the total inflow during the 1980
water year. Minor inflows were from Mud Creek (16 per-
cent), and Lost and Pondtown Creeks and from precipita-
tion (12 percent). Agricultural use of the basin is limited to
grazing of sheep (15,000 head) and cattle (5,000 head).
Ninety-three percent of the drainage basin is forest or
range land.
The reservoir supplies water for domestic use to 20,000
people downstream, and it provides recreational opportu-
nities for about 70,000 visitors per year. There are 1,100
recreational camping or cabin units on the shores. Sani-
tary facilities for the camps, cabins, and the town of Sco-
field (200 people) are limited to pit privies and septic tanks.
ALGAL BLOOMS AND FISHKILLS
Periodic data for algal populations and blooms are availa-
ble since 1975. Colonies of objectionable blue-green al-
gae were identified in the reservoir in 1975, but their num-
bers did not reach bloom proportions, most likely because
of increased tributary flow and high water in the reservoir.
During most wet years, algal populations are quite di-
verse, with few blooms of blue-green algae. Considerable
data on the algal community have been collected since
1981, and this may be summarized by examining the fluc-
tuations of two blue-green algae, Aphanizomenon flos-
aquae and Anabaena flos-aquae, and the diatom
Stephanodiscus minutula, which are good indicators of eu-
trophication in reservoirs in Utah. The density of these
blue-green algae and the percentage of the diatom (rela-
tive to all diatoms) are plotted in Figure 2 for 1981-84.
Although both groups of organisms are indicators of eu-
trophication, their densities typically are inverse. In-
creased densities of the blue-green algae indicate blooms
during August 1981 and October 1982. Possible bloom
conditions for the diatom are indicated during early Au-
gust 1981, September 1981, and August 1983. The phyto-
plankton bloom in August-September 1981, resulted in a
considerable fishkill. The bloom of blue-green algae in
October 1982 did not result in a major fishkill, because
water storage had reached an historic high in June and
was still relatively high during October. Similarly the
diatom bloom of August 1983 occurred at a time when the
water storage historically was very large.
Fishkills in Scorield Reservoir have been reportea dur-
ing 6 years since 1960, and most were associated with
small discharges from the principal inflowing stream or
decreased reservoir water storage (Table 1). During most
years, Fish Creek contributes about 70 percent of the an-
nual inflow to the reservoir. The Utah Division of Wildlife
Resources (Donaldson, 1984) has found that fishkills oc-
cur in years when the annual value for mean daily dis-
charge of Fish Creek is less than 1.13 m3/sec. This obser-
vation has been verified by all data since 1960, with the
exception of data from 1963 and 1966. In 1963, the flow
from Fish Creek decreased to 63 percent of the 25-year
average daily flow of 1.5 m3/sec, and the minimum water
storage in the reservoir was only 52 percent of the 25-year
average. The flushing rate (total inflow divided by Price
River outflow) in 1963 was about 1, which may have pre-
vented a fishkill. During 1966, the average daily flow of
Rsh Creek was 67 percent of the 25-year average, but
142
-------
LAKE QUALITY
/ > /V
i ; '" "' \
UTAH
DRAINAGE-BASIN BOUNDARY
01234 5 KILOMETERS
Figure 1.—Location of Scofield Reservoir and associated streams.
minimum water storage in the reservoir was 143 percent
of the average.
Analysis of published (Waddell et al. 1983; Denton et al.
1983) and calculated water budgets since 1960 indicate
that fishkills occur 80 percent of the time the annual flush-
ing rate for the reservoir is less than 0.85. No fishkills have
been reported during years when the flushing rate was
DENSITY OF BLUE-GREEN ALGAE.IN CUBIC CENTIMETERS PER UTER.AND THE
DIATOM. IN RELATIVE PERCENTAGE OF THE TOTAL POPULATION OF DIATOMS
-BULK-GREW AlCAC
-DIATOM
b in
a
B
s
S
Figure 2.—Fluctuations of the blue-green algae Anabaena
//os-aquae, Aphanlzomenon flos-aquae, and the diatom
Step/ianodVscus mlnutula In Scofield Reservoir, 1981-84.
greater than 1.1. It is likely that fishkills result from a com-
bination of environmental factors that are intensified by
decreased tributary inflow. These factors are large popula-
tions of blue-green algae that release biological toxins,
oxygen demands resulting from respiration and decompo-
sition of algal populations, increased rate of warming of
lake water, and increased volume of the anaerobic hypo-
limnion.
Considerable reservoir data were collected before and
after the 1981 summer fishkill (Utah Dep. of Health, 1984),
which indicate the relationship between total reservoir wa-
ter storage and the actual volume of fish-habitable water
(Fig. 3). Total water storage began decreasing in June,
and during August it decreased 15 percent. The volume of
fish-habitable water containing a minimum dissolved oxy-
gen concentration of 5.5 mg/L, which is sufficient to sup-
port a fishery, decreased at about the same rate until Au-
gust. During the first half of August, the habitable water
decreased about 60 percent. This was caused by an ex-
pansion of the anaerobic hypolimnion and decreased vol-
ume of total storage, which greatly decreased the living
area for fish and resulted in an estimated fishkill of
200,000.
TROPHIC STATE OF THE RESERVOIR
Because of recurring algal blooms and large concentra-
tions of nutrients, Scofield Reservoir has been classified
as eutrophic (Denton, 1980) to meso-eutrophic (Waddell et
al. 1983; Denton et al. 1983). Data collected by the U.S.
Geological Survey since 1979 indicate that 47 percent of
all measurements of total phosphorus in the epilimnion
143
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 1.—Fishkills and minimum water storage in Scotieid Reservoir, ratio of inflow to outflow, and discharge of Fish Creek.
Percentage
Water year
1960
1961
1972
1976
1977
1981
Type and time
of fishkill
major summerkill
major winterkill
minor summerkill
minor summerkill
major summerkill
major summerkill
Minimum useable
water storage
(thousands of cm3)
2,070
680
24,040
38,350
21,840
9,390
of 25-year
average
minimum
storage
7
2
80
127
73
31
Total inflow
Total outflow
0.95
1.09
.74
.84
.49
.73
Mean daily
discharge of
Fish Creek
(mVsec)
0.88
.42
.91
1.01
.29
.89
1981
Figure 3.—Total and fish-habitable water stored In Scofleld
Reservoir during the 1981 fishkill.
and 69 percent of phosphorus measurements in the hypo
limnion have exceeded the Utah water quality standard of
0.025 mg/L. The seasonal cycling of total phosphorus,
using data collected by the Utah Division of Environmental
Health and the U.S. Geological Survey, is shown in Figure
4. Phosphorus concentrations typically increase in the hy-
polimnion during winter stratification because of release
from the sediments in the reservoir. During turnover in
May or June, concentrations generally decrease in the
hypolimnion and increase in the epilimnion. The increase
in algal and bacterial populations during the summer,
coupled with increased stratification and development of
an anaerobic hypolimnion, decreases the phosphorus
concentration in the epilimnion. In July or August, when
summer stratification is at a maximum, release of phos-
phorus from the anaerobic sediments results in large con-
centrations in the hypolimnion. During fall turnover, in late
August or September, the phosphorus accumulatedjn the
hypolimnion is released to the overlying waters and sur-
face concentrations of phosphorus typically increase. It is
during the August-September turnover that recycled
phosphorus is made available to the blue-green algae and
late summer blooms can occur.
Denton et al. (1983) reported that ratios of nitrogen to
phosphorus mostly were greater than 15 during 1981, indi-
cating phosphorus limitation of algal populations. During
1982, ratios in the spring indicated phosphorus limitation
and summer ratios generally identified a nitrogen limita-
tion. Bioassay tests of algal growth potential indicated that
on June 27, 1984, when the ratio of nitrogen to phos-
phorus in the reservoir water was 43 to 1, the water was
primarily phosphorus limited. Additions of small concen-
trations of nitrogen (as nitrate), as well as phosphorus (as
orthophosphate), however, stimulated nearly four times as
much algal growth as phosphorus alone. On July 31,
1984, when the nitrogen-to-phosphorus ratio was 40 to 1,
the addition of nitrogen as well as phosphorus to the test
...
~ s ~ = s - s ;
Figure 4.— Total phosphorus in surface and deep water In
Scofield Reservoir, 1979-84.
algae resulted in nearly 17 times the growth of phos-
phorus alone. By Sept. 19, the ratio was 50 to 1, but the
addition of nitrogen resulted in only a twofold increase in
algal mass when compared with the addition of phos-
phorus alone. It is likely that phosphorus is limiting during
most years, particularly when large populations of blue-
green algae, such as Aphanizomenon and Anabaena, are
present, because such algae are capable of fixing atmos-
pheric nitrogen into a form available to organisms.
SOURCES OF NUTRIENTS
Because of basin slopes of 0.38 in Boardinghouse and
Eccles Canyons, and the construction of extensive roads
and coal mine portals in Eccles Canyon, large quantities
of sediment are transported into Mud Creek and then into
Scofield Reservoir. Suspended sediment loads in Mud
Creek during July-September storms averaged 18,000 kg
during 1983-84. Loads during thunderstorms in spring
were believed to be even larger. Soils in the Mud and Fish
Creek drainages contain considerable quantities of natu-
rally occurring phosphorus. Soils collected in Boarding-
house and Eccles Canyons during 1984 contained 300-
500 mg of total phosphorus per kilogram. Suspended
stream sediments from the Fish Creek drainage were re-
ported to contain 440-900 mg of total phosphorus per
kilogram, with biologically available phosphorus averag-
ing about 49 percent (Denton et al. 1983).
Considerable loads of nitrogen and phosphorus are dis-
charged to the reservoir by Mud and Fish Creeks, the
major inflowing streams to Scofield Reservoir (Table 2).
Loads of nitrogen and phosphorus increased considerably
during 1984 compared with 1983, although the total inflow
to the reservoir was nearly the same. The distribution of
loads also changed considerably during 1984, showing a
greater proportion of nutrients, relative to the discharge,
entering from the Fish Creek drainage. During 1980 and
1983, the proportion of total nutrient loads, relative to dis-
charge, was greater for the Mud Creek drainage. The in-
creased loads in Mud Creek during 1984 may be due, in
144
-------
part, to increased construction in the Eccles Canyon.
There is no evidence to explain the increased loads during
1984 in Fish Creek.
Clyde et al. (1981) reported that wells in the Scofield
area yielded water with an average phosphorus concen-
tration of 170 /ig/L, except in several of the subdivisions
adjacent to the lake. In these areas, the water from several
shallow wells contained mean total phosphorus concen-
trations greater than 1,400 /tg/L. Relatively large concen-
trations of nitrate nitrogen also were common in the same.
wells, with concentrations ranging from 1,000 to 14,000
pg/L Contamination of the ground water by disposal of
domestic waste was suspected.
PHOSPHORUS IN LAKE SEDIMENTS
Selective chemical extraction of phosphorus from sedi-
ments in Scofield Reservoir by Messer and Ihnat (1983)
indicated that nearly 200 mg of phosphorus per kilogram
of sediment were potentially available for biological up-
take. Although the quantity of phosphorus in the sedi-
ments is quite large, the actual quantity released is small
because of the large concentrations of iron in the sedi-
ments. The phosphorus is in the form of iron-oxide gels,
which release phosphorus when the iron in the gel is re-
duced to the ferrous form during anaerobic periods of
stratification.
Release rates of available orthophosphorus from intact
sediment cores were determined under aerobic and an-
aerobic conditions by the authors. Some differences in
release rates depended on the sampling location of the
core. Under anaerobic conditions at a typical hypolimnion
temperature of 15°C, however, phosphorus release gen-
erally peaked within 15 days, at a rate of about 2 mg/m2
per day and decreased to about 1.3 mg/m2 per day for the
remaining 17 days of the experiment. Under aerobic con-
ditions, at a typical epilimnion temperature of 20°C, the
maximum release rates were about 0.6 mg/m2 per day,
decreasing to about 0.3 mg/m2 per day after 28 days.
These rates are slightly less than those reported for Deer
Creek Reservoir (75 km to the northwest) by Messer and
Ihnat (1983).
The quantity of orthophosphorus released from the sed-
iments during summer stratification was calculated using
data obtained during 1983. The reservoir was stratified
from the end of July to the end of September, about 70
days. During this time, water covering about 380 ha was
anaerobic. Using a release rate of 1.3 mg/m2 per day, a
total of 360 kg of phosphorus could have been released
from the sediments to the hypolimnetic water. This indi-
cates that phosphorus release from anaerobic lake sedi-
ments is of minor importance when compared with loads
of total phosphorus entering from streams (Table 2).
TRACE METALS
The presence of an active mining industry and large sedi-
ment loads created by construction of road and mine por-
tals in the basin indicated that trace metals may be a
LAKE QUALITY
pollution problem in the reservoir. Prior to November
1984, the State water quality standards for the protection
of aquatic wildlife did not incorporate water hardness into
the allowable concentrations for trace metals. Because of
the accumulation of literature indicating that trace metal
toxicity to aquatic wildlife is dependent on water hardness,
trace metal standards were revised to less stringent con-
centrations. This necessitated revisions in the interpreta-
tion of the effects of trace metals on Scofield Reservoir,
particularly for mercury, which had been identified previ-
ously as a potential pollutant problem (Clyde et al. 1981).
Under the old standards, 39 of 46 analyses for total mer-
cury in reservoir waters indicated concentrations in ex-
cess of State standards of 0.05 /ig/L The mean of all the
violations was 0.17 ^g/L. Under the revised standard of 1
figIL, none of the samples exceeded the standard. Analy-
ses of fish tissue also indicated the existence of small
concentrations of mercury in the reservoir waters, but the
concentrations were considerably less than the allowable
limit for edible fish (Denton et al. 1983). The small concen-
trations of mercury probably originate from coal particles
transported into the reservoir. The composition of coal in
the area indicates that the mercury content may be as
large as 0.25 part per million by weight (Smith, 1981).
Manganese and zinc are the only other trace metals
that are potential pollutants in the reservoir. The standards
of the U.S. Environmental Protection Agency (1976) for a
public water supply prohibit exceeding a total manganese
concentration of 50>g/L. During periods of stratification,
concentrations of total manganese may range from 500 to
700 uglL in the anoxic hypolimnion in Scofield Reservoir.
The concentrations increase considerably in the hypolim-
, nion during summer stratification and abruptly decrease
during spring and late summer turnover. It is during these
turnover periods that the concentrations increase in the
epilimnion because of mixing with hypolimnetic water en-
riched with manganese. Concentrations of manganese in
the epilimnion rarely exceeded the standard during 1979-
82, but during the late summer turnover in 1983 and 1984,
near-surface concentrations tended to range from 80 to
120 /xg/L Total, and to a lesser extent, dissolved iron con-
centrations were similar in magnitude to those of total
manganese during periods of stratification and mixing.
Concentrations of total zinc in the reservoir occasionally
exceed 150 pg/L but the dissolved form rarely exceeds
the State standard of 50 pg/L Some cycling of both forms
of zinc occurs in the surface and bottom waters, but it
does not conform to well-defined periods of stratification
and mixing as is the case with manganese.
EXTRACTIONS FROM SEDIMENT CORES
Sediment cores were collected at five sites in the reser-
voir, radioisotope methods were used to determine sedi-
mentation rates, and chemical extractions were per-
formed (Skei and Paus, 1979) to provide an estimate of
the ease with which trace metals may be removed from
the sediment. In general, isotope-dating methods indi-
cated that sedimentation rates within the reservoir were
Table 2.—Loads of nitrogen and phosphorus entering Scofield Reservoir in gauged streams during 1980,1983, and 1984
water years. Numbers in parentheses are the percent of total.
Water year
Fish Creek
Mud Creek
Total
Inflow
(million
tn>)
74.9 (82)
16.2 (18)
91.1
1980
Dissolved
nitrogen
(XT)
57 (79)
15 (21)
72
Total
phosphorus
(MT)
4.0 (71)
1.6 (29)
5.6
Inflow
(million
.m3)
100.9 (81)
23.9 (19)
124.8
1983
Dissolved
nitrogen
(MT)
37 (79)
10 (21)
47
Total
phosphorus
(MT)
15 (73)
5 (27)
20
Inflow
(million
m*)
96.6 (78)
27.5 (22)
124.1
1984
Dissolved'
nitrogen
(MT)
110 (81)
25 (19)
123
Total
phosphorus
(MT)
79 (86)
13 (14)
92
145
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
not uniform. Sedimentation rates near the major sources
of inflow could not be determined accurately, but they
were believed to be greater than rates in the central part of
the reservoir. Sediments accumulating in the shallow ar-
eas periodically move toward the center of the lake, where
the sedimentation rate is about 1 cm per year. An even
smaller rate of deposition, 0.3 cm per year, has been de-
termined for an area near the dam, where the velocity of
the current is slowest.
Concentrations of arsenic, cadmium, and cobalt in all
cores were very small, and these elements primarily were
present in a tightly bound form. Copper concentrations in
the sediments ranged from 15 to 20 ^g/g. Most of the
copper was present in the nonsilica fraction, most likely
tightly bound to clays or in sulfide compounds. The quan-
tity of copper potentially available to the biota was usually
about 5 ng/g. Concentrations of manganese at most sites
showed a gradual decrease from the sediment surface to
the bottom of the 20-40 cm core. Concentrations of total
manganese varied from 400 to 700 ^g/g in the surface
segments and decreased about one-third in the bottom
segments of the core. This is a typical pattern for manga-
nese in lake sediments. The oxidized forms of manganese
are deposited in the sediments and gradually buried. As
oxygen is excluded from the cores and reducing condi-
tions are established, the oxides are reduced and made
soluble, releasing manganese with a + 2 valence which
migrates upward in the pore water (Skei and Paus, 1979).
Mercury concentrations in the sediments generally
were less than 0.02 uglg, particularly in the shallow, re-
cently deposited alluvium from Mud and Fish Creeks.
Sediments from deeper sites had considerably larger con-
centrations of mercury, with a maximum concentration of
6 uglg. At sites where concentrations were larger, the con-
centrations were present in the least soluble (silicate)
form. Concentrations in a biologically available form did
not exceed 0.02 /ig/g, and they generally were present in
concentrations less than 0.01
CONTROL OF EUTROPHICATION
THROUGH MANAGEMENT
The State of Utah has proposed several management
practices for controlling nonpoint and point sources of nu-
trients (Denton et al. 1983). These methods consist of
integrating public education programs with physical,
chemical, and biological management practices to de-
crease the rate of eutrophication and restore reservoir
quality. These practices may be categorized as lakeshore
and reservoir management.
Lakeshore management would decrease the quantity of
sewage-associated nutrients entering the lake from non-
point sources, such as holding tanks and leach fields. It
would prohibit the installation of additional sewage-hold-
ing tanks in subdivisions around the lake shore, require
sanitary dump stations for recreational vehicle parks, and
require centralized septic tanks with drain fields located
far from the lake shore in several subdivisions. Extensive
wetlands would be created near the mouth of Mud Creek,
which would be managed to preclude animal grazing and
to encourage plant growth so as to decrease nutrient con-
centrations in the inflowing water. The plant growth would
be periodically harvested and removed from the area. Wil-
lows planted along the creek drainage area near the reser-
voir would stabilize stream banks. The wetlands proposal
could not be funded, and it has been temporarily sus-
pended from consideration.
Reservoir management is designed to decrease the ef-
fects of eutrophication by controlling nutrients after they
reach the reservoir. Fishery practices, such as no-limit or
commercial fishing, would decrease the nutrient load
caused by chemical treatment periodically used to control
rough fish. Fish cleaning, with return of the entrails to the
reservoir, is estimated to contribute more than 180 kg of
phosphorus annually (Denton et al. 1983). Prohibition of
fish cleaning on the reservoir and along the shore, and
construction of fish-cleaning stations in the camping areas
would decrease this source of nutrients. Public education
and more restroom facilities would be used to decrease
waste disposal by boaters. Other water management
practices considered but rejected as being too costly or in
conflict with water users included: alum treatment to pre-
cipitate phosphorus, followed by fly ash addition to seal
the nutrients in the sediment ($160-$600 per hectare);
aerators to decrease anaerobic conditions under which
phosphorus leaches from the sediments ($6,000-$12,000
annually); dredging to remove sediment ($1,000 per hect-
are); installation of a large pipe, which would allow hypo-
limnetic water to be discharged during the summer (more
than $43,000); water drawdown during, midsummer to en-
courage complete water mixing and prevent formation of
an anoxic hypolimnion.
Several practices had been implemented by 1985. An
improved waste disposal system for recreational vehicles
and a containerized fish-cleaning station with grinder and
waste system connections are in use at the State park
campground. Posting of antipollution signs has increased
public awareness, and the channel has been stabilized
along the downstream reach of Mud Creek. Subsequent
monitoring and study is planned by the State to determine
the effectiveness of these control practices.
All lakes are different, and the effect of nonpoint source
pollutants cannot be predicted unless the chemistry and
biology of each lake are understood. Management prac-
tices to control eutrophication need to be deferred until the
dynamics of specific lake systems are fully known.
REFERENCES
Clyde, C.G., et al. 1981. Water quality in Pleasant Valley, Utah.
Rep. WG-281, Utah Water Res. Lab., Logan.
Denton, R.L. 1980. Water Quality of Selected Utah Impound-
ments. Bur. Water Pollut. Control, Utah Dep. Health. Salt Lake
City.
Denton, R.L., M.I. Cox, and L.B. Merritt. 1983. Scofield Reser-
voir Phase 1 Clean Lakes Study. Bur. Water Pollut. Control,
Utah Dep. Health. Salt Lake City.
Donaldson, W. 1984. Personal communication. Utah Div. Wildl.
Resour., Helper, UT.
Messer, J.J., and J.M. Ihnat. 1983. Reconnaissance of sedi-
ment-phosphorus relationships in some Utah reservoirs.
Rep. UWRL/Q-83/03. Utah Water Res. Lab., Logan.
Skei, J., and P. Paus. 1979. Surface metal enrichment and parti-
tioning of metals in a dated sediment core from a Norwegian
fjord. Geochim. Cosmochim. Acta. 43: 239-46.
Smith, A.D. 1981. Muddy Creek coal drilling project, Wasatch
Plateau, Utah. Spec. Stud. 55. Utah Geolog. Mineral Surv.,
Salt Lake City.
U.S. Environmental Protection Agency. 1976. Quality Criteria for
Water. Off. Water and Hazardous Mater, U.S. Govt. Printing
Off., Washington, D.C.
Utah Department of Health. 1984. Scofield Lake Phase 1 Study-
Water Chemistry and Phytoplankton Data. Bur. Water Pollu-
tion Control, Salt Lake City.
Waddell, K.M., D.W. Darby, and S.M. Theobald. 1983. Chemical
and physical characteristics of water and sediment in Scofield
Reservoir, Carbon County, Utah. Open-file Rep. 83-252. U.S.
Geolog. Surv.
146
-------
TROPHIC STATE RESPONSE TO NONPOINT POLLUTION CONTROL:
APPLICATION OF COUPLED MICROCOMPUTER MODELS
TO THE GREAT LAKES
MARTIN I AUER
Department of Civil Engineering
Michigan Technological University
Houghton, Michigan
THOMAS M. HEIDTKE
Department of Civil Engineering
Wayne State University
Detroit, Michigan
RAYMOND R CANALE
Department of Civil Engineering
University of Michigan
Ann Arbor, Michigan
ABSTRACT
Phosphorus loading from the Fox River (Wisconsin) pro-
duces a gradient in trophic state along the major axis of
Green Bay (Lake Michigan) ranging from hypereutrophic
to oligotrophic. Water quality problems associated with
the-gradient include high turbidity, excessive algal
growth, and dissolved oxygen depletion. The Fox River
contributes 78 percent of the tributary total phosphorus
load to Green Bay; more than half of that load originates
from nonpoint sources. A unit area load (UAL) based mi-
crocomputer model is used to generate estimates of non-
point total phosphorus loads as a function of land use and
soil texture in the Fox River watershed. Phosphorus loads
are input to a water quality microcomputer model which
calculates the total phosphorus and chlorophyll concen-
trations, water transparency, and trophic state corres-
ponding to that load. Changes in water quality and trophic
state are examined under existing conditions and two
hypothetical land use scenarios: 100 percent woodland
and 100 percent high tillage cropland. The basin is well
suited to such a demonstration because of the domi-
nance of the cropland land use classification (65 percent
of total basin land acreage). Water quality and trophic
state changes associated with the two hypothetical sce-
narios are dramatic, demonstrating the utility of the ap-
proach in providing a basinwide overview of the potential
impact of nonpoint management programs in the water-
shed.
POINT AND NONPOINT POLLUTANT
SOURCES
Deterioration in the water quality of the Great Lakes and
concommilant changes in the biota of that system have
paralleled the cultural development of the region (Beeton,
1970). Long-term alterations in land use from a primarily
forested condition to intensive urban and agricultural uses
have played an important role in the water quality degra-
dation process (Int. Joint Comm., 1980). One manifesta-
tion of cultural development in the basin is accelerated
eutrophication of these waters resulting from point and
nonpoint loads of plant nutrients and oxygen-demanding
materials.
Actions taken in response to the Federal Water Pollution
Control Act Amendments of 1972 (RL. 92-500), the Munic-
ipal Wastewater Treatment Construction Grant Amend-
ments of 1981 (RL. 97-117), and establishment of the Na-
tional Pollution Discharge Elimination System (NPDES,
Sec. 402 of RL. 92-500) have reduced point source loads
of conventional pollutants to the Nation's waters.
The 1978 Great Lakes Water Quality Agreement estab-
lished phosphorus target loads designed to maintain or
improve the trophic status of the Great Lakes' waters.
Point source phosphorus management strategies (e.g.,
upgraded treatment, effluent limitations, detergent phos-
phorus ban) have reduced phosphorus loads to the Great
Lakes. Hartig and Horvath (1982) concluded that reduc-
tions in taste and odor problems and decreased chloro-
phyll levels in Saginaw Bay (Lake Huron) occurred at least
partly in response to improved treatment efficiencies and
the Michigan phosphorus detergent ban. Bierman et al.
(1984) reported, however, that almost half of the observed
loading reduction to Saginaw Bay was caused by altered
nonpoint contributions associated with decreased tribu-
tary flow.
Nonpoint source loads of phosphorus are important
across the basin. The Pollution from Land Use Reference
Group (PLUARG) reported that contributions of phos-
phorus from land use sources in 1976 accounted for one-
half of the total load to Lakes Superior, Huron, and Erie,
and one-third of the total load to Lakes Michigan and On-
tario (Int. Joint Comm., 1980). Nonpoint sources of pollu-
tion differ from point sources in that they represent the
cumulative effect of a large number of diffuse sources that
are difficult to identify, monitor, and control.
MANAGEMENT OF NONPOiNT
POLLUTION
During the last decade, comprehensive studies of water-
quality degradation in the Great Lakes and their major
embayments have led to increased concern over the sig-
nificance of nonpoint sources of pollution as contributors
to deteriorating conditions (Great Lakes Basin Comm.,
1981; PLUARG, 1978; U.S. Army Corps Eng., 1982). This
concern is evidenced at all levels of government: at the
international level through the U.S./Canadian Great Lakes
Water Quality Agreement, at the national level through the
Rural Clean Water Program and the Nationwide Urban
Runoff Program, and at the State level through the Wis-
147
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
consin Nonpoint Source Water Pollution Abatement Pro-
gram.
Mathematical models play a critical role in the water
quality management process by linking stated water qual-
ity objectives with the treatment or control options re-
quired to meet those objectives (Thomann, 1972). As the
state of the art has advanced, mathematical models have
been more extensively used to gain insight into the dy-
namics of water quality degradation, and models are re-
lied on to forecast the impact of reduced contaminant
stress. The development of areawide water quality man-
agement plans in response to Section 208 of P.L. 92-500
has extended the potential role of models within the plan-
ning process. -\
The need for quantification of the role of nonpoint
sources of pollution (rural and urban runoff) in determining
surface water quality has led to the development of math-
ematical models of varying degrees of complexity and so-
phistication (SWMM, Huber et al. 1975; STORM, U.S.
Army Corp. Eng., 1975; CREAMS, Knisel, 1980; and AN-
SWERS, Beasely and Muggins, 1980). Models such as
these have the capability of estimating pollutant loads cor-
responding to land use, soil classification, and climatologi-
cal conditions. Successful implementation of a water qual-
ity management plan requires that the impact of those
loads on the system under study be quantified as well.
This manuscript demonstrates the application of coupled
microcomputer models for a preliminary analysis of non-
point pollutant loading and water quality response within a
drainage of interest. Responses in trophic state and water
quality conditions are demonstrated using the lower Fox
River-Green Bay drainage (Wisconsin, Lake Michigan) as
an example.
THE GREEN BAY SYSTEM
Green Bay is a large gulf located in the northwest corner
of Lake Michigan (Fig. 1). The bay is approximately
160 km long and 22 km wide, has a mean depth of
15.8 m, a volume of 67 km3, and a hydraulic retention time
of 6 years (Mortimer, 1978). For many years, industrial
(pulp and paper) and municipal discharges of oxygen-de-
manding substances and plant nutrients have contributed
to severe dissolved oxygen depletion in the lower Fox
River and extreme southern Green Bay (Wis. State
Comm. Water Pollut., 1939; Epstein et al. 1974). Reduc-
tions in municipal and industrial pollutant discharges over
the past decade have led to improved water quality condi-
tions in the river and lower bay. Significant residual water
quality problems related to agricultural runoff (high turbid-
ity) and point and nonpoint sources of phosphorus (exces-
sive algal growth and reduced water clarity) remain. Hypo-
limnetic oxygen depletion in Green Bay at sites far
removed from point source discharges of organic material
occurs, apparently in response to secondary enrichment
(e.g., eutrophication).
Strong longitudinal gradients in trophic status are set up
along the major axis of the bay as a result of phosphorus
loaded from the Fox River. Hypereutrophic conditions ex-
ist in the southernmost region near the Fox River mouth,
while oligotrophic conditions prevail in the northern
reaches near the junction with Lake Michigan. The Fox
River is the single greatest hydrologic and pollutant
source for Green Bay, contributing 45 percent of the an-
nual tributary flow and 78 percent of the annual tributary
phosphorus load (Roznowski and Auer, 1984). The lower
Fox River drainage basin is highly agricultural: nonpoint
sources (Lake Winnebago and urban and rural runoff) ac-
counted for 55 percent of the annual Fox River total phos-
phorus discharge to Green Bay in 1970-71 (Sager and
Wiersma, 1975). That nonpoint fraction has probably in-
Figure 1.—Major morphological features of the Green Bay-
Lower Fox River system and their location with respect to
Lake Michigan.
creased with upgraded point source treatment efficiencies
in the basin achieved over the past 15 years. For example,
Roznowski and Auer (1984) calculated that the Green Bay
Metropolitan Sewerage District contributes only 3 percent
of the total annual Fox River total phosphorus load.
OBJECTIVE AND APPROACH
The objective of this paper is to demonstrate the applica-
tion of coupled, interactive, microcomputer models in
evaluating the sensitivity of the system to changes in land
use patterns. A unit area load (UAL)-based microcompu-
ter model is used to generate estimates of nonpoint total
phosphorus loading as a function of land use and soil
texture distributions. Loads calculated by the nonpoint
model drive a steady-state water quality model that de-
scribes the spatial distribution of phosphorus within the
system. Corresponding water quality conditions and
trophic levels may be assigned through the use of water
quality and trophic state indices. When costs are provided
for various levels of land use management, the coupled
models provide an overview of the cost effectiveness, as
well as the remedial impact of nonpoint pollution control
scenarios.
THE NONPOINT LOADING MODEL
The microcomputer model used to calculate average an-
nual total phosphorus loadings from nonpoint sources
within a drainage basin uses a unit area load approach.
The model assumes that the total phosphorus load result-
ing from surface runoff at a given site is characteristic of
the predominant land use and soil texture of the area. The
148
-------
LAKE QUALITY
UAL (mass of phosphorus per unit area per unit time)
reflects nonpoint loadings for an "average" year of wet-
ness.
To evaluate the annual nonpoint source total phos-
phorus loading, the area under study is divided into a set
of subbasins or subwatersheds representing the major hy-
drologic units within the system. A matrix is constructed
for each subbasin to reflect all possible combinations of
land use and soil texture—each of which has a character-
istic UAL The area associated with each point on the land
use-soil texture matrix (e.g., hectares of fine textured
cropland) is multiplied by the appropriate UAL (e.g., kgP
per hectare per year) to yield the site-specific phosphorus
load (kgP per year). Contributions for each point on the
land use-soil texture matrix are summed across the sub-
basin and then subbasin totals are added to yield the
annual nonpoint phosphorus load for the drainage; the
procedure is summarized in Figure 2.
For purposes of this demonstration, the lower Fox River
watershed is divided into three subbasins: Lower Fox
River, Lake Winnebago, and River-Lakes (Fig. 3). Informa-
tion on land use and soil classifications within the water-
shed were obtained from the Fox Valley Water Quality
Planning Agency and Soil Conservation Service offices
for the five-county area that constitutes the Fox River wa-
tershed. Eight land use classifications were considered,
with cropland predominating in all subbasins. High tillage
cropland (complete soil inversion) was not differentiated
from low tillage cropland (noninversion soil conditioning);
for demonstration purposes, it was assumed that high and
low tillage cropland practices are equally represented in
each subbasin. Soil classifications were grouped into
three major soil texture categories: coarse, medium, and
fine; fine soils predominated in all subbasins. Major land
uses and soil textures are summarized in Figure 3.
The nonpoint source total phosphorus contribution for
each land use-soil texture combination is calculated using
UALs derived from studies of systems having similar cli-
matological, land use, and soil texture conditions (see
Heidtke et al. 1985). The UALs applied in this demonstra-
tion (Table 1) are intended to represent annual total phos-
phorus contributions from surface runoff events within the
lower Fox River watershed during an average year of total
precipitation. Contributions from interflow-baseflow are
not included in the UAL values, but are treated as con-
stant at a rate of 0.02 kgP/ha per year over the entire
watershed. The reader is referred to Heidtke et al. (1985)
for additional details on the nonpoint source microcompu-
ter model.
THE WATER QUALITY MODEL
The water quality microcomputer model is a steady-state,
mass balance model for phosphorus (Auer and Canale,
1985). The model quantifies phosphorus sources (tribu-
tary loads) and sinks (net settling, mass transport) and
calculates the steady-state, summer average phosphorus
concentration for each of 12 model cells oriented along
the major (northeast/southwest) axis of Green Bay. Tribu-
tary loads (annual average and summer average) used in
model calibration and verification were calculated from
tributary monitoring data, USGS flow information, and
point source discharge permit reports (Roznowski and
Auer, 1984). Net settling losses are estimated from mea-
sured sedimentation rates and the phosphorus content of
the sediments and overlying water. Mass transport (dis-
persion) is quantified by calibration to cell-specific, sum-
mer average chloride concentrations. Model output is a
profile of steady-state phosphorus concentration in Green
Bay as a function of distance from the Fox River mouth.
NONPOINT SOURCE MODEL
Watershed
Soil Texture
Soil Texture
'LOAD
for all
sub-basins
Figure 2.—Schematic for the nonpoint model calculation
procedure.
FOX RIVER WATERSHED
Sub-Basin
LOWER FOX RIVER
64% Cropland
17% Residual
8% Residential^
35%"course~Soiis
20% Medium Sails
45% Fine Soils
Sub-Basin
LAKE WINNEBAGO
69% Cropland
18% Residual
6% Residential
Sub-Basin
RIVER LAKES
66% Cropland
25% Residual
4% Residential
97 %~F»ie Soils"
99% Fine Soils
Figure 3.—Land use, soil texture, and major subbasins of
the Lower Fox River watershed. Residual land use classifi-
cation includes woodlands, wetlands, and pasture.
Table 1.—Total phosphorus UAL matrix (kgP/ha • year).
Soil Texture
Land Use Coarse Medium Fine
Residential
High tillage cropland
Low tillage cropland
Pasture
Woodland/residual
0.06
0.55
0.22
0.06
0.13
0.74
0.29
0.07
0.02
0.18
0.92
0.37
0.09
0.04
Empirically derived trophic state and water quality indi-
ces are used to evaluate water quality conditions for each
model cell. Relationships published by Dillon and Rigler
(1975) are used to estimate water clarity (Secchi disk
transparency) and chlorophyll concentration from steady-
state phosphorus levels. Chlorophyll levels reflect the phy-
toplankton standing crop and will respond to phosphorus
loading reductions in phosphorus-limited systems (see
Bierman et al. 1984). Secchi disk transparency is influ-
enced by dissolved color, chlorophyll, and suspended par-
ticulate matter and will respond to phosphorus loading
reductions where phytoplankton play a dominant role in
light extinction. Secchi disk transparency is closely tied to
the public perception of water quality (Shapiro et al. 1975)
and thus provided useful input for cost benefit analyses.
Trophic state is evaluated in a similar fashion using the
criteria of Chapra and Dobson (1981) where a total phos-
phorus concentration, < 11.0 jtgP/L, indicates oligotrophy,
the range 11.0-21.7 /igP/L indicates mesotrophy, and
>21.7 /xgP/L indicates eutrophy. The Chapra and Dobson
trophic classification may be also used to estimate hypo-
149
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
WATER QUALITY MODEL
Total Phosphorus Loading
(from nonpoint model)
Steady State Phosphorus Model
(mass balance - sources and sinks)
yields spatial phosphorus profiles
Figure 4.—Schematic for linking nonpoint and water quality
models to generate Information on response In water quality
and trophic state.
Table 2.—Annual average and summer total
phosphorus loads.
Land Use Scenario
Baseline
100% woodland
100% high tillage
Annual
Average Load
(kgP/day)
2361
252
4591
Summer
Average Load
(kgP/day)
959
102
1865
limnetic oxygen depletion rates. Linkages between non-
point phosphorus loads, the steady-state model and
trophic state and water quality indices are illustrated in
Figure 4.
NONPOINT MODEL OUTPUT AND
HYPOTHETICAL MANAGEMENT
SCENARIOS
The annual average nonpoint total phosphorus loading for
the lower Fox River under existing land use conditions, as
calculated by the nonpoint source model, is 2,361 kgP/
day. This value agrees well with estimates for point plus
nonpoint loads reported by Roznowski and Auer (1984:
2,174 kgP/day) and Sager and Wiersma (1975: 2,008 kgP/
day). The measured summer average total phosphorus
load for the Fox River is 959 kgP/day (Auer and Canale,
1985); loads calculated by the nonpoint model are normal-
ized to summer average levels for input to the water qual-
ity model using the ratio of annual average to summer
average loads (959/2,361 = 0.406).
The nonpoint source microcomputer model is used to
estimate the average annual total phosphorus loads for
the Fox River for baseline (= existing) conditions and two
hypothetical land use scenarios: 100 percent woodland
and 100 percent high tillage cropland. The annual aver-
age total phosphorus loads and the normalized summer
average loads for the three scenarios are presented in
Table 2. The baseline scenario represents existing condi-
tions, the woodland scenario, a best case condition and
the high tillage scenario, a worst case condition. These
test cases are not intended to characterize specific land
use trends within the basin, but rather represent a range
of conditions that demonstrate the value of the model in
obtaining a rapid, macroscopic estimate of nonpoint load-
ings and corresponding trophic state. The test cases fur-
ther demonstrate the model's utility for examining water
quality improvements resulting from remedial actions
within the watershed.
TROPHIC STATE AND WATER QUALITY
RESPONSE
Summer average total phosphorus loads for each land
use scenario are input to the steady-state mass balance
model and used to calculate corresponding spatial distri-
butions for phosphorus in Green Bay. Figure 5 illustrates
spatial patterns for total phosphorus, chlorophyll, Secchi
disk transparency, and trophic state for each land use
scenario. Total phosphorus concentrations in Green Bay
range from 117 to 10 ngPIL under the baseline scenario,
12 to 8 pgP/L under the woodland scenario, and 191-11
/tgP/L under the high tillage scenario. When compared
with baseline conditions, the worst case, high tillage sce-
nario results in a transparency reduction of approximately
1 m over much of the bay; actual conditions may be worse
over the first 15-20 km of the gradient because of in-
creased suspended sediment load and the shallow nature
of that part of the system. Chlorophyll levels increase by
2-5 uglL in the mid- and outer-bay regions and by 20-50
nO/L near the river mouth. Chlorophyll increases near the
river mouth may be overestimated because of nutrient
saturation and reduced light penetration (self-shading).
the best case, woodland scenario generates striking im-
provements over existing water quality conditions
throughout the bay. An increase in Secchi disk transpar-
ency of 2-3 m and a reduction in chlorophyll of 1-5 figIL
are characteristic of much of Green Bay. Again, the shal-
low nature of the extreme southern portion of the bay may
lead to an overestimate of water clarity and chlorophyll
levels.
Under baseline conditions, the four southernmost
model cells (approximately 20 km) are classified as eu-
trophic, the mid-bay region as mesotrophic, and the outer
reaches as oligotrophic. The most dramatic change in
trophic status is that for the best case woodland scenario
where oligotrophic conditions prevail over the entire
length of the bay. Output from this scenario may offer an
indication of trophic conditions in the bay prior to cultural
development. Under the worse case, high tillage scenario,
eutrophic conditions persist for approximately 65 km out
into the bay and overlay regions most susceptible to hypo-
limnetic oxygen depletion.
An example of the impact of land use changes on dis-
solved oxygen depletion may be developed using the
Thienemann Index as presented in Equation 17 of Chapra
and Dobson (1981). Volumetric oxygen depletion rates,
calculated for an initial hypolimnetic oxygen concentration
of 11 mg/L and a duration of stratification of 4 months,
would yield anaerobic conditions in 2.2-4.4 months for
eutrophic waters, 4.4-8.8 months for mesotrophic waters,
and >8.8 months for oligotrophic waters. The portion of
Green Bay classified as eutrophic under baseline condi-
tions is shallow and does not permanently stratify; thus,
severe dissolved oxygen depletion is rare in that region.
The trophic state pattern under the high tillage scenario
extends eutrophic conditions over a substantial portion of
the bay that does thermally stratify. For the 3-4 months
stratification period characteristic of Green Bay, this re-
gion could experience severe hypolimnetic oxygen deple-
tion.
150
-------
LAKE QUALITY
BASELINE SCENARIO
WOODLAND SCENARIO
HIGH TILLAGE SCENARIO
fM?—. * r. ... *'&$:, f i
DOHgolraphic
D MMOtropfck
Eutroptiic
UOIigotrophic
Q MMOtropMc
• Eutropftic
UOIigotroptiic
D MwotropMc
Eutropfctc
0 20 40 6O 80 100
DISTANCE FROM FOX R. MOUTH (km)
Figure 5.—Examples of water quality model
tions, (b) 100 percent woodland, and (c) 100
0 20 40 60 80 100
DISTANCE FROM FOX R MOUTH (km)
0 20 40 6O 80
DISTANCE FROM FOX R MOUTH (km)
output for three hypothetical land use scenarios: (a) baseline (existing) condl-
percent high tillage cropland.
SUMMARY
Coupled microcomputer models for nonpoint source phos-
phorus loads and water quality are used to evaluate the
sensitivity of water quality conditions and trophic state to
changes in land use in the watershed. The approach is
demonstrated for three land use scenarios in the Fox
River-Green Bay drainage, a highly agricultural water-
shed in Wisconsin. The demonstration compares water
quality and trophic state for loads representing best and
worst case scenarios (woodland and high tillage crop-
land). Water quality response was dramatic in both cases
because of the significant load from agricultural sources
under the baseline and high tillage scenarios. The exer-
cise demonstrates the utility of a coupled microcomputer
model for nonpoint pollutant loads and water quality as an
interactive, user-friendly planning and management tool.
REFERENCES
Auer, M.T., and R.R Canale. 1985. A phosphorus budget for
Green Bay. Rep. to U.S. Environ. Prot. Agency, Environ. Res.
Lab., Duluth, MN (In prep.)
Beasely, D.B., and L.F. Muggins. 1980. ANSWERS Users Man-
ual. Agric. Eng. Dep., Purdue Univ., West Lafayette, IN.
Beeton, A.M. 1970. Changes in the environment and biota of the
Great Lakes. Pages 150-87 in Eutrophication: Causes, Con-
sequences and Correctives. Nat. Acad. Sci., Washington, DC.
Bierman, V.J., Jr., D.M. Dolan, R. Kasprzyk, and J.L. Clark.
1984. Retrospective analysis of the response of Saginaw Bay,
Lake Huron, to reductions in phosphorus loadings. Environ.
Sci. Technol. 18(1): 23-31.
Chapra, S.C., and H.F.H. Dobson. 1981. Quantification of the
lake trophic typologies of Naumann (surface quality) and
Thienemann (oxygen) with reference to the Great Lakes. J.
Great Lakes Res. 7(2): 182-93.
Dillon, P.J., and F.H. Rigler. 1975. A simple method for predicting
the capacity of a lake for development based on lake trophic
status. J. Fish. Res. Board Can. 32(9): 1519-31.
Epstein, E., M. Bryans, D. Mezel, and D. Patterson. 1974. Lower
Green Bay: an evaluation of existing and historical conditions.
Rep. for U.S. Environ. Prot. Agency, EPA-905/9/74-006. Wis.
Dep. Nat. Resour., Madison.
Great Lakes Basin Commission. 1981. Great Lakes environ-
mental planning study. Ann Arbor, Ml.
Hartig, J.H., and F.J. Horvath. 1982. A preliminary assessment
of Michigan's phosphorus detergent ban. J. Water Pollut.
Control Fed. 54(2): 193-97.
Heidtke, T.M., M.T. Auer, and R.R Canale. 1985. Coupling non-
point pollution and water quality models: an example for the
Green Bay-Fox River watershed. Proc. Nonpoint Pollution
Symp., Milwaukee, Wl, April.
Huber, W.C., et al. 1975. Stormwater Management Model Users
Manual: version II. EPA 670/2-75-017. U.S. Environ. Prot.
Agency, Washington, DC.
International Joint Commission. 1980. Pollution in the Great
Lakes Basin from Land Use Activities. Windsor, Ontario.
Knisel, W.G., ed. 1980. CREAMS: a Field Scale Model for Chem-
icals, Runoff, and Erosion from Agricultural Management Sys-
tems. Conserv. Res. Rep. No. 26. U.S. Dep. Agric., Washing-
ton, DC.
Mortimer, C.H. 1978. Water movement, mixing and transport in
Green Bay. In Green Bay Workshop Proc., Univ. Wisconsin
Sea Grant Publ. No. WIS-SG-78-234, Madison.
Pollution from Land Use Activities Reference Group (PLUARG).
1978. Environmental Management Strategy for the Great
Lakes System. Final Rep. Int. Joint Comm., Windsor, Ontario.
Roznowski, D.M., and M.T. Auer. 1984. Tributary loadings to
Green Bay: a mass balance approach. Rep. to U.S. Environ.
Prot. Agency, Environ. Res. Lab., Duluth, MN.
Sager, RE., and J.H. Wiersma. 1975. Phosphorus sources for
lower Green Bay, Lake Michigan. J. Water Pollut. Control Fed.
47(3): 504-14.
Shapiro, J., J.B. Lundquist, and R.E. Carlson. 1975. Involving
the public in limnology—an approach to communication. Verh.
151
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Int. Verein. Limnol. 19: 866-74. 1982. Lake Erie Wastewater Management Study Fi-
Thomann, R.V. 1972. Systems Analysis and Water Quality Man- nal Report. Buffalo District, Buffalo, NY.
agement. McGraw-Hill, New York. Wisconsin State Committee on Water Pollution. 1939. An inves-
U.S. Army Corps of Engineers. 1975. Urban stormwater run- tigation of the Fox and East Rivers and of Green Bay in the
off—computer program 723-SB-L2520. Hydrologic Eng. Cen- vicinity of the city of Green Bay. Madison.
ter, Davis, CA.
152
-------
A PROJECT TO MANAGE AGRICULTURE WASTES HAS IMPROVED
THE QUALITY OF VERMONT'S LAKE PARKER
RICHARD J. CROFT
U.S. Department of Agriculture
Soil Conservation Service
Winooski, Vermont
ABSTRACT
Lake Parker's quality declined in the 1970's. This 83 ha
(206 acre), northeastern Vermont Lake suffered from
weeds, algae, and bacteria growths. Trout fishing and
recreation were no longer an attraction. The Vermont De-
partment of Water Resources determined that the lake's
problems were caused by excessive phosphorous and
bacteria loads from the 11 dairy farms in the watershed.
The town of Glover and other sponsors joined with the
Soil Conservation Service to implement a Resource Con-
servation and Development Project. Eight of the 11 farms
had critical waste management problems. All eight partic-
ipated in the project. Treatment included proper utilization
and disposal of wastes through manure storage, barn-
yard runoff control, and milkhouse waste management.
The project was started in January 1981 and completed
in June 1982. In 1983 and 1984 Lake Parker has im-
proved markedly.
Pollution and eutrophication of lakes and ponds is a major
problem in Vermont. In its water quality management plan-
ning, Vermont has recognized agriculture as the single
most significant nonpoint source of nutrients and other
pollutants reaching many of these waters.
Vermont completed its original State Water Quality Plan
for Controlling Agricultural Pollution in 1978. The plan pro-
vides a priority list of lake and pond watersheds for project
assistance. The State of Vermont, the USDA, and local
sponsors have been successful in jointly implementing
agricultural nonpoint source management projects in
eight of the priority watersheds. The Lake Parker Re-
source Conservation and Development Measure is one of
these and provides the focus for this report.
The Lake Parker Measure is particularly noteworthy be-
cause it was requested, planned, funded, designed, and
installed in 2 years; it demonstrates the cooperation
needed among Federal, State, and local entities and land-
owners to expedite such a project; and it quickly improved
lake water quality.
LAKE PARKER AND ITS WATERSHED
Lake Parker is a glacially formed water body at 44° 40'
north latitude nestled in the steep, rolling hills of Orleans
County in northeastern Vermont (Fig. 1). Table 1 provides
a summary of the lake and watershed features.
Two perennial streams flow eastward into the lake. Of
the 11 dairy farms within the drainage area, two are only
partially in the area. The lake shoreline includes 110 sea-
sonal or year-round cottages. Table 2 summarizes land
use in the watershed.
The town of Glover (population 790), lakeshore resi-
dents (seasonal population 350), nearby communities,
and Vermont tourists and fishermen all have used Lake
Parker extensively for recreation.
The lake has supported both cold- and warmwater fish
in the past, including brown, lake, and rainbow trout;
smallmouth bass; pickerel; and yellow perch. The trout
fishery has declined in the past decade.
The predominant soil association in the watershed is
Buckland-Cabot. These silt loams are poorly drained to
moderately well drained. They are glacial tills and often
have a hardpan (U.S. Department of Agriculture, 1980).
Climate strongly influences activities in the area. Winter
low temperatures of -40°C and summer highs of 32°C
are common. Average annual precipitation is 109 cm (30
percent as snow) and runoff is 61 cm. Average flushing
rate for the lake is about once every 6.4 months (Vermont
Agency of Environmental Conservation, 1980).
WATER QUALITY PROBLEMS
IN THE LAKE
Weeds were documented as a threat to the lake as early
as 1966, when the Vermont Department of Fish and Wild-
life surveyed a fishery and noted extensive growths of
aquatic plants, especially Eloden (Vermont Agency of En-
viron. Conserv. 1974). At that time, the Department cited
the possibility that agricultural runoff was stimulating the
plant growth.
By the 1970's weed growth had become worse, particu-
larly near the stream inlets. Dominant species were Pota-
mogeton richardsonii, Eloden canandensis, Chara vulgaris,
and Potamogeton anplifolus (Warren, 1985). Weed growth
was so dense that motor boats could not negotiate much
of the area shallower than 3 m. The weeds served as a
refuge for yellow perch, rock bass, and stunted bait fish
from the larger predators—trout and bass. Decaying
weeds consumed dissolved oxygen, especially in the
deeper lake segments.
The Lake Parker Association has conducted a weed
harvesting program since the early 1970's. In 1980 the
weed problem became so severe that the Association was
able to gain assistance from the Vermont Department of
Water Resources to purchase and operate a weed har-
vester. Since 1980, the Department of Water Resources
and the Association have spent over $32,600 and count-
less hours of volunteer time trying to control the weeds
through a harvesting and disposal program (Garrison,
1984,1985).
Shoreline and town residents were alarmed by the
changing character of the lake. It was no longer aestheti-
cally pleasing. Boating and fishing were not up to par.
The Department's water quality testing (Table 3)
showed reason for concern. Though years of data are
necessary to draw reasonable conclusions on water qual-
ity trends, a key parameter, mean springtime total phos-
phorus, seemed to be on the increase as did algal density
(chlorophyll a) prior to 1982. Bacteria were not a wide-
spread problem throughout the lake, but high fecal coli-
form counts were observed on the west side of the lake
between the two perennial stream inlets. The decaying
plant material and other organic sediments entering the
lake have consumed available dissolved oxygen (DO).
Deeper portions of the lake (below 7 m), where the trout
tend to congregate during July and August, were found to
have inadequate levels (less than 5 mg/L) of dissolved
153
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
WATERSHED BOUNDARY
SCALE
Figure 1.—A location map of the lake and watershed.
Table 1.—Selected physical features of Lake Parker and Its watershed.
Size
Feature
Normal lake surface
Normal lake level
Major axis . . . .
Minor axis
Maximum depth
Mean depth . .
Normal lake volume
Shoreline length
Watershed area
Highest watershed elevation
Metric
830 ha
396 0 m
1 677 0 m
854 0 m
12 8 m
8.5 m
7 154200.0m3
4 880.0 m
2 1 25 0 ha
658 5 m
(U.S. Customary)
206 ac
1 299 ft NGVD
5 500ft
2 800ft
42ft
28ft
5,800 ac-ft
16,000ft
5 250 ac
2 160ft NGVD
Source: VI. Agency of Environ. Conserv. 1974.
154
-------
LAKE QUALITY
Table 2.—Land use of the Lake Parker watershed.
Size
Percent of
Type
Pasture and hayland
Cropland in rotation
Mixed hardwoods and conifers
Water surfaces
Residential
Road surface
Other
Totals
Hectares Acres
971 2,400
24 60
903 2,230
89 220
45 110
57 140
36 90
2,125 5,250
Total
45.7
1.1
42.5
4.2
2.1
2.7
1.7
100.0
oxygen. In 1974, five of six DO samples in this depth
range were under 5 mg/L DO. This, along with the inac-
cessibility of prey to fish because of dense weed growth,
may partially explain the decline of the trout fishery during
the 1970's.
AGRICULTURAL NONPOINT SOURCES
A 1980 study by the Orleans County Natural Resources
Conservation District for the Department of Water Re-
sources found that soil erosion was not a significant non-
point source (Orleans County Nat. Resour. Conservation
Dist. 1980). Instead it found that agricultural waste runoff
from 8 of the 11 farms was reaching water courses leading
to the lake. The town of Glover diverted runoff from one of
three other farms to preclude this waste from entering the
lake. Selected characteristics of the eight farms with sig-
nificant runoff are provided in Table 4.
Several reasons accounted for the farm runoff prob-
lems: barns, feedlots, barnyards, manure stacks, and
milking center waste outfalls were located close to
streams (most within 60 m); the landscape is steep and
includes stream gradients (fast and flushing runoff); winter
manure storage space was inadequate, resulting in winter
spreading and improper stacking; and above-site drain-
age was not diverted, allowing quantities of clean water to
mix with and carry wastes away to the streams.
In developing the Rural Clean Water Program Measure
plan, the watershed's agriculture was estimated to con-
tribute 85 percent (4,310 kg annually) of the total phos-
phorus load to the lake. Studies by the Lake Parker Asso-
ciation and the Department of Water Resources found that
lake contamination from human wastes and other cultural
activities was not a problem. Therefore, agricultural runoff
management was key to reducing total phosphorus loads
to the lake (Dunbar, 1983).
PROJECT DEVELOPMENT
AND INSTALLATION
The Resource Conservation and Development program
provides technical and financial assistance to accelerate
resource development and environmental protection in
multicounty areas. Local sponsors plan, implement, oper-
ate, and maintain projects in cooperation with USDA's Soil
Conservation Service.
In August 1980, the Soil Conservation Service (SCS)
received an application for assistance to control agricul-
ture-related pollutants in the Lake Parker watershed.
Sponsors were the Northern Vermont Resource Conser-
vation and Development Council, Orleans County Natural
Resources Conservation District, and the town of Glover.
The Soil Conservation Service developed a plan to im-
prove agricultural waste management on eight farms. The
plan called for negotiating voluntary long-term contracts
(3-10 years) with the farmers. This plan was unique to the
Resource Conservation and Development program in that
the federally assisted long-term contracts were to be lo-
cally administered by the town of Glover. The farmers
would install, operate, and maintain proper waste man-
agement practices in return for cost-share and technical
assistance.
In all, the planned waste management systems in-
cluded 427 m of diversions, 0.8 ha of waterways, 0.8 ha of
filter strips, 305 m of fencing, and eight waste storage
structures. The plan estimated these could be installed for
$178,950, with $131,720 cost-share and $32,720 techni-
cal and administrative assistance through SCS's Re-
sourcie Conservation and Development program. The
measure was funded in February 1981.
Farmers favored the measure. They recognized that
their operations could cause problems. They liked the
idea of a voluntary rather than a regulatory program. They
appreciated the technical expertise available to help them
develop efficient waste management systems on their
farms. The Soil Conservation Service planned, surveyed,
and designed practices for each of the farms in the early
part of 1981. All eight farmers undertook long-term con-
tracts. By July 1982, all the practices were installed and
operating on all eight farms (Dunbar, 1983). Table 5 sum-
marizes the project.
The construction cost for the project amounted to
$162,750. Of this, SCS cost-shared $108,800. The
farmers now operate and maintain their practices, follow-
ing operation and maintenance plans prepared for them
as a part of the long-term contract. The Orleans County
Natural Resources Conservation District and the Soil Con-
servation Service provide followup assistance as needs
arise.
Project implementation was successful because of the
cooperation of the farmers, the town of Glover, the Or-
leans County District, Lake Parker Association, the De-
partment of Water Resources, and the Soil Conservation
Service.
LAKE PARKER NOW
Two summers have passed since the Lake Parker Mea-
sure was completed. A longer period of evaluation will be
needed to establish trends of the lake's response to the
waste management. So far, indications have been promis-
ing.
From the Lake Parker Association's viewpoint the proj-
ect has been highly successful. Leo Millette, who heads
the lake's weed harvesting program, maintains that "ma-
nure management has really helped—you can see the
difference in the lake." In July of 1983 and 1984, little
weed cutting was needed, in comparison with prior years.
Weeds still require harvesting in August. Gerald Ander-
son, president of the Lake Parker Association, has lived by
the lake for 9 years. Anderson claims that the Lake has
improved "drastically" since the project. In fact, the asso-
ciation will request the Department of Fish and Wildlife's
assistance in restocking the lake with trout. Department of
Water Resources officials also are pleased with the lake's
response so far.
Water quality data available for the lake in 1983 and
1984 (Table 3) show that averages for summer chlorophyll
a and Secchi disk transparency values were about the
same as in earlier years (Warren, 1985). However, spring-
time total phosphorus concentrations are declining (Vt.
Agency Environ. Conserv. 1978-1983).
The lake's aquatic weed growth appears to be diminish-
ing. Plankton activities may be less sensitive or respond
more slowly to the declining concentrations of total phos-
phorus. The presence of nuisance bacteria needs further
evaluation but is not perceived to be a problem. Recycling
of various forms of phosphorus from the lake's sediments
155
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 3.—Selected water quality data from testing of Lake Parker by the Vermont Department of Water Resources.
Dates 19—
Unit 74 77 78 79 15 81 82 83 84 Remarks
Parameter
Mean summer
total phosphorus pQl\ 44 — —
Mean spring
total phosphorus /ig/1 — 14 10
Summer average
chlorophyll a i*gl\ — 4.8 —
Summer average Secchi disk
transparency m — 3.5 —
Fecal coliform
spring value
— — — — — — not available
16 20 21 17 15 12
6.1 7.2 5.9 5.2 5.6 5.8
3.3 3.7 3.2 4.0
None Available
4.2 3.6
count 700
per 100ml
Dissolved oxygen (percent
observations below 5 mg/l at or
more than mean depth) % 42 None Available
•high of
observations.
Range 8-700 at
various locations
Source: Vt. Dept. Water Resources.
Totals
Table 4.—Animal and waste management characteristics of eight Lake Parker watershed farms.
Cropland Milkhouse
Farm No.
1
2
4
5
6
7
9
10
Animal
Units
80
68
60
77
80
80
65
65
Annual
manure
tons
1,440
1,225
1,080
1,385
1,440
1,440
1,170
1,170
receiving
manure
acres
55
68
110
70
84
99
80
80
effluent
could enter
watercourse
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Barnyard
runoff to
watercourse
Yes
No
Yes
No
No
Yes
Yes
No
575
10,350
646
Table 5.—Lake Parker resource conservation and development (RC&D)—practices installed.
Practice
Type
Quantity
Unit
Total Cost $
Waste storage structure
Waste storage structure
Waste storage pond
Waste utilization
Diversion
Milking center
absorption trench and grass filter
Grass filter strip
Farm road relocation
Heavy use area protection
Rein, concrete
Timber wall
Earth
—
Grassed
—
—
—
Barnyard
2
4
2
8
185(607)
6
0.7(1.8)
1
1
no.
no.
no.
no.
m(ft)
no.
ha (ac)
no
no
60,170
75,180
19,460
840
4,300
310
750
1,740
Total
162,750
should not pose a critical problem because only minor
amounts of sediment reach the lake and because the lake
is flushed rapidly.
CONCLUSION
The Lake Parker Measure is an example of what can be
done through a program of voluntary participation. Gov-
ernment agencies at various levels and the local citizenry
participated. All farmers with significant on-farm nonpoint
sources participated. Lake users are already pleased with
the results. Continued monitoring is needed to establish
long-term water quality trends.
REFERENCES
Dunbar, H.H., Jr. 1983. A measure to save Lake Parker—from
plan to completion in eighteen months. New England Country
Folks. 111(1): 4-5.
Garrison, V. 1984. Pers. commun. Supervisor, Lake and Pond
Unit, Vt. Agency. Environ. Conserv. Dep. Water Resour. Envi-
ron. Eng.
1985. Pers. commun. Supervisor, Lake and Pond
Unit, Vt. Agency. Environ. Conserv. Dep. Water Resour. Envi-
ron. Eng.
Orleans County Natural Resources Conservation District. 1980.
A contract to document soil erosion and fertilizer management
needed to control nonpoint sources in the Lake Parker water-
shed. Newport, VT.
U.S. Department of Agriculture. 1980. Lake Parker agriculture
related pollutant control resource conservation and develop-
ment measure plan. Soil Conservation Service. Burlington,
VT.
Vermont Agency of Environmental Conservation. 1974. Lake
Parker water quality report. Lake Eutrophication. Ser. No. 4.
Dep. Water Resour. Environ. Eng.
. 1978. Vermont Lakes and Ponds Program. Lake Eu-
trophication. Ser. No. 5. Dep. Water Resour. Environ. Eng.
_. 1979. Vermont lay monitoring report. Vol. I. Water
Quality Surveil. Ser. No. 9. Dep. Water Resour. Environ. Eng.
156
-------
LAKE QUALITY
1980a. Vermont Lake Classification Survey. Water Quality Surveil. Ser. No. 12. Dep. Water Resour. Environ. Eng.
Quality Surveil. Ser. No. 8. Dep. Water Resour. Environ. Eng. 1983. Vermont lay monitoring report. Vol. II. Water
1980b. Vermont lay monitoring report. Vol. II. Water Quality Surveil. Ser. No. 13. Dep. Water Resour. Environ. Eng.
Quality Surveil. Ser. No. 10. Dep. Water Resour. Environ. Eng. 1982b. Water Quality Assessment. Section 305B
_. 1981. Vermont lay monitoring report. Vol. II. Water rep. Dep. Water Resour. Environ. Eng.
Quality Surveil. Ser. No. 11. Dep. Water Resour. Environ. Eng. Warren, S. 1985. Pers. commun. Aquatic biologist, Vt. Agency
1982. Vermont lay monitoring report. Vol. II. Water Environ. Conserv. Dep. Water Resour. Environ. Eng.
157
-------
Estuarine Quality
URBAN RUNOFF POLLUTANT INPUTS TO NARRAGANSETT BAY:
COMPARISON TO POINT SOURCES
EVA J.HOFFMAN
State Coordinator
Narragansett Bay Project
Rhode Island Department of Environmental Management
Providence, Rhode Island
ABSTRACT
Urban runoff samples were collected from four drains,
each serving a different land use: residential, commer-
cial, highway, and industrial. Twenty-one storm events
were monitored to establish mass discharge rates of wa-
ter volume, suspended solids, petroleum hydrocarbons,
polycyclic aromatic hydrocarbons, and a variety of trace
metals, as a function of storm rainfall and land use. These
loading rates were combined with local rainfall and land
use records to estimate annual urban runoff inputs to the
Narragansett Bay watershed. For comparison, we com-
piled a point source inventory for the same components
though self-monitoring reports and past monitoring stud-
ies conducted at the university, augmented with addi-
tional analyses as required. Urban runoff was found to be
the source of 48 percent of the petroleum hydrocarbons,
3 percent of the lower molecular weight (2 ring) polycyclic
aromatic hydrocarbons, 44 percent of the higher molecu-
lar weight polycyclic aromatic hydrocarbons, 65 percent
of the lead, 56 percent of the zinc, and 5 percent of the
copper entering the Narragansett Bay watershed annu-
ally. The application of the urban runoff loading rates was
tested on one of the Narragansett Bay tributaries, the
Pawtuxet River. The wet-weather related mass discharge
rates for these constituents in the river, as monitored dur-
ing and following one storm event, was estimated within a
factor of 2 using our loading factors with the rainfall and
local land use data. The fate and transport of wet-weather
components in the Narragansett Bay estuary will be ex-
amined as part of the Narragansett Bay Project of the
EPA National Estuaries Program.
Narragansett Bay is one of the best studied estuaries in
the world. The University of Rhode Island's Graduate
School of Oceanography, Brown University, Roger Wil-
liams College, and neighboring institutions such as the
Massachusetts Institute of Technology and Woods Hole
Oceanographic Institution have used the bay as a re-
search laboratory. In 1979, Rhode Island's Coastal Re-
sources Center published The Bay Bib, containing over
1,800 references to literature on this estuary. The Center
then made an attempt to examine these data in order to
answer the question, "Where do the various pollutants in
Narragansett Bay come from?" One conclusion of this
study states simply that "Sufficient data do not exist to
assess the relative importance of the many sources of
pollution in the upper bay's watershed. Data comparable
to that available on effluents from sewage treatment
plants and industrial sources do not exist for flows result-
ing from runoff and other nonpoint sources" (Olsen and
Lee, 1979).
URBAN RUNOFF
As a first step in evaluating the annual pollutant loads
generated by urban runoff, it is necessary to have loading
rates (such as mass/drainage area/time) that can be ap-
plied with some degree of confidence to the drainage area
in question. Although appropriate urban runoff loading
factors exist for metals generated by the National Urban
Runoff Program (NURP) (U.S. Environ. Prot. Agency,
1984), the urban runoff data on hydrocarbons and PAHs
were minimal. Because we were particularly interested in
these organic components, we found it necessary to con-
duct an urban runoff study of our own (for more detail see
Hoffman et al. 1983a, 1982,1985, 1984, 1983b). The ex-
periment was designed to examine hydrocarbons and
PAHs in runoff as a function of land use in a manner
similarly used for other components in the NURP studies.
The results of our study, derived from 21 storm events for
organics, and 12 storm events for metals, are given in
159
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 1. Where they are available, runoff loading factors
generated by the NURP studies are included for compari-
son.
Inspection of our data reveal a strong dependence of
urban runoff pollutant loading with land use. Often differ-
ences by several orders of magnitude are involved. The
urban runoff loadings for PAHs with three or more rings
Fe, Mn, Cu, Pb, Cd, Zn, and suspended solids (TSS) were
highest at the interstate highway location. Even though
highways represent only a very small proportion of the
land use in some locations, they become more important
near urban areas. Since the loading factors are high, the
highway land use can become an important part of the
total urban runoff loads to urban water bodies. Highways
were not studied separately in the NURP program.
Loadings for petroleum hydrocarbons and PAHs with
two rings were highest at the industrial location. Our col-
lection site, admittedly, could be termed "heavy indus-
trial," since it was located in the Port of Providence area.
These values, then, would not be typical of newly devel-
oped industrial parks, which would have loadings similar
to our commercial location. (The commercial land use and
industrial land use were combined in the NURP studies in
1984 which, in our view, would be satisfactory for light
industry, but inappropriate for heavy industrial areas as
illustrated in Table 1.) -
The next step is the combination of the urban runoff
loading factors with land use data for the specific drainage
basin of interest. This would seem, at first inspection, to
be a trivial matter, but hidden pitfalls exist for the unwary
scientist. To give only two examples: (1) poor choice of
land-use categories (categories for urban planning pur-
poses'may not be the best for urban runoff studies be-
cause the utility category can include both power line
right-of-ways (open land) and power plants (heavy indus-
try)); (2) land uses as a function of drainage basin are
most frequently derived using topographical maps which
may not represent where the storm sewers actually carry
the water.
Once we had determined loading factors and found
land use statistics, we could then calculate urban runoff
loads to the waterbody of interest, for areas which are
newly developed. However, the situation in Providence,
and in other cities of the Northeast where sewer systems
collect both wastewater and urban runoff, leads to compli-
cations in the calculations. In the 1890s, at the time of its
original construction, the combined system in Providence
was considered innovative because it collected urban run-
off, recognized even then as contributing to water pollu-
tion. At that time the runoff did not contain automotive-
related pollutants, but horse-related ones. A schematic of
a typical combined sewer system is given in Figure 1.
In these systems, urban runoff can take any of three
routes: it can travel down the street to the nearest water-
body via overland transport; it can travel to a catch basin
tied into a separate storm sewer which usually takes the
runoff to the nearest waterbody; or it can travel to a catch
basin tied into a combined sewer system. Once in a com-
bined system, it can travel to a sewage treatment plant,
which may not be in the same drainage basin, or can
overflow the system via a combined sewer overflow, usu-
ally in the drainage basin of origin.
As a first step, it is necessary to subdivide the land use
statistics into subdrainage areas, so that loading rates for
the areas served by storm drains can be calculated inde-
pendent of areas served by combined sewers. For Provi-
dence, this was done using a land use map superimposed
on a city sewer map (Martin and Robadue, 1983). It is not
difficult to estimate the amount going into combined
sewers, once the land use characteristics for these areas
are available. The more difficult question is where does
the runoff go once it gets into the system? Does it overflow
the system close to the source? Does it go all the way to
and through the treatment plant? Does it go to the treat-
ment plant only to be bypassed around the plant? Once
the runoff goes into a combined system it is mixed with
unknown proportions of raw sewage; how much of this
sewage overflows along with the runoff during rain
events?
There are two basic approaches to answering these
questions. One can monitor each overflow individually or
model the system. The city of Providence has been di-
vided into nine combined sewer overflow (CSO) drainage
districts. Preliminary design projects for two of these dis-
tricts have been contracted and include flow monitoring of
each CSO in these two districts and some pollutant deter-
minations on selected CSOs. These two projects cost in
excess of $1.2 million. Although we now have some con-
ception of the nature of CSO discharge in two districts, the
data are not useful in assessing the problems in the other
seven districts of the city. The monitoring of each of the 65
overflows in Providence would be logistically difficult and
very expensive. Modeling of the sewer system is a much
less costly way to estimate how important CSOs are in
context with other sources. It is also an inexpensive
method of assessing whether expensive design and moni-
toring studies are warranted.
Three models have been attempted for Providence's
combined sewer system: one model estimates CSOs by
difference between total flows entering the system and the
amount that gets all the way to the plant (Hoffman, 1983);
two other models estimate CSOs by calculating the sew-
age and runoff flows in each district sending all of it to the
plant until the capacity of the connector pipes in the dis-
Pollutant
Petroleum hydrocarbons (HC)
LMW-PAHs
HMW-PAHs
Fe
Mn
Cu
Pb
Cd
Zn
Suspended solids (TSS)
Table 1.— Urban runoff loading factors as a function of land use.
Residential1 Commercial1
(single family (shopping Industrial1
suburban) mall) (heavy)
180
0.009
0.258
135
49.6
3.0 (8)
22.4 (36)
0.18
43.5 (34)
4400 (12200)
580 14000
0.100 2.42
0.589 3.97
166 856
8.6 65.8
3.0 (22) 35.3
43.6 (82) 166
0.69 0.85
n.d. (177) 639
32400 (54300) 548000
Highway1
(8 lane)
7800
1.220
16.9
915
513
146
2250
2.48
7020
424000
'(kg/km2 of land use/yr)
Annual rainfall « 121 cm/yr
n.d. not determined;
Values in parentheses are loading factors as projected from National Urban Runoff Program (NURP).
160
-------
ESTUARINE QUALITY
URBAN RUNOFF
overland transport
separate storm sewers
combined
sewers
DOMESTIC and
INDUSTRIAL
WASTEWATER
combined
• ewers
combined sewer
river
separate sanitary
sewers
±11^1
SEWAQE
TREATMENT
FACILITY
river
sludge
Figure 1.—Water pathways in a combined sanitary-storm water sewerage system during rainy conditions.
trict is reached, the rest being discharged by the local
CSO (Martin and Robadue, 1983; Metcalf and Eddy,
1983).
All three of the system models predict that some frac-
tion of the runoff goes to the treatment plant, although the
absolute magnitude varies. We monitored the influent and
the effluent of this plant during three rainstorms to evalu-
ate the impact of urban runoff in the plant (Hoffman et al.
1985). Urban runoff was found to affect the plant in two
ways: first by increasing the loads of pollutants during
storms and then by producing elevated flow rates which
are sometimes sufficient to produce hydraulic overload-
ings of the secondary treatment system. When combined,
these produce higher mass discharges from the plant in
wet weather than during analogous dry periods. It is likely
that each treatment plant receiving stormwater dis-
charges will behave differently in this aspect.
In summary, to produce urban runoff estimates for Nar-
ragansett Bay, we monitored storm drains serving differ-
ent land uses; we modified land use data, when neces-
sary to make them useful for water quality planning; we
estimated how much urban runoff never went to the drain-
age basin of origin but went to a treatment plant; and we
estimated how much runoff mixed with sewage and was
discharged by CSOs. For example, we calculated that, on
an annual basis in Providence, 47 metric tons of hydrocar-
bons were discharged by separate storm drains, 20 metric
tons were discharged via combined sewer overflows, 100
metric tons went to the treatment plant during rainy condi-
tions, and 222 metric tons went to the treatment plant
during dry conditions. Similarly, we calculated the urban
runoff expected from each of the 36 cities and towns sur-
rounding the bay. These total urban runoff Narragansett
Bay watershed calculations for a variety of different pollu-
tants are compared with other sources later.
WASTE CRANKCASE OIL DUMPING
The improper disposal of used crankcase oil down sewers
has been cited by numerous authors as a potential contri-
bution to the oil content of sewage and receiving waters.
The impact of this disposal method is impossible to assess
directly, since it is done surreptitiously. Often evidence is
seen—empty oil cans in rivers and on streets, large oil
blotches around catch basins—but the magnitude of the
problem has been the subject only of speculation. To ad-
dress this question, we designed a survey that we mailed
to 1,000 Providence residents. Under the guise of asking
whether they would participate in a used oil recycling pro-
gram, we slipped in a question about their current dis-
161
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 2.—Used crenttcese of! dispose! practices.
Population density
Percent of oil changed by owners
Disposal method used by owners:
Give it to service station
Put in garbage
Store at home
Pour it out or bury it in backyard
Pour it on the road
Pour it down sewer
Take to dump
Other
Urban
>3000/mi2
33.5%
6.9
40.7
4.1
29.7
4.8
7.6.
2.8
3.5
Suburban
3000-500/mi2
39.9%
Percentage of oil volume
10.4
23.4
6.5
39.0
4.0
2.6
3.9
14.3
Rural
48.5%
3.0
14.0
5.0
38.0
0
1.0
9.0
24.0
posal practices (Hoffman et al. 1980). Following this study,
the same questionnaire was used again in a South Caro-
lina legislative study, querying South Carolinians about
their habits in this regard (Marchand et al. 1980). These
two data sets give us an idea of what urban, suburban,
and rural residents do with their waste oil. A summary of
the survey results is given in Table 2.
The joint study (Hoffman et al. 1981) found that: on the
average, car owners changed their crankcase oil in their
vehicles twice a year, regardless of population density; as
the population density increased, the percentage of do-it-
yourself oil changers decreased; the disposal methods
used are a function of demographic parameters; and the
specific practices of pouring the used oil on the road or
pouring it down catch basins is clearly more common in
highly urban areas where catch basins are convenient.
We used the survey results to predict waste oil contribu-
tions of each city and town in the Narragansett Bay drain-
age basin. First, we classified each town into one of three
categories (urban, suburban, and rural) by population den-
sity criteria to determine which of the data sets were the
most appropriate for each town. Then we calculated the
amount of waste oil dumped down sewers or poured on
roads per town, using the number of vehicle registrations
in each town. The other waste oil disposal methods could
also eventually result in surface or ground water contami-
nation, but this process would take longer and some deg-
radation is possible. Leaks from underground storage
tanks used for waste oil in gas stations are also a potential
water pollution problem. However, when oil is dumped
down a sewer, its transportation to receiving waters is
rapid. Our waste oil dumping estimates are based only on
the amount dumped down sewers and represent a con-
servative value if other methods of oil disposal also con-
tribute to water pollution.
Because used crankcase oil contains metals and PAHs,
we estimated the loadings expected for these constituents
using literature data about the composition of used crank-
case oil (Pruell, 1983; Brinkman et al. 1981).
A word of caution on assembly of the final pollutant inven-
tory: The dangers of double accounting must be recog-
nized. This is a particular hazard with combined systems
(i.e., urban runoff going to a sewage treatment facility
could be put in either the urban runoff category or the
sewage category). For the purposes of these calculations,
we have made the following assumptions: (a) urban runoff
going to sewage treatment plants becomes part of the
sewage values and is no longer part of the urban runoff
category; (b) urban runoff or sewage going out of a CSO
becomes part of the CSO values; (c) atmospheric fallout
on land is a part of urban runoff, and only atmospheric
fallout on water is listed separately; and (d) industrial dis-
charges going to sewage treatment plants are a part of the
sewage values, and only industries discharging directly
onto waters are listed separately.
The nature of annual pollutant input inventories should
be kept in mind. There are no completely steady dis-
charges into the bay. Municipal plants receive more flow
and higher concentrations during the day than during the
night; industrial sources discharge more during the day;
urban runoff occurs only during and following rain events;
the time and location of oil spills cannot be predicted. The
nature of these spatial and temporal variabilities of each
input constitutes an important consideration for several
management decisions.
Graphic presentations of the various sources of organic
contaminants, such as petroleum hydrocarbons and poly-
cyclic aromatic hydrocarbons, and of selected metals, are
given in Figure 2. It becomes obvious very quickly that
only one general statement can be made about the
sources of toxic pollutants to the aquatic environment:
each pollutant has different major sources. We have
shown three classes of hydrocarbons in Figure 2: total
petroleum hydrocarbons, lower molecular weight (two
rings) polycyclic aromatic hydrocarbons, and higher mo-
lecular weight (three rings or more) polycyclic aromatic
hydrocarbons. For each of these hydrocarbon classes, the
major entry pathway is different in the Narragansett Bay
watershed. While urban runoff accounts for 48 percent of
the total hydrocarbons, it accounts for only 3 percent of
the two-ring PAHs. We have observed that two-ring PAHs,
while found in significant concentrations in used crank-
case oil and presumably also in drips of crankcase oil on
the street surface, seem to be lost by weathering on the
street prior to incorporation in urban runoff. These lower
molecular weight PAHs in petroleum products discharged
to the sewer system are not exposed to such weathering;
thus, the major sources of two-ring PAHs represent fresh,
unweathered oil in sewage effluent. The PAHs with three
or more rings formed during combustion of fossil fuels are
not Ipstvia weathering—at least not to the same extent as
the lower molecular weight compounds found—but are in
lower concentrations in used crankcase oil and sewage
effluents. Atmospheric deposition becomes more impor-
tant for these PAHs than for the other hydrocarbons. Pre-
liminary calculations suggest that atmospheric deposition
on land surfaces can account for 50 percent of the PAHs
with three or more rings in urban runoff and, thus, about
10 percent of these PAHs in sewage. Fallout of PAHs with
three or more rings from the atmosphere can directly or
indirectly account for over half the entry of such PAHs to
Narragansett Bay.
The metals also have varied sources (see Fig. 2). The
primary source of lead in Narragansett Bay is from urban
runoff, presumably due to the use of leaded fuel in auU>
mobiles. The lead is emitted through the exhaust system.
162
-------
ESTUARINE QUALITY
ATMOSPHERIC DEPOS!YiON
• INDUSTRY
WASTE OIL-, I I i OIL SPILLS
CSOs
PETROLEUM HYDROCARBONS
11726 tons Syr I
ATMOSPHERIC
DEPOSITION
.URBAN
RUNOFF
ATMOSPHERIC
DEPOSITION
•WASTE OIL
Pb I79.3tont/yr)
ATMOSPHERIC
DEPOSITION
URBAN RUNOFF
LUW-PAHt
( 4.O6 tout/yr>
HMW PA Hi
(1.46 tont/yrl
ATMOSPHERIC
INDUSTRY-, ("DEPOSITION
URBAN
SEWAGE } RUNQFF
Cu fl23tons/yr>
Zn ( 3O8 lont/yr)
Figure 2.—Pathways of pollutant entry Into the Narragansett Bay watershed.
When it is incorporated in crankcase oil, it is a component
of the oil drips. While this source of hydrocarbons is the
predominant contributor to hydrocarbons in urban runoff,
oil drips, per se, are only a minor source (15 percent) of
the lead in runoff (Latimer, 1984). Copper entering Narra-
gansett Bay comes from sewage treatment plants, with
the Providence plant contributing over half of the bay's
copper content. The copper comes from industrial dis-
charges to the sewer system from metal-finishing and
electroplating industries. For Zn, both sewage treatment
plants and urban runoff are important sources.
WATER BODY VERIFICATION
Recently, we conducted an experiment to determine the
impact of a rain event on the water quality of the Pawtuxet
River. The rain event also afforded us the opportunity to
properly evaluate the application of urban runoff loading
factors developed in our earlier study. We combined our
urban runoff data with Pawtuxet River land use data to
estimate the urban runoff loads we anticipated for this
storm. A comparison of the predicted urban runoff load to
the river with the actual load we observed through river
monitoring is given in Table 3. The actual and predicted
discharge rates agreed within a factor of 2 for 9 of the 12
components we examined. All of the rates agreed within a
factor of 3.
These data also allowed us to evaluate how important
urban runoff components are to the water quality of the
river during storms. The background discharge rates (re-
sulting from point sources) were minor in comparison with
the wet weather contributions for most of the PAHs, HC,
Pb, and Zn. Concentrations of Cd and Cu were not greatly
affected by stormwater inputs. During this storm, 85 per-
cent of the PAH's, 79 percent of the hydrocarbons, 82
percent of the Pb, and 63 percent of the Zn were due to
wet weather inputs.
In summary, on an annual basis, urban runoff was the
major source of hydrocarbons and lead to Narragansett
Bay, and a significant source of PAHs and zinc to this
estuary. The urban runoff loading rates we determined
were later found to predict accurately the actual wet
weather inputs to one of the bay's tributaries. Changes in
tributary discharge rates during wet weather conditions
can be substantial.
Table 3.—Comparison of actual Pawtuxet River discharge rates with predicted urban runoff loads (Nov. 3-4,1983,1.39 cm
rainfall, river station #9).
Pb
Zn
Cd
Cu
HC
PAH
Actual
discharge
3770 gm
258 kg
455 gm
11.9kg
101 kg
240 gm
Background
dry weather
discharge
667 gm
96.4 kg
369 gm
9.8 gm
20.9 kg
36.7 gm
Urban runoff
from monitoring
data
3110 gm
162 kg
86 gm
2.1 kg
80.0kg
204 gm
Predicted
urban runoff
rate from land
use data
6230 gm
106 kg
46 gm
3.8kg
200 kg
267 gm
Ratio of
actual to
predicted
rate
0.50
1.52
1.82
0.55
0.40
0.76
163
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
REFERENCES
Brinkman, D.W., A.L. Whisman, N.J. Weinstein, and H.R. Em-
merson. 1981. Environmental Resource Conservation and Ec-
onomic Aspects of Used Oil Recycling. DOE/BE TC/RI-80/11.
U.S. Dep. Energy, Washington, DC.
Hoffman, E.J. 1983. A simple model to predict CSO discharges
from urban runoff and treatment plant influent data. Unpubl.
mss.
Hoffman, E.J., E. Askins, J.G. Quinn, and J. Marchand. 1981.
Used crankcase oil disposal practices: implications to recy-
cling programs. Int. Conf. Energy Ed., Providence, Rl. Aug. 4-
7,1981.
Hoffman, E.J., C.G. Carey, G.L. Mills, and J.G. Quinn. 1984. The
magnitude and effect of wet weather pollutant inputs to a mu-
nicipal wastewater treatment facility served by a combined
stormwater sewage collection system. Unpubl. mss.
Hoffman, E.J.. A.M. Falke, and J.G. Quinn. 1980. Waste lubri-
cating oil disposal practice in Providence, Rhode Island: po-
tential significance to coastal water quality. Coastal Zone Man-
age. J. 8: 337.
Hoffman, E.J. et al. 1985. Stormwater runoff from highways:
chemical and physical characteristics and implications for
treatment. Water Air Soil Pollut. In press.
Hoffman, E.J., J.S. Latimer, C.D. Hunt, and J.G. Quinn. 1983.
Inputs of pollutants into Rhode Island rivers via urban runoff.
Rep. Rhode Island Dep. Environ. Manage. Div. Water Resour.
Providence.
Hoffman, E.J., J.S. Latimer, G.L.. Mills, and J.G. Quinn. 1982.
Petroleum hydrocarbons in urban runoff from a commercial
land use area. J. Water Pollut. Control Fed. 54: 1517-25.
Hoffman, E.J., G.L Mills, J.S. Latimer, and J.G. Quinn. 1983.
Annual input of petroleum hydrocarbons to the coastal envi-
ronment via urban runoff. Can. J. Fish. Aquat. Sci. 40 (Suppl.
2) 41-53.
Hoffman, E.J. et al. 1984. Urban runoff as a source of polycyclic
aromatic hydrocarbons to coastal waters. Environ. Sci. Tech-
nol. 18: 580-7.
Latimer, J.S. 1984. Characterization of the sources of hydrocar-
bons in urban runoff from relationships of organic distributions
and metal content. M.S. Thesis. Dep. Chemistry, Univ. Rhode
Island.
Marchand, J.P., E. Askins, L. LeFebvre, and M. Lehder. 1980.
Used Oil Recovery in South Carolina. S.C. Joint Leg. Comm.
Energy, Columbia.
Martin, B.K., and D. Robadue. 1983. Estimates of combined
sewage and storm water flows from the city of Providence.
Coastal Resour. Center, Univ. Rl, Narragansett.
Metcalf and Eddy, Inc. 1983. Demonstration of Water Quality
Benefits for Rhode Island Combined Sewer Overflow Control
Projects. Rep. to R.I. Dep. Environ. Manage., Providence.
Olsen, S., and V. Lee. 1979. A Summary and Preliminary Evalu-
ation of Data Pertaining to the Water Quality of Narragansett
Bay. Coastal Resour. Center, Univ. R.I., Narragansett.
Pruell, R.J. 1983. Personal comm. Graduate School Oceanogr,
Univ. Rhode Island, Narragansett.
U.S. Environmental Protection Agency. 1984. Water Plann. Div.
Final Rep. Nationwide Urban Runoff Prog. Washington, DC.
164
-------
CHESAPEAKE BAY NONPOINT SOURCE POLLUTION
JOSEPH MACKNIS
Chesapeake Bay Liaison Office
Annapolis, Maryland
ABSTRACT
In 1976, the EPA was directed by Congress to conduct an
in-depth study of the Chesapeake Bay, its resources and
its management. The goal was "to protect and preserve
the quality of the Chesapeake Bay by effectively manag-
ing its uses and resources." In completing the $27 million
study, the EPA Chesapeake Bay Program developed a
watershed model to estimate point and nonpoint source
loadings to the Bay and to evaluate management strate-
gies in reducing nutrient loadings. Model production runs
indicate that nonpoint sources contribute between 31 and
64 percent of the phosphorous load and between 62 and
81 percent of the nitrogen load to the Bay system depend-
ing upon annual rainfall conditions. Most of the phospho-
rous loadings to Chesapeake Bay come from point
sources which are concentrated close to tidal waters,
while most of the nitrogen enters the Bay from nonpoint
sources located throughout the basin, primarily runoff
from agricultural croplands. Model simulations indicate
that a Level II best management practice such as conser-
vation tillage is a cost-effective management alternative.
In response to the findings of the Chesapeake Bay Pro-
gram, the Bay States of Maryland, Pennsylvania, and Vir-
ginia initiated agricultural (and urban) nonpoint source
control programs that increase technical and financial as-
sistance to farmers and augment demonstration projects
and education efforts. The Program is tracking these ef-
forts and attempting to evaluate their effectiveness in
controlling nonpoint source pollution.
INTRODUCTION
In 1976, the U.S. Environmental Protection Agency (EPA)
conducted an in-depth study of the Chesapeake Bay, its
resources and management "to protect and preserve the
quality of the Chesapeake Bay by effectively managing its
resources." EPA fulfilled this Congressional mandate
through the Chesapeake Bay Program (CBP), which doc-
umented declines in living resources such as submerged
aquatic vegetation (SAV), striped bass, shad, oysters, and
clams. These declines parallel changes in water quality
which include increases in nutrient concentrations, chloro-
phyl a, turbidity, and toxic chemicals and decreases in
levels of dissolved oxygen.
Specifically, submerged aquatic vegetation has de-
clined dramatically throughout the Bay; landings for fresh-
water spawning fish, such as shad, alewife and striped
bass, have decreased in recent years; oyster spat set also
has declined significantly in the past 10 years. Nutrient
increases (primarily nitrogen and phosphorus) in many
areas of the Bay have led to declining water quality Ele-
vated levels of heavy metals and toxic organic compounds
are found in Bay water and sediment; and the amount of
Bay water showing low (or no) dissolved oxygen in the
summer is estimated to have increased 15-fold in the last
30 years.
The $27 million research study attributes the decline to
excessive nutrients and, to a lesser degree, toxic effluents
and sedimentation. The nutrients, primarily from munici-
pal waste discharges and agricultural runoff, spur the
growth of algae that deplete oxygen from the water and
prevent sunlight from reaching the submerged aquatic
vegetation that provides critical habitat to the Bay's living
resources. Sediment and toxic effluents also directly af-
fect vegetation and fish. This paper focuses primarily on
nutrient pollution.
SOURCES OF NUTRIENTS
The sources of nutrient loadings to the Chesapeake Bay
are influenced by population growth and land use within
the 64,000 mi2 catchment area that includes portions of
six States and the District of Columbia (Fig. 1). For man-
agement purposes the area was divided into the eight
major drainage basins also shown in Figure 1. Basinwide,
the population grew 49 percent between 1950 and 1980
and is projected to grow an additional 15 percent by the
year 2000, to a total of 14.6 million. Population growth
contributes to the major point source of nutrients to the
Bay, sewage treatment plants. The other major type of
point source in the basin is industrial wastewater.
In addition to increasing sewage treatment plant dis-
charge volume, population increases drive changes in
land use. The percentage of land in urban and residential
usage has increased 282 percent since 1950 and, al-
though agriculture land use has declined somewhat, the
agricultural and livestock practices employed have inten-
sified.
CHESAPEAKE BAY WATERSHED MODEL
Estimates of sources and loadings of nutrients to the Bay
as well as the efficacy of management strategies to con-
trol them, were determined with the assistance of a Bay-
MAJOR BASINS
1. Susquehanna
2. Eastern Shore
3. West Chesapeake
4. Patuxent
5. Potomac
6. Rappahannock
7. York
8 James
Figure 1 .—The Chesapeake Bay drainage basin..
165
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
wide water quality model. The model (Hartigan et. al.
1983), simulated nonpoint source loadings between
March 1 and Oct. 31, the period most important in terms
of algal growth in the Chesapeake Bay. Model input data
were based on 1980 point source loadings and land use.
The U.S. Geological Survey (USGS) rainfall records from
a wet year (1975), dry year (1966), and an average year of
rainfall (1974), were used to estimate the nonpoint loads.
The basin model routed nonpoint (and point) source loads
to the fall line, simulating the physical, chemical, and bio-
logical processes that transform the pollutants as they are
transported downstream.
Model production runs indicate that the total nutrient
load to the Bay varies according to rainfall conditions (Fig.
2) and that the relative amounts of point and nonpoint
source loadings to the Bay similarly change with rainfall
conditions. Baywide, nonpoint sources contribute from 31
to 64 percent of the phosphorus load (39 percent under
average rainfall conditions) and from 62 to 81 percent of
the nitrogen load (67 percent = average). Point sources
contribute from 36 to 69 percent of the phosphorus load
and 19 to 38 percent of the nitrogen load, depending upon
the annual rainfall conditions; under average conditions,
point sources contribute 61 percent of the phosphorus
load and 33 percent of the nitrogen.
Figure 3 illustrates the point and nonpoint source load-
ings from each of the major basins discharging to Chesa-
peake Bay during an average rainfall year. Collectively, the
three major tributaries to the Bay, James, Potomac, and
Susquehanna contribute 30 percent of the nonpoint
source load and 70 percent of the total phosphorus load.
For nitrogen, they contribute 55 percent of the nonpoint
Phosphorus
Dry Year
Wet Year
Average Year
5,500,000 kgs.
6,300,000 kgs.
Nitrogen
Dry Year
10,800,000 kgs.
Wet Year
Average Year
55,966,000 kgs.
66,465,000 kgs.
119,669,000 kgs.
Point Sources
Non-
point Sou roes
Figure 2.—Bay-wide nutrient loadings (March to October) under dry, average, and wet conditions.
166
-------
ESTUARINE QUALITY
source load and 78 percent of the total load during aver-
age rainfall conditions. The James is dominated by point
sources while the Susquehanna is dominated by nonpoint
sources; the Potomac has a more balanced mixture. To be
effective, nutrient control strategies must recognize the
unique nature of each basin and the relative contributions
of point and nonpoint sources of nutrients within each.
Figure 4 illustrates which basins are dominated by point
and nonpoint sources. It clearly shows that point sources
are concentrated in sub-basins adjacent to Chesapeake
Bay; essentially the urban corridor between Baltimore,
Maryland, and Washington, D.C., and the fall line city of
Richmond, Virginia. These point-source-dominated areas
have high population densities and consequently, large
volumes of wastewater discharged from sewage treat-
ment plants. Model estimates of point and nonpoint
source loads for each major drainage basin from above
and below the fall line are summarized in Tables 1 and 2.
They indicate that croplands generate a large portion of
the total nutrient load and are by far the major nonpoint
source basinwide. Croplands contribute from 27 to 53 per-
cent of the total phosphorus load and from 60 to 75 per-
cent of the total nitrogen load in average and wet years
respectively.
In contrast, "other" nonpoint sources, which include
runoff from pasture, urban, and forest lands, contribute
only 11 to 12 percent of the total phosphorus and 6 to 7
percent of the total nitrogen load under similar rainfall
conditions. However, the low percentages do not neces-
sarily indicate that these nonpoint sources, especially ur-
ban sources, are not a problem in Bay waters. In urban
areas adjacent to critical habitats such as tidal freshwater
spawning grounds, the accumulated pollutants flushed
from streets and residential areas during wet weather con-
tribute significant quantities of both conventional and toxic
pollutants.
EVALUATION OF MANAGEMENT
ALTERNATIVES
In addition to estimating point and nonpoint source nutri-
ent loadings, the model evaluated the relative effective-
ness of point and nonpoint source controls and estimated
Phosphorous
James
Susquehanna
year 2000 loads. The point source strategies simulated by
the model were primarily technology-based controls that
limit the effluent concentration of nitrogen and phos-
phorus. A phosphorus ban and future (year 2000) loadings
were also evaluated.
For nonpoint sources, the model estimated the impact
of changes in tillage practices and, in the lower Susque-
hanna, the simultaneous strip cropping and conversion of
all conventional tillage cropland in each basin to conserva-
tion tillage. The factor in the model that represents vegeta-
tive cover was the primary adjustment made to simulate
this option. A point source effluent limitation of 2 nig/ total
phosphorus was also tested under existing and future
conditions. Agricultural land use was assumed to remain
unchanged in the year 2000 model simulations.
Table 3 contains the estimated reductions in nutrient
loads, by major basin, achieved in the conservation tillage
model simulation during average and wet rainfall condi-
tions. Conservation tillage is more effective in reducing
phosphorus than nitrogen loads because phosphorus is
transported in the particulate form adsorbed to sediment
particles. Conservation tillage minimizes disturbances of
the soil surface and significantly reduces soil loss.
Nitrogen, however, is mostly soluble and what does not
wash off is taken up by plants or transformed to gas and
percolates down into the ground water, some of which
flows into adjacent waterbodies. The complicated nutrient
forms and pathways, along with diverse crop and pasture-
land management systems, illustrate the need to imple-
ment separate best management practices (BMP's) to
control both nitrogen and phosphorus.
The effectiveness of conservation tillage is related to
current cropping practices, soil type, slope, and other fac-
tors that vary among river basins. In some areas, physical
conditions preclude its use. Furthermore, the benefits of
conservation tillage in preventing sediment and nutrient
losses must be weighed against the increased use of her-
bicides associated with this practice and other farm man-
agement considerations.
Data from model simulations in the lower Susquehanna
indicate that the simultaneous implementation of conser-
vation tillage basinwide and strip cropping in the lower
Susquehanna would reduce existing (1980) total, phos-
Point
Non-point
Potomac
West
Chesapeake
Eastern
Shore Others
16
28%
•21%
21%
17%
«6%-H
Kilograms
Nitrogen
James
2,000.000
Susquehanna
I
4.000.000
Potomac
I I
6.000.000
West Eastern
Chesapeake Shore Others
36
13
2 3
14% '
40%
• 24%
•11%
T
I
T
T
T
I I
Kilograms 20,000,000 40,000,000 60,000,000
Figure 3.—Percentages of nutrient loadings (March to October) by major basin under average rainfall conditions.
167
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Nitrogen
Phosphorus
Point source dominated
Non-point source dominated
Figure 4.—Relative importance of point and nonpoint source of nutrients within major basins.
Basin
Table 1. — Phosphorus loadii
Phosphorus (kg)
igsto
so
conti
Chesa
Point
urce
•Ibutlon
peake Bay by major basin (March-October)
% Cropland % Other
load source load
contribution source contrlb.
t Total
nonpoint
contribution
Dry Avg. Wet Dry Avg. Wet Dry Avg. Wet Dry Avg. Wet Dry Avg. Wet
Part A: at the fall line
Susquehanna
Patuxent
Potomac
Rappahannock
York
James
TOTAL
941,000 1,318,000 2,864,000
156,000 149,000 174,000
326,000 388,000 1,077,000
49,000 47,000 130,000
30,000 35,000 151,000
299,000 349,000 690,000
1,801,000 2,286,000 5,086,000
24
92
27
1
7
45
33
23
90
15
1
7
36
28
12
76
7
1
2
21
14
60
7
52
58
74
46
53
77
19
72
75
86
63
72
16
3
33
41
19
18
19
11
5
21
24
12
16
14
76
8
73
99
93
55
67
76
10
85
99
93
64
72
88
24
93
99
98
79
86
Part B: to tidal waters (below the fall line)
W. Chesapeake
Patuxent
Potomac
Rappahannock
York
James
Eastern Shore
TOTAL
Part C: Part A +
Susquehanna
Patuxent
Potomac
Rappahannock
York
James
W. Chesapeake
Eastern Shore
TOTAL
988,000 1,087,000 1,384,000
59,000 68,000 130,000
882,000 915,000 1,263,000
54,000 79,000 221 ,000
39,000 65,000 208,000
1,325,000 1,374,000 1,570,000
345,000 379,000 962,000
3,692,000 3,967,000 5,738,000
PartB
941,000 1,318,000 2,864,000
215,000 217,000 304,000
1,208,000 1,303,000 2,304,000
103,000 126,000 350,000
69,000 100,000 359,000
1,624,000 1,723,000 2,259,000
988,000 1,087,000 1,384,000
345,000 379,000 962,000
5,493,000 6,253,00010,786,000
93
79
82
89
84
96
44
87
24
88
67
47
50
86
93
44
69
85
69
79
61
50
93
40
81
23
83
59
39
35
81
85
40
61
67
36
57
22
16
81
16
56
12
58
34
14
10
63
67
16
36
8
19
10
27
27
3
50
12
60
10
23
39
44
12
8
50
27
25
51
31
69
68
14
79
36
77
33
50
71
76
29
25
79
53
7
12
11
12
10
4
10
7
17
7
18
22
6
7
7
10
12
8
13
12
9
8
5
5
8
11
9
16
15
14
8
8
5
11
7
21
18
11
16
4
56
13
76
12
33
53
50
14
7
56
31
15
31
21
39
50
7
60
19
77
17
41
61
65
19
15
60
39
23
64
43
78
84
19
84
44
88
42
66
86
90
37
23
84
64
168
-------
ESTUARINE QUALITY
Table 2.—Nitrogen loadings to Chesapeake Bay by major basin (March-October)
Basin
% Point % Cropland % Other f Total
source load source load nonpolnt
Nitrogen (kg) contribution contribution source contrib. contribution
Dry Avg. Wet Dry Avg. Wet Dry Avg. Wet Dry Avg. Wet Dry Avg. Wet
Part A: at the fall line
Susquehanna
Patuxent
Potomac
Rappahannock
York
James
TOTAL
21 ,500,00026,500,00048,000,000
580,000 536,000 809,000
6,270,000 7,500,00017,800,000
695,000 727,000 1,680,000
380,000 370,000 1,264,000
1,760,000 2,300,000 5,030,000
31 ,185,00037,933,00074,559,000
10
71
10
10
10
10
11
10
65
10
10
10
9
11
5
41
10
10
10
8
7
Part B: to tidal waters (below the fall line)
W. Chesapeake
Patuxent
Potomac
Rappahannock
York
James
Eastern Shore
TOTAL
Part C: Part A +
Susquehanna
Patuxent
Potomac
Rappahannock
York
James
W. Chesapeake
Eastern Shore
TOTAL
6,179,000 7,265,00010,038,000
439,000 596,000 1,278,000
8,094,000 8,399,000 1 1 ,394,000
279,000 611,000 2,047,000
315,000 688,000 2,255,000
6,272,000 7,013,000 8,913,000
. 3,269,000 3,973,000 9,500,000
24,847,00028,565,00045,425,000
PartB
21 ,500,00026,455,00047,727,000
1,015,000 1,133,000 2,088,000
1 4,367,000 1 5,944,000 29, 1 67,000
975,000 1,339,000 3,734,000
630,000 1,058,000 3,492,000
8,032,000 9,320,00013,945,000
6,179,000 7,265,00010,038,000
3,269,000 3,973,000 9,500,000
55,967,00066,487,0001 1 9,691 ,000
85
48
77
37
34
88
13
72
10
61
48
17
22
71
85
13
38
72
35
74
17
15
79
10
62
10
49
44
13
13
62
72
10
33
52
16
55
5
5
62
4
39
5
26
28
7
7
43
52
4
19
85
29
83
72
78
73
83
20
55
17
73
76
15
83
30
85
43
48
72
77
29
20
83
60
91
53
84
78
82
78
88
40
75
37
89
90
32
92
54
91
66
66
84
87
49
40
92
75
5
6
7
18
12
18
6
8
10
9
10
9
6
7
4
6
6
12
8
14
5
8
9
8
6
5
6
4
90
29
90
90
90
90
89
15
52
23
63
66
12
87
90
35
90
90
90
91
89
28
65
26
83
85
21
90
95
59
90
90
90
92
93
48
84
45
95
95
3
96
28
38
61
5
8
8
15
10
9
8
7
7
4
8
6
9
6
8
8
4
6
90
39
52
83
78
29
15
87
62
90
51
55
87
87
38
28
90
67
95
74
72
93
93
57
48
96
81
Table 3.—Estimated nutrient reductions achieved In level two model simulation under average and wet conditions
(March-October).
% Phosphorus load reduction % Nitrogen load reduction
Basin
(kg reduction)
Avg. year Wet year
(kg reduction)
Avg. year Wet year
Susquehanna
West Chesapeake
Eastern Shore
Patuxent
Potomac
Rappahannock
York
James
Basin-wide
16.0
(211,000)
2.3
(25,000)
14.3
(54,000)
1.1
(2,000)
4.3
(56,000)
5.1
(6,000)
6.7
(8,000)
0.8
(15,000)
6.5
(377,000)
32.0
(916,000)
14.4
(200,000)
43.7
(421,000)
14.2
(43,000)
25.4
(594,000)
35.0
(122,000)
37.0
(141,000)
9.5
(214,000)
24.5
(2,651,000)
1.3
(356,000)
1.7
(120,000)
6.3
(250,000)
0.8
(9,000)
1.3
(207,000)
1.9
(25,000)
2.5
(26,000)
0.5
(49,000)
1.6
(1,042,000)
8.0
(3,818,000)
10.9
(1 ,098,000)
23.9
(2,273,000)
11.6
(241,000)
11.1
(3,228,000)
18.0
(669,000)
20.0
(436,000)
7.6
(1,066,000)
10.7
(12,612,000)
phorus and nitrogen loads from the Susquehanna 22 and
5 percent respectively. This indicates that significant ba-
sinwide reductions in nutrient loadings, including nitro-
gen, can be achieved through appropriate BMP's. Final
decisions, however, should consider agricultural strate-
gies that leave the specific BMP's to the discretion of
farmers and soil conservationists.
CHESAPEAKE BAY NONPOINT SOURCE
RECOMMENDATIONS (NUTRIENTS)
The watershed model showed nutrient loads to the Chesa-
peake can be reduced through control strategies. The Bay
community supports reductions to improve Bay condition.
Although it is very difficult to predict with confidence water
quality or ecological response in the Bay, enough is known
today to call for limiting nutrient loads to Bay waters.
In 1983 the Chesapeake Bay Program developed the
following specific recommendations to control and reduce
nonpoint pollution (Tippe et a!. 1983).
• The States and EPA, through the Management Com-
mittee, should develop a detailed nonpoint source control
implementation program as part of a basinwide water
quality management plan.
• The U.S. Department of Agriculture and the EPA, in
consultation with the Management Committee, should
strengthen and coordinate their efforts to reduce agricul-
tural nonpoint source pollution to improve water quality in
Chesapeake Bay.
• Federal agencies, States, and counties should de-
velop incentive policies by July 1, 1984, that encourage
farmers to implement BMP's.
• The State, counties, and municipalities located in
169
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
subbasins adjacent to tidal-fresh and estuarine segments
of Chesapeake Bay and its tributaries should implement
fully and enforce existing urban stormwater runoff control
programs.
° The States of Maryland and Virginia and local gov-
ernments should consider strengthening wetland protec-
tion laws to include nontidal wetlands because of their
value as nutrient buffers and living resource habitat.
Following the publication of the Chesapeake Bay Program
findings and results, a conference was convened by the
Governors of Virginia, Maryland and Pennsylvania, the
Mayor of the District of Columbia, the EPA Administrator,
and the Chesapeake Bay Commission. The conference
marked the beginning of a coordinated visible effort to
correct problems identified by Chesapeake Bay Program
reports.
The centerpiece of the commitments made by the spon-
sors was the "Chesapeake Bay Agreement of 1983"
which recognized the need for a regional management
structure to support and enhance a regional cooperative
approach for the environmental management of the Bay.
The Agreement provided the authority to establish an Ex-
ecutive Council, an Implementation Committee, and a
Chesapeake Bay Liaison Office. The Executive Council is
to assess and oversee the implementation of coordinated
plans to improve and protect the water quality and living
resources of the Chesapeake Bay estuarine system. The
Implementation Committee will coordinate technical mat-
ters and develop and evaluate management plans. The
Committee has established subcommittees for Planning,
Monitoring, Modeling and Research, and Data Manage-
ment. The Liaison Office will advise and support the Coun-
cil and Committee.
The Liaison Office has assumed the lead in coordinat-
ing Federal clean-up efforts and has negotiated Memo-
randa of Understanding (MOD) with five other Federal
agencies whose activities impact Bay resources and water
quality. These agencies include: U.S. Fish and Wildlife
Service (F&WS), the Soil Conservation Service (SCS), the
National Oceanic and Atmospheric Administration
(NOAA), the U.S. Army Corps of Engineers (COE), and the
U.S. Geological Survey (USGS).
All of the MOU agencies pledge cooperation in areas of
mutual interest and support of the goals of the Chesa-
peake Bay Agreement. The SCS has deployed 10 addi-
tional people to work specifically in the Chesapeake Bay
drainage basin to help train State and Federal agency
personnel in the application of best management prac-
tices to control nonpoint source pollution from agricultural
lands.
NOAA will work with EPA in monitoring trends in the
Bay. USGS will work with other agencies in developing
mapping techniques and evaluating impacts of ground-
water pollution on the Bay. F&WS will work with other
agencies to evaluate certain wetlands activities and assist
with monitoring trends of contaminants in fish.
The Corps will provide particular help with modeling the
Bay and tributaries, and work with other agencies while
conducting its recently authorized Chesapeake Bay Ero-
sion Control Study. In addition to the above MOUs the EPA
has signed a Joint Resolution on Pollution Abatement in
the Chesapeake Bay with the Department of Defense
(DoD). The DoD has pledged to give priority consideration
to funding pollution control projects and studies affecting
the Bay.
Complementing these Federal efforts, the District of Co-
lumbia and the States of Maryland, Pennsylvania, and
Virginia have each initiated programs to reduce pollutant
loadings and to protect and restore Bay resources and
habitat within their jurisdiction. For example, the District of
Columbia has developed initiatives to deal with problems
in the Upper Potomac Estuary which may be contributing
to the decline of the Bay. The program covers point and
nonpoint source pollution, provides resource manage-
ment, and encourages regional cooperation. Moreover,
the District of Columbia is developing a stormwater regu-
latory program to control new development and redevel-
opment after construction. A BMP manual and a home-
owners BMP guidebook will complement the regulations.
These products will not only reduce loadings of pollutants,
but will also improve public understanding of the need to
abate nonpoint source pollution.
The Maryland General Assembly appropriated $36 mil-
lion in FY 1985 for a variety of point and nonpoint source
pollution control strategies including $2 million cost shar-
ing to implement agricultural BMP's and $1.4 million to
hire 42 new employees to provide technical assistance to
landowners in designing appropriate BMP's. Existing cost
share program grants have already totaled $5 million
since July, 1983. Another important component of the agri-
cultural conservation program is an intensive informa-
tional and educational program to encourage farmer par-
ticipation in pollution control activities. The overall goal of
the Maryland agricultural initiative is to have conservation
plans in place on farms located in "priority" areas having
direct impacts on Chesapeake Bay water quality within 5
years.
Other nonpoint pollution abatement actions the State of
Maryland has undertaken include:
o Increasing enforcement of the State stormwater con-
trol law that requires that streams be just as clean after
nearby construction as they were before construction;
o Transferring authority for enforcing sediment and ero-
sion control laws to the State unless counties can demon-
strate they can do the job;
o Establishing rules and regulations requiring efficient
design, construction, operation and maintenance of agri-
cultural drainage projects;
° Providing grants to local governments for a forest
buffer program;
° Providing construction funds for shoreline erosion
control; and
° Increasing appropriations for the conservation ease-
ment program.
For the 1984-86 biennium, the 1984 General Assembly
of Virginia appropriated $10.4 million for Chesapeake Bay
initiatives, including $2.5 million for an agricultural pollu-
tion control plan. The largest single element in the plan is
a program to cost share the installation of BMP's with
farmers.
This program employs a multiple level targeting strat-
egy. At the first level, all farmers within Virginia's portion of
the Bay watershed are eligible for cost-sharing assistance
on certain, specified, water-quality-related BMP's. The
second level targets cost-sharing funds for certain prac-
tices to subbasin areas with intensified, cropland and ani-
mal waste practices. The third level, a demonstration proj-
ect, targets a small agricultural watershed for an intensive
BMP promotion program. Continuous water quality moni-
toring at the site should give an indication of the water
quality impacts of the BMP program over time.
The Virginia agricultural control plan also established a
process for identifying priority areas where technical as-
sistance, demonstration projects and education programs
will be targeted. The goal of the program is to increase
implementation of BMP's by farmers and land developers
within the Chesapeake Bay drainage basin. In addition to
the agricultural initiative, the Commonwealth has estab-
lished other nonpoint initiatives demonstrating pollutant
170
-------
ESTUARINE QUALITY
control from urban areas and assisting low income shore-
line residents with sanitation deficiencies to install septic
tanks and other facilities.
The Commonwealth of Pennsylvania has also initiated a
comprehensive agricultural nonpoint source control pro-
gram with the commitment of $2 million in State and Fed-
eral funds in its fiscal 1985 budget. One million dollars in
financial assistance is available to assist Pennsylvania
farmers implement BMP's to control soil and nutrient loss.
Educational programs will help Pennsylvanians under-
stand the Bay's problems, their contributions to those
problems, and explain ways to mitigate those problems.
Additional educational programs, particularly for farmers,
will stress the importance and potential savings from nutri-
ent management.
, tillage demonstration projects will compare
different practices and show proper tillage
s. A pesticides management program will pro-
ftformation on appropriate projects and a manure
FTagement program will stress on and off site use of
Manure as a resource. The program's goal is to accelerate
rthe implementation of best management practices on agri-
cultural land. It focuses on animal waste and nutrient man-
agement. The initial phase targets seven watersheds in
the lower Susquehanna River with high livestock density
and intensive cropping practices. The program will later
be extended to other watersheds.
IMPLEMENTATION GRANTS TO THE
STATES
To assist the States and the District of Columbia in devel-
oping programs to improve Bay water quality and re-
sources, EPA awarded $3 million in implementation grants
in 1984. It is anticipated that the current Administration will
provide $10 million for each of the next 4 years. Approxi-
mately $7.2 million will be available annually for State im-
plementation grants. Although various types of projects
are eligible for funding, FY1985 grant criteria require that
75 percent of the grant amount be applied towards non-
point source controls. Structural, educational, and demon-
stration projects which address a significant pollution
source in geographic areas of concern will also receive
priority.
The States and EPA have been further directed that in
selecting projects to be funded by Chesapeake Bay Imple-
mentation grants, they must consider the following crite-
ria:
• The project's potential contribution to reductions in
pollutant loadings or improvements in resource habitat;
• The appropriateness and cost-effectiveness of the
project. Higher priority should be given to projects located
in designated critical watersheds;
• The potential beneficial effect of the project on eco-
logically important areas in the Bay;
• The unavailability of other Federal funding. For exam-
ple, projects that can be funded through EPA's construc-
tion grants program should not be considered;
• The project should be included in the Restoration and
Protection Plan of 1985.
• The project represents an incremental step in a
phased long-term commitment to determine effective new
programs or is part of a comprehensive abatement pro-
gram in a specific hydrologic unit or watershed.
• NPS implementation efforts should be concentrated
in targeted hydrologic units or targeted to types of sources
for which solutions are not known.
OUTLOOK
The Chesapeake Bay Program findings clearly indicate
that the Bay's water and sediment quality have degraded
and many of its important living resources have declined.
Given the increasing environmental stress projected to be
placed on the Bay resulting from population increases and
land use changes, it will be difficult to halt this decline and
even more difficult to reverse it. It is generally agreed,
however, that reducing the nutrient loadings to the Bay
from point and nonpoint sources will begin to restore the
environmental quality of the Bay.
Fortunately the States, EPA, and other Federal agen-
cies already have begun control efforts to address ob-
served Bay problems. While scientists, however, cannot
predict with confidence how much the current and pro-
posed initiatives will reduce nutrient (and toxic) loadings to
the Bay nor how quickly or extensively the Bay will re-
spond, it is generally agreed that a long-term strategy is
necessary to restore and protect the Chesapeake Bay.
So that mid-course correction in control strategies can
be made, the effectiveness of agricultural nonpoint source
programs must be assessed. It is therefore necessary that
a monitoring and tracking system be established. The
monitoring system should include both water quality and
biological monitoring and provide input for model develop-
ment to project results from BMP implementation. Effec-
tive monitoring will identify areas where BMP implementa-
tion measurably improved water quality. The tracking
system will help document where and under what condi-
tions specific BMP's were implemented and allow calcula-
tion of their cost effectiveness. Data gathered from these
parallel efforts, along with results of specific programs and
projects, will help to guide the cooperative Federal and
State efforts to restore and protect the Chesapeake Bay.
REFERENCES
Tippe, V.K., et al. 1983. Chesapeake Bay: a framework for
action. Chesapeake Bay Progr. U.S. Environ. Prot. Agency.
Natl. Tech. Inf. Serv., Springfield, VA.
Hartigan, J., et al. 1983. Chesapeake Bay Basin Model. Final
rep. N. Va. Plann. Oist. Comm., Annandale, VA.
171
-------
THE INFLUENCE OF NFS POLLUTION IN FLORIDA ESTUARIES:
A CASE STUDY
JOE RYAN
J. H. COX
Department of Environmental Regulation
Tallahassee, Florida
ABSTRACT
A study designed to characterize the pollutant climate of
13 major bays and estuaries in Florida was carried out by
examining sediment chemistry. This study provided im-
proved interpretive tools that were used to distinguish
natural versus anthropogenic metal concentrations and
to help identify nonpoint sources. Results confirmed that
although elevated metals and synthetic organic com-
pounds were present in the water column, the concentra-
tion of these constituents was well below State and Envi-
ronmental Protection Agency water quality standards.
These data provide a clearer understanding of pollutant
trends and revealed encroachment of metal contamina-
tion in several major estuaries. The highest levels of
metals and synthetic organics were found in sediments
from the Miami River and Biscayne Bay. The river re-
ceives pollutants—particularly Ag, Cd, Cu, Hg, Pb, and
Zn—from a variety of nonpoint sources originating from
the adjacent city of Miami. The river essentially becomes
a point source discharging moving contaminants into the
Bay. Results from the study were used as a basis for
making recommendations to State and Federal agencies
for cleaning up nonpoint sources entering the Miami
River.
9
INTRODUCTION
Nonpoint sources, especially urban stormwater, are a ma-
jor source of pollution to bays and estuaries along devel-
oped coastal areas. Although the need to protect these
productive environments is widely recognized, there are
many deficiencies in traditional regulatory approaches.
This paper illustrates how a better understanding of
sediment chemistry provides more meaningful informa-
tion for assessing and managing nonpoint source dis-
charges of metals and organic compounds. It also outlines
how these techniques have recently been used iflgtate
and local attempts to eradicate the effects of
stormwater outfalls in the Miami River and Bisca^
in South Florida.
Studies of contamination in coastal areas have
ally focused on water quality studies in which results ;
compared with a State or federally established water qual-"1
ity standard. This preoccupation with water quality criteria
is counterproductive for three fundamental reasons. First,
traditional approaches relying on water quality information
fail to adequately consider environmental geochemistry.
Second, the use of a water quality standard alone offers
little protection to the estuarine biota, most of which are
linked to the sediment through food webs. Finally, the cost
of carrying out water quality studies in large urbanized
areas with complex nonpoint source problems can be con-
siderable and still not provide meaningful measures of
pollution. With the limited funds available to study non-
point pollution, improvements are needed to provide the
best information for the least money.
GEOCHEMICAL CONSIDERATIONS
In contrast to lake and ocean systems, the processes that
affect the distribution of chemical constituents in an estu-
ary are complex and often poorly understood. As a result,
undue emphasis is often placed on inappropriate and mis-
leading pollution indicators. Reliable interpretive tools for
assessing the degree of estuarine contamination relative
to background conditions are virtually nonexistent.
In Florida, eight of the 10 largest cities are surrounded
by marine or brackish waters that receive a wide variety of
nonpoint source discharges. As the aqueous nonpoint
sources mix with brackish or marine waters in these ar-
eas, many of the materials previously suspended or dis-
Table 1.
Arsenic
Antimony
Cadmium
Chromium
Copper
Fluoride
Iron
Lead
Mercury
Nickel
Silver
Zinc
—Metal and fluoride concentrations reported for water from other regions (In ^g/L)
World S.E. U.S. Hudson Average S.E. U.S.
rivers1 rivers2 estuary3 ocean water1 coastal waters2
1.70
1.00
0.02
1.00
1.50
—
40
0.10
—
0.50
0.30
30
0.04-0.65
—
0.002-0.02
—
0.25-0.77
50-100
22-120
0.02-0.51
0.01-0.04
0.11-0.57
—
0.21-2.0
—
—
0.1-0.5
—
1.0-7.0
—
5-96
—
—
0.8-11
—
.3-33
1.50
0.24
0.01
0.30
0.10
—
2.00
0.003
—
0.20
0.04
0.10
0.70-1 .60*
—
0.01-0.03
—
0.06-0.45
1.00-1.50
0.30-5.60
0.012-0.083
0.004-0.046
0.29-0.47
—
0.06-0.32
Florida
standard
50
—
5
50
15
5,000
300
30
0.10
100
0.05
1,000
'From Martin and Whitfield (1983)
'From Wlndom and Smith (1984); Wlndom et al. (1984); Windom and Taylor (1979);
Waslenchuk and Windom (1978); Windom (1971)
'From Klinkhammer and Bender (1981)
•From Waslenchuk (1978)
All references are given In the bibliography in the Manual (Ryan et al. 1984).
172
-------
ESTUARINE QUALITY
60-i
50-
I40"
20-
10-
0
.-4
Slope- 4x10 ..
U.SxlO'4)
I I I I
5678
|^100-
Slope- 9x10 .
(8.5x10 )
Aluminum |
Figure 1 .—Relationship between lead and aluminum, and
zinc and aluminum observed in natural sediments from the
southeastern United States (Wlndom, 1984) Slope of each
line reflects the calculated metal-to-metal ratio. Values in
parentheses are average ratios reported in the scientific lit-
erature (referenced in the text).
solved in freshwater are rapidly incorporated into bottom
sediments by physical and chemical processes such as
flocculation, precipitation and coprecipitation with scav-
enging. As a result, the estuarine water column shows
extremely low concentrations of trace metals and organic
compounds. Indeed, many of the constituents remaining
in the complex saltwater matrix approach levels at or be-
low the analytical detection limits of most chemical labora-
tories. More importantly, such low concentrations enhance
the potential for sample contamination, leading to spuri-
ous results.
Table 1 shows the range of concentrations for several
trace metals in waters throughout the United States
(Klinkhammer and Bender, 1981; Martin and Whitfield,
1983; Waslenchuk, 1978; Waslenchuk and Windom,
1978; Windom and Taylor, 1979; Windom and Smith,
1984; Windom et al. 1984). The last column shows State
of Florida water quality standards which are in many re-
spects the same as EPA's. This table shows that metal
levels in the water column rarely approach water quality
standards. Except for sampling in the plume of a dis-
charge, violations of water quality standards for trace
metals or organic compounds in marine or brackish wa-
ters are difficult to find. It follows, then, that traditional
water quality standards for metals and organic com-
pounds—originally developed for drinking water—are in-
appropriate in the marine environment.
Table 1 and findings by other investigators (Pavlou and
Weston, 1983; Talbot, 1983; Williams et al. 1978) indicate
that bottom sediments, not the water column, are the real
indicators of pollution in coastal environments. The con-
cept that sediments reflect the degree to which an estuary
is contaminated is straightforward, but understanding the
levels of contamination is more difficult. This is especially
true for metals, since they occur both naturally and as a
result of man's activities. At present, there is no consen-
sus on a reliable tool for judging the extent of metal con-
tamination. Such tools must be developed on a regional,
rather than on a national basis.
The complexities of understanding the relevance of
metal levels as they occur over heterogeneous substrates
of varying grain size make it extremely difficult to interpret
the degree of pollution based on absolute concentrations
alone (Ackerman et al. 1983; Forstner and Salomons,
1981). While many tools have been used to interpret sedi-
ment data (Brieri et al. 1975; Helz et al. 1975; Nishida et
al. 1982), we have found that the ratio of a trace metal to
aluminum is quite useful for interpreting the degree to
which sediments are enriched with metals in Florida (Ryan
and Windom, in prep.). Sediment data from 13 bays in
Florida indicate that up to 70 percent of the variance in
observed metal concentrations can be explained by alumi-
num.
Figure 1 shows this relationship between lead, zinc, and
aluminum in over 1,100 uncontaminated marine sediment
samples off the southeastern U.S. coast (Windom, un-
publ.). As the concentration of aluminum increases, so do
observed concentrations of lead and zinc. Deviations from
the plotted line suggest that certain sediments are en-
riched in lead and zinc. In essence, these findings provide
a method for normalizing the complex relationships be-
tween metal concentrations and grain size, as well as dis-
tinguishing natural from polluted sediments.
Figure 2 illustrates a broader regulatory use of the me-
tahaluminum relationship that we have employed to deter-
mine metal enrichment in sediments. Metakaluminum ra-
tios are calculated from raw data (in this case copper to
aluminum) and plotted against the absolute metal concen-
trations and ratios reported in unpolluted sediments. If the
point(s) falls in the shaded area the sediments are
deemed to contain natural copper concentrations. Points
•Average Composition of Crustal Material
0.10
Natural
Ratio
OjOOOl
20
25
Copper Concentration in Sediments (ppm)
Figure 2.—Graph depicting copper concentration versus
copper-to-aluminum ratio in natural sediments. Points plot-
ted from empirical data falling within the shaded area are
considered natural. Outliers indicate copper enrichment in
the sediment sample.
173
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
International
airport
Figure 3.—Map showing relative station locations In the Miami River, Tamiami Canal, and Blscayne Bay. Major nonpolnt
source Inputs are shown by arrows (B.R. = Boat Repair facilities).
falling outside the shaded area indicate copper contami-
nation from human activities. Thus, the problem of com-
paring chemical data from areas of different grain sizes is
diminished.
Using this approach, the Florida Department of Environ-
mental Regulation (DER) studied the environmental
chemistry of Florida's bays and estuaries. The study of
Biscayne Bay and adjacent Port of Miami ship channels
offers an example of how pollution trends can be ob-
served more clearly and used to provide a basis for man-
agement of nonpoint source problems in a complex urban
setting. In addition, an application of the previously dis-
cussed interpretive approach is demonstrated.
BISCAYNE BAY AND MIAMI RIVER
STUDY
Biscayne Bay (Fig. 3) is a shallow, tropical lagoon approxi-
mately 48.27 km (35 miles) long and up to 16.09 km (10
miles) wide, with average depth of 3.66 m (12 feet). Sev-
eral features make this bay unique among other urban-
ized coastal areas of the United States. The bay, while
essentially estuarine in character, was tormea as rising
sea level filled in a rigid pre-existing limestone depression,
rather than being formed as a drowned river valley like
many other estuaries. Unlike other estuaries receiving
sediments from river inflows or oceanic processes, fresh-
water inflows from numerous flood control canals carry
little mineral detritus to the bay. Instead, most of the sedi-
ments in Biscayne Bay are produced by the local biota
(Wanless, 1976).
Because of this unique arrangement, Biscayne Bay has
little capacity to dilute or sequester anthropogenic con-
taminants that enter the South Florida coastal environ-
ment. Most pollutants enter the bay from nonpoint sources
in urban Miami, traveling to the bay through canal sys-
tems, stormwater discharge pipes or ground water.
Study results indicated that port sediments were con-
siderably enriched with trace metals, polynuclear aromatic
hydrocarbons (RAM's), and polychlorinated biphenyls
(PCB's). Because the most obvious source of these con-
taminants was a large canal, the Miami River (Fig. 3), an
additional study examined the environmental chemistry of
the river and its tributaries.
Figure 4 shows the results of the bulk sediment analy-
ses for trace metals in the Miami River and in Biscayne
Bay and compares metal concentrations (based on analy-
ses of triplicate samples) with those collected in 10 other
bays and estuaries in Florida. As the graph shows, metal
levels in Biscayne Bay and in the Miami River are signifi-
cantly greater (p < .01) than the average values recorded
at all other study areas in Florida. Trace metal concentra-
tions are normalized for grain size using the metal-to-alu-
minum ratio as discussed earlier. The Miami River is a
major source of trace metals to the bay as shown by the
decreasing concentration of metals from right (river) to left
(bay) in Figure 4. Sediments also followed the same pat-
tern for arsenic and chromium, not shown on this figure.
While metal levels appear to gradually decrease from the
river to the bay, these six metals in Biscayne Bay are still
significantly higher (p < .05) than levels encountered at
10 other bays in the State.
A diverse group of synthetic organic compounds was
also detected in the river, but few appear to have moved
into the bay PCBs (Arochlor 1254) were detected in all
river sediments sampled while four of the 10 PAH's exam-
ined were also ubiquitous throughout the river. In their 2-
year study of the Biscayne Bay system, Corcoran et al.
(1975) found that the highest synthetic organic concentra-
tions occured in the Miami River. While concentrations of
trace metals and synthetic organic compounds are so
high that no benthic organisms were observed during the
sampling program, no violations of water quality stand-
ards were detected.
Sources of Pollution
Numerous potential nonpoint sources of pollution to the
river were identified. For example, RAM's in the Tamiami
Canal apear to originate from activities at the adjacent
Miami International Airport complex.
174
-------
ESTUARINE QUALITY
Cadmium, copper, lead, silver, chromium and zinc con-
centrations were significantly greater (p < .05) at MIR-8
than at any other site in the river. This site is directly adja-
cent to a large boat-building and repair facility on the Ta-
miami Canal.
Silver and arsenic concentrations were also high in sed-
iments from the middle portion of the river and in Seybold
Canal. One possible source is a large hospital complex
that discharges wastes into the canal. Silver could origi-
nate from releasing X-ray wash waters into the canal. The
source of arsenic is not known. Silver appears to be fairly
mobile as reflected by the enriched sediments down-
stream from the canal at the mouth of the river and in
adjacent areas of Biscayne Bay.
Metal enrichment was also encountered in the vicinity of
known stormwater discharges. Surprisingly, stormwater
discharge areas draining the city bus repair facility and a
large scrap metal yard, originally thought to be a source of
metal contamination, showed no metal enrichment along
the immediate shoreline. Because this site receives peri-
odic freshwater discharges from a salinity control barrier
during heavy rainfall, these sediments could be remobi-
lized and moved downriver. High metal levels were found
approximately 500 m below the scrapyard.
In sum, the Miami River receives pollutant inputs from
numerous sources and acts as a temporary repository for
these wastes. The river converts many nonpoint pollution
sources into one large point source discharging into Bis-
cayne Bay. While the total flow of contaminants from Mi-
ami's urban area to the Miami River many be no greater
than for other major cities, the Miami River lacks the large
volume of natural sedimentary materials that accompa-
nies lotic inflows into other estuaries. In other urban areas
such material can more effectively dilute anthropogenic
inputs and sequester pollutants. Thus, the importance of
understanding nonpoint pollution on a regional basis can-
not be overemphasized.
NONPQINT MANAGEMENT ISSUES IN
SOUTH FLORIDA
Water quality in the Miami River has been deteriorating for
over 50 years. Until recently, little if any recognition has
been given to the river/canal as a potentially important
historical, commercial, and recreational resource. After
years of neglect this attitude is changing at both the State
and local levels.
KEY:
o Port oi Miami
41 Miami River
A Berths
a Inner Navigational Channels
• Outer Navigational Channels
1 10 100 0.1 LO 10 10 100 1000
Concentrations (ppm) Concentrations (ppm) Concentrations (ppm)
Figure 4.—Comparison of metal levels In Biscayne Bay and the Miami River with those encountered In sediments of 10 bay-
estuarine systems in Florida (these are average values for Inner and outer portions of the bays and for port berthing areas,
which represent worst case sediments).
175
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
The Miami River Management Committee, established
by executive order of Governor Graham on Dec. 15,1983,
has completed more than 1 full year of operation. The
committee has presented a report setting specific, "do-
able" goals and objectives for restoring and enhancing
the Miami River.
Largely as a result of the sediment analyses discussed
previously, scientists and agency personnel who have
studied the river now believe that stormwater outfalls are a
major, if not the major, source of pollutants entering the
river today. Fifty-five stormwater outfalls greater than
30.48 cm (12 inches) in diameter drain roadways and the
urban and industrial areas that abut the river. In addition,
an unknown number of smaller outfalls, overland runoff,
inwater, and upstream sources contribute to the poor wa-
ter and sediment quality in the river.
Two key problems have emerged as a result of this
study: (1) what to do about the movement of pollutants into
Biscayne Bay from the existing Miami River sediments,
and (2) how to contain nonpoint discharges currently en-
tering the river.
Dredging contaminated river sediments is one option.
However, pollutant levels are so high that disposal of the
dredged material is difficult. The high cost of land in South
Florida prohibits upland disposal. Offshore disposal
seems unlikely because of the nature of the river sedi-
ments. Other land disposal options are severely limited by
the shallowness of the Biscayne Aquifer that supplies
most of South Florida's drinking water.
Regardless of whether the river is dredged, water pollu-
tion sources are being eliminated, particularly from
stormwater outfalls. Priority outfalls are being redesigned
by the city of Miami to percolate the first inch of runoff as a
part of its current $30 million stormwater renovation pro-
gram. Retrofitting those outfalls not scheduled for immedi-
ate reconstruction by the city was given a very high priority
by the Committee, which has asked the State for addi-
tional funds to help eliminate or redesign the remaining
outfalls that cannot be upgraded with available local
funds.
The Department is concerned with the discharge of in-
adequately treated stormwater runoff into State waters.
The agency is seeking information on potential control
techniques for retrofitting or renovating existing stormwa-
ter pollution sources in heavily developed urban areas. To
this end, the Department and the Committee propose to
demonstrate innovative storm drain design and manage-
ment practices in the lower Miami River watershed.
Because of the high cost of storm drain renovation, a
prioritization process was developed to help make the
most efficient use of the available funds. Sediment analy-
ses will be used to pinpoint priority nonpoint source areas
for cleanup.
REFERENCES
Ackerman, F, H. Bergmann, and V. Schleichert. 1983. Monitor-
ing of heavy metals in coastal and estuarine sediment—a
question of grain size. Environ. Technol. Lett. 4: 317-28.
Brieri, R. et al. 1982. Part III: Toxic substances in Chesapeake
Bay Program Technical Studies: A Synthesis. U.S. Environ.
Prot. Agency, Washington, DC.
Corcoran, E. 1984. Biscayne Bay Hydrocarbon Study. Rep. to
Dade County Environ. Res. Managem.
Forstner, U., and W. Salomons. 1981. Trace metal analysis on
polluted sediments—Part I. Environ. Technol Lett. 1: 494-517.
Helz, G.R., R.J. Huggett, and J.M Hill. 1975. Behavior of Mn, Fe,
Cu, Zn, Cd, and Pb discharged from a wastewater treatment
plant into an estuarine environment. Water Res. 9(7): 631-6.
Klinkhammer, G.P., and M.L Bender. 1981. Trace metal distribu-
tions in the Hudson River Estuary. Estuar. Coast. Shelf Sci.
12: 624-43.
Martin, J.M., and M. Whitfield. 1983. The significance of the
river input of chemical elements to the ocean. Pages 265-98
in C.S. Wong et al. Trace Metals in Sea Water, Plenum Press,
New York and London.
Pavlou, S.P., and D.P. Weston. 1983. Initial evaluation of alterna-
tives for development of sediment related criteria for toxic
contaminants in marine waters (Puget Sound). Final Rep. to
Region X, U.S. Environ. Prot. Agency, Seattle.
Nishida, H., M. Miyai, F. Tada, and S. Suzuki. 1982. Computation
of the index of pollution caused by heavy metals in river sedi-
ment. Environ. Pollut. Ser. B. 4(4): 241-8.
Talbot, V. 1983. Lead and other trace metals in sediments and
selected biota of Princess Royal Harbour, Albany, Western
Australia. Environ. Pollut. (Ser. B) 5: 35-49.
Wanless, H.R. 1976. Geologic setting and recent sediments of
the Biscayne Bay Region, Florida. Pages 1-33 in A.
Thorhaug, ed. Biscayne Bay: Past/Present/Future. Special
Rep. No. 5.
Waslenchuk, D.C. 1978. The budget and geochemistry of arse-
nic in a continental shelf environment. Mar. Chem. 7: 39-52.
Waslenchuk, D.C., and H.L. Windom. 1978. Factors controlling
estuarine chemistry of arsenic. Estuar. Coast. Mar. Sci. 7:
455-62.
Williams, S.C., H.J. Simpson, C.R. Olsen, and B.F. Bopp. 1978.
Sources of heavy metals in sediments of the Hudson River
estuary. Mar. Chem. 6:195-213.
Windom, H.L., and R.G. Smith. 1984. Factors influencing the
concentrations and distribution of trace metals in the South
Atlantic Bight. J. Geophys. Res. In press.
Windom, H.L., and FE. Taylor. 1979. The flux of mercury in the
South Atlantic Bight. Deep Sea Res. 26A: 283-92.
Windom, H.L., R.G. Smith, and M. Maeda. 1984. The geochem-
istry of lead in the South Atlantic Bight. Mar. Chem. (in rev.).
176
-------
NONPOINT SOURCE POLLUTION CONTROL IN SMALL BAYS OF
PUGET SOUND
BOB SAUNDERS
Shorelands Division
Washington Department of Ecology
Olympia, Washington
ABSTRACT
In the last 4 years, five commercial shellfish growing ar-
eas in Puget Sound have been closed because of non-
point bacterial contamination. These have been in rural
areas characterized by small acreage, semi-recreational
farms, rural residential development, and moderate resi-
dential density on the saltwater frontage. With one excep-
tion, they are notable for the absence of point discharges,
large commercial farming, and urban stormwater dis-
charges. The Department of Ecology conducted a year-
long study of the water quality in two of these estuaries—
Minter Bay and Burley Lagoon. The data correlated
stream segment pollution levels with surrounding land
use. Agricultural sources appeared to be the major prob-
lem, with failing septic tanks the suspect in some areas.
The Department funded the two counties in which the
watersheds occur to develop a basin plan for controlling
the problem. A consultant developed three ordinances to
address animal keeping practices, onsite waste disposal,
and erosion control. The issues are complicated by the
watershed's overlapping two counties with somewhat dif-
ferent sets of land use ordinances in place. Also, commu-
nity reaction to the initial proposals has been adverse.
The proposals are undergoing community review and
may undergo considerable revision. Other efforts are be-
ing funded to continue to develop farm management
plans on a voluntary basis, pending completion of a basin
planning program.
Shellfish in Washington State are important. Washington
is the fifth largest producer of oysters in the United States,
the eighth largest producer of clams, and the only pro-
ducer of the giant geoduck clams, which average 2 Ibs
apiece and can reach 10 Ibs. Puget Sound mussels have
been the winners for the last 2 years in national taste test
competitions. In addition to the commercial importance,
Puget Sound supports 441,000-user trips/year of recrea-
tional clam digging. The availability of freshly dug or pur-
chased shellfish is a significant feature of traditional Puget
Sound lifestyle.
Concern over the health of this resource began in 1982
after the third decertification of a commercial growing
area. Oyster growers' concerns and pressures led to the
initiation of a shellfish protection planning effort by the
Department of Ecology (WDOE), the State agency respon-
sible for water pollution laws and for shoreline manage-
ment. A year and a half later, when the agency's Shellfish
Protection Strategy was completed, the decertified areas
had grown to six and the problem was getting front-page
coverage in Sunday issues of the largest papers in the
State.
Based on these closures, four of which were due to
nonpoint sources, the Shellfish Protection Strategy identi-
fied nonpoint source pollution in watersheds draining to
areas with shellfish resources as the major problem, and
called for a program of basin planning to control the prob-
lem. The concept was to develop a pilot basin plan or
nonpoint pollution control program in one area and then to
promote the adoption and adaptation of this model in
other watersheds.
The pilot program area chosen was two small lagoons
called Minter Bay and Burley Lagoon. Both are classical
lagoons partially enclosed by a sand spit formed by littoral
accretion across their mouths. Burley Lagoon is 92 ha
(230 acres) and Minter Bay is 32 ha (80 acres); both flush
fairly well.
The watersheds, about 4,000 ha (10,000 acres) each,
are characterized by rural residential uses. Small-scale
farming for pleasure and supplemental income are com-
mon; commercial-scale agriculture is infrequent. The larg-
est herd in Minter/Burley is a small dairy with 40 head.
Residential and agricultural uses tend to be concentrated
in the stream valleys with heavily forested hilly terrain
higher up the watershed. Population in the two water-
sheds is about 10,000. A small commercial node exists
next to Burley Lagoon. About 50 percent of the soils are
poorly drained clays derived from glacial till.
The poor soils, rural residential use, and small-scale
agriculture are typical of Puget Sound, although some
areas do have more commercial farming.
A three-pronged approach addressed the nonpoint
problem:
Water quality investigation. Using coastal zone man-
agement funds, the WDOE water quality investigation sec-
tion conducted a year-long evaluation of water quality in
the two estuaries to identify more clearly the sources of
contamination and to provide a basis for developing and
justifying a control program.
Farm management. The local conservation district was
funded to begin a program of identifying farms with animal
waste problems, to begin informational and educational
programs, and to develop farm management plans.
Planning and land use. Planning grants were awarded
to the two counties in which the watersheds occur to de-
velop a pollution control program. A respected consulting
firm was retained to prepare an evaluation of alternative
strategies and ordinances that would institute appropriate
controls.
These efforts were coordinated through a technical ad-
visory committee, which had been previously established
by Pierce County to develop ordinances to control the
problem. After some initial tightening of the on-site waste
disposal regulations, the committee had begun to lose
momentum and focus. The use of this area as a pilot study
was intended to strengthen the committee's performance
of their original mission.
The results of this three-part program follow. Conclu-
sions rather than methodology are emphasized for brevity.
WATER QUALITY SURVEY
The main features of the methodology were
1. bimonthly ambient sampling at 20 stations
2. two rain event samplings
3. correlation of bacterial loads, loads, and land uses
along various reaches of the stream
4. special studies on time of travel, sediment, seepage,
and seabird populations
177
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
The major conclusions from the extensive data collected
were that all land uses throughout the entire watershed
were contributing to the problem, in particular:
1. Stream segments with only agricultural use showed
high loading.
2. Stream segments with only residential use showed
high coliform loading.
3. Undeveloped control streams met the water quality
standards.
4. Conforms could easily survive in the fresh water for 3
days, long enough to reach the estuary from the farthest
headwaters.
5. Shoreline residences could not account for the estu-
ary loads.
6. In Burley Lagoon, correlation between estuarine
conditions and stream loading was very high. In Minter
Bay it was not as high. In Burley as little as Vio of an inch
of rain could cause violations in estuarine water quality.
7. Sediments appeared in some areas to act as a reser-
voir of bacteria. Disturbance of sediments produced large
increases in downstream counts. This was a possible fac-
tor in explaining lower correlations between stream load-
ing and estuarine loads in Minter Bay.
8. No correlation at all developed between coliform lev-
els and bird counts.
9. Rainfall events produced rapid increases of 6-10
times the typically observed loads during ambient sam-
pling. Investigators concluded that we learned most from
sampling while the system was rainfall-stressed and that
future investigations should de-emphasize ambient sam-
pling.
The second part of the program inventoried and identified
farm ownerships. About 30 farm management plans were
developed. A handbook was developed describing agri-
cultural best management practices (BMP's) that were
most applicable to small farms. An unplanned, but signifi-
cant, followup to this phase of the work came when
WDOE secured a construction crew funded by a State
jobs program that provided free (to the homeowner) labor
for building improvements called for in the farm plans.
Some 2,400 m (8,000 ft) of fences were built, plus a num-
ber of bridges and stock watering areas.
The local governments hired a respected consulting firm
to evaluate alternative approaches to controlling the prob-
lem. Based on the WDOE water quality study, the consult-
ants recommended concentrating on animal waste man-
agement practices, failing on-site waste systems, and
erosion control. Since the area is quite rural, they recom-
mended controlling these primary sources rather than ad-
dressing collected storm water. In one area of Puget
Sound, urban storm water drains to a commercial shellfish
culture area, and typically high bacteria counts (900/
100 ml) have been found. Most shellfish cultures, how-
ever, occur in rural areas where infiltration is still high and
the recommended approach is to keep densities low to
avoid creating more serious stormwater problems.
The report recommended developing ordinances to
control these three activities. The ordinance approach
seemed necessary to ensure the long-run protection of
the area. Only 20 to 30 percent of the watersheds were
developed, so ignoring new development could quickly
undo current corrections. The local governments had pre-
viously added political support by directing the staffs to
develop ordinances to control the problems. This ap-
proach also suited the State interest in developing a pro-
gram that was integrated into local land use controls. Be-
cause of the small scale nature of the farms, it has not
been considered possible for direct state programs to ef-
fectively address the problem.
The recommended ordinances contained the following
features:
Fairm management Each farm in the watershed would
be required to have a farm management plan. The plan, to
be prepared by conservation district or Soil Conservation
Service (SCSj personnel, would conform to general poli-
cies, but considerable site-specific flexibility in .the applica-
tion of BMP's would be allowed. Farm plans were to be
submitted to the Health Department as a condition of ob-
taining health or building permits.
On-SKa Waste Management This ordinance was mod-
eled after a California county (Merced) ordinance. It re-
quired inspections of on-site systems periodically (2-year
intervals were proposed), annual permits to help fund the
system, and mandatory pumping if inspections failed.
Clearing and Grading. This ordinance set up a fairly
standard permit system to authorize clearing and grading
activities that result in excavation or fill in excess of a
minimum amount. In Washington such permits are com-
mon in cities, but rare in counties.
The development of these ordinances has not been
completed. While the technical committee was still en-
gaged in developing a system for funding and implement-
ing these programs, draft ordinances began to circulate in
the local community. Opposition to the proposals grew
very rapidly, culminating in a public meeting where 350
residents demanded a halt to the process.
Citizens expressed considerable resistance to a regula-
tory approach, to permit fees, and to various specific pro-
visions of the draft proposals. There was refusal to admit a
serious problem existed, demands to know the actual
health risks, demands to let the oystermen go somewhere
else, queries about depurating the oysters, and considera-
ble finger pointing between farmers and residential users
regarding who was most at fault. Despite a good data
base, the political heat derailed the original proposals and
resulted in a more lengthy and formal citizen advisory
committee process being developed.
Although the ordinances were sidetracked, the results
were not all negative. The controversy produced far higher
awareness and interest than previous educational meet-
ings. After the initial stormy meeting, large numbers of
people began to show up at the conservation district office
seeking farm plans. Cooler heads have generally been
appointed to committees and a program will likely be de-
veloped that is less regulatory and more assistance ori-
ented.
The general publicity and interest in Puget Sound water
quality and shellfish contamination also spawned a num-
ber of good bills in the legislature. One of these, HB 1068
provided for a comprehensive approach for planning to
control nonpoint pollution. The bill may not survive the
dual problems of a State budget crunch and some local
government resistance, but I would like to conclude this
story with a brief description of it because it embodies the
type of program that is needed to reverse the trend to-
wards decertifications.
The bill required a cooperative State/local effort to ad-
dress shellfish contamination. WDOE was directed to
identify "closed correctable" and "highly threatened"
commercial and recreational shellfish culture areas.
WDOE was also to prepare minimum standards for land-
use based nonpoint control programs. Local government
was given a year to prepare local plans to control the
sources of the pollution. The local plans would have to
conform to the minimum State standards and would re-
178
-------
quire WDOE approval before any implementation money
would be released.
The bill provided funds for the planning and implemen-
tation for both State and local governments. Significantly,
it also provided State funds for a farm assistance program
parallel to the Agricultural Soil and Conservation Service
program and funds for State Conservation Corps crews to
construct such improvements. The cost for these pro-
grams in the estimated 20 watersheds was projected at $8
million.
This program was modeled to some extent after the
State Coastal Zone Management Program operating
through a similar mechanism of locally developed and ad-
ministered but State approved programs. The bill ad-
dressed the critically important interrelationship between
ESTUARINE QUALITY
local land-use controls and nonpoint pollution and pro-
vided essential resources to develop and implement a pro-
gram. Without such legislative support and direction for a
serious nonpoint pollution control program, the shellfish
problems in Washington will continue to worsen. Even
with such a program, the task is formidable.
REFERENCES
Determan et al. 1984. Sources Affecting Bacteriological Condi-
tions of Water and Shellfish From Winter Bay and Burley La-
goon. No. 84-10. Washington Dep. Ecology, Olympia.
Saunders, B. 1984. Shellfish Protection Strategy. No. 84-4.
Washington Dep. Ecology, Olympia.
179
-------
SHELLFISH SANITATION IN OREGON: CAN IT BE ACHIEVED
THROUGH POLLUTION SOURCE MANAGEMENT?
JOHN E. JACKSON
Oregon Department of Environmei tal Quality
Portland, Oregon
ABSTRACT
Historically, shellfish growing areas are closed as man's
activity pollutes the waters. As these areas close, busi-
nesses and jobs are lost in a local and State economy.
Oregon is taking a different tack to maintain the limited
growing areas available to private industry. Recently com-
pleted fecal waste source management plans in Tilla-
mook Bay demonstrate that safe shellfish harvesting can
exist in the same estuary as nonpoint and point source
discharges—as long as who, what, and when they dis-
charge is known. An overview is presented describing the
process of pollution source identification, management
option determinations, and management plan develop-
ment and implementation.
INTRODUCTION
In recent years, much concern has been directed toward
the water quality conditions of Oregon's estuaries and, in
particular, those bays that receive pressure from commer-
cial and recreational shellfish harvesters. Routine sam-
pling and analysis of these bay waters, in some cases,
show seasonally degraded water quality. In these bays,
safe shellfish harvesting is precarious.
In the United States, when bays become contaminated
by pollution, they are closed to further harvesting of shell-
fish. Closing a bay hurts the local and State economy. The
commercial and recreational harvesters must go else-
where for their shellfish, thus affecting the local economy
This is an unacceptable solution in Oregon.
In the bays threatened by closure due to pollution, Ore-
gon is striving to accomplish a cleanup by achieving the
stated goals of the Federal Water Pollution Control Act
and the National Shellfish Sanitation Program (NSSP). A
goal of the Federal Water Pollution Control Act states that
"... wherever attainable, an interim goal of water quality
which provides for the protection and propagation of fish,
shellfish, and wildlife and provides for recreation in and on
the water be achieved by July 1,1983."
The NSSP goals are "(1) the continued safe use of this
natural resource and (2) active encouragement of water
quality programs which will preserve all possible coastal
areas for this beneficial use." The natural resource re-
ferred to by the NSSP goals is shellfish: "Shellfish are a
renewable, manageable natural resource of significant ec-
onomical value to many coastal communities, and which
should be managed as carefully as are other natural re-
sources such as forests, water and agricultural lands."
In Oregon, shellfish propagation and harvesting come
under the headings of "Resident Fish and Aquatic Life"
and "Fishing" and are considered beneficial uses as
stated in Oregon Administrative Rules 340-41-205 (Table
1). Oregon sets water quality standards to protect these
nonprioritized beneficial uses of the water. One specific
standard stated in OAR 340-41-205 is: "Bacterial pollution
or other conditions deleterious to waters used for domes-
tic purposes, livestock watering, irrigation, bathing, or
shellfish propagation, or otherwise injurious to public
health shall not be allowed."
The goals of the Clean Water Act and the National
Shellfish Sanitation Program, coupled with Oregon's re-
quirements to protect beneficial uses through abiding by
water quality standards, provide that, if a water quality
problem affects the shellfish resources, then action must
be taken to correct the problem. In meeting the goals and
correcting the pollution problem, Oregon is keeping bays
open to shellfish harvesting, even when nonpoint and
point source discharges exist in the water basin. This is
accomplished by managing the pollution sources. The
process is effective only when the sources, types, and
frequency of pollution discharge are known.
Tillamook Bay and Coos Bay, two separate estuaries in
Oregon, have been threatened by closures to shellfish
harvesting in the past. This paper describes the process
of pollution source identification, source management op-
tions and plan development, and, the 4 years of success-
ful management in keeping these bays open for safe shell-
fish harvesting. For the sake of clarity, only the Tillamook
Table 1.—Recognized beneficial uses of Tillamook Bay and tributaries.
Public domestic water supply
Private domestic water supply
Industrial water supply
Irrigation
Livestock watering
Anadromous fish passage
Salmonid fish rearing
Salmonid fish spawning
Resident fish & aquatic life
Wildlife & hunting
Fishing
Boating
Water contact recreation
Aesthetic quality
Hydropower
Commercial navigation & transportation
Estuary and
adjacent marine
waters
X
X
X
X
X
X
X
X
X
X
X
Columbia River
mouth to RM 86
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
All other
streams and
tributaries thereto
X
X
X
X
X
X
X
X
X
X
X
X
X
X
180
-------
ESTUARINE QUALITY
Bay effort conducted by the Oregon Department of Envi-
ronmental Quality (DEQ) will be discussed here.
THE TILLAMOOK BAY BACTERIA STUDY
In 1979, the Tillamook Bay Bacteria Study was initiated by
the DEQ to specifically identify the sources and extent of
fecal pollution occurring in the bay and its watershed.
From the study, it was proposed that corrective actions
would be developed to reduce the principal sources of
fecal contamination to acceptable levels, so as to elimi-
nate the potential health risk that occurs when a person
accidentally ingests the water while swimming or eats the
raw shellfish harvested from the bay.
The Tillamook Bay Bacteria Study consisted of five
parts: (1) review of the existing data and information, (2)
field investigations to fill gaps in knowledge, (3) definition
of the problem and identification of the pollution sources,
(4) development of a waste-management plan to address
the sources and problem, and (5) adoption and implemen-
tation of the plan.
Throughout the study, the local citizens were kept in-
formed of its progress. Not only was information dissemi-
nated to them, but the meetings, phone calls, and per-
sonal contacts made by the study team were instrumental
in involving the public in the process. If the cleanup effort
was to work, the local citizens had to make their concerns
known, and these concerns had to be incorporated in the
plan. Once the management plan was ready for imple-
mentation, the people knew what had to be done. Thus,
implementation was made more effective and less contro-
versial.
Tillamook Bay and Drainage Basin
The Tillamook Bay drainage basin is located on the north-
ern Oregon coast in Tillamook County, approximately
77 km south of the Columbia River mouth and 96 km west
of Portland. The watershed is 1,425 km2. It id bounded on
the east by the crest of the Coast Mountain Range and on
the west by the Pacific Ocean. Five major river subbasins
drain 97 percent of the total land area into Tillamook Bay.
Ninety percent of the basin is steep mountainous for-
ested terrain. The forested lands are owned and managed
separately by State, Federal, private, county, and munici-
pal agencies in descending order of total ownership.
Eight percent of the land area is devoted to agriculture,
primarily dairy farming. Located in this lowland area are
120 dairies. Total cow population is approximately 19,100
producing 256,360 tonnes of manure annually. The largest
and smallest dairy herd number 400 and 60 cows, respec-
tively. The average dairy holds 150 cows on 40.5 ha.
The population pattern is basically rural. People live
primarily on the alluvial plain and terraces adjoining the
bay. They are found in the towns of Tillamook, Bay City,
Garibaldi, and the unincorporated area of Idaville. Very
little shoreline development has occurred on the bay. How-
ever, many homes line the rivers and small tributaries in-
land. Total permanent population in the Tillamook Bay Ba-
sin for 1980 was 11,305. Recreational population having
residences in the basin adds another 1,700 people to the
total. Approximately 60 percent of the population is served
by three separate sewage collection and treatment facili-
ties. Two additional sewage treatment facilities serve the
industrial areas of the Tillamook airport and the Tillamook
Cheese Factory. All facilities discharge in the basin, with
four discharging directly to the bay or tidal reaches of the
rivers.
The area's climate is characterized by a strong marine
influence, with 70 percent of its precipitation recorded dur-
ing November through March. Winter storms often pro-
duce large amounts of precipitation over short periods of
time, and cause sudden water-level changes in the rivers
and occasional flooding of lowlands. The average annual
rainfall is 229 cm along the coast and 381 cm inland to the
north-central watershed.
Tillamook Bay covers an area of 36 km2 at high tide and
18 km2 at low tide, is 9.6 km long north to south, with a
maximum width of 4.8 km, and acts as a catch basin for
five rivers. The bay is shallow, averaging 1.8 m deep at
high tide. At extreme low tide, the bay water is confined
mostly to the narrow channels.
Shellfishing in Tillamook Bay includes recreational and
commercial clamming, and commercial oyster harvesting.
Clamming occurs throughout the bay. The commercially
grown and harvested Pacific oyster is grown on approxi-
mately 11 km2 of intertidal lands in the middle to upper
bay, using ground-culture methods. Annual harvest ap-
proaches 600,000 clams and 79,546 kg of oysters.
The bay and tributaries also support a good fin fishery
for salmonid fish species. When the fish are migrating, it is
not uncommon to see 50-100 boats in the bay and hun-
dreds of fishermen lining the rivers.
Because of the close proximity of the bay to the metro-
politan area of Portland and the bay's location on the pop-
ular north Oregon coast, the area receives many tourists
during holidays and the tourist season. The attractions are
the aesthetic qualities, camping, biking, fishing, and the
Tillamook Cheese Factory.
Review of Existing Data and Information
A review of past studies, current water quality information,
and discussion with local citizens determined that the ma-
jor problem in the bay was that high fecal coliform bacteria
levels occur during rain events. This indicated that it was
fecal contamination that threatened safe harvest of the
bay's shellfish. The review also concluded that sources of
the contamination had not been accurately identified. Pos-
sible sources included the sewage treatment plants, dairy
wastes, and failing septic tanks.
One certainty in this phase of the bacteria study was
that the people living in the cities thought that the dairies
caused the pollution; the farmers thought the cause was
the sewage treatment plants; and the tourists attributed
the pollution to the seals in the bay or the elk herds and
swimmers in the forested streams.
Field Investigations
Based on the review of existing data and information, the
project collected additional water quality data from
streams and from Tillamook Bay during differing weather
conditions based on rain intensity, ground saturation con-
ditions, and predicted fecal bacteria source discharges.
Four different types of weather situations were selected:
(1) heavy rain on saturated ground, (2) a rain after a period
of dry weather, (3) a dry-weather, low river flow summer
period, and (4) the first freshet storm of the water year.
Because of the lack of definitive information on the loca-
tion of fecal source contributions and the confusion over
the major contributors of the contamination, many fecal
source types had to be considered. To ease the anxiety of
the local citizens that they would be identified as a source,
the fecal sources considered in this phase of the .study
were labeled "potential fecal sources."
Potential fecal source types considered in the sample
design were dairy barnyards, dairy waste disposal meth-
ods on pastures, failing (or lack of) septic tanks, sewage
treatment plants, elk herds, heavy outdoor recreational
use areas, forestry activities, and seals.
Sample site selection was based on: (1) paired water-
sheds; (2) changes in land use; (3) a small watershed
181
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Tillamook River
3/10/801400 Hour to
3/13/801600 Hour
Wilson River
D*e«nb*r 1979
SH»W13*
g
FKal Collfann Wittr \Qiillt» Standard
HT
—I—
T» T» T« T4
River Mile Station Location
Figure 1.—Tillamook River and tributary fecal collform con-
centrations by river mile. RM = River Mile; FA = Forest-
Agriculture common boundary.
having only one or two land uses, such as forestry, for-
estry-agriculture, or forestry-urban; (4) previous sample
stations; (5) potential fecal source locations; and (6) loca-
tion of shellfish growing areas. Seventy-one tributary sites
and 14 bay sites were sampled during each storm. Tribu-
taries were sampled every 8 hours, the bay was sampled
during daylight high and low tides, and oysters during the
low tides.
The water and oyster meat samples were analyzed for
total and fecal conform bacteria. Some selected sites were
also sampled for fecal streptococcus to be used in the
fecal coliform/streptococcus ratio determinations. All anal-
yses used approved standard methods.
The analyses of the data consisted of comparing each
station's data for each storm event against the established
bacteria standard for that type of water. A plot of log mean
fecal concentration versus river mile (Fig. 1) and a plot of
fecal bacteria concentration (organisms/100 mL) versus
time (Fig. 2) were made for each storm and for each sam-
ple station.
Fecal bacteria loading of the bay was determined by
calculating the area under the curve for bacteria concen-
tration and river discharges. Bay loadings were also calcu-
lated using bacteria median values obtained for the sam-
ple period at the farthest downstream sample site in each
major river basin.
Sources contaminating the tributaries and bay under a
given weather condition were identified based on similar
watershed comparisons, the land use immediately sur-
rounding and upstream of each sample site, the magni-
tude of fecal bacteria contributions, and the response
characteristics of each bacteria source type.
To determine the relative impact of each source type on
the shellfish meat bacteria quality, it was necessary to
know the travel speed and circulation pattern of the fresh-
water in the bay. A rhodamine B dye dispersion study of
freshwater entering the upper end of Tillamook Bay was
done by U.S. Food and Drug Administration (FDA), DEQ,
and Oregon State Health Division staff.
Bay circulation patterns were also photographed. The
Oregon National Guard provided thermal infrared photo-
graphs taken approximately 1 hour before an evening low
tide. Contact prints were produced from the film and
s
2/MOO 2/2*00
Time
Figure 2.—Comparison of Wilson River fecal coliform con-
centrations and river flows over time. S = Start of storm; E
= End of storm.
pieced together to make a mosaic of the bay. These pho-
tos and the FDA work were compared and used to deter-
mine the flow pattern of the freshwaters entering the bay.
Pollution Source Identification and
Water-Quality Impacts
Because of the complex interaction of fecal source types
and the five major watersheds discharging into the bay,
the results of the water sampling and analyses were dis-
played in two ways: (1) the impacts of each watershed, its
loading of the bay and the circulation of freshwater in the
bay from each watershed; and (2) the impacts of each
fecal source type (dairy waste, septic tanks, etc.) on the
watershed into which it drains.
Conclusions from the data for the rivers and bay indi-
cated that a potential mechanism for waterborne disease
transmittal by shellfish from animals to man and man to
man exists in the Tillamook Bay Drainage Basin and Tilla-
mook Bay.
Furthermore, bacterial quality of Tillamook Bay as mea-
sured by fecal coliform levels is more degraded shortly
after heavy rains begin. During the winter months, these
rains usually produce turbidity, low salinities, and low tem-
peratures in the bay waters, thus creating suboptimal
feeding conditions for the oysters which, in turn, reduce
the potential of harvesting contaminated oysters. During
the summer rains, the optimum feeding conditions persist
but with a lesser degree of bacterial degradation to water
quality The data suggest, but with minimal confidence
(more oyster meat samples are needed), that the summer
rains may produce fecal bacteria conditions in the bay
water more critical for safe shellfish harvesting than rains
during the winter months.
It was found that most of the fecal coliform bacteria
recovered in the bay originated from dairy animal and
human fecal sources in the river subbasins. The waters
from the Wilson, Trask, and Tillamook Rivers flow over the
oyster beds in the bay on the ebb tide. Waters from the
182
-------
ESTUARINE QUALITY
Miami and possibly the Kilchis River reach the same beds
on a flood tide, but are somewhat diluted by fresh seawa-
ter. The clam beds located throughout the bay have water
from one or more rivers flowing over them during parts of
each tidal cycle.
Finally, small streams in the near bay area also carry
fecal bacteria, but because of their small discharges rela-
tive to the large rivers, they have negligible impact on the
bay
Conclusions from the data for each fecal source type
indicated that:
• Sewage treatment plants have the potential, when
they malfunction, for contaminating the surface waters of
the bay drainage basin, in addition to directly contaminat-
ing the bay None malfunctioned during the study.
• Dairy operations, primarily manure storage and dis-
posal in the barnyards and on the pastures, are contami-
nating the surface waters of the drainage basin with ma-
nure runoff when it rains, or the manure is inadvertently
applied directly to ditches and streams when being spread
on the pastures.
• Inadequate on-site subsurface sewage disposal sys-
tems is also contaminating surface waters of the drainage
basin when it rains, or the lack of such systems is contami-
nating the streams regardless of weather conditions.
• Other fecal sources, such as wild animals, recreation,
forestry activities, and industry, are hot significant contrib-
utors to the fecal contamination of Tillamook Bay and its
drainage basin. It is recognized that a local impact to the
environment could occur near one of these sources if it
should discharge fecal bacteria.
What was known at this point in the study was: (1) how
the bay and rivers interacted hydraulically; (2) who, how,
and when the fecal sources contaminated the surface wa-
ters; and (3) under what weather and runoff conditions the
shellfish in the bay become contaminated. This knowl-
edge formed the basis for developing management op-
tions to control the pollution problems identified.
Development of the Fecal Wastes
Management Plan
Throughout the study, local citizens were actively in-
volved. A group of interested citizens met regularly to re-
view the data collected and analyzed by the DEQ. They
experienced the same accomplishments, defeats, and
frustrations as the study team when arriving at the conclu-
sions from the data. These same citizens developed the
management options to control the problem. At this point,
because of their involvement in the data-analysis phase,
the people were better equipped to suggest solutions to
the problems. The DEQ's role in this effort was to ensure
that the management options addressed the problems.
Dairymen developed the solutions to the dairy problems.
County sanitarians developed control strategies for the
septic tank problems. Sewage treatment plant owners and
operators developed the strategy for minimizing impacts
from their'plants.
Management options that were considered in address-
ing the pollution problem were: (1) closing the bay to har-
vesting of shellfish allowing status quo correction of the
pollution problems from the fecal sources; (2) initiating
new types of corrective actions aimed at reducing the pol-
lution potential of the identified fecal sources and develop-
ing closing-opening criteria for the bay; (3) strengthening
of existing programs responsible for the fecal source types
identified and developing closing-opening criteria for the
bay
The local people wanted an effective plan that would
avoid extensive implementation costs. They knew that the
plan had to reduce the pollution potential in the most eco-
nomical manner feasible.
The last option (strengthening of existing programs and
developing closing-opening criteria) was chosen, primarily
because no new programs had to be instituted for the
sewage treatment plant and subsurface sewage disposal
systems. In agriculture, the industry had made the deci-
sion to develop a pollution abatement plan in Tillamook
County Local citizens did not want to duplicate agency
efforts, but they did want to find a way to make existing
programs more effective, particularly since the mecha-
nisms, effective or not, for correcting the pollution prob-
lems were already in place and operating. To accomplish
this, no additional funds or personnel were needed.
Developing a bay closing-opening criteria, along with
strengthened source control programs, was deemed nec-
essary to ensure safe shellfish harvesting in the interim
while the fecal sources were being corrected. Application
of the closing-opening criteria did not hurt the shellfish
industry, because the industry already had self-imposed,
limited harvesting during critical runoff periods.
The Tillamook Bay Drainage Basin Fecal Wastes Man-
agement Plan is divided into three parts. Each part ad-
dresses a fecal source type, its location, and timing of
discharge from that source. Each part is independent of
the other parts in strategy and implementation schedules.
The whole plan recognizes the legal responsibilities of
each fecal source to eliminate their discharge, but accom-
plishes the cleanup in such a way that it does not force
permanent closure of an activity.
For the problem of malfunctioning sewage treatment
plants, a malfunction notification procedure was devel-
oped. This required that additional alarm and shutdown
equipment be installed in the plants and collection sys-
tems. It also required plant personnel to notify health and
environmental officials immediately in the case of a mal-
function.
For the on-site subsurface sewage disposal problems,
the plan identified critical problem areas and assigned
priorities for correction. This document is used by the
county to prioritize work and for budget preparation.
Dairy waste was found to be the most pervasive prob-
lem in the basin and a continuous source of fecal material
to the bay. Since this was the case, the local agricultural
community, with financial assistance from Section 208
funds from the U.S. Environmental Protection Agency, de-
veloped an extensive plan for the basin. The Tillamook
Bay Drainage Basin Agricultural Nonpoint Source Pollu-
tion Abatement Plan cleanup strategy simply stated is: (1)
Keep clean rainwater and surface waters from coming into
contact with manure and if that is not possible; (2) prevent
the contaminated water from entering streams, rivers and
ditches by intercepting, storing and disposing of the ma-
nure contaminated water in a sanitary manner.
The plan developed from this strategy directs each
farmer to develop individual farm water quality plans.
Each plan addresses the water quality problems of that
farm and displays a 3-15 year schedule for implementa-
tion of best management practices specifically designed
for the farm, so that the practices fit the established farm
management scheme.
Since we recognize the long-term nature of the cleanup
and the need for immediate action to safeguard public
health, tradeoffs between management of fecal sources
and harvesting of shellfish had to occur immediately To
this end, a bay closing and opening procedure was
adopted, based on criteria developed from the study of the
interaction of fecal sources, river to bay hydraulics, and
oyster meat bacterial quality
The procedure dictates temporary bay closures for sew-
age treatment malfunctions, for first and second major
183
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
rainfall events in the fall, for winter storms that cause rapid
increases in river flows and the resultant possible flood-
ing, and for summer storms that cause moderate levels
but a rapid rise in river flows.
Fecal Wastes Management Plan
Implementation
'The plan and bay closure criteria were adopted by local
and State agencies and organizations in July 1981. Today,
the spring of 1985, implementation of the Plan continues
to be successful. This success can be attributed to at least
two factors: (1) the involvement of local citizens through-
out all phases of the study fostered local pride in the ac-
complishments and, more important, fostered a pride in
the liability of the Tillamook area; and (2) funding to ac-
complish the implementation of the plan. Whether it is
appropriate to recognize the fact, money still solves prob-
lems. A change in public attitude toward a problem can
accomplish a lot, but in many cases, money is needed to
effect such a change in attitude. For example, on a farm,
the attitude change that manure is an asset to be applied
to pastures as fertilizer rather than a liability to be piled
next to the creek and washed away, accomplishes a lot
toward achieving a cleanup. Yet without storage areas in
the barnyard away from the rain, the farmer has trans-
ferred the problem away from the stream and has placed
the problem where a ditch now transports the accumulat-
ing rainwater and manure to the stream. It takes longer to
get to the bay this way, but the problem has not been
solved without roofed or curbed storage areas.
What has been accomplished to date? The sewage
treatment facilities now have alarms and shutdown de-
vices that operate when critical equipment malfunctions.
Each plant has, as a part of its discharge permit, the
requirement to initiate the malfunction notification proce-
dure.
The problem areas for septic tanks are still under inves-
tigation, with corrective measures being instituted. One
severe problem area previously identified has now had
sewers installed.
As for the 120 dairies, the Tillamook Soil and Water
Conservation District has received more than $3 million to
assist dairy owners in cleaning up priority problem dairy
farms. The farmers have also committed more than $1.8
million of their own resources to th«j cleanup effort. This
work is being carried out on over 70 farms.
What is happening to the water quality trends in the
rivers and bay with all this activity on the land? Preliminary
indications are that an improving trend in bacterial water
quality is occurring in the rivers. An unequivocal state-
ment of water quality trend cannot be made at this time,
however.
SUMMARY
Can shellfish sanitation be achieved in Oregon through
pollution source management? Yes. Tillamook Bay, dis-
cussed in this paper, and Coos Bay, mentioned here but
not discussed, are both important estuaries for the shell-
fish and recreational industries in Oregon. Neither has
been closed permanently to shellfishing even though both
receive wastes from other industries and at one time had
poor shellfish sanitation characteristics. With regard to Til-
lamook, industries and dairy farming are still open for
business. People still live in the watersheds to the bay, and
they still flush their toilets. What has changed to improve
the water quality situation?
The most important factor is that people's attitudes to-
ward how they live have changed. People now realize that
how they handle wastes in their homes and businesses
will have an effect on some other person's business, liveli-
hood, and recreation.
Along with this attitude change have come tools to help
people prevent and control water pollution. A new area of
the city of Tillamook has been sewered. There is a red
light in a police station to alert someone that the sewage
treatment plant is not working in the middle of the night.
Concrete tanks now store manure and curbs around the
dairy barnyards control manure runoff, where once bare,
sloping ground was covered with piles of manure in the
rain.
With knowledge of who, how, and when sources of pol-
lution operate and discharge in a watershed and bay, point
and nonpoint source discharges can coexist with the shell-
fish industry. This can occur to the point that no industry or
person is hurt—least of all the public that loves shellfish.
184
-------
Streams and Rivers
MONITORING CHANGES IN AGRICULTURAL RUNOFF QUALITY IN
THE LAPLATTE RIVER WATERSHED, VERMONT
DONALD W. MEALS
Vermont Water Resources Research Center
School of Natural Resources
University of Vermont
Burlington, Vermont
ABSTRACT
The LaPlatte River Watershed in northwestern Vermont is
the focus of an intensive program of land treatment to
control agricultural runoff. Best Management Practices
for controlling dairy manure and cropland erosion have
been implemented by the USDA-SCS on 90 percent of
the priority areas in the watershed. A long-term monitor-
ing program is being conducted to evaluate the effective-
ness of BMP application in improving water quality. The
monitoring program includes precipitation and stream
discharge recording and water sampling for suspended
solids, phosphorus, and nitrogen analysis. A concurrent
land use monitoring program is collecting information re-
quired to couple changes in agricultural practices with
changes in stream water quality. The water quality moni-
toring program is outlined. Application of several statisti-
cal trend analysis techniques to 5 years of record from
four watersheds is described and some results are dis-
cussed.
INTRODUCTION
Agricultural activities have long been identified as poten-
tial nonpoint sources of water pollution (Loehr, 1974;
Omernik, 1976). Best management practices are widely
believed to provide effective control of agricultural pollu-
tion sources (Loehr, et al., 1979). However, the effective-
ness of many BMP's has not been well documented on a
watershed scale (Baker and Johnson, 1983).
Dairy farming is a major industry in Vermont. In the late
1970s, Vermont's 208 planning process concluded that
excessive cropland erosion, lack of fall cover, and im-
proper animal waste management were significant agri-
cultural pollution sources to the State's waters (Vt. Agency
Environ. Conserv., 1978). At that time, the LaPlatte River
watershed was identified as a major source of sediment
and phosphorus to Shelburne Bay, a heavily used eu-
trophic bay in Lake Champlain (Vt. Agency Environ. Con-
serv., 1977; Soil Conserv. Serv., 1978).
In response to these problems, the LaPlatte River Wa-
tershed Plan was developed in 1979 by the Soil Conserva-
tion Service, in cooperation with the Winooski Natural Re-
sources Conservation District, the Vermont Department of
Agriculture, the Vermont Agency of Environmental Con-
servation, and the University of Vermont (Soil Conserv.
Serv., 1979). The plan, prepared under the authority of
P.L. 83-566, has the objective of controlling water quality
problems by installing best management practices for ani-
mal waste management and conservation land treatment.
Implementation of the plan involves the preparation of
conservation plans and contracts with individual land-
owners. The BMP's are financed on a cost-share basis.
Associated with the plan is a long-term, comprehensive
water quality monitoring program. The primary objective
of the monitoring effort is to document changes in surface
water quality resulting from the BMP's.
This paper will describe the monitoring program and
briefly outline the application of two trend-analysis tech-
niques to several years of monitoring data.
STUDY AREA
The 13,815 ha LaPlatte River watershed lies just east of
Lake Champlain in northwestern Vermont (Fig. 1). About
50 percent of the watershed is devoted to agriculture, for-
ests cover 40 percent of the area, and residential areas
encompass 8 percent. There are 60 active farms, mostly
dairy, in the watershed, averaging 117 ha in size. Average
herd size is about 120.
Soils in the watershed have developed on lacustrine
deposits, glacial till, or bedrock of the Green Mountains.
185
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Over 44 percent of the watershed is underlain by lacus-
trine sands, silts, or clays, and most of the agricultural
activities are concentrated on these soils.
The climate of the LaPlatte River watershed is of the
cool, continental type. Mean annual temperature is 6.7°C;
normal annual precipitation for the watershed is 85.6 cm
(Nat. Oceanogr. Atmos. Admin., 1983). Average annual
snowfall is 179cm.
The 26-km LaPlatte River drains the watershed from
east to west to Shelburne Bay of Lake Champlain. There
are four major tributaries to the river and numerous inter-
mittent streams. One point source discharges to the La-
Platte River 8 km above its mouth: a 0.01 m3/sec aerated
lagoon wastewater treatment plant serving the village of
Hinesburg. The LaPlatte River flows through an extensive
forested wetland before emptying into Shelburne Bay.
METHODS
The 11-year monitoring program includes long-term rou-
tine stream monitoring, several short-term studies, and
intensive land use monitoring.
Long-term Monitoring
Four watersheds are being monitored on a long-term ba-
sis (Fig. 1). Watershed 1, drained by the LaPlatte River,
includes 80 percent of the entire watershed and is used to
document overall water quality changes (Table 1). Water-
shed 2 is highly agricultural and has received extensive
BMP implementation. Watershed 3 serves as a control;
agriculture is nonintensive and no BMP's are needed or
planned. Watershed 4 contains portions of three farms,
each of which has contracted for BMP implementation.
The effluent of the Hinesburg treatment plant, discharging
within watershed 1, is also monitored.
At each of the monitoring stations, stream stage is con-
tinuously recorded by ISCO Model 1870 bubbler-type
stage recorders. Stage records are later digitized, and dis-
charge is calculated from site-specific ratings. Watershed
precipitation is measured by a network of 20-cm weighing
bucket recording precipitation gauges.
Water samples are collected automatically by ISCO
Model 1680 samplers and stored in refrigeration units.
Samples are collected at 8-hour intervals and combined to
yield four 24-hour and one 72-hour composite samples
each week. During the spring snowmelt and some individ-
ual storm events, discrete samples are collected at more
frequent intervals, typically 2-8 hours apart.
Samples are routinely analyzed for turbidity, total and
volatile suspended solids, total phosphorus (persulfate di-
gestion, ascorbic acid), dissolved inorganic phosphorus
(ascorbic acid), total Kjeldahl nitrogen (macro digestion,
distillation), and ammonia nitrogen (distillation, nessleriza-
tion). Twice weekly, in situ measurements are made for
temperature, dissolved oxygen, and specific conduc-
tance; grab samples are taken for laboratory analyses of
pH, fecal coliform, and fecal streptococcus. All water qual-
ity analyses are conducted by accepted methods (Stand.
Methods, 1980; U.S. Environ. Prot. Agency, 1983).
Short-term Studies
Several 2-3-year studies have been conducted or are un-
derway to supplement routine monitoring. The attenuation
of phosphorus in the LaPlatte River below the Hinesburg
Table 1.—Characteristics of monitored watersheds, LaPlatte River Watershed Project.
Watershed 1 Watershed 2 Watersheds Watershed 4
Area (ha)
Land Use (%)
Agricultural
Forested
Other
11,395
42
39
19
1,682
74
20
6
165
77
0
23
182
77
0
23
onitoring Areas
&
Facilities
LAPLATTE RIVER WATERSHED
A [//] ATM monitored by
V. tl£J Trend Station
[ | Monitored tub-waterehed*
Figure 1.—Map of LaPlatte River Watershed, Vermont.
186
-------
STREAMS AND RIVERS
treatment plant has been investigated to determine the
relative impacts of point and nonpoint sources of phos-
phorus (Cassell and Arris, 1983). A baseline biological
inventory of fish, invertebrates, and periphyton has been
completed, and a postimplementation reinventory is
planned for the future (LaBar, 1982). A paired watershed
study comparing traditional and BMP manure manage-
ment on two cornfields has been completed (Broido,
1983). A verification study of the CREAMS model (Knisel,
1980) has been performed on the same fields (Jamieson,
1985).
Currently, studies of barnyard runoff and milkhouse
waste are underway. These studies will assess the magni-
tude of these sources and evaluate the treatment effec-
tiveness of several BMP's at the source.
Land-use Monitoring
To achieve the objectives of the project, changes in water
quality must be related to changes in land use. To this end,
land use and agricultural activities in the watershed are
being monitored on an intensive, field-by-field basis. Data
are being collected on activities such as time, location,
and magnitude of manure application, land tillage, and
crop harvesting. These data are being processed through
a geographic information system (GIS), a computer-based
mapping overlay system with highly specialized capabili-
ties for the analysis and display of spatial data.
RESULTS
BMP Implementation
Implementation of BMP's began in 1980 and is now nearly
complete. A summary of contract and implementation sta-
tus is shown in Table 2. All expected contracts have been
signed, bringing 2,851 ha of farmland under contract. The
26 manure storage structures completed can collectively
store 70 percent of the manure generated in the water-
shed. All projected conservation cropping systems and
most barnyard and milkhouse waste management sys-
tems have been contracted; some installation work on
these BMP's remains to be done. Over $700,000 of Fed-
eral cost-share funds have been paid to date, of which 80
percent has been applied to manure management BMP's.
Water Quality
Water quality in the monitored watersheds has been
highly variable. Concentrations of sediment and nutrients
tend to differ significantly among the watersheds and
show strong seasonal variation. Concentrations peak dur-
ing periods of high runoff, a pattern characteristic of non-
point source activity (Novotny and Chesters, 1981). An-
nual mean total phosphorus concentrations of 0.125 mg/L
in watershed 2 and 0.130 mg/L in watershed 4 are higher
than the average 0.082 mg/L of total phosphorus in
streams draining agricultural watersheds in the northeast
reported by Omernik (1976).
Mass export of sediment and nutrients from the moni-
tored watersheds has also been quite variable. As is typi-
cal of nonpoint runoff, 50-70 percent of the annual load
tends to be exported during the spring runoff period, usu-
ally in a few major storm/melt events. Significant export
also occurs during stormflow in the fall.
Ranges of annual area! export from two monitored wa-
tersheds in the LaPlatte River watershed are shown in
Table 3. Export of total P and dissolved inorganic P gener-
ally tends to be higher than loadings reported by Omernik
(1976) for agricultural watersheds in the northeast and
within the range observed in watershed studies in the
Great Lakes basin (PLUARG, 1978). Sediment export has
been generally very low, while nitrogen export has been
within the lower range of values reported elsewhere.
Thus, nutrient export, particularly phosphorus, appears
to be the primary nonpoint problem in the LaPlatte River
watershed, and while sediment loads appear relatively mi-
nor, soil loss from individual fields may be significant.
DISCUSSION
Detecting significant changes in water quality attributable
to land use changes on a watershed scale is a difficult
task. In a recent study, Persson et al. (1983) observed that
the incremental nature of BMP implementation, together
with climatic variation, tended to obscure obvious
changes in water quality. This study is no exception. Fur-
thermore, the absence of preproject water quality data
makes a relatively straightforward before-after compari-
son impossible. Thus, more complex techniques of trend
analysis are required.
Table 2.—Contract and Implementation summary, LaPlatte River Watershed Project.
Reporting year
1980
1981
1982
1983
1984
Totals to date
Projected
No. of
farms
4
16
6
1
0
27
27
Ha
656
1309
655
231
0
2851
2851
No. Of
manure
storage
0
9
11
5
1
28
26
Waste
utilization1
(ha)
0
641
543
372
184
1740
1740
•impiemeniei
No. of
barn-
yards
0
1
2
2
4
9
14
No. of
milk-
Houses
0
5
4
7
1
17
19
Conservation
cropping
(ha)
0
0
387
170
386
943
1001
'Area receiving manure according to BMP.
Table 3.—Range of annual areal export observed from two monitored watersheds compared with literature values.
Watershed Watershed Omernik PLUARG
Export 3 4 (1976) (1978)
Total suspended solids
Total phosphorus
Dissolved inorganic phosphorus
Total Kjeldahl nitrogen
Ammonia nitrogen
28-65
0.3-0.4
0.1-0.2
3.1-3.6
0.3-0.4
11-96
0.2-2.3
0.1-0.8
1 .4-20.4
0.1-2.3
—
0.2
0.1
6.31
3.7
3-5600
0.1-9.1
0.1-0.6
0.6-421
—
'Total nitrogen.
187
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Four approaches to preliminary trend analysis in the
LaPlatte River watershed have been employed. These
are: (1) least-squares fit over time, where the slope of a
regression line through the data may be suggestive of a
trend; (2) comparison of annual means, where a two-sam-
ple t-test is used to confirm significant differences be-
tween years; (3) comparison of frequency distributions;
and (4) paired watershed analysis. Complete results of
preliminary trend analysis are presented in the most re-
cent annual project report (Meals, 1984). The latter two
methods will be discussed in greater detail.
Frequency Distribution
The cumulative frequency distribution is useful for basic
data description and for assessing year-to-year changes
in water quality. Such distributions, where the proportion
of observations less than or equal to the value of each
observation is plotted as a percentage of the total number
of observations, may be used to evaluate extreme values,
such as peak concentrations during runoff periods. Fur-
thermore, the probability of exceeding some specified
level in a given year can be easily derived from the distri-
bution.
Frequency distributions of concentration, discharge,
and export data are generated for each year. Figure 2
shows a set of cumulative frequency plots of weekly mean
log total Kjeldahl nitrogen (TKN) concentration data
grouped by project year. Shifts in the plots between years
can indicate trends in the data; differences between the
distributions are tested for significance using the Kolmo-
gorov-Smirnov 2-Sample Test. In Figure 2, the curves
generally shift to the left (i.e., toward lower concentrations)
with succeeding years, suggesting a decreasing tendency
for TKN concentrations in watershed 1. Statistically signifi-
cant differences (P < 0.10) were confirmed between each
of the distributions except between years 4 and 5.
Extremes in the data may be evaluated by comparing
the probability of exceeding (Pe) a particular critical value
in succeeding years, where Pe is defined as 100 percent
minus the cumulative frequency percentage of the critical
value. Probabilities of exceeding particular critical values
are plotted for each year, and the shape of the curve may
indicate a trend. The critical values selected may be water
quality standards or some other specific concentration or
load. In this case, critical values have been arbitrarily set
as the long-term mean of values with a Pe of 5 percent in
the control watershed (watershed 3). It should be noted
that the general shape of the resulting curve is indepen-
dent of the particular critical values selected.
OJMUUnVE FSHXENCr OBIHBUHON
WATTRSKD 1 kCAM WEEXUf TKN CONCEKTOAI10N
YDWS2-5
UdATTE RIVER WATDGKD PROJECT
Ligend
& imn
X IU«J_
O TUH4
• mst.
-«J 04 OJ
l*g 1KN CONCODWmN (WVU
An example of such a plot is shown in Figure 3 for mean
weekly discharge, total suspended solids (TSS), total
phosphorus (TP), and TKN export from watershed 2.
Probabilities of exceeding all four parameters appear to
have increased significantly (P < 0.10) through year 4;
that is, the likelihood of exceeding the critical values has
increased. Differences in Pe for discharge, and for TP and
TKN export between years 4 and 5 were not significant,
but the lack of significant increase may itself be important.
This approach is quite sensitive to climatic variation; the
increasing Pe values for export in Figure 3 parallel increas-
ing Pe for stream discharge.
Paired Watershed Analysis
This approach compares the behavior of two watersheds,
where one is assumed to be the control (untreated) and
the other the experimental (treated) (Hewlett and Pienaar,
1973). A linear regression is performed on paired data
when both watersheds are treated similarly (the calibration
period), with the control watershed serving as the inde-
pendent variable. Following treatment of the experimental
watershed, a second regression is performed. Significant
differences between the calibration and treatment regres-
sions indicate the treatment effects. The major advantage
of this technique is that climatic and hydrologic variability
is controlled by the inclusion of the control watershed.
For this analysis, watershed 3 is taken as the control
watershed (no BMP's) against the treated watersheds (ex-
tensive BMP's). Because of the lack of preimplementation
data, there is no clear calibration period; rather, succes-
sive years' paired regressions may be examined for
trends.
Three years of paired watershed regression lines for TP
export from watershed 2 and 3 are shown in Figure 4.
Each of the regression lines is statistically significant (P <
0.001). Testing for differences between lines by analysis of
variance of regression coefficients shows that the lines are
significantly different from each other (P < 0.01). From
Figure 4 it can be seen that, for a TP export of 100 kg/
week from the control watershed, the regressions predict
TP exports from watershed 2 of 600 kg/week in year 2,
315 kg/week in year 3, and 150 kg/week in year 5. Thus,
there is a tendency for decreasing TP export from water-
shed 2 relative to the control.
Net change between years, based on the paired water-
shed analysis, may be estimated by examining the devia-
tion of the current year's data from predictions derived
from earlier regressions (Hibbert, 1969; Hornbeck et al.
1970). For example, year 5 export values from the control
PROBAHLTTY OF EXCEEDANCE - DISCHARGE AND EXPORT
WATERSHED 2
LAPLATTE RIVER WATERSHED PROJECT
1
YEM
Figure 2.—Example of cumulative frequency plots grouped
by project year.
Figure 3.—Plot of exceedance probabilities for discharge
and export from watershed 2.
188
-------
watershed 3 are inserted into earlier regression equations
to determine "untreated" predictions of watershed 2 ex-
port; that is, predictions of what export would be in year 5
in the absence of change. These predicted values are
subtracted from observed year 5 watershed 2 export, and
the differences are summed to yield an estimate of net
change. Furthermore, when individual weekly differences
are plotted against time, the timing of deviations emerges.
Such a pattern may characterize seasonal BMP perform-
ance.
Some results of this procedure are shown in Table 4.
Export of most constituents appeared to increase, but
there was a net decrease in TSS and TP export from
watershed 2 between years 3 and 5. When weekly devia-
tions in TSS export are plotted (Fig. 5), it is evident that
positive deviations (export higher than predicted) tended
to occur during warm weather storm periods, while nega-
tive deviations (export lower than predicted) occurred
mainly during the spring runoff period. Deviations are sig-
nificant if they exceed the 90 percent confidence intervals
placed around the calibration regressions (Cl = ±1.4 kg/
week in Fig. 5).
CONCLUSIONS
Agricultural nonpoint sources contribute significant quan-
tities of nutrients to the surface waters of the LaPlatte
River watershed. Extensive application of BMP's through-
out the watershed, focusing on animal waste manage-
ment, began in 1980 and is now nearly complete. A de-
tailed water quality data base has been established
through a comprehensive monitoring program and will
continue through 1991. Variability in climate, streamflow,
and water quality, and the lack of a preimplementation
data base have obscured clear changes in water quality
resulting from BMP implementation. However, preliminary
trend analysis of monitoring data suggest that sediment
"fable 4.—Net difference: observed minus predicted export
from watershed 2, LaPlatte River Watershed Project.
Parameter
Year
3 vs. 4
Year
4 vs. 5
Year
3 vs. 5
TSS export
TP export
TKN export
. kg/ha/yr.
-409355 +346365 -118936
+ 33 +312 -59
+ 1013 +1877 +1055
+ - > Predicted
- - < Predicted
TOOL PHOSPHORUS EXPORT
WOEKSKD 3 v> WAJDGHD 2
YEARS 3AAND5
UFLOTE RMER WOTRSO PROJECT
A tnmi
X TIM*
vaaa& i TOW. p cxroRTl?VwtEK)
STREAMS AND RIVERS
and nutrient loads may be decreasing in some parts of the
watershed. Continued monitoring, more rigorous trend
analysis, and the results of an intensive land use monitor-
ing effort are expected to provide a more definitive assess-
ment of the effects of BMP's on water quality.
ACKNOWLEDGEMENTS: This study is being conducted by the
Vermont Water Resources Research Center. Funding for this
project is provided by the USDA-Soil Conservation Service and
the School of Natural Resources, University of Vermont. The
cooperative efforts of the Winooski Natural Resources Conser-
vation District, the Vermont Agency of Environmental Conserva-
tion, and the farmers of the LaPlatte River watershed are grate-
fully acknowledged.
REFERENCES
1980. Standard Methods for the Examination of Water and
Wastewater. 1980. 15th ed. Am. Pub. Health Ass., Washing-
ton, DC.
Baker, J.L., and M.R Johnson. 1983. Evaluating the effective-
ness of BMP's from field studies. Pages 281-304 in F.W.
Schaller and G.W. Bailey, eds. Agricultural Management and
Water Quality. Iowa State Univ. Press, Ames.
Broido, T.J. 1983. Calibration and characterization of three agri-
cultural field-sized watersheds. M.S. thesis, School Nat. Re-
sour. Univ. of Vermont, Burlington.
Cassell, E.A., and L.L. Arris. 1983. Attenuation and transport of
phosphorus and suspended solids in the LaPlatte River. Com-
ple. Rep. Vermont Water Resour. Res. Center, School Nat.
Resour. Univ. of Vermont, Burlington.
Hewlett, J.D., and L. Pienaar. 1973. Design and analysis of the
catchment experiment. Pages 88-106. in E. M. White, ed.
Proc. Symp. Use of Small Watersheds in Determining Effects
of Forest Land on Water Quality. Univ. of Kentucky, Lexington.
Hibbert, A.R. 1969. Water yield changes after converting a for-
ested catchment to grass. Water Reour. Res. 5(3): 634-40.
Hornbeck, J.W., R.S. Pierce, and C.A. Federer. 1970. Stream-
flow changes after forest clearing in New England. Water Re-
sour. Res 6(4): 1124-32.
Jamieson, C.A. 1985. Verification of the model CREAMS on two
agricultural fields in Vermont. M.S. thesis, School Nat. Re-
sour. Univ. of Vermont, Burlington.
Knisel, W.G., ed. 1980. CREAMS: A Field-scale Model for
Chemicals, Runoff, and Erosion from Agricultural Manage-
ment Systems. Conserv. Res. Rep. No. 26. U.S. Dep. Agric.,
Washington, DC.
LaBar, G.W. 1982. LaPlatte River fisheries/benthos evaluation
phase I: 1980-1981. Comple. Rep. Vermont Water Resour.
Res. Center, School Nat. Resour. Univ. of Vermont,
Burlington.
Loehr, R.C. 1974. Characteristics and comparative magnitude of
non-point sources. J. Water Pollut. Control Fed. 46(8): 1849-
72.
OBSERVED - PREDICTED TSS EXPORT
WATERSHED 2 YEARS
UPIATTE RIVER WATERSHED PROJECT
•CREASED EXPORT
OCCREA8ED EXPORT
Figure 4.—Paired-watershed regression lines for total phos- Figure 5.—Plot of weekly deviations of observed from pre
phorus export, watersheds 2 and 3. dieted total suspended solids export, watershed 2.
189
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Loehr, R.C., D.A. Haith, M.F. Walter, and C.S. Martin, eds. 1979.
Best Management Practices for Agriculture and Silviculture.
Ann Arbor Science, Ann Arbor, Ml.
Meals, D.W. 1984. LaPlatte River watershed water quality moni-
toring and analysis program, proj. year 5. Progr. Rep. No. 6.
Vermont Water Resour. Res. Center. Univ. of Vermont,
Burlington.
National Oceanographic and Atmospheric Administration. 1983.
Local climatological data for Burlington, Vermont. Nat. Cli-
matic Data Center, Ashville, NC.
Novotny, V, and G. Chesters. 1981. Handbook of Nonpoint Pol-
lution, Sources and Management. Van Nostrand Reinhold,
New York.
Omernik, J.M. 1976. The influence of land use on stream nutri-
ent levels. EPA-600/3-76-014. U.S. Environ. Prot. Agency, Cor-
vallis Environ. Res. Lab., Corvallis, OR.
Persson, L.A., J.O. Peterson, and F.W. Madison. 1983. Evalua-
tion of sediment and phosphorus management practices in
the White Clay Lake watershed. Water Res. Bull. 19(5): 753-
62!
Pollution from Land Use Activities Reference Group (PLUARG).
1978. Environmental Management Strategy for the Great
Lakes System. Final rep. Int. Joint Comm. Windsor, Ontario.
Soil Conservation Service. 1978. Erosion and sedimentation
nonpoint pollution sources and controls. LaPlatte River Water-
shed. LCBS-19. U.S. Dep. Agric. Natl. Tech. Inform. Serv.,
Springfield, VA.
:. 1979. Watershed Plan for LaPlatte River Watershed.
Burlington, VT.
U.S. Environmental Protection Agency. 1983. Methods for
Chemical Analysis of Water and Wastes. EPA-600/4-79-020.
Off. Res. Develop., Cincinnati, OH.
Vermont Agency of Environmental Conservation. 1977. Nutrient
loading to Shelburne Bay and St. Albans Bay, Lake Cham-
plain, Vermont 1975-1976. Montpelier, VT.
1978. A State Water Quality Plan for Controlling Agri-
cultural Pollution. Montpelier, VT.
190
-------
NONPOINT SOURCE POLLUTION IN THE RICE CREEK WATERSHED
DISTRICT—THE RESULTS OF 10 YEARS OF WATER QUALITY
MONITORING
PETER R. WILLENBRING
EUGENE A. HICKOK
WILLIAM D. WEIDENBACHER
E.A. Hickok and Associates
Wayzata, Minnesota •
ABSTRACT
A summary of nonpoint source water quality data gener-
ated by a monitoring program undertaken by the Rice
Creek Watershed District from 1974 to 1984 is presented.
The monitoring program focused on establishing the
characteristics of the nonpoint source runoff generated
from 21 subwatersheds that were delineated, within the
472.7 sq. km watershed district. For each of the subwa-
tersheds, area! loadings and flow-weighted mean con-
centrations of total Kjeldahl nitrogen, total phosphorus,
orthophosphorus, total suspended solids, and chlorides
are presented. Detailed land use and hydrologic informa-
tion for each subwatershed are also presented and ana-
lyzed along with the water quality information to allow the
nonpoint source water quality data to be used to model
nonpoint source runoff characteristics of other similar
subwatersheds.
INTRODUCTION AND BACKGROUND
The Rice Creek Watershed District is located immediately
north of the cities of Minneapolis and St. Paul, Minnesota
(Fig. 1). The District has 31 municipalities or townships
within its 473 km2 tributary drainage area. Land use
ranges from agriculture and open space to high-density
commercial and residential development. Nonpoint
source water quality has been monitored within the Dis-
trict from 1974 to 1984. The monitoring programs were
developed to provide managers with information on the
quality of the nonpoint source runoff generated from 21
subwatersheds within the District. This information was
subsequently used to identify and isolate water quality
problems, and to develop solutions to them. This informa-
tion can also be used to estimate typical nonpoint source
pollution concentrations and loadings from subwa-
tersheds with similar areas, soils, and land use, but for
which no water quality information is available.
DESCRIPTION OF MONITORING
PROGRAM
Two nonpoint source water quality monitoring programs
were employed by the District during the past 10 years.
They include the Rice Creek Watershed District Stream
Monitoring Program from 1974 to 1984, and the Long
Lake Chain of Lakes Stream Monitoring Program, which
was completed from 1977 to 1984.
The Rice Creek Watershed District Stream Monitoring
Program generally consisted of measuring the flow at 13
stream stations within the District 12 times/yr (Fig. 1). The
stream stations were R1, R2, R5, R6, R7, R7A, R8, R8A,
R9, H1, H2, C1, and C2. Samples were collected during
six of these 12 flow-measuring trips.
The Long Lake Chain of Lakes Monitoring Programs
provided water quality information for stream stations B,
C, D, G1, F, H1J, N, and JL3. For these stations, flow was
gauged and samples collected approximately 12 times/yr
from 1977 to 1983.
All samples were analyzed for total Kjeldahl nitrogen
(TKN), total phosphorus (TP), orthophosphorus (OP), total
suspended solids (TSS), and chlorides. The analysis pro-
cedures were those outlined in the most recent edition of
Standard Methods for the Examination of Water and
Wastewater, published by the American Public Health
Assn., Washington, DC.
These nonpoint source water quality data were further
examined by two methods. In one case, the discharge rate
of the stream and its pollutant concentrations determined
the flow-weighted mean concentration of the pollutants in
the stream. In the second case, the pollutant concentra-
tions and flow determined the pollutant loadings. It was
necessary to evaluate the data using both these consider-
ations to accurately interpret water quality in the areas
investigated.
DISCUSSION AND RESULTS
The flow-weighted mean concentrations of various param-
eters at each monitoring station are tabulated in Table 1.
The water quality at each station is also ranked relative to
other subwatersheds in the District and shown in the right-
hand column of this table. A ranking of 0 indicates rela-
tively good water quality and a ranking of 5 indicates that
the water quality was among the poorest. This ranking
was derived by comparing the flow-weighted mean con-
centration of a given parameter at a given station to the
average for all stations. The number of parameters that
exceed the average is listed in the far right column to
provide a relative measure of water quality for each
stream station.
It is not surprising that at each of the three watersheds
with the best water quality based on a flow-weighted mean
concentration, a lake was located at the subwatershed
outlet. These lakes provided treatment for the stormwater
discharged from the subwatershed. The highest concen-
tration of nutrients and solids in the stormwater runoff
appears to be present at monitoring stations B, G1, and D.
These stations monitor runoff generated from highly ur-
banized areas.
Table 2 provides hydraulic and nutrient loadings from
local subwatersheds. This information was derived by tak-
ing the hydraulic and nutrient loadings observed at the
outlet of a subwatershed and subtracting from it any hy-
draulic and nutrient loading that was generated from sub-
watersheds upstream. This allowed a segmental hydraulic
and nutrient loading to be determined that was then di-
vided by the area to get an areal loading in kg/kmz/day (Ib/
m2/day).
The subwatersheds with the highest nutrient loadings
on an areal basis are 14C and D, 16 + 13C, and 15. It
C--.191
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Figure 1.—Rice Creek Watershed District water quality monitoring stations.
Table 1.—Flow-weighted mean concentration (mg/L) (based on monitoring data generated from 1974-1984).
Ranking
Monitoring
station
R9
R8A
H1
H2
C1
C2
R7A
R8
R7
R6
R5
B
N
H1J
G1
F
D
JL3
C
R2
R1
Watershed
number
1
3
4 + 5
6
8
9
11
12A
12B
128
13C
13C
14A
14B
14D
14E
14C
15A
15B
16
16
TKN
2.77
1.20
1.52
1.63
1.49
1.83
2.62
1.95
2.05
2.20
2.04
1.99
1.56
1.15
1.63
1.28
1.76
1.27
1.49
1.77
1.82
TP
.21
.08
.15
.16
.07
.11
.11
.21
.11
.23
.20
.19
.11
.08
.20
.16
.21
.16
.12
.14
.15
OP
.057
.055
.110
.22
.023
.059
.06
.114
.040
.149
.101
.089
.067
.038
.098
.102
.227
.125
.044
.053
.062
TSS
21.5
3.3
4.7
9.2
3.9
10.8
8.0
6.0
6.5
10.4
10.5
15.4
11.0
7.0
13.8
6.6
38.5
9.2
15.9
16.9
12.6
Chlorides
22.01
11.02
9.47
10.55
22.98
24.40
15.06
15.70
15.70
17.10
18.84
19.58
30.50
72.5
59.08
67.15
77.41
74.0
61.73
32.42
32.55
0— Best
5— Worst
3
0
1
2
0
1
1
3
1
3
3
4
0
1
4
3
4
3
1
2
2
Average
1.76
0.15
0.85
11.50
33.8
192
-------
should be noted that stations G1 and D, which were previ-
ously identified in Table 1 as having high flow-weighted
mean concentrations, are also located in these subwa-
tersheds.
Table 2 also shows that subwatersheds 2 + 3+12A,
10 + 12B, and 13A +13B plus part of 13C have a negative
loading of nutrients and solids. This negative sign indi-
cates that the subwatershed removes more of a particular
pollutant from the water passing through than it is pres-
ently generating. These subwatersheds act as stormwater
treatment areas, with the treatment generally taking place
along the chain of lakes from Baldwin Lake to Howard
Lake. Land use and soils information for each of the sub-
watersheds are shown in Tables 3 and 4.
STREAMS AND RIVERS
SUMMARY AND CONCLUSIONS
Nonpoint source runoff from 21 subwatersheds within a
473-km2 watershed was monitored from 1974 to 1984.
The results of this monitoring reinforce many expected
nonpoint source water quality relationships, including:
1. The best surface water quality was found immedi-
ately downstream of lake outlets.
2. Lakes and wetlands have the ability to remove nutri-
ents and solids from water routed through them.
3. The most concentrated solids and nutrient loadings
originate from higher density urban areas.
The information presented here can expand the non-
point source runoff data base, and assist in modeling non-
point source pollution in watersheds similar to Rice Creek
Watershed District.
liable 2.— Hydraulic and nutrient loadings from Individual watersheds.
Flow
WMmhed Number Area Rate TKN TP OP TSS Chlorides
or Description (ml2) (cfs/ml2) (Ibs/mP/day) (Ibs/mi2/day) (lbs/miz/day) (Ibs/"ni2/day) (Ibs/mP/day)
1
.Part of WS 3 tributary to
Rondeau Lake
4 and 5
61
7 and 8
&
11
2 + 3 + 12A3
10 + 12B4
13A + 13B + Part of 13CS
14A
14B"
14C + 140
Part of 14E Upstream of Co. Rd P
14E8
15
16 + part of 13C9
TOTAL
AVERAGE
10.75
1.3
19.52
8.71
31.39
12.35
10.28
23.64
9.96
7.5
1.31
4.25
3.25
1.31
2.01
7.67
10.52
165.72
9.75
1.16
5.85
0.73
1.03
0.37
0.69
0.59
0.73
0.69
0.70
0.99
1.88
1.69
1.22
1.99
1.42
1.05
22.78
1.34
18.78
33.38
5.18
9.29
2.97
8.42
7.16
0.15
18.67
2.32
9.92
9.78
13.25
-0.53
50.0
10.89
75.33
274.96
16.17
1.38
2.69
0.46
0.82
0.15
0.55
0.37
0.93
-2.10
-3.4
0.53
0.56
2.18
0.53
5.07
1.19
5.45
17.36
1.02
0.31
2.00
0.33
2.88
0.05
0.33
0.15
-0.28
-2.08
-6.05
0.38
0.11
1.08
1.52
2.33
0.38
-3.21
.23
.0135
131.20
88.85
15.26
72.32
6.15
66.93
18.21
-74.3
58.75
58.68
55.34
66.6
112.55
454.6
1.620.4
174.51
441.6
3,367.5
198.09
134.50
343.46
33.92
64.60
49.33
65.78
34.03
6.87
13.06
87.64
145.80
453.62
620.98
661.7
1.788.7
324.50
-283.7
4,544.78
267.34
Ranking
0-Best
6-Worst
3
5
1
1
1
1
1
0
1
0
1
3
4
3
6
4
3
Note: Underlined values are higher than the arithmetic average.
A negative sign (-) Indicates the watershed removes more of that particular pollutant
from water passing through It than it generates.
'Total loadings discharged from watersheds 4 and 5 were subtracted from those
discharged from watershed 6.
*lbtal loadings discharged from watersheds 7 and 8 were subtracted from those
discharged from watershed 9.
•fetal loadings discharged from watersheds 1, 6, and 9 were subtracted from those
discharged from watershed 12A.
*Tbtal loadings discharged from watersheds 11 and 12A were subtracted from those
discharged from watershed 12B.
'Total loadings discharged from watershed 12B were subtracted from the loadings
present at Rice Creek upstream of Lexington Avenue (Monitoring Station R5).
6Total loadings discharged from watershed 14A were subtracted from those dis-
charged from watershed 14B.
'Total loadings discharged from watersheds 14 and 14E were subtracted from load-
ings discharged upstream of County Road F (Station F).
*Total loadings discharged from watersheds 14B and 140 were subtracted from those
discharged from watershed 14E.
*Total loadings discharged from watersheds 14E, 1 SB, and the part of watershed 13C
upstream of Lexington Avenue (Monitoring Station R5) were subtracted from those
discharged from watershed 16.
193
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 3.—Land use within watershed.
W&tefshMJ
No.
1
2
3
4 .
5
6
7
8
9
10
11
12a
12b
13a
13b
13c
14a
14b
14c
14d
14e
15a
15b
16
Area
(acres)
6,884
4,019
8,170
9,507
2,992
5,573
11,545
8,544
7,904
4,234
6,582
2,944
12,256
1,458
832
7,286
842
2,718
529
1,547
836
2,310
2,599
4,701
Area
(mi2)
10.75
6.28
12.76
14.85
4.67
8.71
18.04
13.35
12.35
6.62
10.28
4.60
19.15
2.28
1.3
11.38
1.31
4.25
.83
2.42
1.31
3.61
4.06
7.35
Wetlands
20
18
42
13
25
7
5
17
6
7
19
13
21
26
1
9
8
1
1
1
33
1
1
2
Open
water
14
0
12
3
0
2
27
17
5
0
1
32
14
1
45
3
14
9
23
14
1
3
7
8
Single
family
residential
9
1
1
1
2
10
22
20
2
5
6
2
8
11
50
60
60
62
17
3
25
48
60
67
— kano use (nj —
Multiple Comiher-
family cfal/
residential retail
1
0
0
1
0
0
1
0
0
0
1
0
0
6
0
1
0
1
0
0
0
3
8
2
2
1
0
0
0
1
1
1
1
1
1
1
1
4
0
2
0
19
0
10
0
1
10
5
Industrial/
manufac- Agricul-
hiring tural
0
2
0
1
0
1
2
0
8
0
2
31
25
15
32
2
10
25
39
24
50
58
60
23
29
66
80
60
25
20
0
0
1
0
0
0
0
0
0
0
0
Undeveloped SCS
vacant and runoff
open space curve no.
28
40
20
31
15
19
20
15
18
7
11
27
35
50
4
16
18
5
28
47
26
12
12
6
76
82
78
64
80
76
81
81
74
76
77
88
79
74
87
71
77
79
79
76
75
78
77
76
Time of
concentration
(hours)
5.0
23.0
26.4
25.0
7.7
15.0
9.2
10.0
6.0
18.5
12.5
4.9
22.2
2.0
1.0
4.5
2.5
3.9
1.3
1.3
1.2
4.5
1.9
2.8
194
-------
STREAMS AND RIVERS
Table 4.—General soil associations: Rice Creek Watershed District, by subwatershed.
Soil
No. Association
1 Hubbard-Nymore
2 Zimmerman-lsanti-Lino
3 Rifle-lsanti
4 Antigo-Chetek-Mahtomedi
5 Zimmerman-Urban and Rifle
6 Urbanland-Chetek Mahtomedi
7 ' Hayder-Kingsley-Hayden
8 Nessel-Dundas-Webster
9 Hayden-Kingsley
10 Hayden-Nessel-Dundas
1 1 Santiago-Kingsley
12 Demontreville-Kingsley
13 Hayden-Urbanland
14 Kingsley-Urbanland
15 Antigo-Comstock
16 Urbanland-Waukegan
Co.1
A
A/W
A
W
R
R
A
A/W
W
W
W
W
R
R
W
R
Topography
Nearly level to
gently sloping
Nearly level to
undulating
Nearly level
Nearly level to
steep
Level to gently
rolling
Level to very
steep
Gently
undulating to
steep
Nearly level to
gently sloping
Undulating to
steep
Level to gently
rolling
Undulating to
steep
Undulating to
steep
Undulating to
steep
Undulating to
steep
Level to
moderate
Nearly level
to very steep
Drainage
Excessively
drained
Excessively
drained
Very poorly
drained
Well drained
to excessively
drained
Excessively
drained and very
poorly drained
Somewhat
drained to
excessively
drained
Well drained
Moderately well
drained to poorly
drained
Well drained
Moderately well
drained to poorly
drained
Well drained
Well drained
Well drained
Well drained
Well drained to
somewhat poorly
drained
Well drained
and somewhat
excessively
drained
Texture
Sandy
throughout
Fine sand
Organic
material and
fine sand
Moderately
coarse and
coarse
Coarse and
organic
Moderately
coarse and
coarse
Loamy
Loamy
Coarse
Moderately
coarse
Medium to
moderately
coarse
Coarse to
moderately
coarse
Moderately
coarse
Moderately
coarse
Medium
Medium
Parent
material
Outwash
Outwash
Outwash
Outwash
terraces
Outwash
Outwash
Glacial till
Glacial till
Glacial till
Glacial till
Glacial till
Glacial till
Glacial till
Glacial till
Lacustrine
Outwash
'A- Anoka County
W - Washington County
R - Ramsey County
195
-------
MICHAEL K. BUTLER
JOSEPH A. ARRUDA
Division of Environment
Kansas Department of Health and Environment
Topeka, Kansas
Routine monitoring of Kansas surface waters began in
1973 with a network of 11 sites and continued in 1984
with over 100 fixed stream sites and over 40 rotating lake
sites. A change in detection limits in 1977 led to a dra-
matic increase in the number of pesticides detected.
Since 1977 the most consistently found pesticides in
stream samples have been atrazine (17 percent), alachlor
(5 percent), 2,4-D (4 percent) and Dual (4 percent). Lake
sampling sites revealed the same pattern. Although re-
cent numbers of detections and concentration levels are
higher than earlier in the program, a clear temporal pat-
tern is not evident. There are, however, clear patterns in
the location where the pesticides are routinely found.
Eastern Kansas streams and lakes show the most con-
sistent patterns of pesticide detection, probably related to
use and runoff conditions. Pesticides have also been
found in treated drinking water that originated from lake
water. Treatment reduced concentrations about one-third
from those found in the raw lake water.
ularly. Additional sampling of some Federal lakes occurred
during 1984. Sample sites were generally on the main
body of the lake, with occasional sampling of inflows and
outflows of Federal lakes. All pesticides and other data are
stored on STORET, EPA's water quality data base.
Water Sypp>ly Late Stado@s
During 1983, pesticide samples were taken from 19 water
supply lakes in eastern Kansas from spring to fall. For
these 19 lakes, historical pesticide data from the treated
water (part of routine triannual organic screening begun in
1977 for these communities) also are summarized. In
1984, KDHE surveyed pesticide concentrations in raw and
treated lake water from three communities suspected to
have pesticides in raw and treated water. Water samples
were taken in May and September.
Water samples were collected in solvent-rinsed gallon
dark glass jugs, filled by emersion or by pouring from a
stainless steel bucket. Analysis is by GC/EC by the Office
of Laboratories, KDHE, according to EPA procedures.
The Kansas Department of Health and Environment
(KDHE) has maintained a water quality monitoring net-
work in streams and lakes in Kansas for many years. Pes-
ticide data have been collected routinely from stream sites
since 1973 and from lake sites since 1975. The objective
of this report is to summarize the pesticide monitoring
data and to evaluate its water quality significance.
The Ambient Stream Water Quality Network began moni-
toring for pesticides in 1973. The distribution of the Net-
work sites across Kansas reflects both hydrological char-
acteristics (rainfall and runoff) and pollution source and
potential (population centers, point and nonpoint sources).
Network sampling began with about 40 stations, 11 of
which were sampled for pesticides, some monthly. As the
network expanded in the 1970's to 100-120 sites the pes-
ticide sampling frequency dropped first to semiannual,
then to annual. The sampling schedule has generally
been spread out throughout the year, with more emphasis
in the ice-free periods. All pesticide and other data are
stored on STORET, EPA's water quality data base.
The Lake Monitoring Program sampled 58 lakes 1 to 6
times from 1975-1982. Lakes were sampled anytime from
April to October. The lakes included 22 Federal lakes (sur-
face area ranging from 440 to 6400 surface hectares)
sampled on a 3-year cycle and other smaller lakes (sur-
face area from 10 to 300 surface hectares) sampled irreg-
Over 700 samples were collected from the Stream Net-
work before 1977 without detecting any pesticides. In
1977, the detection limits for a number of pesticides were
lowered and pesticides began to be detected (Table 1).
Five different pesticides were initially detected in 1977.
Since 1977, a total of 21 different pesticides have been
found above their detection limits (Table 1). The number of
different pesticides detected in any one Water Year has
been as low as 5 (in 1973) and as high as 17 (in 1983).
Over the period of record, 5 of the 21 pesticides (atrazine,
alachlor, Dual, 2,4-D and Sencor) have accounted for 77
percent of the total detections (Table 1, Fig. 1). The aver-
age rates of samples (detected plus undetected) having
one of these five pesticides found in them since 1977 were
17 percent with atrazine, 5 percent with alachlor, 4 percent
with 2,4-D, 4 percent with Dual, and 2 percent with Sencor.
Atrazine has always been the single most frequently
found pesticide. In 1977, atrazine accounted for 13 of 23
(53 percent) pesticide detections, and appeared in 6 per-
cent of all samples (Table 1). Atrazine detections have
been as much as 77 percent of the total pesticide detec-
tions (in 1978) and atrazine has been detected in up to 33
percent of the sites sampled (in 1983) (Table 1).
No pesticides were detected in the lakes from the start
of the lake monitoring program in 1975 to 1976. Since
1977, pesticides have been detected in samples from 19
of the 58 lakes. Atrazine was the most commonly detected
substance, ranging from 1.4 to 23.0 /xg/L in the Federal
lakes, and 1.2 to 2.8 /*g/L in the other lakes (Table 2).
Alachlor and Dual, the next most frequently detected sub-
stances, were present at lower concentrations. Detections
of Sencor, Ramrod, and 2,4-D occurred less frequently.
196
-------
STREAMS AND RIVERS
The same general pattern holds for the water supply lakes
sampled in 1983 (Table 3).
Atrazine was detected nearly as frequently with alachlor
(25 times) as alone (31 times) in the LMP, while of the 33
occurrences of alachlor, only two were without atrazine.
The mean concentration of atrazine was higher when
alachlor also was present (7.0 ^g/L with alachlor, 2.3 ^g/L
without alachlor, T-test p < 0.001).
The 1983 water supply lake and historical treated water
data (Table 3) show that those pesticides commonly found
in lake water in 1983 had also been detected in the treated
water over the period of analysis of the treated water
(1977-present). Atrazine, alachlor, and Dual were the pes-
ticides found in the three water supply lakes sampled in
1984 (Table 4). Only one lake had detected pesticides in
May, but all three had pesticides in September. Atrazine
removal, estimated from the September data, ranged from
6.1 percent to 40.6 percent. Based on only one data point
each, alachlor removal was 19.2 percent, and Dual re-
moval was 41.7 percent.
DISCUSSION
Temporal Distribution
Despite the shortcomings of data taken semiannually or
annually, a few temporal trends emerge from the stream
data. The frequency of detection of several pesticides
showed increases over time (atrazine, alachlor, Dual, Sen-
car, and 2,4-D) or decreases (Dachthal) (Fig. 2). The num-
ber of detections fluctuates widely, however. This is partly
due to the relationship between the time of sampling and
rainfall and the exact time of pesticide application. The
DUAL
(10%)
I.HYDROXYCHLORDENE
Figure 1.—Overall frequency of detection of.major pesti-
cides from Kansas Ambient Stream Water Quality Network
(1977-1984).
overall temporal trend is towards increased numbers of
detections (Fig. 2). Although at some stations, the concen-
trations of atrazine may be increasing, no clear temporal
trend in concentrations exists (Fig. 3).
Geographic Distribution
Physical factors such as soil type, land topography, rainfall
amount, distribution, and intensity largely determine po-
tential crop types which, in turn, determine pesticide us-
age. These same physical factors determine the need and
Table 1.—Frequency of detections of pesticides from 1977 to 1984 from the Amuient Stream Water Quality Network.
Substance 1977 1978 1979 1980 1981 1982 1983 1984 Total
Alachlor
Aldrin
Alpha-bhc
Atrazine
Chlordane
Dachthal
DDE
Diazanon
Dieldrin
Dual
Dursban
HCB
Lindane
Malathion
Sencor
Propazine
Ramrod
1 -Hydroxychlordene
2,4-D
2,4,5-T
2,4,5-TP
Total detections
Total samples
2
0
0
13
0
5
0
0
0
0
0
2
0
1
0
0
0
0
0
0
0
23
209
6
0
0
28
0
1
0
0
0
0
1
0
0
1
4
0
1
0
0
0
0
42
161
4
0
0
18
0
3
0
0
0
0
0
1
1
0
1
1
0
3
4
2
0
38
119
2
0
0
9
1
1
4
0
0
5
0
1
1
1
1
0
0
6
4
1
0
37
103
3
0
3
14
2
1
1
2
0
5
0
0
4
1
1
3
2
4
9
1
0
56
106
7
1
0
17
0
0
0
0
1
9
0
0
2
0
5
0
3
0
6
2
0
53
100
15
1
0
39
1
1
0
1
1
12
0
0
2
1
3
1
3
1
10
1
1
94
120
11
0
0
34
0
1
0
0
0
12
0
0
5
2
9
0
2
0
7
1
0
84
117
50
2
3
172
4
13
5
3
2
43
1
4
15
7
24
5
11
14
40
8
1
427
1035
Table 2.—Summary of pesticide data from lakes sampled in the Lake Monitoring Program from 1979 to 1984.
Data are in Mg/L.
Federal lakes Small lakes
Substance
Alachlor
Atrazine
Dual
Sencor
Ramrod
2,4-D
Propazine
Range
0.10-3.1
1 .4-23.0
0.26-2.6
0.05-0.31
0.25-2.90
0.69-2.4
—
Mean
0.82
4.8
0.74
0.18
1.00
1.37
2.6
N
21
43
19
8
5
4
1
Range
—
1.2-2.8
—
—
0.27-1.3
0.42-0.48
—
Mean
0.36
2.0
—
0.21
0.79
0.45
—
N
1
5
—
1
2
2
—
197
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
feasibility of large reservoirs. These large reservoirs (the
Federal lakes), patterns of pesticide usage, and physical
factors combine to influence downstream patterns of pes-
ticide occurrence.
The Stream Network sites above the Federal lakes
show only sporadic pesticide occurrence, while those
sites below the Federal lakes show consistent pesticide
detections, at least in the case of atrazine at these major
sites (Rg. 3). Thus, the reservoirs seem to be sequester-
ing pesticides loaded during runoff events, and slowly
"metering" them out as water is released from the lake.
so
CO 40
O
O
LU 30
r-
LU
Q
CC.
LU
CD
Z) ,o
78
84
WATER YEAR
Figure 2.—Summary of number of detections of most fre-
quently found pesticides from 1977-1984 from the Kansas
Ambient Stream Water Quality Network. Total detections x
0.5 (hexagons), atrazine (open circles), alachlor (solid
squares), Dual (open squares), 2,4-D (solid triangles), Sen-
cor (open triangles), Dachthal (closed circles).
Table 3.—Summary of pesticide data from 1983 survey of
community water supply lakes, with historical treated water
data. Data are In pg/L.
Lake Treated water
Substance
Range N Range N
Alachlor
Atrazine
Dual
2,4-D
O'-P'-DDE
P'-P'-DDE
2,4,5-T
0.28-0.36
1.4-4.0
0.54
0.51
0.12
0.12
0.21
2 0.51-1.10 3
4 1.2-4.8 10
1 0.33 1
1 0.8-3.2 2
1 — —
1 — —
1 0.22 1
Two other general geographic trends are present in the
Stream Network data. The first trend involves sites drain-
ing urban environments. Although detections are not con-
sistent, a wide range of pesticides are found, including
lindane, chlordane, dieldrin, propazine, and others not
found regularly at the other network sites. The second
trend is the almost exclusive presence of one hydroxy-
chlordene at network sites in the western, primarily wheat-
producing part of the State.
Use Inventory, Use Impairment, and Water
Quality Criteria
Aquatic life can be impaired by these pesticides, espe-
cially atrazine, for which the best data are available. Labo-
ratory and field studies have shown that atrazine can af-
fect phytoplankton successional changes, photosynthetic
rates, and growth rates (DeNoyelles et al. 1982; Kosinski
and Merkle, 1984). Some effects may occur at concentra-
tions in the range 1-20 pg/L as observed by DeNoyelles et
al. (1982) and O'Kelley and Deason (1976). Brockway et
al. (1984) indicated that atrazine concentrations of 50 ^g/L
or less could negatively effect phytoplankton communi-
ties.
16
CT>
LU
2
^^
^^
cr
<
77
78
79
80
81
83
84
WATER YFAR
Figure 3.—Temporal trend of concentrations of atrazine In
stream water above and below two major Federal lakes In
Kansas. Data from the Kansas Ambient Stream Water Qual-
ity Network (1977-1984). Solid symbols are data from below
Federal lakes, open symbols are data from above Federal
lakes. Squares are Tuttle Creek Lake, and circles are Mllford
Lake.
Table 4.—Summary of pesticide data from synoptic 1984 survey of three water supply lakes and their treated (tap) water.
Data are In /»g/L.
LAKE #1 LAKE 02 LAKE #3
Substance
Source
May
Sept.
May
Sept.
May
Sept.
Alachlor
Atrazine
Dual
Lake
Tap
Lake
Tap
Lake
Tap
—
—
2.1
—
0.38
—
—
—
3.3
3.1
1.2
0.70
— —
— 4.2
— 3.7
— —
— 2.6
— 2.1
— 16.0
— 9.5
— 1.1
— —
198
-------
STREAMS AND RIVERS
Klaassen and Kadoum (1979), Lynch et al. (1982) and
Brockway et al. (1984) indicate that atrazine has a low
potential for bioconcentration and that it degrades slowly
(>50 days). Klaassen and Kadoum (1979) applied atra-
zine at 300 /ig/L to a previously unexposed pond and
observed quick uptake of atrazine by the biota and the
sediments. A year later, there were no biological residues,
but residues were found in the water and mud. Although
uptake of atrazine is weak and not long-lasting, it does
accumulate in the biota.
There currently are no aquatic life criteria or guidelines
in the United States or Canada (McNeely et al. 1979) for
these commonly detected pesticides (atrazine, alachlor,
Dual, 2,4-D, Ramrod, Sencor). The data suggest that
slight or moderate aquatic life impairments would occur
upon continuous exposure to atrazine. The long-term ef-
fects of short exposures to atrazine are now known, but
would likely vary with the affected community and the
concentration. If criteria for safe levels of these pesticides
were developed, such criteria might fall into the range of
concentrations commonly detected in streams and lakes
downstream of direct runoff and would likely fall into the
range of concentrations found in direct runoff.
Drinking Water Supply
Water from some of the lakes and certain stream seg-
ments with detected pesticides is used for domestic con-
sumption after treatment. Our water supply lake data indi-
cate that pesticides are also found in the treated drinking
water. Unfortunately, the available toxicity data may be
insufficient to suggest firm drinking water criteria for some
of these substances. There are no proposed drinking wa-
. ter criteria or maximum contaminant levels for the most
common pesticides found. However, Canada (McNeely et
al. 1979) has set a maximum acceptable level (MAL) of
100 ftg/L and an objective level (goal) of not detected for
atrazine and alachlor. For 2,4-D, the MAL is 100 jig/L and
the objective level is 1 /xg/L.
"Drinking Water and Health" (Safe Drinking Water
Committee, 1977), provided "suggested no-adverse lev-
els," or SNARLS, for three of the frequently found pesti-
cides. Two SNARLS, one allotting 20 percent of total ac-
ceptable daily intake (ADI) to water and one allotting 1
percent were given. For atrazine, the SNARLS were 150
Mg/L (20 percent ADI) and 7.5 ^g/L (1 percent ADI), for
alachlor they were 700 uglL (20 percent ADI) and 35.0 ^g/
L (1 percent ADI), for 2,4-D they were 87.5/*g/L (20 percent
ADI) and 4.4 /ig/L (1 percent ADI).
The recent EPA restrictions on the use of alachlor focus
attention on the data for carcinogenicity of these sub-
stances. The restrictions were based on new data sug-
gesting that alachlor is a carcinogen. This new develop-
ment may result in future revisions in the SNARL
estimates or other criteria development.
The drinking water data suggest that the levels of these
pesticides currently being found should not cause chronic
health problems. However, for the three sets of SNARLS
available at the time of publication, the uncertainty factor
used in calculating doses was 1000, indicating limited
chronic data. To more firmly confirm or deny the potential
for carcinogenicity and long-term health problems, further
research and development of criteria are needed.
CONCLUSIONS
Atrazine, alachlor, Dual, Sencor, and 2,4-D are the pesti-
cides most commonly found in the surface waters of Kan-
sas. Their occurrences correspond with agricultural land
use, rainfall, and the potential for runoff. Large Federal
lakes may act as a buffer, receiving large pulsed inputs of
Figure 4.—Statewide distribution of pesticides in Kansas
based on data from Ambient Stream Water Quality Network
(circles) and lakes (squares) over the period of record. Black
symbols indicate frequent occurrence (>7 detections) at
stream sites or any occurrence at lake sites. Stippled sym-
bols indicate moderate occurrence (5-7 detections) at
stream sites. Open circles indicate low occurrence (<5 de-
tections) at stream sites and no occurrence at lake sites.
pesticides and releasing them later over longer time peri-
ods.
Based on laboratory and field research, the concentra-
tions of atrazine, the pesticide found most frequently and
at the highest concentrations, may be sufficient to be im-
pacting aquatic life. Phytoplankton may be the primarily
affected nontarget organism.
Pesticides are found in raw water sources and in the
final treated drinking water. The concentrations of the pes-
ticides found (atrazine, alachlor, and Dual) are lower than
those that would cause human health problems based on
available data. However, further data should be collected
to establish firmer criteria.
Routine monitoring data have been adequate to assess
the distribution and concentrations of these pesticides,
and to suggest further water quality management needs.
Criteria development for these pesticides will be neces-
sary in order to provide water quality managers with better
information for assessing potential water quality prob-
lems. Further field research is needed in areas with
pulsed or continuous concentrations of these pesticides in
order to determine actual on-site impacts. The results of
research on the effects of pesticides in water and on ac-
tual on-site use impairments must be made known to the
199
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
agricultural community. Their support will be needed if the
benefits of modified pesticide application procedures or
other land treatment practices are to be realized.
REFERENCES
Brockway, D.L., P.O. Smith, and RE. Stancil. 1984. Fate and
effects of atrazine in small aquatic microcosms. Bull. Environ.
Contam. Toxicol. 32: 345-353.
DeNoyelles, P., W.D. Kettle, and D.E. Sinn. 1982. The response
of plankton communities in experimental ponds to atrazine,
the most heavily used pesticide in the United States. Ecology
63: 1285-93.
Glotfelty, D.E., A.W. Taylor, A.R. Isenee, J. Jersey, and S. Gienn.
1984. Atrazine and simazine movement to Wye River estuary.
J. Environ. Qual. 13:115-21.
Kansas Department of Health and Environment. 1983. Water
Quality Data for Kansas Water Year 1983.
Klaassen, H.K., and A.M. Kadoum. 1979. Distribution and reten-
tion of atrazine and carbofuran in farm pond ecosystems.
Arch. Environ. Contam. Toxicol. 8: 345-53.
Kosinski, R.J., and M.G. Merkle. 1984. The effect of four terres-
trial herbicides on the productivity of artificial stream algal
communities. J. Environ. Qual. 13: 75-82.
Lynch, T.R., H.E. Johnson, and W.J. Adams. 1982. The fate of
atrazine and a hexachlorobiphenyl isomer in natural-derived
model stream ecosystems. Environ. Toxicol. Chem. 1: 179-
92.
McNeely, R.N., V.P. Neimanis, and L. Dwyer. 1979. Water Quality
Sourcebook, a Guide to Water Quality Parameters. Inland Wa-
ters Directorate, Water Quality Branch, Environment Canada,
Ottowa, Canada.
O'Kelley, J.C., and G.R. Deason. 1976. Degradation of Pesti-
cides by Algae. EPA-600/376-022. U.S. Environ. Prot.
Agency, Athens, GA.
Safe Drinking Water Committee. 1977. Drinking Water and
Health. Natl. Acad. Scien., Washington, DC.
200
-------
EFFECTS OF INTENSIVE AGRICULTURAL LAND USE ON REGIONAL
WATER QUALITY IN NORTHWESTERN OHIO
DAVID B. BAKER
KENNETH A. KRIEGER
R. PETER RICHARDS
JACK W.KRAMER
Water Quality Laboratory
Heidelberg College
Tiffin, Ohio
ABSTRACT
Cropland comprises 80 percent of the land surface in
northwestern Ohio. Runoff of sediments, nutrients, and
pesticides from this cropland adversely affects water
quality in area rivers, in associated estuaries and bays,
and in Lake Erie. Although the terrain is relatively flat, the
combination of fine textured soils, extensive tile drainage,
and cropping practices results in unusually large
amounts of agricultural pollution. Ten years of detailed
studies on streams and rivers of this area reveal several
basic characteristics of nonpoint pollution transport. Peak
concentrations of paniculate and soluble pollutants occur
during different parts of the storm hydrograph. Storm to
storm and year to year variability is very large. Effects of
season and watershed size are also evident. An aware-
ness of these characteristics may be useful in evaluating
less detailed studies.
The streams and rivers that drain into the western and
central basins of Lake Erie have been the object of unusu-
ally detailed and long-term water quality studies (Baker,
1985). These studies were undertaken for a variety of pur-
poses, including: the provision of baseline data for major
water quality planning studies (U.S. Army Corps Eng.
1982; Ohio Environ. Prot. Agency, 1979); the study of pol-
lutant transport in river systems (Baker, 1984; Verhoff et al.
1978); the assessment of conservation tillage demonstra-
tion projects (Honey Creek Joint Board of Super. 1982;
Defiance County Soil Water Dist. 1983); and the determi-
nation of pollutant loads from major tributaries into Lake
Erie (Baker, 1983a).
The Water Quality Laboratory at Heidelberg College
has been the focus of the stream monitoring programs in
this region. The laboratory's river and stream sampling
program is summarized in Table 1, with the locations of
the sampling stations shown in Figure 1. More than
43,000 samples have been collected and analyzed as part
of these studies. All samples have been analyzed for sus-
pended solids, total phosphorus, soluble reactive phos-
phorus, nitrate plus nitrite nitrogen, chloride, silica, and
conductivity. Total Kjeldahl nitrogen has been measured in
approximately 50 percent of the samples. Since 1980,
subsets of samples have been analyzed for additional
forms of bioavailable phosphorus and many currently
used pesticides. For all of the samples, stream flow data
are available from U.S. Geological Survey stream gauges
(stations 1-18) or from the laboratory (station 19).
For many of the study watersheds, pollutant loadings
are dominated by runoff from cropland. Consequently,
these studies provide a detailed illustration of the impacts
of cropland runoff on water quality in streams and rivers.
The term "regional water quality" in the title of this paper
is used to emphasize that these studies deal with condi-
tions in streams and rivers as opposed to runoff water
from research plots or individual fields. Most data on the
impacts of agriculture on water quality are derived from
edge-of-field studies (Wauchope, 1978).
In this paper, we will use some of our data to illustrate
several basic characteristics of the concentrations and
transport of sediments, nutrients and pesticides in area
streams and rivers. In particular, we will describe:
1. The patterns of concentration changes during runoff
events;
2. The variability in material transport from storm to
storm and year to year;
3. Seasonal aspects of concentrations and loadings;
and
4. The effects of watershed size on pollutant concentra-
tions and loadings.
In a companion paper in this volume (Baker et al.
1985b) we summarize the unit area exports of nutrients
and sediments from these watersheds and compare these
exports with gross erosion rates and land use.
THE STUDY AREA
As part of the Lake Erie Wastewater Management Study,
the U.S. Army Corps of Engineers developed a cellular
data file of the soil types, slopes, and land use for all the
major U.S. watersheds of the Lake Erie Basin (Adams et
al. 1982). For northwestern Ohio, including the Maumee,
Portage, Sandusky, and Huron river basins, cropland oc-
cupies, on the average, 77 percent of the land surface,
pasture 3 percent, forest 8 percent, and water or wetlands
3 percent. Most of the area is relatively flat. The fine- to
medium-textured soils are derived from glacial till depos-
ited as ground or end moraines. Some glacial lake de-
Figure 1.—Locations of river sampling stations in North-
western Ohio. (Identification numbers refer to the listing in
Table 1.).
201
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
posits are also present. Average gross erosion rates are
relatively low, falling in the 5.0 to 8.3 metric tons/hectare/
year range (2.2 to 3.7 tons/acre/year) (Logan et al. 1982).
Extensive tile drainage systems are common, particularly
in the flatter fields. Corn and soybeans are the dominant
crops.
STUDY METHODS
All of our stream samples are collected at stream gauging
stations, most of which are operated by the U.S. Geologi-
cal Survey. All-weather pumping systems have been in-
stalled at the gauge houses and automatic sampling
equipment is used to subsample from the continuously
operating pumps. At least four samples per day are ana-
lyzed during storm events and one sample per day at
other times on a year-round basis. The sample collection
procedures have been described in detail by Baker (1984).
The analytical procedures for suspended solids, total
phosphorus, soluble reactive phosphorus, nitrate-nitro-
gen, ammonia-nitrogen, total Kjeldahl nitrogen, conductiv-
ity, chloride, and silica are adapted from those described
in the U.S. Environmental Protection Agency's Methods
for Chemical Analysis of Water and Wastes (1979). Our
modifications to the automated analytical methods are de-
scribed by Baker (1984). Our pesticide method involves
methylene chloride extraction and transfer to /so-octane,
followed by capillary gas chromatography using nitrogen-
phosphorus detectors (Kramer and Baker, 1985).
CONCENTRATION PATTERNS DURING
RUNOFF EVENTS
Nutrient and sediment transport in area rivers follows a
typical pattern (Figure 2). During a runoff event, stream
flow increases very rapidly on the rising limb of the hydro-
graph, reaches a peak value, and then decreases more
slowly on the falling limb of the hydrograph. Sediment
concentration peaks early in the runoff event and is al-
ready decreasing well before the peak discharge occurs.
This advanced peak of sediment concentrations is fre-
quently observed. Since most of the phosphorus trans-
ported during storms is attached to sediment, the phos-
phorus concentration closely follows the concentration
pattern for suspended sediments. During the falling por-
tion of the hydrograph the phosphorus concentration does
not decline as rapidly as the sediment concentration. This
can be attributed both to the presence of soluble phos-
phorus in the streams and to an increasing ratio of panicu-
late phosphorus to sediment as the sediment concentra-
tion decreases (Baker, 1984). The latter effect probably
results from decreasing average particle size accompany-
ing decreasing sediment concentration (Baker et al.
1979).
Nitrate concentration increases during the falling limb of
the hydrograph. In this area, most of the nitrate enters
streams as part of the tile drainage and interflow (Johnson
and Baker, 1984). Waters from these sources comprise a
larger proportion of the total flow during the falling limb of
the hydrograph. The concentration patterns of soluble
herbicides, such as atrazine, are distinct from both the
sediments and the nitrates. Atrazine apparently moves off
the fields with surface runoff water. We observed these
general patterns of concentration changes during storms
for both small watersheds (e.g., Lost Creek, 8.8 km2) and
the large river basins (e.g., the Maumee River,
16,395km2).
The large changes in sediment and nutrient concentra-
tions with stream flow necessitate using two distinct meth-
ods for averaging concentration data. If the intent is to
characterize pollutant concentrations relative to pollutant
loading, a flux-weighted average concentration should be
calculated using Equation 1. If the intent is to characterize
the pollutant concentrations relative to ambient water
quality at the sampling station, a time-weighted average
concentration should be calculated using Equation 2. For
most pollutants derived from cropland runoff, the flux-
weighted average and time-weighted average concentra-
tions differ greatly. Examples of these differences for
northwestern Ohio watersheds are shown in Table 2.
Table 1 .—Summary of river transport studies conducted by the Water Quality Laboratory at Heidelberg College.
U.S. Geological Watershed
Survey area Study Samples
Location Station no. km2 period analyzed*
Sandusky River station
1. Fremont
2. Mexico
3. Upper Sandusky
4. Bucyrus
Sandusky River tributaries
5. Wolf Creek East
6. Wolf Creek West
7. Honey Creek, Melmore
8. Honey Creek, New Wash.
9. Tymochtee Creek
10. Broken Sword Creek
11. Rock Creek
Other Lake Erie rivers
12. Maumee
13. Raisin (Michigan)
14. Cuyahoga
15. Portage
16. Huron
17. Ottawa
18. Bean Creek
19. Lost Creek
04198000
04197000
04196500
04196000
04192450
04197300
04197100
04197020
04196800
04196200
04197170
04193500
04176500
04208000
04195500
04199000
04187500
04184500
3,240
2,005
722
230
213
171.5
386
44.0
593
271
88
16,395
2,699
1,831
1,109
961
414
534
8.8
1974-
1974-81
1974-81
1974-81
1976-81
1976-81
1976-
1979-81:1983-
1974-81
1976-81
1982-
1975-80; 1982-
1982-
1981-
1974-78
1974-79
1980-81
1980-81
1982-
4590 +
2178
2973
2998
2425
2419
4595 +
2271 +
2471
2512
982 +
3154 +
805 +
1380 +
1856
2027
267
365
1747 +
* Number of samples analyzed through 1984 water year.
+ Stations Included in 1985 water year program.
i- + Number not yet assigned.
202
-------
STREAMS AND RIVERS
9. 10.
JUNE
14.
9. 10. 11. 12. 13. 14.
JUNE 1981
OO)
o
o
UJ
CO
CO
a:
ui
o
H
DO
O
O
G>
CM
CO
a:
i-
M
7. 8. 9. 10. 11. 12.
JUNE 1981
13. 14.
9. 10. 11. 12..
JUNE 1981
13. 14.
Figure 2.—Concentration patterns for sediments (A), total P (B), nitrate-N (C), and atrazlne (D) relative to stream discharge
(solid line) for a typical runoff event in the Honey Creek watershed.
Table 2.—Comparison between flux-weighted average and time-weighted average concentrations of
nutrients and sediments at representative stream transport stations.
Transport station
Maumee River
Flux-weighted
Time-weighted
Sandusky River, Fremont
Flux-weighted
Time-weighted
Honey Creek, Melmore
Flux-weighted
Time-weighted
Cuyahoga River
Flux-weighted
Time-weighted
Raisin River
Flux-weighted
Time-weighted
Suspended
solids
mg/L
218
98.5
249
85.9
198
60.0
188
94.1
81.1
41.9
Total
phosphorus
mg/L
0.474
0.307
0.468
0.222
0.413
0.196
0.428
0.405
0.238
0.178
Soluble
reactive
phosphorus
mg/L
0.090
0.091
0.084
0.059
0.074
0.068
0.105
0.165
0.046
0.048
Nitrates-
nitrite
nitrogen
mg/L
5.13
3.95
4.61
3.34
4.79
4.03
1.82
2.53
3.51
2.58
Total
Kjeldahl
nitrogen
mg/L
1.85
1.49
1.73
1.11
1.79
1.12
1.36
1.27
1.23
0.96
Conduc-
tivity
pmhos
484
622
464
684
378
582
674
770
528
668
203
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Average Flux-Weighted Cone.
Average Time-Weighted Cone.
(D
(2)
where 5 = concentration of Ith sample
q, = instantaneous discharge when ith sample
was collected
t( = time ith sample was used to characterize
stream system (ranges from 2 to 24 hr)
VARIABILITY IN CONCENTRATION
AND LOADS
Storm-to-Storm Variability
Considerable effort goes into characterizing nutrient and
pesticide export during a single storm event. What makes
studies of agricultural runoff so challenging is that every
storm event is different. One way to characterize concen-
trations during a single storm is to calculate flux-weighted
average concentrations for that storm. We calculated flux-
weighted average total phosphorus concentrations for 52
storm events at the Upper Sandusky gauging station. In
Figure 3 concentrations for individual storms are plotted in
relationship to the peak discharge. Considerable variabil-
ity is evident in phosphorus concentrations, even for
storms with similar peak discharges. Part of the variability
observed in the storms is associated with seasonal ef-
fects. Similar storm to storm variability was observed for
nitrates and suspended solids (Baker, 1984).
Annual Variability
Nutrient loads derived from cropland runoff generally have
large year to year variability as illustrated by phosphorus
loading for the Sandusky River for the 1975-84 period.
These loads along with the total annual discharges are
shown in Table 3. In general, years with high discharges
have high loads. The highest load did not coincide, how-
ever, with the highest discharge. In 1981 we had large
storm events in June which resulted in very high sediment
and phosphorus concentrations and the highest annual
phosphorus load we have observed. In 1978 there was a
large annual discharge. Most of the discharge occurred in
winter runoff events that were accompanied by low phos-
phorus and sediment concentrations. Consequently, we
had relatively low total phosphorus loads that year.
The large amounts of annual variability in loading from
nonpoint sources complicate comparing point source and
a
o
O. a>
M
55
. •
•• •
••
0.6 1.0 1.6 1.0
Lof (P«ak Flow) lit3/8
2.6
Figure 3.—Variability In flux-weighted average concentra-
tions of total phosphorus for 52 Individual runoff events at
the Sandusky River, Upper Sandusky sampling station.
nonpoint source loads. It is important to estimate a mean
annual load from cropland sources, rather than using
loads for any single year.
SEASONAL VARIATIONS
As noted, seasonal variations in concentrations and load-
ing can account for some of the variability in cropland
runoff. The most obvious example of seasonal effects is
shown by the pesticides. The highest pesticide concentra-
tions in streams and rivers, as well as the highest loads,
generally occur during the first runoff event following pes-
ticide application (Baker et al. 1981; Baker, 1983b). In this
respect, streams reflect the characteristics of edge-of-field
losses (Wauchope, 1978). This first runoff event normally
occurs in May or June. Subsequent storm events have
much lower pesticide concentrations. For much of the
year, most pesticides are undetectable in the rivers. Atra-
zine is an exception, and traces are present throughout
the year.
The seasonal occurrences of herbicides in stream sys-
tems also result in seasonal pesticide exposures in drink-
ing water derived from river systems (Baker, 1983c). Ex-
amples of peak observed pesticide concentrations in
finished tap water from three northwestern Ohio cities are
shown in Table 4. Most soluble herbicides pass through
conventional water treatment plants with very little attenu-
ation in concentration.
Seasonal patterns are also present for nitrate concen-
trations. Year after year, the highest nitrate concentrations
occur during runoff events in May, June, and early July
(Baker, 1985). In the Sandusky River at Fremont, concen-
trations of nitrate-N exceed the drinking water standard of
10 mg/L an average of 4.1 percent of the time on an an-
nual basis and 16 percent of the time in the May through
July period. Relatively high nitrate concentrations are also
present during winter and early spring runoff events.
Since these events are generally larger in volume, the
bulk of the nitrate loading occurs during the winter and
early spring period.
Table 3.—Annual variability in discharge and total phosphorus
loads at the Sandusky River, Fremont station.
Water
year
1975
. 1976
1977
1978
1979
1980
1981
1982
1983
1984
Average
Discharge
108m3/year
1,048
755
636
1 ;385
1,107
1,211
1,249
1,353
649
1,940
1,133
Total
phosphorus load
metric tons/year
423
240
259
462
476
691
792
633
200
759
504 (1.55 kg/ha)
Table 4.—Peak herbicide concentrations In tap water at
Tiffin, Fremont, and Bowling Green, Ohio.
Tiffin Fremont
Bowling
Green
Tiffin
tap water tap water tap water tap water
Simazine
Atrazine
Alachlor
Metolachlor
Linuron
Cyanazine
1983
P9/I
0.63
7.64
2.73
13.65
0.61
1.49
1983
0.13
1.22
0.47
1.33
0.39
1980-82
M9/I
0.35
5.20
5.91
4.75
0.39
1.92
1980-82
1.90
30.0
14.3
24.2
2.40
204
-------
STREAMS AND RIVERS
In the case of total phosphorus and suspended solids,
seasonal patterns are affected by watershed size. For
Honey Creek, a 386 km2 watershed, winter runoff events
have much lower total phosphorus concentrations than do
summer runoff events (Baker, 1983a). Snowmelt events
generally have lower sediment concentrations than events
associated with rainfall. However, this seasonal pattern is
much less evident in larger watersheds such as the Mau-
mee River Basin.
We have summarized our long-term records for the flux-
weighted sediment concentrations for the December
through February period and for the March through July
period (Table 5). For the Maumee River, the concentra-
tions were very similar for both periods. As the watershed
sizes become smaller, the ratios of March-July to Decem-
ber-February sediment concentrations increase. The
smaller watersheds more closely reflect the timing of ero-
sion events on the landscape, while the larger rivers re-
flect river transport effects, including deposition and re-
suspension (McGuinness et al. 1971).
Since the months December through March generally
have the highest discharges, pollutant loads are also very
high for these months. Seasonal aspects of pollutant ex-
port are summarized in Table 6.
EFFECTS OF WATERSHED SIZE
One of the most obvious effects of watershed size is on
the peak concentrations of chemicals. As watershed size
decreases, everything else being equal, the peak concen-
trations of sediments, nutrients, and pesticides increase.
However, the concentrations also decrease to baseline
values more quickly as watershed size decreases. This is
illustrated in Figure 4 for atrazine concentrations in Lost
Creek and the Maumee River. These patterns are primar-
ily a consequence of the routing of runoff water through
stream systems. When conducting toxicological studies
for pesticides, organisms characteristic of low-order
streams need to be exposed to much higher pesticide
concentrations than organisms from higher-order
streams.
One convenient way to describe concentration patterns
at a station is to use concentration duration curves (Fig. 5).
These curves depict the percent of time particular concen-
trations are exceeded. For example, a nitrate concentra-
tion of 10 mg/L is exceeded 4.3 percent of the time at our
station on Honey Creek (Fig. 5a). The nitrate concentra-
tion duration curves for the Maumee River and Honey
Creek are very different (Fig. 5b). Honey Creek has much
higher peak concentrations than does the Maumee River.
However, at intermediate duration values, the concentra-
tions for the Maumee are higher than for Honey Creek.
LOST CREEK
MAUMEE RIVER
I TJ I
231. 237. 243. 249. 255. 261. 267. 273.
DAY OF UATFR YEAR 1884
279. 285.
Figure 4.—Comparison of atrazine concentrations at the
Lost Creek and Maumee River sampling stations.
Table 5.—Seasonal distribution of flux-weighted sediment
concentrations In relationship to watershed size.
Season
December-
February
March-July
Ratio
M-J/D-F
Drainage
area km2
Flux weight suspended solids concentration,
mg/L
Maumee R.
227
224
0.99
16,395
Sandusky
R.
148
340
2.30
3,240
Honey Cr.
95
284
2.99
386
Lost Cr.
88
380
4.31
9.7
A CONCENTRATION OF 10 MG/L IS EXCEEDED
1.31 OF THE TIKE
A CONCENTRATION OF 6 MG/L IS EXCEEDED
161 OF THE TIME
0. 18, 20 38. 48. 58. 60. 78.
DURATION CX>
88. 98. 188.
X
Z
U
-HONEY CREEK
,MAUMEE RIVER
8. 18
28. 38. 48. 58. 68. 78.
DURATION
98. 188.
MAUMEE RIVER
SANDUSKY RIVER
UPPER HONEY CREEK
HONEY CREEK
48. SB. 60.
DURATION
98. 188.
Figure 5.—Representative concentration duration curves at
sampling stations: A. nitrates at Honey Cr., Melmore;
B. comparison of nitrates at Honey Creek, Melmore, and
Maumee River; C. suspended sediments at 4 stations.
205
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Sediment concentration-duration curves also show sys-
tematic differences with watershed size (Fig. 5c). These
curves reflect the observation that small streams clear up
much more quickly than large rivers after the passage of a
storm event. This is probably a consequence of both flood
routing through the drainage system and the higher aver-
age current velocities in larger rivers (Leopold, 1974). The
higher velocities reduce deposition of suspended solids.
In this region, watershed size has a much greater effect
on concentration patterns than on unit area loading. For
streams of the sizes we are studying, there is little evi-
dence of long term delivery losses to stream bottoms or
flood plains. These stream systems lack deep reservoirs
that could act as permanent sinks for suspended solids
and attached pollutants.
Watershed size does affect the proportion of the annual
loads accounted for by fluxes exceeded fixed percentages
of the time. For example, suspended sediment fluxes
which were exceeded 2 percent of the time accounted for
41.8 percent of the total load for the Maumee River, 52.6
percent of the load for the Sandusky River, and 66.4 per-
cent of the total load for Honey Creek at Melmore. Sus-
pended solids and total phosphorus loads associated with
fluxes that were exceeded various percentages of time are
shown in Table 7.
It is also very likely that annual variability in loading
increases as watershed size decreases. This complicates
the task of assessing the effectiveness of cropland best
management practices, since small watersheds are gen-
erally chosen for demonstration programs to facilitate
achieving a high proportion of implementation. Accurate
loading data are more difficult to obtain for small water-
sheds than for large watersheds.
CONCLUSIONS
These characteristics of nutrient and pesticide transport
are based on our observations of streams and rivers in the
Lake Erie Basin. It is likely that these same characteristics
will also apply in other regions, although the absolute con-
centrations and loads may be different in other regions
with more erodible soils or different land uses.
We need to keep these characteristics of nutrient and
pesticide transport in mind as we work to improve our
understanding of the offsite impacts of cropland runoff, as
we compare data on offsite impacts from various regions
and establish priorities for control programs, and as we
establish programs to document the effectiveness of crop-
land runoff control programs.
The basic problem of quantifying cropland impacts in
regional water quality is one of accurately characterizing
highly variable systems. Documenting the effects of crop-
Table 6.—Percent of total observed loads exported In various seasons for three long-term transport stations.
Transport station and season
Suspended
solids
Total
phosphorus
Soluble reactive
phosphorus
Nitrate
nitrogen
Maumee River
December-March
April-July
August-November
Sandusky River, Fremont
December-March
April-July
August-November
Honey Creek, Melmore
December-March
April-July
August-November
54
42
4
46
48
6
32
62
6
56
38
6
54
37
9
44
46
10
61
31
7
67
22
11
55
27
17
50
43
6
52
40
18
47
44
10
Table 7.—The role of periods of high flux rates In the export of suspended solids for watersheds of various sizes.
Percent of the suspended solids export over the period of record
associated with fluxes exceeded:
Transport station
Maumee River (16,395 km2)
Sandusky River, Fremont (3,240 km2)
Honey Creek, Melmore (386 km2)
2% of
the time
41.8
52.6
66.4
5% Of
the time
64.2
73.8
80.8
10% Of
the time
79.5
87.9
90.3
20% Of
the time
91.2
96.1
96.8
Table 8.—The role of periods of high flux rates in the export of total phosphorus for watersheds of various sizes.
Percent of the total phosphorus export over the period of record
associated with fluxes exceeded:
Transport station
Maumee River (16,395 km2)
Sandusky River, Fremont (3,240 km2)
Honey Creek, Melmore (386 km2)
2% of
the time
29.6
39.1
45.6
5% Of
the time
50.5
62.4
64.3
10% of
the time
67.6
80.0
79.9
20% Of
the time
83.6
92.0
91.8
206
-------
land BMP's requires detecting management related
changes in the presence of this natural variability.
ACKNOWLEDGEMENTS: This paper was prepared as part of
grants from the Great Lakes National Program Office of the U.S.
EPA, the Soil and Water Conservation Service, pesticide manu-
facturers, and the Soap and Detergent Association.
REFERENCES
Adams, J.R., et al. 1982. A Land Resources Information System
(LRIS) for water quality management in the Lake Erie Basin. J.
Soil Water Conserv. 37: 45-50.
Baker, D.B. 1983a. Studies of sediment, nutrient and pesticide
loading in selected Lake Erie and Lake Ontario tributaries.
Part V. Sediment and nutrient loading summary. U.S. Environ.
Prot. Agency, Region V, Chicago, IL.
1983b. Studies of sediment, nutrient and pesticide
loading in selected Lake Erie and Lake Ontario tributaries.
Part IV. Pesticide concentrations and loading in selected Lake
Erie tributaries—1982. U.S. Environ. Prot. Agency, Region V,
Chicago, IL.
_. 1983c. Herbicide contamination in municipal water
supplies of northwestern Ohio. Heidelberg College, Tiffin, OH.
_. 1984. Fluvial transport and processing of sediments
and nutrients in large agricultural river basins. EPA-600/8-83-
054. U.S. Environ. Prot. Agency, Athens, GA.
_. 1985. Regional water quality impacts of intensive
row-crop agriculture: A Lake Erie Basin case study. J. Soil
Water Conserv. 40(1): 125-32.
Baker, D.B., K.A. Krieger and J.V. Setzler. 1981. The concentra-
tions and transport of pesticides in northwestern Ohio rivers—
1981. U.S. Army Corps. Eng., Buffalo, NY.
Baker, D.B., K.A. Krieger, J.W. Kramer and R.P. Richards. 1985.
Gross erosion rates, sediment yields and pollutant yields for
Lake Erie tributaries: implications for targeting. This volume.
Baker, J.L., H.P. Johnson, M.A. Borcherding, and W.R. Payne.
1979. Nutrient and pesticide movement from field to stream: A
field study/Pages 213-45 in R.C. Loehr, D.A. Haith, M.F. Wal-
ter, and C.S. Martin, eds. Best Management Practices for Ag-
STREAMS AND RIVERS
riculture and Silviculture. Ann Arbor Science Publ. Inc., Ann
Arbor, Ml.
Defiance County Soil and Water District. 1983. Defiance
County-Lost Creek demonstration project 1982 demonstra-
tion report. Defiance County, Ohio.
Honey Creek Joint Board of Supervisors. 1982. Honey Creek
watershed project final program evaluation report 1979-1981.
U.S. Army Corps Eng., Buffalo, NY.
Johnson, H.P., and J.L. Baker. 1984. Field-to-stream transport of
agricultural chemicals and sediment in an Iowa watershed:
Part II. Data base for model testing (1979-80). EPA-600/S3-
84-055. U.S. Environ. Prot. Agency, Athens, GA.
Kramer, J.W., and D.B. Baker. In Press. An analytical method
and quality control program for studies of currently used pesti-
cides in surface waters. In J.K. Taylor and T.W. Stanley, eds.
Quality Assurance for Environmental Measurements. ASTM
STP 867. A. Soc. Test. Mater., Philadelphia.
Leopold, L.B. 1974. Water: A Primer. W.H. Freeman & Co., San
Francisco, CA.
Logan, T.J., D.R. Urban, J.R. Adams, and S.M. Yaksich. 1982.
Erosion control potential with conservation tillage in the Lake
Erie Basin: Estimates using the Universal Soil Loss equation
and the Land Resource Information System (LRIS). J. Soil
Water Conserv. 37(1): 50-5.
McGuinness. J.L., L.L. Harrold, and W.M. Edwards. 1971. Rela-
tion of rainfall energy and streamflow to sediment yield from
small and large watersheds. J. Soil Water Conserv. 26(6):
233-5.
Ohio Environmental Protection Agency. 1979. Initial water qual-
ity management plan: Sandusky River Basin. Columbus.
U.S. Army Corps of Engineers. 1982. Lake Erie wastewater
management study. Final rep. Buffalo, NY
U.S. Environmental Protection Agency. 1979. Methods for
chemical analysis of water and wastes. EPA-600/4-79-020.
EMSL, Cincinnati, OH.
Verhoff, F.H., D.A. Melfl, S.M. Yaksich, and D.B. Baker. 1978.
Phosphorus transport in rivers. U.S. Army Corps Eng., Buf-
falo, NY.
Wauchope, R.D. 1978. The pesticide content of surface water
draining from agricultural fields—a review. J. Environ. Qua). 7:
459-72.
207
-------
Livestock Waste
Management
WHAT DO YOU DO WITH A REGULATION?
MARY BURKE
Chairman, Subcommittee for Water Rights and Resources
National Cattlemen's Association
Cle Elum, Washington
The most obvious step in deciding what to do with a regu-
lation is to determine what the regulation will do with you.
This is important whether you are a private citizen about to
be regulated or a regulator about to adopt a regulation.
Take a quick look at the effect of regulations by asking a
number of questions. Some are outlined in Table 1.
The first question of critical importance in water quality
is what will be the result of the regulation? Is the desired
result the maintenance of existing water quality? What is
the existing quality? Be sure these criteria are defined and
that everyone understands the definition. Much confusion
arose within one committee on how to explain to cattle-
men "land which was rural but not agricultural having no
row crops but only sagebrush, grasses, and cows."
• If the regulation is designed to increase water quality,
how and up to what standard? Will this require the cessa-
tion of existing activities; are you willing as an owner to
cease some of your farming or ranching? Can you fence
all your cattle off from their live water source? Water
freezes at 0°C (32°F) and gets rather solid at -20°F.
Some shoreline acts exempt cattle wintering operations
from permit systems so that cattle have access to running
or live water.
Can you as a regulator show the scientific justification
for the regulation? For example, if you prohibit all row
cropping or timber clear-cutting within a watershed to im-
prove the sediment load level and improve water quality
for a dying fishery, are you certain that forbidding those
activities will give the desired result (more fish)? If the fish
are also being overharvested and diseased, such meth-
ods may be inappropriate. The bottom line should be:
Does the regulation produce the desired product?
Is the regulation possible, technically and administra-
tively? If the regulation reads that one culvert shall be
placed every 273 m (300 yds), there would be five within
every mile; however, if the terrain is broken or if one dis-
charges into an erosive area, perhaps the regulation
should read, "Five culverts should be placed within every
mile." The regulation must be adaptable to the geography,
for the geography will not change to meet a regulation.
Does the regulation conform to the administratively estab-
lished procedures already in place? A great deal of grief
has descended upon the heads of regulators because of a
simple omission for a system of due process or appeal in
the adoption of a regulation. When this occurs in the legis-
lative process, constitutional questions are raised. Re-
sponsible agency people go to laudable lengths to include
existing agency procedures in new regulations, saving
their agencies much public relations and court time.
Does the regulation follow its legal parent? The regula-
tion should cite the antecedent law or act specifically, by
section. The enforcement and appeal-processes should
be set out and, if possible, the legislative intent or desired
result should be given.
Is the regulation compatible with other laws and require-
ments? As a Nation of free people with a strong ethic for
individual civil rights, we have assumed that laws and
regulations are constitutional, contain due process proce-
dures, and assure compensation for private property
taken for public uses. With the invention of "the public
trust doctrine," which purports to precede constitutional
rights, these assumptions are naive if not stupid. The most
controversy and litigation occurs between regulators and
the regulated over this point. In addition to the loss of
property, the failure to address this issue before the regu-
lation is adopted leads to administrative nightmares.
In Washington State it is now illegal to drive a mechan-
ical vehicle through any wetland. How is that compatible
with good farming practices, building settling ponds or
manure lagoons for 208 Best Management Practices, or
with protecting your property "in a responsible and dili-
gent manner" for FEMA or your insurer, not to mention
your banker?
In addition to these conflicts, if farming is precluded, the
209
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
resultant subdivision is even less desirable in trying to
create a "quality environment." Also in the interest of wa-
ter quality for fishable waters, Washington water rights
owners can use only hand-held tools to divert water from
water courses without a 45-day permit. Great Lakes
States concerned with downstream desires for their water
should note that this is the best downstream water steal
yet devised.
The last question, where the most practical problems
arise, is, what is the cost? As a regulator, perhaps you
should balance the first consideration, "does the regula-
tion produce the product," against the man-hours needed
to administer and enforce the regulation. Sometimes
many hours are spent in hearings, courts, and in-house
meetings to regulate the easy activities while the more
difficult problems are ignored. A lack of scientific research
may account for some of these instances. An excellent
example is demonstrated by various interest groups' per-
ceptions of the problem or value of livestock waste. Wes
Jackson, in New Roots for Agriculture, writes that,
Livestock manure is of tremendous value in holding the
soil. Its spongy nature absorbs the blows of the rain and
water itself. Sixteen tons of manure on a 9 percent slope
in Iowa reduced the soil loss in 1 year by over 17 tons. It is
clear that organic matter is meant to be left in the field
and in no way regarded as waste ...
Many environmental groups believe that manure should
never be used when it has the potential for reaching a
waterway or wetland. A recent article in a major Seattle
daily cites small animal keeping operations' failure to
fence in animals as a major cause of shellfish pollution. So
are septic tank failures. These causes of pollution, if true,
lend themselves best to the nonregulatory approach of the
best management practices in the development of State
208 plans under the Clean Water Act. Under this ap-
proach public acceptance is quicker and more supportive,
and administrative cost is much less.
The second cost is to the landowner, the regulated. This
can be a requirement for outlay of capital such as settling
ponds or manure lagoons, lost production, and lost jobs in
secondary industries such as food processing, machinery
sales, the support of rural towns, and so on. There is some
support for strict punitive regulation because it creates
bureaucratic jobs. I have never seen a job for job equality
here, and it must be kept in mind that you are weighing a
production activity against a consuming activity.
The last cost and the most difficult task for both regu-
lated and regulator is to assure that the beneficiary bears
the cost. This problem is thousands of years old but given
new impetus because of the current feeling that the uni-
dentified public has rights to a pure environment superior
to all rights of life, liberty, and property. In the 1500's peas-
ants in France "were forbidden to weed their fields or to
mow hay at certain times of the year for fear that they
would disturb nesting partridges or destroy their eggs"
(Blum, 1982). In 1984 the Washington State Game Depart-
ment prepared a pamphlet, The Path Between Habitat
and Development, that suggests farming practices that
can benefit wildlife, including the "timing of farming oper-
ations to avoid nesting, brood periods; particularly benefi-
cial is delaying the first cutting on alfalfa for 1 to 3 weeks."
Since these regulations provide cost-free benefit to the
hunters, the nobility, or the affluent, and no mention is
made in either case of compensation to the peasants for
deferral or loss of crops, we come to the last—and grow-
ing—problem.
Is there just compensation? Is there any compensation
at all? If a landowner is required to lose a right to use his
property for his productive purposes and that right is given
to others, how do you pay him for that loss? Of course the
public trust theory claims that the public has a superior
right similar to that of the French nobility, but in the United
States is that either defensible, necessary, or useful in
solving our environmental problems?
Perhaps we should ask one more question: is there an
alternative to regulation? Certainly in the management
and uses of livestock waste, a biodegradable product, we
have learned a great deal through the participation of the
public, the regulators, and the regulated-to-be, in the
adoption of 208 nonpoint source programs (the majority of
which are voluntary.)
Voluntary best management practices or cooperative
ventures have many practical advantages over regulation.
In seeking improved water quality, regulators can work
with water and land-owners on a site-specific basis, using
the geography to determine the BMP's that will most rap-
idly improve water quality at the least cost to the land-
owner and the administrative agency. Much of the techni-
cal information gathered had the chance to be tried on the
ground. Some of the potential practices, such as zero
runoff, have the opportunity to be abandoned before be-
ing cast in bureaucratic concrete. Thousands of farmers,
ranchers, and agency persons possess a vast information
and technical base we ought to put to use. We ought not to
regulate our land and people as much as we should be
taught by them.
Table 1.—What do you do with a regulation?
I. What is meant to be the product of the regulation?
A. Maintain the status quo
B. Increase water quality
C. Control activity
D. Eliminate current uses
II. Is the regulation administratively and technically
possible?
A. Will it fit on the ground
B. Will it fit existing administrative procedures
III. Does the regulation follow its legal parent?
IV. Is the regulation compatible'with other laws and
requirements?
A. Constitutional
B. Other laws
C. Management requirements
V. What will be the cost of the regulation?
A. To administer and enforce
B. To the land or water owner
C. In lost production or jobs
D. Do the beneficiaries bear the cost
REFERENCES
Blum, J. 1982. Our Forgotten Past. Thames and Hudson, Ltd.
London.
Jackson, W. 1980. New Roots for Agriculture, Friends of the
Earth. San Francisco.
Washington State Department of Game. 1984. The Path Be-
tween Habitat and Development. State Printing Plant. Olym-
pia.
210
-------
A NATIONAL PERSPECTIVE FOR LIVESTOCK WASTE
MANAGEMENT
RONALD A. MICHIELI
Director, Natural Resources
National Cattlemen's Association
Washington, D.C.
Thirteen years have passed since enactment of PL. 92-
500, the Federal Pollution Control Act of 1972, and al-
though the level of public awareness has increased, pro-
gress in controlling nonpoint source pollution has been
slow, largely because of the wrong perceptions created by
inaccurate and unreliable information.
Policy makers need credible information and data to
make decisions that will evolve into workable manage-
ment plans. Plans that will in turn respond to meet de-
signed objectives—in this case, clean water.
National planning and policy directives many times fall
short of their goal because of a lack of practical and appli-
cable techniques that assure implementation responsive
to local needs.
On the other hand, local decisions many times fail be-
cause of an unwillingness to address the "big picture"
that interacts with the social, political, and economic inter-
ests at the national level.
Simply stated, this means best management practices
should be implemented only when a state or an areawide
agency has properly identified the problem, made an as-
sessment that has examined all of the alternative prac-
tices, and has had appropriate public participation.
Identification of the problem is important. It must recog-
nize that pollution from nonpoint sources can be attribut-
able to a number of activities. Included among these ac-
tivities are:
1. Agriculture—cropland, pastures, rangeland, wood-
lands, and small livestock and poultry feeding operations;
2. Silviculture—forest growing stock, logging, and for-
est road building;
3. Construction—urban or commercial development
and highway construction;
4. Surface mining;
5. Terrestrial disposal of agricultural, industrial, com-
mercial and municipal waste and wastewaters; and
6. Stormwater drainage from urban areas.
Erosion, runoff, and water quality effects must consider
local soil, vegetation, aquatic, geology, hydrology and land
use relationships.
Here is where the "big picture" group collides with the
"local" group. To illustrate this point, let's use the "big
picture" syndrome approach that is used in Washington
when an issue of this magnitude is debated in Congress.
Proponents for controlling nonpoint source pollution
create the perception that millions of acres of our cropland
are being washed away annually "Any day you can stand
on the bridge overlooking the Mississippi River and watch
five farms float away!" What a preposterous exaggeration
of fact! And yet, this exaggeration is what stirs the emo-
tional level of the public into "doing something" to stop
this abuse of our soil and water resources.
The real mission here is to create a proper perspective
to encourage practical solutions to these problems. Using
this example, let us create a more realistic perception
based upon reliable data (Assessing Erosion, 1984). Let's
take a close look at the land erosion classes, their makeup
and some practical solutions:
Noneroslve: about 37 percent (63 million ha) of U.S.
cropland. Its rate of soil erosion will always be less than
4.5 metric tons per hectare per year under any manage-
ment. Operators of 53 percent of such land, some of them
encouraged by Federal programs, use one or more con-
servation practices to control their minimal erosion prob-
lems.
Moderately erosive, but within tolerable levels:
about 40 percent (69 million ha) of U.S. cropland. This
land has the potential to erode above this tolerable level of
4.5 metric tons per hectare per year, but the operators, by
using crop rotations, contour plowing, minimum tillage,
and terraces, keep their erosion below that level.
Moderately erosive, but above the tolerable level:
about 15 percent (25.5 million ha) of U.S. cropland. With
good farm management, this land could also be worked to
keep its rate of erosion below the tolerable 4.5 metric tons
per hectare per year. But the type of management prac-
ticed causes topsoil to wash away, in some places exceed-
ing 22.7 metric tons per hectare per year. About half of the
operators of such land apparently make no effort to stem
their losses by applying conservation practice. This land
and these owners should be targeted for Federal conser-
vation programs.
Highly erosive: about 8 percent (13.4 ha) of U.S. crop-
land. It will erode by more than 4.5 metric tons per hectare
per year with any kind of cultivation. The only way to pre-
vent erosion on this land is to put it in permanent sod or
convert it to another less intensive land use. More than
two-thirds of this land is planted to row crops like corn and
soybeans, which cause serious erosion problems. Fur-
thermore, operators of nearly half of this land have applied
no conservation practices.
These figures hardly support the perception that five
farms per day float down the Mississippi River! My point is
simply this, we cannot afford the luxury of dealing with
misperceptions based on the misapplications of data. Al-
ready we have made too many decisions based upon
faulty information that has cost U.S. taxpayers dearly. Our
current Federal deficit is part and parcel of a misapplica-
tion of taxpayers' dollars for perceived water quality goals
that were never achievable.
Washington, D.C., the mecca of all policy, is a little town
within a big city where a lot of little fish in a big pond
struggle for clean water. The goal is to achieve fishable,
swimmable waters. To achieve that goal, we must balance
the social, political, and economic values innate to our
society, apply the lessons of history, and make every effort
to interact with the various publics whose interests vary
from practical to absurd.
Let us keep in mind that the problem of water pollution
is not new. Richard Graber in his publication, Agricultural
Animals and the Environment, noted that "early travellers
and settlers often experienced difficulty in locating a
stream of potable water because of the activities of large
herds of buffalo. The buffalo found the rivers and streams
211
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
not only a source of drinking water but a haven from the REFERENCES
heat and a source of mud to protect themselves from bit- . ._. ..««.....,..
ina insects As a consenuence the streams flowed richlv Assessing Erosion on U.S. Cropland, Land Management and
ing insects. AS a consequence, tne streams npwea ncmy Physicai Features. 1984. USDA/ERS Ag. Economic Report
wrth manure, unne and mud. This srtuation is acknowl- #51y3 us_ Dept Agrjc Econ Res serv.?Washington, Da
edged in the name the Pawnee Indians gave the Republi- Qraber, R. No date. Agricultural Animals and the Environment.
can River in Nebraska. The English translation is some- Rep. for Feedlot Waste Manage. Reg. Ext. Serv. Oklahoma
thing close to "Buffalo Manure Creek." (Graber, no date). state Univ., Stillwater.
212
-------
ECOPSYCHORRHEA
TOM HOVENDON
Idaho Cattle Association
Boise, Idaho
William Ruckelshaus is a distinguished public servant
who has twice administered our top environmental
agency. Upon his recent departure from public service he
released a comprehensive statement on the need for risk
assessment and risk management policies for the man-
agement of our environmental concerns and regulatory
actions (Ruckelshaus, 1985).
Ruckelshaus outlines the sudden appearance of envi-
ronmentalism on the American scene in the mid-sixties
and Congress' rapid response to regulate matters after
1970. In typical legal jargon Congress wrote laws with
unattainable goals, poor time schedules, conflicting direc-
tions, and absolutely no understanding of the total envi-
ronment. Congress made promises that could not be kept.
Senator Daniel Moynihan (D-N.Y.) said of this, "The mal-
aise of over promising derives almost wholly, in my experi-
ence, from the failure of executives and legislators to un-
derstand what is risked when promises are made . . .
When things don't work out as promised it is all too easy to
suspect that someone intends that they should not." (Em-
phasis added.)
Ruckelshaus feels that after 15 years environmentalism
has changed, that the U.S. Environmental Protection
Agency is getting on a better track—that of perceiving the
big picture as President Nixon described it when he set
about reorganizing the government to cope with the prob-
lems of burning rivers, raw sewage, and brown air.
The big concern now is toxic substance exposure and
carcinogens that brings us into a new area of controversy,
one that raises more questions than we have answers for.
Edith Efron's book, The Apocalyptics, reveals a miserable
track record in the handling of carcinogen research by the
regulatory scientists. Efron's work has drawn high praise,
making one suspect all that is said about science in gov-
ernment hands. One must question the motives of Rachel
Carson, Barry Commoner, Ralph Nader, and others who
predicted such dire happenings in this area 20 years ago,
predictions that time has proven to be false.
This brings us to the topic of this paper: ecopsychor-
rhea. Ruckelshaus referred to the new terms coined for
the American language by the environmental revolution. I
crafted ecopsychorrhea. It comes from three Greek
words: Eco meaning the environment; psycho referring to
the mental processes; and rrhea meaning a continuing
flowing through. Ecopsychorrhea infects those people
who constantly talk about the environment without any
basic knowledge of how it actually works on a continuing
basis.
Ecopsychorrhea was epidemic in the sixties. Brown air
was highly visible, raw sewage in rivers was quite obvious.
Corrections were needed. How well I remember the day
when Dr. Carver of the Idaho Health Department called
me to his office. "Hovenden," he said, "after we get all of
our cities on sewage treatment plants, your cattle feeders
will be next! Got it?" I got it. No big environmental impact
statement or expensive study, just get yourself in gear and
get going. That one-minute conversation changed my life.
Along with the Idaho Cattle Feeders Association I was
committed to finding answers. There were no answers in
1968. By 1969 we were getting calls from the League of
Women Voters and American Association of University
Women. They would inform me that they were "ecolo-
gists" and that a feedlot of 10,000 cattle was the equiva-
lent of a city of 50,000 people with no sewage treatment
plant. These were acute cases of Ecopsychorrhea!
In 1969 the directors of the Cattle Feeders Association
agreed to participate in a feedlot study by the Federal
Water Quality Administration. I helped the engineer desig-
nated for the study to devise a questionnaire that was
mailed to about 120 feedlot operators. Forty-five people
responded. The response was very useful as the answers
came from throughout the principal cattle feeding areas of
Idaho.
The resulting report that we received in July of 1970
was frightening. The recommendations would have re-
moved all of the beef and dairy cattle from the Boise Valley
and other areas with high water tables. I was launched
into a career of finding answers to feedlot pollution.
The greatest set of answers came from the Second In-
ternational Symposium on Livestock Wastes at Ohio State
University April 19 to 22,1971. The program was arranged
by Dr. J. R. Miner of Iowa State University; local arrange-
ments were under the direction of Dr. E. R Taiganides of
Ohio State. At this conference I made valuable contacts
with those two individuals as well as Dr. Tom McCalla,
distinguished microbiologist at USDA's Agricultural Re-
search Service station at the University of Nebraska at
Lincoln. All I know about manure and pollution I have
learned from these men and from John Sweeten of Texas
A&M.
In 1973 the EPA, through its Robert S. Kerr Laboratories
in Oklahoma, approached the American National Cattle-
men's Association with an offer to hold a conference be-
tween the cattle industry, a number of top EPA officials
and members of the scientific community. The agenda
contained the research results of the last 5 years. George
Spencer of the Cattlemen's Association immediately bor-
rowed me from the Idaho Association because our associ-
ation was out in front in seeking answers and working with
research personnel. No other cattle association, including
the national, had an executive with the amount of experi-
ence that I possessed at that time.
Lynn Schuyler of the Kerr Laboratory at Ada, Okla-
homa, and I arranged the program for this Action Confer-
ence. Rep. Morris Udall (D-AZ) accepted our invitation to
be the keynote speaker. From the EPA came Michael
Glenn, Acting Deputy Assistant Administrator for Water
Enforcement; Albert Prinz, Director of the Permit Pro-
gram; and Harold Coughlin of the Effluent Standards Divi-
sion.
Speakers from the academic and research community
included Dr. Paul Taiganides who spoke on the life history
of a river and Dr. Dan Wells of Texas Tech who spoke on
the subject of Manure, How it Works.
We presented a manual to participants and later offered
it for sale to livestock associations as well as individuals
around the country. The second day of the program was a
discussion of the manual. The attendees were divided into
three groups: Arid Areas, the Corn Belt and the South-
east.
On the policy side, we talked of the ultimate require-
ment of controlling the runoff from a storm of 24-hour
duration and 25-year frequency. In the final wrapup, I
asked the representatives from the Corn Belt and South-
213
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
east if they could live with this requirement. In these latter
two areas, annual rainfall exceeds annual precipitation,
giving them a much larger runoff control problem. They
agreed that this was within reason. Those who created
this guideline understood that there was a limit to the
degree of control that could be asked of any individual
operation. It is far more equitable than the "no discharge"
standard adopted by some States. This means zero. We
learned in other controversies that someting as small as
10 parts per trillion was more than zero.
Before all of this information was available, I wrote a
pamphlet for my Idaho people. Entitled Total Retention,
the piece advocated keeping all runoff out of the irrigation
ditches, streams, and rivers. I soon learned that this was
the wrong approach.
Others made errors too. Kansas adopted a rule for feed-
lots that called for cleaning all manure from the pens
every 80 days. It required cleaning the manure right down
to the bare ground. Later research by Dr. Tom McCalla
and others proved that this was the worst thing they could
do. It destroyed the seal between the earth and the or-
ganic material in the manure. This seal prevented most of
the liquids from percolating into the soil underneath the
feedlot.
My position as Action Conference chairman conferred
upon me the responsibility of responding to the Hamilton
Standard study of feedlots. The study was done for EPA in
preparation for issuing feedlot guidelines in the Federal
Register. The scientific community came forth in great
numbers at a special meeting in Lincoln, Nebraska, to
prepare comments. We were able to study one of the few
available copies of the report. I wrote the ANCA's re-
sponse to the proposed guideline based on our group
discussions that day on the Nebraska campus.
I also represented the national association before the
EPA and Congressional committees on the subject of
feedlot guidelines. I drafted the responses and mailed
them to the volunteers on our consulting committee. They
called me with their comments before I delivered the testi-
mony to Washington.
In early 1974 we went through the National Pollutant
Discharge Elimination System program with the EPA. The
Natural Resources Defense Council sued in Federal Court
and obtained a judgment against EPA on the guidelines.
In March 1976, EPA issued new feedlot criteria saying that
"Those feedlots that only discharge in the event of a 25
year or greater storm do not need permits." I commented
to Permit Program Director Albert Prinz at the time of this
new release, "Sounds to me like if you now have a permit
that you no longer need one!" He concurred.
Basically, a feedlot operator with the will to do so can
control runoff. A good educational program is most help-
ful; Bill Ruckelshaus feels that this is a proper role for EPA.
As we have learned through practical experience, a
great amount of the feedlot pollution only existed in peo-
ple's minds. They tend to associate this fermented bypro-
duct with a vulgar four letter word that is used in less than
polite circles. It is this perception that leads to the develop-
ment of ecopsychorrhea.
One of the strongest forces working for us can be peer
pressure, which is 90 to 95 percent effective. Our regula-
tions and police power are only needed for that small per-
centage of people who believe in doing what they like
without regard for their neighbors.
I was appointed to the Idaho Advisory Committee autho-
rized by section 208 of the 1977 Clean Water Act. I served
for the full life of this committee, from 1976 until 1983. For
the last 4 years, I chaired the agricultural subcommittee.
When we started trying to control nonpoint source pollu-
tion, agricultural return flows were the big item in an agri-
cultural State like Idaho. We were told to adopt best man-
agement practices. EPA's instructions called for
mandatory compliance and regulations. I objected, hold-
ing that if we could show our farmers how to keep their
$3,000 per acre land from being washed into the Snake
River they would willingly cooperate. We held steadfast to
this position and the EPA in Region X finally agreed to our
program.
Most of the 208 funds allocated to Idaho were used to
build demonstration projects. A project was created along
the L-Q Drain just west of the city of Twin Falls. The next
year the Balanced Rock Soil Conservation District took
notice of this. Their photographs of the silt-laden water of
nearby Deep Creek (a return flow artery in western Twin
Falls County) in the spring sparked voluntary action.
By the following year this District had many manage-
ment practices in place. They took colored slides of their
projects and of the improvement of the water in Deep
Creek. Voluntary cooperation was working.
In the spring of 1983 the EPA sent a man from Seattle to
meet with our 208 Advisory Committee. He came to praise
us. Idaho was far ahead of any other western State in
developing its nonpoint source program. The praise was
so extensive that it was almost, but not quite, frightening.
Peer pressure and demonstration projects had prevailed.
The people in Idaho are proud of their accomplishments.
Farmers and ranchers can be true environmentalists even
if they don't carry cards in the Sierra Club or similar orga-
nizations.
Our efforts did close down two poorly-run feedlots on
the Payette River. These lots were built on gravel beds. No
one wanted to use manure with a lot of rocks in it. Conse-
quently, the operators simply pushed their manure into the
river for easy disposal. Their actions gave the rest of the
cattle feeders a bad image. They contributed to ecopsy-
chorrhea.
We must continue to work with those great tools of hu-
mankind, education and peer pressure.
When we embarked upon a search for practical solu-
tions to the discharge permit requirements, I realized that
man had lived close to cattle for many thousands of years
and that they had not hurt him. I did not panic at the
thoughts being circulated by people who wanted to com-
pare manure to human wastes. Instead, I agreed that we
could abide by information developed by sound, scientific
investigation. Our great soil scientists and microbiologists
found the answers.
Manure is a friendly product. It arrives on the feedlot at
about 85 percent moisture. It is loaded with bacteria that
immediately begin degrading it, giving it a half-life of 120
days. Coliform bacteria find it hard to survive in such an
environment. After all, manure is only alfalfa and grain that
have been through a time proven fermentation process.
When it is dry it has no odor. It can readily be washed from
one's bare feet or hands.
I love it. It smells like money to me.
Ruckelshaus, William. 1985. Risk, science and democracy.
Page 30 in Issues in Science and Technology. Spring 1985.
ed., National Academy Press, Washington, DC.
214
-------
CONTROLLING WATER POLLUTION FROM NONPOINT SOURCE
LIVESTOCK OPERATIONS
JOHN M. SWEETEN
Agricultural Engineering Department
Texas Agricultural Extension Service
The Texas A&M University System
College Station, Texas
STEWART W. MELVIN
Agricultural Engineering Department
Cooperative Extension Service
Iowa State University
Ames, Iowa
CONFINED CATTLE OPERATIONS
The 1972 Federal Water Pollution Control Act Amend-
ments and the 1977 Clean Water Act created a system of
Federal effluent guidelines, performance standards for
new sources, and permits for point sources that include
cattle feedlots with 1,000-head capacity or more. The
point source water pollution control program encom-
passes most of the 12 million head of cattle on feed for
slaughter in the Nation. Since the early 1970's, virtually all
the large cattle feedlots have installed water pollution
abatement systems that prevent discharge of pollutants.
These feedlots now comply with the Federal and State
regulations. Feedlot runoff holding ponds must be de-
signed to collect all runoff from the 25-year frequency, 24-
hour duration rainfall event or process-generated waste-
water, followed by land disposal by irrigation or
evaporation.
UNCONFINED CATTLE OPERATIONS
Approximately 100 million cattle are raised on 360 million
ha (900 million acres), or about 40 percent of the land area
in the United States, by hundreds of thousands of individ-
ual producers. The size of cow-calf herds varies with re-
gion, but averages less than 100 cows nationwide on pas-
tures or rangeland. This unconfined cattle production
accounts for half the almost 109 million metric tons (120
million dry tons) of animal manure generated each year in
the United States. Cattle stocking densities on range and
pasture lands vary by 1000-fold or more, from up to 10
head/ha with temporary grazing of irrigated, fertilized pas-
tures and grainfields, to an average of only 1 head/160 ha
on certain arid native rangelands (Sweeten and Reddell,
1978).
Unconfined cattle operations with these stocking densi-
ties clearly satisfy part of the definition of nonpoint
sources as "diffuse or multiple outlets," but in many cases
it is doubtful whether the presence of unconfined cattle on
range or pasturelands can be discerned from natural or
background levels of water quality parameters.
Research in recent years has determined the effects of
cattle grazing operations on runoff quantity and water
quality in streams. This research has determined that un-
confined livestock production is an environmentally sound
water quality management practice.
The most common change in stream water quality from
unconfined livestock production is elevated counts of indi-
cator bacteria and sediment concentrations (Milne, 1976;
Saxton et al., 1983). Chemical pollutant concentrations
are sometimes increased slightly, but they seldom exceed
stream quality standards. The available data have shown
that any detectable water pollution from unconfined cattle
operations may not be related to cattle numbers or ma-
nure quantity involved, but rather to hydrogeological fac-
tors that contribute to rapid surface runoff or sediment
movement (Dixon, 1983).
AGRICULTURAL NONPOINT SOURCES
Agricultural nonpoint sources of water pollution, under
section 208 of the Federal Clean Water Act, include runoff
from dryland and irrigated cropland, livestock production
on range and pastureland, small-scale livestock confine-
ment facilities, and manure disposal areas. Water pollu-
tants from agricultural nonpoint sources may include sedi-
ment, nutrients, salts, organic matter, certain indicator
bacteria, or pesticides.
Pollutants from nonpoint sources are entrained and
conveyed by rainfall runoff, and may not be traceable to
individual operations. In those areas where agricultural
activities contribute a major share of the nonpoint source
pollution load, the vastness of the land areas involved
relative to other land uses, rather than acute problems, is
primarily accountable.
Nonpoint source pollutants can generally be controlled
by management techniques known as best management
practices (BMP's) instead of wastewater treatment meth-
ods. BMP's reduce the volume and concentration of runoff
from nonpoint sources. For example, BMP's for cropland
include maintaining vegetative cover, conservation tillage,
furrow diking, contour plowing, and terracing to conserve
rainfall and reduce loss of soil, nutrients, and pesticides in
runoff. These practices save the farmer or rancher both
cash inputs and irreplaceable resources. Using marginal
croplands for hay production or cattle grazing is also con-
sistent with good soil and water conservation practices,
and with nonpoint source pollution abatement principles.
CATTLE GRAZING OPERATIONS AS
NONPOINT SOURCES
Watersheds containing cattle grazing sometimes show in-
creased concentrations in adjacent streams of bacterial
indicator organisms, primarily coliforms and streptococ-
cus (Dixon, 1983; Milne, 1976). Reported effects, how-
ever, are erratic and detectable only for short distances
downstream. Fecal deposits along drainage ways may
contribute a disproportionate share of the bacteria from
grazed watersheds. Often the effects of cattle are indistin-
215
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
guishable from the effects of wildlife within the watershed
(Doran et al. 1881; Dixon, 1983).
Water quality standards for coiiform indicator orga-
nisms, developed for point sources, have questionable
value for assessing water quality effects of cattle grazing
operations, according to U.S. EPA research on grazing
watersheds (Doran et al. 1981; White et al. 1983; Saxton
et al. 1983). The amount of nitrogen and phosphorus car-
ried away each year from grazed pastureland (Tables 1
and 2) is far less than from feedlots (Loehr, 1974; Doran et
al. 1983). In fact, runoff from livestock pastures often does
not exceed nutrient levels in runoff from ungrazed pas-
turelands, forests, dryland farms, or even precipitation
(Saxton et al. 1983). Nutrient losses and runoff amounts
are usually greater for overgrazed pastures than for prop-
erty managed grazing systems.
Within a pasture, manure deposits may cover only 1-20
percent of the surface area depending on stocking density
and duration (Sweeten and Reddell, 1978). Dung deposits
are greatest on cattle bedgrounds and resting areas,
which typically are on well-drained soils (Powell et al.
1983). Direct movement of dung deposits into stream
channels is minimal because standing vegetation and
grass litter serve as filters.
Unconfined livestock may decrease vegetative cover
and increase runoff, erosion, and transport of sediment,
plant nutrients, and oxygen demand. At high-impact feed-
ing and watering sites, sediment load can be minimized by
management practices that include protecting fragile
stream banks, maintaining vegetative cover, low or moder-
ate stocking levels, distributing salt and water, and provid-
ing feed, salt, or water away from streams.
Small feedlots (below 1,000 animal units or beef cattle
equivalents) are regarded as nonpoint sources of pollution
in most states. The distinction between feeding operations
that are point sources and those that are nonpoint sources
varies among States, but it is typically based on number of
head, animal spacing, proximity to streams, and likelihood
of wastewater discharge.
BMP's for nonpoint source water pollution control at
small feedlots include:
1. Locating the feeding facility away from a stream or
drainage channel;
2. Diverting outside runoff away from the feedlot sur-
face using diversion terraces and roof gutters;
3. Collecting solids carried off the feedlot surface by
runoff water; solids should be settled out in channels, de-
bris basins, or grass waterways where they can be re-
moved and disposed of properly on land;
4. Installing a grass filter strip at least twice as large as
the feedlot, where a small feeding site is close to a water-
body, to improve runoff quality before it enters the water;
5. Installing a runoff holding pond if the water quality
risk is high and the location of a feedlot prevents the use
of a vegetated filter. The collected runoff should be dis-
posed of by irrigation onto nearby crop or pasture land;
and
6. Making the best use of nutrients in the manure to
improve the soil's physical properties by applying manure
to cropland.
A great deal of research has been conducted to help cat-
tlemen and farmers properly use manure. The proper ma-
nure application rate is normally determined based on soil
and plant requirements for nitrogen (Gilbertson, 1983).
Phosphorus and salt content are sometimes limiting fac-
tors also.
Manure application rates are usually about 22 metric
tons/ha/yr (10 tons/acre/yr) for irrigated corn, grain sor-
ghum, wheat, vegetables and hay crops, and about half
this amount or 12 metric tons/ha/yr (5-6 tons/acre/yr) for
dryland crops. Application rates should be selected based
on soil sample and manure nutrient analysis, using the
advice of a professional agronomist.
Solid manure should be spread evenly with a spreader
truck or tractordrawn box spreader and disked into the soil
promptly to conserve nutrients. Soil injection of liquid ma-
nure will conserve nutrients and prevent runoff, odors,
and flies. Site-specific factors influence the type of liquid
manure application equipment that is most advantageous.
Tatolo 1. — Annual ylold and concentrations of nitrogen and p
Total nitrogen
Souro®
Precipitation
Forested land
Cropland runoff
Irrigated cropland in western U.S. (surface flow)
Urban land drainage
Feedlot runoff
Ib/acire
5.0-8.9
2.7-11.6
0.1-11.6
2.7-24.1
6.3-8.0
89.3-1430
PPM
1.2-1.3
0.3-1.8
9
0.6-2.2
3
920-2100
hosphorus1.
Total phi
Ib/Bcro
0.04-0.05
0.03-0.8
0.05-2.6
0.9-3.9
1.0-5.0
8.9-554
»sphonus
PPM
0.02-0.04
0.01-0.11
0.02-1.7
0.2-0.4
0.2-1.1
290-360
'Data do not reflect extreme values caused by improper waste management or extreme storm conditions.
'Parts per million
Source: Loehr, 1974.
Table 2.—Average annual nutrient yields In runoff from some pastureland.
Location
Nebraska
Oklahoma
Ohio
Minnesota
Management system
Rotation graz.
Rotation graz.
Continuous graz.
Rotation graz.
Rotation graz.
Rotation graz.
Prairie
Total nitrogen
Ib/acre
2.5
1.7
8.7
1.9
4.6
0.5-2.9
0.7
Total phosphorus
Ib/acire
0.62
0.17
4.11
1.16
0.04
0.3-1.15
0.10
Source: Doran, Schepers and Swanson, 1981.
216
-------
LIVESTOCK WASTE MANAGEMENT
Livestock producers and farmers need to leave a vege-
tated buffer strip of 33 m (100 ft) or more while surface-
applying manure near streams, lakes, or drainage struc-
tures to prevent direct runoff into a waterbody (Gilbertson,
1983). If land is frozen or snow covered, manure generally
should be applied on predominantly flat ground to mini-
mize direct runoff. In spreading manure, avoid permeable
soils with high water table, shallow creviced bedrock, wa-
terways, floodplains, wet soils, and excessive application
rates.
BMP'S FOR CATTLE ON RANGE OR
PASTURE
EPA's 1984 report to Congress recommended several
BMP's for nonpoint source pollution control for unconfined
cattle production. They include (U.S. EPA, 1984):
1. Adopt an effective erosion control program;
2. Tailor the grazing programs and stocking rates to the
microclimate, soil, vegetation, topography, and geology of
the particular area;
3. Locate necessary animal holding pens or high-den-
sity grazing at hydrologically remote places (away from
streams);
4. Disperse feeding facilities, watering sites, and shel-
ters to reduce manure accumulation, soil compaction, and
erodible paths;
5. Maintain grass cover downslope from sites where
animals congregate and along stream banks to provide a
vegetative filter;
6. Maintain good forage and ground cover to decrease
volume and rate of runoff, prevent erosion, entrap manure
and other organic matter, and use fertilizer nutrients; and
7. In special situations, consider more drastic mea-
sures, including (a) using tillage to break up or incorporate
manure deposits, (b) modifying runoff drainage pathways,
or (c) restricting animal access to critical areas.
FEDERAL AND STATE PROGRAMS FOR
NONPOINT SOURCE POLLUTION
CONTROL
In a 1978 review of water pollution effects of unconfined
animal production EPA recommended against any regula-
tory programs that would discourage or restrict livestock
production on pasture or rangeland, because of the low
level of water pollution associated with unconfined animal
production (Bobbins, 1978). Successful agricultural non-
point source water pollution strategy requires clearly de-
fining solvable problems within a specific watershed and
targeting specific sites for BMP application (Duda and
Rnan, 1983).
The 1984 EPA report to Congress favored management
of nonpoint source pollution control at the State level in-
stead of a national strategy that cannot be flexible enough
to identify priority water quality problems nor target man-
agement control efforts at those problems.
Barriers to BMP adoption by ranchers and farmers can
be categorized as economic, educational, and institu-
tional. Ways to increase BMP adoption include:
1. Demonstrate the economic viability of each practice;
2. Provide cost-sharing incentives for those water pollu-
tion control practices that are not economical. (19 States
now offer their farmers cost sharing incentives to adopt
conservation measures or BMP's);
3. Provide educational programs through existing
agencies, such as the Agricultural Extension Service, Soil
and Water Conservation Districts, and producer groups;
and
4. Provide stable agricultural policies that encourage
investment in pollution control and resource management
projects.
SUMMARY AND CONCLUSIONS
Instead of producing surplus grain on marginal land with
increased potential for erosion and water degradation, cat-
tle operations recycle nutrients back to land to better man-
age the soil. Cattlemen can use alternative BMP's to in-
crease forage and beef production while preventing
nonpoint source water pollution. When good management
is practiced, cattle production enhances environmental
quality and is an important asset to this Nation's soil and
water resources.
REFERENCES
Dixon, J.E. 1983. Controlling water pollution from cattle grazing
and pasture feeding operations. Pages 1-7 in J.H. Smits, ed.
Profit Potential of Environmental Protection Practices of Cat-
tlemen. Natl. Cattlemen's Assoc., Englewood, CO.
Dixon, J.E. et al. 1983. Comparison of runoffquality from cattle
feeding on winter pastures. Trans. Am. Soc. Agric. Eng. 26(4):
1146-9.
Doran, J.W., J.S. Schepers, and N.P. Swanson. 1981. Chemical
and Bacteriological Quality of Pasture Runoff. J. Soil and Wa-
ter Conserv. 36(3): 166-71.
Duda, A.M., and D.S. Finan, 1983. Influence of Livestock on
Nonpoint Source Nutrient Levels in Streams. Trans. Am. Soc.
Agric. Eng., 26(6): 1710-1716.
Gilbertson, C.B. 1983. Controlling Water Pollution from Land
Application of Cattle Manure. Pages 9-13 in J.H. Smits, ed.
Profit Potential of Environmental Protection Practices of Cat
tlemen. Natl. Cattlemen's Assoc. of Englewood, CO.
Loehr, R.C. 1974. Characteristics and Comparative Magnitude
of Nonpoint Sources. J. Water Pollut. Cont. Fed., 46(8): 1849.
Milne, C.MTT976. Effect of a Livestock Watering Operation on a
Western Mountain Stream. Trans. Am. Soc. Agric. Eng. 19(4):
749-52.
Powell, J., F.R. Crow, and D.G. Wagner. 1983. Rangeland Wa-
tershed Water Budget and Grazing Cattle Waste Nutrient Re-
cycling: Project Summay. EPA-600/S2-83-017. U.S. Environ.
Prot. Agency. Ada, OK.
Robbins, J.W., 1978. Environmental Impact Resulting from Un-
confined Animal Production. EPA-600/2-78-046. U.S. Environ.
Prot. Agency. Ada, OK.
Saxton, K.E., L.F. Elliott, R.I. Papendick, M.D. Jawson, and D.H.
Fortier. 1983. Effect of Animal Grazing on Water Quality of
Nonpoint Runoff in the Pacific Northwest: Project Summary.
EPA-600/SZ-82-071. U.S. Environ. Prot.Agency Ada, OK.
Sweeten, J.M., and D.L. Reddell. 1978. Nonpoint Sources:
State-of-the-Art Overview. Trans. Am. Soc. Agric. Eng., 21(3):
474-83.
U.S. Environmental Protection Agency. 1984. Report to Con-
gress; Nonpoint Source Pollution in the U.S. Environ. Prot.
Agency, Off. of Water Program Oper., Washington, DC.
White, R.K., R.W. Van Keuren, L.B. Owens, W.M. Edwards, and
R.H. Miller. 1983. Effects of Livestock Pasturing on Nonpoint
Surface Runoff: Project Summary. EPA-600/S2-83-011. U.S.
Environ. Prot. Agency, Ada, OK.
217
-------
National Fertilizer Development Center
Tennessee Valley Authority
Department of Animal and Dairy Sciences
Auburn University
Department of Agricultural Engineering
Auburn University
L L BB
National Fertilizer Development Center
Tennessee Valley Authority
Recycling livestock and poultry waste through biological
nafeeding systems or through energy recovery systems
may provide an economic incentive for farmers to con-
sider alternative waste management systems (Martin and
Madewell, 1971). This paper briefly describes several re-
feeding/energy recovery systems including the use of
poultry house waste (broiler litter) in several beef cattle
production systems, and recycling swine wastes through
an aquaculture production (algae-fish-water chestnut-
biogas) system.
Ruminant animals such as cattle have no parallel in the
role of scavenging and, consequently, no parallel in re-
source recovery from agricultural wastes. Their unique
digestive system includes a microbial fermentation stage
that enables crop residues, certain industrial byproducts,
and livestock and poultry wastes to be used as feedstuffs
and converted into meat. The Food and Drug Administra-
tion leaves the regulatory responsibilities for this practice
to individual States. However, those States that have ap-
proved livestock and poultry wastes as feedstuffs for cattle
have developed regulatory standards similar to the model
regulations for processed animal waste products as ani-
mal feed ingredients, available from the Association of
American Feed Control Officials (Minyard, 1978).
Broiler litter is a mixture of broiler manure, bedding ma-
terial, waste feed, and feathers. Wood shavings, sawdust,
and peanut hulls are the main bedding materials used in
broiler houses in the Southeast. The kind of bedding ap-
parently does not affect the quality of litter removed from a
broiler house. However, broiler litter from different houses
and management systems varies in nutrient content.
The Alabama Cooperative Extension Service and the
Tennessee Valley Authority collected and analyzed litter
samples from 31 broiler houses in north and central Ala-
bama (Table 1). Average crude protein content of litter
samples was 23.9 percent, which is not as high as the
percent quoted by other States (Fontenot, 1978). A greater
loss of nonprotein nitrogen (NPN), caused by higher hu-
midity and heat, may account for the lower average. NPN
protein equivalent was 5.7 percent on a dry weight basis.
Thus, the NPN makes up only 24 percent of the total
crude protein. Litter is a rich source of calcium and phos-
phorus. Trace minerals are also present in more than ade-
quate amounts for cattle when broiler litter is fed properly
When litter makes up more than 30 percent of a ration,
minerals need not be added.
The Total Digestible Nutrients (TON) found in broiler
litter indicates a relatively low-energy feed (Ruffin and
Martin, 1981). Calculated TON values range from 26 to 64
percent. Even dry brood cows need more energy from
such grain as corn and wheat when wintered on broiler
litter. The average TDN level of litter is similar to average
quality hay grown in the Southeast.
Broiler litter should be processed so that harmful agents
like salmonella and coliform bacteria are destroyed (Ruffin
and Martin, 1981). Processing litter will also increase its
acceptability to cattle. Most on-farm processed litter is
deep stacked in a shed or stored outside and covered with
heavy duty polyethylene. It should be stacked at least 2 to
2.5 m (about 6-8 ft) deep so that heat will destroy poten-
tial pathogens. Usually after ensiling 4-6 wks the litter is
ready for feeding. Properly stored litter will lose its typical
manure smell and will be much more acceptable to the
cattle. Broiler litter can also be processed in a pit or bunk
silo.
Table 1.—Nutrient content off broiler litter from 31 broiler
houses In north and central Alabama (Ruffffin, 1978).
Composition
Dry matter
Composition of dry matter
TDN (calculated)
Crude protein
Crude fiber
N-P-N (protein equiv.)
Ash
Calcium
Phosphorus
Potassium
Magnesium
Sulfur
Copper
Arsenic
78.3
55.0
23.9
26.9
5.7
21.5
2.1
1.6
1.7
0.44
0.21
0.036
0.0036
69-84
26-64
13-31
14-46
1.0-11
10-47
1.0-3.5
1.1-1.9
1.3-2.1
0.3-2.1
0.1-0.41
0.0011-0.060
0.0018-0.0062
218
-------
LIVESTOCK WASTE MANAGEMENT
Table 2.—Suggested rations including broiler litter.
Ration Number
Ingredients
1
Broiler litter
Cracked yellow corn
Ground limestone
% Corn/% litter
Dry matter
TON
Crude protein
Crude fiber
Calcium
Phosphorus
364
90
20/80
80.5
62.6
18.1
21.2
1.60
1.30
L
300
154
34/66
'9
227
227
50/50
• calculated analysis1
82.2
68.3
16.4
17.2
1.27
1.11
83.8
73.8
14.7
13.6
0.96
0.93
157
295
2
66/34
85.4
78.9
13.1
9.9
0.78
0.74
'Based on data presented In Table 1.
Table 2 suggests rations of varying broiler litter-corn
mixtures. These should be used only as a guide because
the nutrient levels in broiler litter vary (Ruffin and Martin,
1981). Ration No. 1 is calculated for use as the major
ration for dry beef cows. Hay or some other roughage
should be provided to maintain normal rumen function.
About 1 kg (2.2 Ibs) of long hay fed every 2 or 3 days will
be adequate. A 454-kg (1,000-lb) dry brood cow will need
about 9-11 kg (20-24 Ibs) of ration No. 1 for maintenance
during winter months. Corn mixed with broiler litter should
be cracked or ground.
The No. 2 ration is formulated for lactating brood cows.
About 11 kg (24 Ibs) daily will furnish adequate nutrients
during the winter months. Some long hay, as with ration
No. 1, will be needed for normal function of the rumen.
The No. 3 ration is formulated for growing stacker cat-
tle. Stacker cattle weighing about 227 kg (500 Ibs) will
consume abut 6.3 kg (14 Ibs) of this ration. Healthy stack-
ers that have been wormed, vaccinated, implanted, and
otherwise managed as recommended by the Alabama Ex-
tension Service should gain 0.9 kg (2 Ibs) or more daily
consuming this ration.
The No. 4 ration can be fed to cattle weighing about
341 kg (750 Ibs) or more. Consumption should be about
11-13 kg (25-28 Ibs) daily for maximum gain. Long hay or
a small amount of oat or wheat straw will maintain normal
rumen function for cattle consuming finely ground rations
such as the No. 4 mixture.
EFFECTS OF FEEDING SYSTEMS ON
CHEMICAL CONTENT OF
CATTLE MANURE
The precise composition of cattle manure varies accord-
ing to the type of cattle operation and particular feeding
system. Broiler litter contains much higher concentrations
of elements such as phosphorus, potassium, and calcium
than are required by cattle; therefore, its use in feeding
generally increases the fertilizer value of the cattle ma-
nure.
Nine typical on-farm feeding systems (five of which in-
clude broiler litter for four different types of cattle opera-
tions) are given in Table 3 (Ruffin and Martin, 1983). Ma-
nure samples from these farms were air-dried to simulate
pasture drying conditions, then oven-dried and analyzed
for nitrogen, phosphate, potash, calcium, and magne-
sium.
In three of the systems on Farms 1 through 7 represent-
ing brood cows, replacement dairy heifers, and growing
stacker cattle, the manure was deposited on the land by
the grazing animals. Results in Table 4 indicate that ma-
nure from cattle on rations containing broiler litter contrib-
uted more to the fertility of a pasture than did manure from
conventional rations. Obviously, part of the manure was
concentrated in loafing areas. Within each system, ma-
nure from cattle consuming broiler litter contained more
phosphate than did the manure from cattle on hay or si-
lage. For example, manure from brood cows on a 20 per-
cent corn plus 80 percent broiler litter ration (Farm 1) con-
tained more than double the amount of phosphate than
did the manure from cows fed good quality coastal bermu-
dagrass hay (Farm 3) and three times more phosphate
than the manure from cows fed low-quality fescue hay
(Farm 4). The phosphate content of manure from feedlot
cattle on a 66 percent corn plus 34 percent broiler liner
ration was 2.98 percent; the phosphate content of feedlot
manure from a conventional corn silage ration was 1.53
percent.
Table 4.—Effects of different cattle feeding systems on
plant nutrient content of manure.
Farm
Number
1'
21
3
4
51
6
71
81
9
isomp
N
1.45
1.31
1.12
1.32
1.21
1.84
1.72
1.93
2.03
losmon OT dry came manu
P2O, K2O Ca
Brood cow maintenance
2.13 0.71 1.76
1.17 0.84 1.20
1.01 0.84 0.83
0.66 1.07 0.92
Replacement dairy heifers
1.19 0.61 1.20
Growing stocker cattle
1.28 1.16 0.82
1.65 1.70 0.85
Finishing slaughter cattle
2.98 2.56 1.49
1.53 1.97 0.71
re, w
Mg
0.30
0.30
0.22
0.31
0.33
0.52
0.27
0.51
0.43
'Contained broiler litter.
Farm
Number
Table 3.—Cattle feeding systems In north and central Alabama.
Type of Cattle
Operation Feeding Systems
1
2
3
4
Brood cow maintenance
20% corn + 80% litter
20% wheat + 80% litter + wheat
pasture + coastal bermudagrass hay
Coastal bermudagrass hay (good quality)
Fescue hay (poor quality)
Replacement dairy heifers
50% corn + 50% litter + hay + fescue pasture
6
7
Growing stocker cattle
Wheat pasture + cottonseed + hay
50% corn + 50% litter + fescue pasture
8
9
Finishing slaughter cattle
65% corn + 35% litter
Corn silage + cottonseed meal
219
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
A 5-year study was conducted at Muscle Shoals, Ala-
bama, of the recovery of plant nutrients from swine ma-
nure in an integrated algae-fish-water chestnut produc-
tion system (Behrends et al. 1982, 1983; Maddox et al.
1982). Swine manure fertilizes the algae; filter-feeding fish
consume the algae; and effluent from the fish pond sup-
plies nutrients to a water chestnut bed. One phase of this
study also included biogas production from swine manure
in an anaerobic digester; the digester waste then fertilized
the aquatic farming system.
The swine manure is flushed directly into the fish-algae
culture ponds with virtually no odor because enough oxy-
gen is maintained by the growing algae. A specified level
of manure loading rates assures proper oxygen levels for
growing and maintaining fish. In static water systems
where water is added only to replace evaporational/seep-
age losses, manure loading rates should not exceed
40 kg/ha/day (dry matter basis). The flowing water system
should maintain a waterflow rate sufficient to replace all
fish water in 10 days, giving a flow rate of 655 L/ha (70 gal/
acre) min continuously. The waste from 5,000 kg of pigs
applied to a hectare 91 cm deep (11,000 Ib/acre at 3-ft
depth) at this flow rate should stimulate a lush growth of
algae which will provide adequate feed for filter-feeding
fish.
Fish such as tilapia (77/ap/a spp.) can be grown in this
type system. They should be stocked into the aquatic
farming system at a density equivalent to 10,000 to
15,000/ha (3,240 to 4,080/acre) with fish weighing 30-
80 g (1-2 oz) each. Accessory species, such as silver
carp (Hypophthalmichthys molitrix) and bighead carp (Aris-
tfchthys nobilis), can be stocked at a rate equivalent to
2,500/ha (1,000/acre). Silver carp can be substituted for
tilapia in this stocking ratio if they are the principal fish
grown in the system, and bighead carp can be stocked up
to 25 percent of the total population. Annual fish yields of
5,600-7,850 kg/ha (5,000-7,000 Ib/acre) can be expected
from this system if the proper size fish are stocked and
grown for 150-180 days. Tilapia cannot survive water tem-
peratures below 13°C (55°F) but may be overwintered in
power plant water with raceway facilities or in artesian
springs where water temperature is at least 15°C.
Wide diurnal fluctuations in dissolved oxygen and pH
make survival impossible for many of the pathogenic bac-
teria such as Salmonella spp. (Baker, 1981). Thus fish
grown in manure-fertilized systems are free of many of the
pathogenic bacteria associated with the manure.
Water discharged from the fish pond can be irrigated
onto a sand bed filter to grow Chinese water chestnuts
(Eleocheris dulcis) with an area equal to approximately
one-half the water surface area of the fish system (1:2
ratio). Chinese water chestnuts should be planted on 51-
cm (20-in) spacings early in the spring, and the bed should
be flooded 5-10 cm (2-4 in) deep. Plant tops (shoots) can
be cut and baled for hay after the first frost. The water
chestnuts can be harvested with modified root harvesting
equipment from the dry fields. Yields of 15 metric tons/ha
(6.7 tons/acre) of dry hay and nearly 40 metric tons/ha
(17.8 tons/acre) of water chestnuts can be expected. The
quality of the water leaving the sand beds would meet
tertiary wastewater treatment standards during the grow-
ing season, thus minimizing the pollution potential of the
swine wastes.
The water chestnut hay is suitable for cattle feed and
the water chestnuts can be sold for gourmet cooking.
Also, the water chestnut is a sugar and starch crop suit-
able for animal feed or use as a possible source of carbo-
hydrates for alcohol fermentation.
Anaerobic digester waste has proved to be suitable as
an aquatic fertilizer. Reduced oxygen demand of the
waste as a result of the pretreatment process permits
higher waste loading to the aquatic system and reduces
the land area required to recover and treat a given amount
of swine waste.
Baker, D.A. 1981. Longevity of Salmonella typhimuriuun in 77/a-
p/a aurea and water receiving swine waste and the antigenic
response of the fish. Master's thesis. Auburn Univ. AL.
Behrends, L.L., J.B. Kingsley, J.J. Maddox, and E.L. Waddell, Jr.
1982. Integrated agriculture/aquaculture systems. Pap. No.
5031. Pres. 1982 summer meet. Am. Soc. Agric. Eng. June
27-30. Univ. Wis. Madison.
1983. Fish production and community metabolism in
an organically fertilized fish pond. J. World Maricul. Soc. 14:
510-22.
Fontenot, J.P. 1978. Broiler litter as a feed ingredient for rumi-
nants. Pages 34-9. in Alternate Nitrogen Sources for Rumi-
nants. Proc. Conf. Atlanta, GA. Nov. 9-11, 1977. Bull. Y-130.
Nat. Fertilizer Develop. Center, Tenn. Valley Author. Muscle
Shoals, AL.
Maddox, J.J. et al. 1982. Optimization of biological recycling of
plant nutrients in livestock wastes by utilizing was" heat from
cooling towers. EPA-600/7-82-041. U.S. En .on. Prot.
Agency, Nat. Tech. Info. Serv. Springfield, VA.
Martin, J.B., and C.E. Madewell. 1971. Environmental and eco-
nomic aspects of recycling livestock waste products. South-
ern J. Agric. Econ. 137-42.
Minyard, J.P, Jr. 1978. Regulation of recycled wastes as animal
feeds in Mississippi. Pages 82-5 in Alternate Nitrogen
Sources for Ruminants. Proc. Conf. Atlanta, Ga. Nov. 9-11,
1977. Bull. Y-130. Nat. Fertilizer Develop. Center, Tenn. Valley
Author. Muscle Shoals, AL.
Ruffin, B.G. 1978. Broiler litter in cattle feed. Pages 69-71 in
Alternate Nitrogen Sources for Ruminants. Proc. Conf. At-
lanta, Ga. Nov. 9-11. 1977. Bull. Y-130. Nat. Fertilizer De-
velop. Center, Tenn. Valley Author., Muscle Shoals, AL.
Ruffin, B.G., and J. Martin. 1981. Feeding broiler litter to beef
cattle. Circ. ANR-280. Ala. Coop. Ext. Serv. Auburn Univ. AL.
1983. Fertilizer value of cattle manure. Circ. ANR-
243. Ala. Coop. Ext. Sen/., Auburn Univ. AL.
220
-------
Nonpoint Programs:
The Status
NONPOINT SOURCE CONTROL PROGRAMS
CHARLES L. BOOTHBY
Executive Vice President
National Association of Conservation Districts
Washington, D.C.
With the passage of the Clean Water Act in 1972, conser-
vation districts decided to become involved in water
cleanup activity partly because of concerns that controls
on agriculture would probably result, and partly from a
belief that conservation districts were an appropriate
agency to handle such controls.
Conservation districts and State soil conservation agen-
cies became involved in the development of 208 plans
and served on the committees considering agricultural,
forestry, construction site, and other types of nonpoint
source pollution control. The conservation districts pro-
vided a link to cooperating landowners. When special pro-
gram funds became available from various sources, the
district could activate to get the job done. A case in point is
money from the Clean Lakes Program. In most instances
the conservation district has served as a link to land-
owners when land treatment was involved.
In most States, the State soil conservation agency or
conservation district has been the designated manage-
ment agency for agricultural nonpoint source control pro-
grams. With encouragement from the States, most con-
servation districts were ready to assume active roles in
1978. Then, as concern about nonpoint sources lessened
in the Federal government, local interest declined as well.
Over these past 2 years, interest has increased, but not to
the degree that was present 4 years ago.
As an association representing its member conserva-
tion districts, NACD has been conducting various studies
as a means of informing districts about success stories in
other States and how to become involved in different pro-
grams. We have prepared papers on conservation district
involvement in the Clean Lakes Program, Hazardous
Waste siting, and, most recently, on Groundwater: The
Hidden Resource.
About 3 years ago, a great need existed for sharing
information about various methods of conservation tillage.
Definitions were not uniform and farmers depended totally
upon sales people for advice. At the urging of the ag
chemical industry, and with a challenge grant from a pri-
vate foundation, NACD established the Conservation Till-
age Information Center. The Center was initially estab-
lished in Washington, DC, with a field office in Ft. Wayne,
Indiana. Subsequently, the entire operation was shifted to
the Ft. Wayne office. The purpose of the Center is to share
available information among those who work directly with
farmers, to put people with questions in touch with those
with answers. The Center does not attempt to reach the
individual farmer, but works with those multipliers of infor-
mation, the Extension Agent, the SCS Soil Conservation-
ist, and other agency people.
The Center does not endorse any particular products,
either chemical or machine. Additionally, it shares infor-
mation about failures, as well as successes. By maintain-
ing an objective approach to issues, it is maintaining credi-
bility with its clients and with its supporting industries. This
credibility is important, as industry is presently supporting
the Center at the rate of about $160,000 per year.
Conservation tillage has been accused of being respon-
sible for increasing the use of chemicals in agriculture.
Although the overall use may be somewhat higher, the mix
of chemicals is different and the management of those
chemicals is, overall, much better. This improved manage-
ment should lead to fewer environmental problems in the
future. Conservation tillage reduces erosion, thus reduc-
ing sedimentation in streams, and also reduces the
amounts of adsorbed fertilizers and pesticides which enter
the waterways.
This year we have been studying the potential need for
a data center comparable to the Conservation Tillage In-
formation Center for the tracking and management of
Best Management Practices (BMP's) in nonpoint source
pollution control programs. If it is found that such a data
221
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
center and information transfer system would benefit local
program managers we will attempt to establish such an
enterprise.
It has been interesting to note the increasing interest
among State governments to control erosion and thus as-
sist with nonpoint source control efforts. Several States
have enacted cost-share programs and provided funding
for them.
In Kansas the Water Resources Cost-Share Program is
designed to reduce pollution from farm wastes, nutrient
loading of surface waters, sediment, and toxic pollutants.
The program provides technical and financial assistance
to landowners for construction of conservation measures.
To initiate this program, an amendment to the State Con-
stitution had to be enacted.
In Missouri, another constitutional amendment was
passed to provide for a percentage to be added to the
State sales tax for use in soil and water conservation ac-
tivities. This required passage of a voter referendum. By
tying it to increased funds for State parks, it received both
rural and urban backing.
Idaho has established a different kind of cost-share pro-
gram, using 4 percent of the funds accruing to the water
pollution control account to fund grants for water quality
priority projects. Stream segments are ranked according
to State priority and BMP's are cost-shared in these water-
sheds. Almost $7 million has been spent since 1981 on
this program.
Wisconsin has four separate programs to help farmers
with the cost of applying BMP's to control nonpoint
sources.
Maryland is using a portion of its Water Quality Bond
issue to cost-share installing BMP's in priority watersheds.
The State also is funding about 40 new technical positions
in conservation districts to design and supervise BMP in-
stallation.
Virginia's cost-share program is not tied to priority wa-
tersheds as closely as is Maryland's. It covers all Chesa-
peake basin watersheds.
Within the last 2 weeks the voters in Alabama have
approved a $2 million cost-share program.
Many of the State developed cost-share programs have
been designed to supplement Federal efforts. Few of
them carry enough additional funds to provide the techni-
cal assistance necessary to implement the cost-share pro-
gram. Therefore, the cuts in the SCS budget proposed by
this Administration would .cripple more than just the Fed-
eral program, they would drastically reduce the State pro-
gram as well. Those of you who can influence Congress to
restore funds to the SCS budget will be assisting the water
quality effort as well as the soil conservation efforts of the
Nation.
With the passage of the Clean Water Act and the recog-
nition of pollution from uncontrolled construction sites,
NACD in cooperation with the Council of State Govern-
ments developed Model State Legislation for Erosion and
Sediment Control. This model act was sent to all States
and a series of seminars were held in many States to
discuss the provisions of the Act.
As a result of these efforts 23 States have enacted
some type of controls on disturbed lands. These controls
normally consist of site plan reviews and approvals by the
conservation district or some other entity of government.
As Congress gears up for another attempt at enacting
legislation to provide Federal assistance for nonpoint
source control programs, conservation districts have an
intense interest in the content of the proposed legislation.
We believe that such legislation should provide for Federal
assistance to the States in carrying out State-developed
programs. Most States have already decided what will and
will not work through the 208 planning process and most
208 plans have been approved by EPA.
The States are ready. Now it is up to the Federal govern-
ment to assist them at least enough to show that the Fed-
eral government is interested.
222
-------
THE STATUS OF SILVICULTURAL NONPOINT SOURCE PROGRAMS
GEORGE G. ICE
National Council for Air and Stream Improvement
Corvallis, Oregon
INTRODUCTION
Since 1979, the National Council of the Paper Industry for
Air and Stream Improvement (NCASI) has been monitor-
ing and surveying State silvicultural nonpoint source
(NFS) control programs (Natl. Counc. Pap. Indus., 1979,
1980, 1983; Ice, 1981). NCASI has examined methods
State agencies use to assess how forest operations may
hinder State attainment of water quality goals, types of
State nonpoint source control programs (voluntary, regula-
tory, or quasi-regulatory), approaches being used to imple-
ment best management practices (BMP's), problems
States face in implementing programs, and the degree of
program success (based on assessments by States).
NONPOINT SOURCE CONTROL
PROGRAMS FOR SILVICULTURE
Initially, a model forest practices act was proposed for use
by States in controlling silvicultural nonpoint source prob-
lems. This was found to be too inflexible for application to
the variety of conditions each State faced. In 1977, guide-
lines were issued by the U.S. Environmental Protection
Agency that allowed States to develop voluntary nonpoint
source control programs if". .. such programs were ade-
quate to achieve desired water quality goals" (U.S. Envi-
ron. Prot. Agency, 1977). This allowed flexible and some-
times creative responses by State agencies to perceived
NPS problems and resulted in "... almost as many differ-
ent varieties of Section 208 programs for silvicultural non-
point source pollution as there are States ..." with forest
operations (Rey, 1980). Figure 1 classifies State programs
as regulatory, quasi-regulatory, or voluntary.
Regulatory
Regulatory programs involve mandatory controls and en-
forcement strategies. Oregon, Washington, Idaho, and
California base much of their NPS control programs for
silviculture on enforcement of Forest Practices Act Regu-
lations. However, Yee (1984) has noted that even for these
programs, the site-specific BMP is not solely a product of
specific Forest Practices Act Rules; rather, the BMP
results from the entire regulatory process. In California
this involves development of a timber harvest plan by a
registered professional forester, public notification of man-
agement plans, review of plans by a multidisciplinary
team, and, in many cases, on-site inspection and plan
modification. Also, the Forest Practices Act may be only
one of several laws influencing water quality protection as
it relates to forest management. State laws concerning
water quality protection in California forests include the
Forest Practices Act, Professional Foresters Law, Califor-
nia Environmental Quality Act, Coastal Act, Wild and Sce-
nic Rivers Act, and others.
Voluntary
Voluntary programs use nonenforcement techniques in-
cluding education, cost-sharing, or other incentives to pro-
mote BMP's. Twenty-six States have adopted voluntary
control programs for silviculture. All of these programs
involve educational efforts to inform operators about
BMP's which have been developed for the State. Even
outside the United States, the importance of education in
protecting water quality is recognized. For example, Lar-
son and Albertin (1984) reviewed erosion and sedimenta-
tion control programs in the forested Panama Canal wa-
tershed, where a "... community education program is
intended to increase public awareness of the harmful ef-
fects of poor land use and to instruct local people in the
techniques of resource conservation." The most effective
State educational programs seem to incorporate BMP
training into normal forestry extension programs.
Consulting by professional foresters has proven to be
one effective means of implementing water protection
practices. In Florida, the highest rate of compliance failure
was found on private nonindustrial lands where there was
no professional guidance (Conner et al. 1982). In Utah, a
study concluded that for". . . timber sales on private land,
the evidence suggests that when technical assistance is
provided by either the State forester or consulting forester,
adverse impacts on water quality are less significant and
occur less frequently than when no technical assistance is
provided" (Kappe et al. 1982).
Cost-sharing programs are not commonly employed for
silvicultural NPS control, but Minnesota spends $150,000
annually to encourage water protection practices in the
erosive southeastern portion of the State. A Rural Clean
Water Act project (the Garvin Brook Experimental Cost-
Sharing Program) also provides support money for imple-
menting BMP's, including forest management BMP's in a
small portion of the State.
In Vermont, the Vermont Timber Truckers and Pro-
ducers Association responds to complaints about forest
operations. When necessary, a committee from the asso-
ciation will visit a site and encourage the logger to resolve
the problem with appropriate practices. The visibility of
this program has had the added advantage of encourag-
ing operators to contact State agencies before forest oper-
ations begin in order to avoid problems. Both State envi-
ronmental officials and industry representatives seem to
find this approach very successful.
Figure 1.—Types of State nonpoint source control programs
for silvicultural activities.
223
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Quasi-regulatory
A third classification, quasi-regulatory, covers those States
that do not have regulatory programs specifically targeting
silvicultural operation. Instead they cover several NPS cat-
egories, including silviculture, through sediment controls
or other stream protection regulations. Pennsylvania regu-
lates all soil-disturbing activities under the Clean Streams
Act of 1937 and under rules developed by the State Envi-
ronmental Quality Board. Specific guidelines are provided
to foresters for compliance with these State regulations.
Nominally, Florida continues to have a voluntary NPS
control program for silviculture. However, a State wetlands
law, administered through water quality management dis-
tricts, mandates the use of State BMP's for much of the
State. Therefore, Florida can now be considered quasi-
regulatory.
ASSESSING ACCOMPLISHMENTS BY
WATER QUALITY PROTECTION
PROGRAMS
Measuring progress toward meeting NPS water quality
goals has long been an objective of the NPS program.
Hurd (1979) stated that "hard outputs" such as water
quality monitoring would be required to demonstrate pro-
gram success. However, this approach has been frus-
trated by the nature of NPS activities and impacts. Forest
management activities are widely scattered in both space
and time. Superimposed on this distribution are natural
variations in water quality.
For small watersheds, 50 percent or more of the total
sediment yield over a 10-year period can result from a
single storm. Measuring changes in discharge and sedi-
ment concentration during such extreme events requires
rapid and numerous sample collection. Rice et al. (1975)
calculated that to detect a 20 percent increase in sediment
load for Caspar Creek (a small experimental watershed in
California), between 73 and 12,327 individual samples
would be required, depending on stream discharge.
For States such as Oregon (where 10,000 notifications
of silvicultural activities are filed annually with the State
Department of Forestry), a priority system is now required
just to insure that Forest Practice Officers concentrate
their inspections on sensitive operations. Given the limited
resources available to State agencies, direct water quality
monitoring programs have proven impossible.
The silvicultural community has had as much experi-
ence as any NPS category in attempting to measure pro-
gram progress. This is because regulatory programs were
already in place (or being considered) in some States
when 208 planning began. Also, voluntary silvicultural
NPS control programs were rapidly adopted and imple-
mented by a number of State forestry agencies, often with
assistance from the Forest Service. Approaches to mea-
suring program progress have followed a logical progres-
sion, depending on NPS program type (regulatory or non-
regulatory) and the status of implementation.
Assessment approaches being used to measure pro-
gram progress (ranked by general degree of sophistica-
tion) include:
1. identifying State agency resources committed;
2. counting the number of public complaints about for-
est activities;
3. gauging recognition of BMP's and user attitudes;
4. conducting field assessments of BMP implementa-
tion;
5. quantifying enforcement actions;
6. conducting surveys of water quality or studies of re-
source protection;
7. installing test watersheds to validate BMP's; and
8. modeling water quality response.
State agency resources. Most State agencies docu-
ment activities that are designed to promote the use of
BMP's. A November 1983 report by the Georgia Forestry
Commission on the Forest Water Quality Management
Program identified several accomplishments since 1981
including the appointment of foresters responsible for wa-
ter protection programs on each State forest district, train-
ing sessions and media announcements, and distribution
of booklets describing recommended BMP's (Mixon and
Thompson, 1983).
Complaints. An indirect indicator of program perform-
ance used by some States is the number of citizen com-
plaints received about silvicultural operations. For exam-
ple, the Alabama Forestry Commission annual report
found "... no complaints involving water quality and for-
estry. ..." Of course this measure can be influenced by
levels of activity (number of operations in the State) and
public attitudes or perceptions. It may or may not reflect
water quality protection.
BMP recognition and user attitudes. Awareness and
acceptance of voluntary programs is important for pro-
gram success. For example, one of the goals of a Georgia
Forestry Commission monitoring program was "... to es-
tablish the degree of success in exposure of the informa-
tion and educational portion of the Forest Water Quality
Program. We wanted to know how many loggers, site
preparation vendors, etc., were aware of the program dur-
ing our initial report, and how many were aware one year
afterwards." Ordemann (1982) noted that in Oklahoma
there has been "cooperation from the forest industry for
... BMP development and use." Other examples of atti-
tude shifts are cited in NCASI Special Report 83-01.
BMP field assessments. A more direct measure of pro-
gram effectiveness is the degree to which BMP's are actu-
ally implemented. Florida provides a good example of a
State that has conducted field surveys of forest operations
in order to determine compliance with recommended
BMPs. These surveys have shown that about 90 percent
of the operations comply with the guidelines. The surveys
have also identified those areas of the State and those
management activities which required additional atten-
tion.
Enforcement records: In States with regulatory pro-
grams, enforcement records for the forest practice rules
provide a measure of program success.
Surveys and studies: Although water quality monitor-
ing for each operation is not feasible, some State agen-
cies have used studies or surveys to measure water pro-
tection success. The Georgia Environmental Protection
Division conduced an NPS impact assessment study di-
rected at identifying water quality changes associated with
NPS activities. In general, for forestry this study demon-
strated that a fixed water quality monitoring scheme (e.g.,
quarterly grab samples) is unable to adequately measure
NPS water quality impacts. Assessments of stream habi-
tat, including counts of fish and macroinvertebrates,
proved to be a more sensitive measure. Where recom-
mended BMP's were not used, stream quality was im-
paired, but impacts were short lived. In California, a soil
erosion study involving 119 10-acre plots was used to as-
sess the adequacy of forest practices rules (Hauge et al.
1979).
Watershed studies: If BMP's can be shown to effec-
tively minimize NPS contributions from forest activities,
and if field assessments demonstrate a high degree of
BMP application, then presumably water quality is being
protected. Most States used existing information to estab-
lish BMP's. However, some States are conducting water-
shed studies to quantify the benefits of BMP's and the
appropriateness of those practices for the State or region.
224
-------
NONPOINT PROGRAMS: THE STATUS
For example, Kentucky has contracted with the University
of Kentucky to conduct watershed BMP studies.
- Models: Modeling as a means of evaluating forest wa-
>ter quality programs has swung on a pendulum of use and
-disfavor. However, most models used in NPS water quality
planning are agriculturally oriented. Even those models
designed for forests have proven difficult to apply in the
field because of the variability of forest sites (Natl. Counc.
Pap. Indus. 1984). The California Erosion Hazard Index,
for example, was found in one study to explain less than 1
percent of the measured erosion (Datzman, 1978).
In summary, direct NPS water quality monitoring on a
statewide basis has not proven feasible. However, numer-
ous surrogate approaches have been applied to silvicul-,
tural NPS programs to measure their success.
PROGRAM ASSESSMENT RESULTS
Several States have conducted assessments of their silvi-
cultural NPS programs, and in general they are satisfied
with their accomplishments. In Arkansas, a survey of 200
sites by district foresters showed an overall good job by
forest operators and an apparent improvement in prac-
tices under the voluntary program (Leister, 1984). In Flor-
ida, two surveys showed that about 90 percent of sites
were in good compliance with State voluntary guidelines
(Conners et al. 1982; Olszewski, 1984). In New Hamp-
shire, a survey of 68 operations found only 10 percent of
these had erosion rates exceeding 2.7 tonnes/ha/yr (3
tons/acre/year) for roads and landings and "... when the
entire harvested area was considered, the overall erosion
rate was less than 1 ton/acre/year in every case..." (New
Hampshire, n.d.). A 3-year NPS assessment by the Geor-
gia Environmental Protection Division found that adverse
impacts did occur from forest operations in about half of
the streams studied, but that these were cases where
BMP's were not employed (Mikalsen, 1984). The study
also found that stream recovery was rapid compared to
other NPS activities. A complementary study, the Georgia
Forestry\Commission Visual Monitoring Program, found
that an estimated 55 percent of the BMP's for Georgia
were being used (Mixon and Thompson, 1983).
Forest practice records for Oregon show that the com-
pliance factor (the number of operations not cited for viola-
tions divided by the number of active operations) has re-
mained at about 98 percent (Ore. State Dep. Forestry,
1984). In Washington, a field assessment was conducted
in 1980; 122 operations involving 219 practices were eval-
uated for water quality impacts and compliance with regu-
lations. The study found that "Water quality was well pro-
tected when forest operations were conducted in
compliance with the regulations" (Sachet et al. 1980).
Where regulations were not followed, water quality was
adversely affected.
CURRENT PROBLEMS
Although many States have assessed their programs as
successful, other States have found problems. A draft as-
sessment of the Idaho Forest Practices Act identifies the
lack of resources (manpower, expertise, funds) as ham-
pering water protection activities. In New Jersey, the NPS
program for private lands was eliminated because of
budget constraints. Another problem still facing sihvicul-
tural NPS control programs involves small nonindustrial
landholdings. Surveys in Colorado, Florida, New Hamp-
shire, Utah, and West Virginia indicate that when profes-
sional planning was provided by State extension agents or
professional foresters, BMP's were used more frequently
and water quality problems were avoided. State assess-
ments are finding that industrial forest operations, which
usually involve professional foresters, tend to use BMP's.
Small nonindustrial private landowners and small inde-
pendent forest operators are responsible for a high pro-
portion of silvicultural water quality problems.
These problems suggest the need for additional fund-
ing, but the outlook for Federal support is not bright. In-
stead, some low-cost or no-cost elements from existing
programs may provide other States with options to pro-
mote BMP implementation. These opportunities include:
industry complaint-response teams (as in Vermont and
West Virginia), cooperative demonstration forests and
roads (as in Florida and Georgia), training sessions spon-
sored by industry or professional organizations (as in New
Hampshire and California), Treasured Forest status (as in
Alabama), operator recognition programs (as in Oregon),
sample timber harvest contracts that incorporate BMP's
(as in Indiana), and master forester programs (as in Ore-
gon).
CONCLUSIONS
The overall status of silvicultural NPS control programs
can be described as good. Many States have conducted
formal field investigations and have found their programs
are working. Other States have made adjustments in re-
sponse to identified problems to improve the effectiveness
of their programs.
REFERENCES
Conner, D., J. Bethea, C. Reinert, and R. Olszewski. 1982.
Results of the 1982 Silvicultural 208 Compliance Survey. Div.
Forestry, Fla. Dep. Agric. and Consumer Serv., Tallahassee,
FL.
Datzman, RA. 1978. The erosion hazard rating system of the
coast forest district: how valid is it as a predictor of erosion
and can a better prediction equation be developed? MS The-
sis. Humbolt State Univ., Arcata, CA.
Hauge, C.J., M.J. Furniss, F.D. Euphrat. 1979. Soil erosion in
California's coastal forest district. Calif. Geol. 32(6): 120-9.
Hurd, M. 1979. Presentation to the Region 10 Water Quality
Management Conference. U.S. Environ. Prot. Agency, Seat-
tle, WA.
Ice, G.G. 1981. Section 208 programs for silvicultural nonpoint
source control—status in the South. Pages 11-18 in Forest
Management Related Water Quality Protection Regulatory
and Investigative Programs in the South. Natl. Counc. Pap.
Indus. Air Stream Improv., Tech. Bull. No. 349. New York.
Kappe, K., D. Hosking, S. Henson. 1982. Utah silvicultural non-
point source assessment report. State of Utah, Nat. Resour.
Energy, Div. State Lands Forest., Salt Lake City, UT.
Larson, C.L., W. Albertin. 1984. Controlling erosion and sedi-
mentation in the Panama Canal watershed. Water Int. 9(4):
161-4.
Leister, R.L. 1984. An assessment of the forestry nonpoint
source control program in Arkansas. Pages 16-23 in Re-
search and Regulatory Programs Related to Southern Forest
Management Practices and Water Quality Protection. Natl.
Counc. Pap. Indus. Air Stream Improve., Tech. Bull. No. 417.
New York.
Mikalsen, T.K. 1984. Preliminary results of Georgia nonpoint
source impact assessment study: commercial forestry. Pages
1-5 in Research on the Effects of Forest Harvesting, Drain-
age, Mechanical Site Preparation, and Prescribed Fire on Wa-
ter Quality. Natl. Counc. Pap. Indus. Air Stream Improv., Tech.
Bull. No. 442, New York.
Mixon, J.W., L.W. Thompson. 1983. Forest water quality man-
agement program. Environ. Prot. Div. Georgia Forest. Comm.
National Council of the Paper Industry for Air and Stream Im-
provement. 1979. A survey of state assessment procedures
for silvicultural nonpoint source control programs. Spec. Rep.
79-02, New York.
1980. Summary of the current status of silvicultural
208 programs—1980. Spec. Rep. 80-12, New York.
225
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
. 1983. Summary of silvicultural nonpoint source pro-
grams—1982. Spec. Rep. 83-01, New York.
_. 1984. A review of models for estimating the effect of
forestry management practices on yield and transport of ma-
terials to receiving waters. Tech. Bull. 420, New York.
New Hampshire Division of Forests and Lands, n.d. Pilot project
on the integration of water quality planning with forest re-
source planning: Final rep. Concord, NH.
Olszewski, R.J. 1984. A Discussion of Florida's silvicultural 208
assessment. In Research and Regulatory Programs Related
to Southern Forest Management Practices and Water Quality
Protection. Natl. Counc. Pap. Indus. Air Stream Improv., Tech.
Bull. No. 417. New York.
Ordemann, K.D. 1982. Pers. Comm. Environmental planner,
Okla. Dep. Pollut. Control, Oklahoma City, OK
Oregon State Department of Forestry. 1984. Statistical data sup-
plement to 1983 Annual Report. Salem.
Rey, M. 1980. The effect of the Clean Water Act on forest prac-
tices. Pages 11-30 in U.S. Forestry and Water Quality: What
course in the 80's? Water Pollut. Control Fed.. Washington,
DC.
Rice, R.M., R. Thomas, and C. Brown. 1975. Sampling water
quality determines the impact of land use on small streams.,
Unpubl. Presented at Am. Soc. Civil Eng. Watershed Manage-
ment Symp. Utah State Univ.
Sachet, J. et al. 1980. An assessment of the adequacy of Wash-
ington's forest practices rules and regulations in-protecting
water quality: summary rep. Wash. Dep. Ecol., Off. Water
Progr. Rep. DOE 80-7. Olympia.
U.S. Environmental Protection Agency. 1977. SAM 31-Regula-
tory Programs for Nonpoint Source Control. Water Plann. Div.,
Washington, DC.
Yee, C.S. 1984. Presentation to the State Water Resources
Board on March 8,1984 in Sacramento, CA.
226
-------
ASSOCIATION OF STATE AND INTERSTATE WATER POLLUTION
CONTROL ADMINISTRATORS
tfOBBI J. SAVAGE
LINDA EICHMILLER
Association of State and Interstate Water Pollution
Control Administrators
Washington, D.C.
ASIWPCA is confident about the future of nonpoint source
pollution. Our status is good. We have the capability to
deal with these environmental problems, but we need to
collectively summon more courage and stamina to follow
through with our commitment in several areas. They in-
volve the need to move from the status quo to
• understanding environmental priorities;
• getting our facts straight;
• making the existing consensus work to our advan-
tage; and
• planning strategically.
UNDERSTANDING ENVIRONMENTAL
PRIORITIES—ASSOCIATION ACTIVITIES
Our ability to perform effectively as an environmental pro-
gram depends directly upon our ability to resolve priority
water quality programs. Ultimately it is against this crite-
rion that we will be judged.
The Association represents pollution control managers.
Our members are responsible for the day to day imple-
mentation of the Clean Water Act which involves writing
permits, compliance and enforcement, actively conduct-
ing water quality monitoring, developing water quality
standards, completing wasteload allocations, overseeing
the construction of municipal treatment plants, managing
municipal funding programs, implementing Federal laws
under delegation, and managing the overall nonpoint
source achievement of clean water goals.
Clean up activities, as you all know so well, are driven
by the need to achieve and maintain designated uses and
to protect public health and welfare. Specifically such
threats as hazardous waste, pesticide contamination, and
erosion are of particular concern. Not surprisingly many of
these problems have a strong nonpoint source element in
many States. However, the public doesn't think well of and
will not support a program of bureaucracies or funding
pots for its own sake. They measure performance in terms
of results, that is the protection of drinking water, fishing,
swimming, and so on. States tailor their program to fit into
this framework.
States firmly believe that implementing nonpoint source
control activities rests primarily with other than Federal
institutions. State and local level governments must be-
come heavily involved. However, State environmental pro-
grams are not the repositories of expertise on control of
agricultural pollution.
The ASIWPCA position on nonpoint source pollution
has been very carefully thought out. It provides that
• Water quality management plans should be the basis
for implementation—that is, existing information is gener-
ally adequate.
• Nonpoint source pollution is primarily a State respon-
sibility.
• States are moving ahead but much remains to be
done.
• Federal funding and technical assistance are needed
to promote that activity.
• States should be held accountable for resolving their
priority water quality problems.
• Federal regulatory programs would not help promote
nonpoint source pollution control.
• Nonpoint source should be an integral part of the
Section 106 program.
GETTING OUR FACTS STRAIGHT
Predictably, we now find ourselves, as States and other
concerned parties, in the position of being concerned
about nonpoint source pollution, wanting to move forward,
but lacking the tools and the resources in several key
areas.
A lot of data exists out there as evidenced by these
proceedings but we are handicapped in several ways:
1. We do not have the quality information properly ar-
rayed for decisionmaking.
2. We do not have a good sense of priorities.
3. It is difficult, therefore, to make good decisions or
maintain the efficiency and effectiveness required by the
public.
States know all too well that this is a highly vulnerable
position to remain in for any length of time. This kind of
inquisitiveness on behalf of the States led us to develop
our 10-year status report, America's Clean Water. In that
report, nonpoint source pollution was targeted clearly as
an issue for the future. For these reasons the Association,
at the U.S. Environmental Protection Agency's (EPA) re-
quest, agreed to undertake a project, soliciting informa-
tion from the States to document the priority water quality
problems, the status and effectiveness of existing pro-
grams in the States, and accomplishments and effective
management techniques.
The results to be published in September 1985 will be
available from the Association. We hope the data made
available as a part of the ASIWPCA report will contribute
to ever more effective decisionmaking. Our report will, in
essence, constitute the baseline of information against
which States and the Nation can track how problems can
be solved on a priority basis, how to document progress,
and how to evaluate or modify programs in the future.
If you think about it, other program elements would also
benefit from such a baseline. It can be used to our definite
advantage in the NPS program.
Moving Forward Based on Consensus
Data are still coming in but we have learned some very
interesting things so far which has served to eliminate the
NPS myths and irrational beliefs, leaving little excuse for
not moving forward with the consensus to implement
nonpoint source controls.
As part of the project, the Association held a retreat of
227
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
national constituencies for NFS involving environmental,
industry, States, local governments, and other institutions.
It was clear from the results of the session, the Association
believes, that there is general agreement on the following:
1. NFS is an important water quality problem but it
does not equate directly to soil erosion.
2. Significant advancements have been made in NFS
management, particularly under 208.
3. Although effective management tools are available,
not enough is being done to manage nonpoint source
pollution.
4. States and local governments are showing savvy
and progress in their management approaches.
5. Targeting resources to specific priority water quality
problems will be necessary—the goals are attainable.
6. At the Federal level, EPA cannot do it alone—Inter-
agency cooperation will be required. Federal agencies
need to do a great deal more to assure their activities are
consistent with water quality goals.
7. The definition of goals is different for nonpoint than
for point sources, not only for water quality but most im-
portantly for institutional development and best manage-
ment practices.
8. It is going to take a long time to measure water qual-
ity results. Therefore, surrogate measures are appropriate
and useful in the interim.
9. Federal money will be limited and that is not neces-
sarily bad. In many areas, other techniques can and
should be used such as private capital where controls are
in the economic interest of the landowner, tax incentives,
and regulation. We have learned enough in the Construc-
tion Grants program to know that is not a model to be used
but one to be avoided.
STRATEGIC DIRECTIONS
We have clearly reached the point where there is little to
debate and little reason not to move forward. The program
has a good outlook but challenges are ahead.
We need to be careful not to end up with a case of the
warm fuzzies (that is, lots of supportive chatter and little to
no action). Nonpoint source control is not the Clean Water
Act's equivalent of a search for happiness. State environ-
mental programs do not run on talk. They run effectively
by
• establishing goals/objectives
• determining water quality priorities
• establishing measures of accountability
• evaluating progress.
In these areas, our collective performance is not yet ac-
ceptable if we expect nonpoint source control to be an
integral part of the national effort to protect the environ-
ment:
1. To be a mature participant in the effort, it is simply
not enough to be a "special interest."
2. The future of NFS in environmental management
depends on the ability to come to the table with other
program elements as partners, using the resources avail-
able to most efficiently solve priority problems.
3. To reach consensus in the legislation before Con-
gress on national policy and commitment is one way we
can all work together. But that commitment cannot stop
with authorizing legislation. To implement a program,
State and local governments will need the financial re-
sources to move ahead expeditiously, which means appro-
priations.
The Association supports the modest authorization in
the House and Senate Bills for nonpoint source, which is a
major milestone for us. While others may run for the hills
when the issue of funding comes up, preferring to rely,
instead, on scraps from the Construction Grants table—a
program that is closing out with needs exceeding the fund-
ing by 40,000 percent—the States are willing to stand up
and recognize the need, even in such difficult budget
times.
A CALL FOR STRATEGIC PLANNING
There is a real question, in light of the funding situation,
whether nonpoint source will be able to achieve the stat-
ure to which many aspire. If we are to achieve progress we
must work together as States and other interested parties.
We must call a halt to the debate over non-issues—we
simply cannot afford to waste the time. We must identify
clearly the goals and objectives of the program as they
relate to the environment then move efficiently to get the
job done, supporting the kind of integrated program activi-
ties necessary to accomplish those objectives. It is dis-
tressing that more interdisciplinary dialogue on NFS, such
as this meeting and our national retreat has not occurred;
we must also support the necessary funding to accom-
plish our goals and objectives—at the State, local and
Federal levels, and target our resources to solve the prior-
ity water quality problems. We should be held accountable
for results—evaluating progress as the program develops.
This is hard to do. It takes time and resources but such
documentation will prove to be the key to our overall suc-
cess. Finally we must support focused technology ex-
change that is so necessary to advance State and local
programs in the future.
We think the tools are there to move forward, but we
collectively must spend time setting strategy if the pro-
gram is to capitalize on its assets, be successful, and
maintain public support. ASIWPCA considers this confer-
ence to be a good start in that direction. We recognize that
resolving NFS problems will be hard work. Our commit-
ment and that of many others is evidenced by the attend-
ance at the Perspectives on Nonpoint Source Pollution
meeting that has brought us together.
228
-------
Economics of Nonpoint
Source Pollution
ECONOMICS: NONPOINT SOURCE POLLUTION IMPACTS
SANDRA S. BATE
Department of Agricultural Economics
Virginia Polytechnic Institute and State University
Blacksburg, Virginia
INTRODUCTION
Too often economics is viewed as relating only to the pri-
vate finances of individuals. With this narrow view, the
economics of nonpoint source pollution is perceived as
identifying the costs to private citizens (for example,
farmers) of controlling nonpoint source pollution prob-
lems. A more sophisticated definition of economics ad-
dresses the costs and benefits of actions to the general
public as well as to the private citizen. An even more
sophisticated definition, however, recognizes that eco-
nomics is concerned with society's values and the role of
the public sector in shaping and reflecting these values.
This broad definition of economics incorporates, as legiti-
mate concerns of the science, the design of institutions,
the evolution and distribution of property rights to re-
sources, and the role of government in reflecting emerg-
ing societal values.
In this paper I will use the broad definition of economics.
I will focus on the costs of nonpoint pollution—both the
costs of "doing something" and the costs of "doing noth-
ing" about nonpoint pollution. Finally, I will relate the dis-
cussion to alternative institutional designs for controlling
nonpoint source pollution impacts.
TYPES OF IMPACTS
Nonpoint source pollution can have several negative im-
pacts. These generally fall into three main headings: in-
stream effects, off-stream effects, and ground water. The
in-stream effects include those to the biology, the recrea-
tional uses, the water storage capacity, and the navigation
service of various water-bodies.
Sediment, for example, can smother spawning areas or
otherwise reduce the usefulness of aquatic habitat for fish,
shellfish, and wildlife. Sediment can carry chemicals
harmful to various aquatic ecosystems. Sediment and al-
gae growth stimulated by sediment-conveyed nutrients
can block sunlight and retard the growth of aquatic plants
on which fish, crustaceans, and other wildlife depend.
Some chemicals can be directly toxic to fish, wildfowl, and
shellfish.
Water-based recreational activities can also be ad-
versely affected by erosion-related pollutants. This is par-
ticularly true with fishing when fish populations are badly
affected by water pollutants. Similarly, the value of boating
or swimming is diminished when it takes place in polluted
lakes or rivers.
Sediment can also interfere with the storage of water in
reservoirs. It is estimated, for example, that 1.4 to 1.5
million acre feet of reservoir and lake capacity is perma-
nently filled each year with sediment (Clark et al. 1985).
Sediment can also necessitate increased dredging of har-
bors and waterways and produce shipping delays and
accidents.
Off-stream damages include flood damages, siltation of
drainage ditches, increased water treatment costs, and
increased costs to landowners because of polluted waters
(for example, increased costs to farmers because of saline
irrigation waters). By changing the shape of stream beds,
sediment can increase the frequency and depth of flood-
ing. Sedimentation of drainage ditches results in signifi-
cant annual maintenance costs. In addition, polluted wa-
ters can increase the costs of treating water from
municipal and industrial uses as more treatment is neces-
sary to remove sediment, pesticides, and other dissolved
contaminants.
Unfortunately, water treatment facilities do not remove
dissolved salts that are frequently associated with non-
point pollution from agricultural sources. "These are esti-
mated to cause $80 million in damages annually to munic-
ipal and industrial users in the lower Colorado River basin
229
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
alone" (Clark et al. 1985). These salts can also increase
costs to farmers who attempt to use highly saline waters
for irrigation purposes.
The third major nonpoint source pollution impact relates
to ground water uses. These effects can include increased
water treatment costs, adverse biological impacts, and the
possible loss of ground water sources where pollution is
severe. In 1981 the Water Resources Council asserted
that "every region in the country experiences ground wa-
ter pollution problems both point and nonpoint ... their
widespread nature supports concern with degradation of
the ground water supplies throughout the Nation"
(Sharefkin et al. 1983). Since ground water is a major
source of drinking water and irrigation water, the effects of
such pollution on human, animal, and plant health and
vitality is of national concern.
MEASUREMENT OF THE COSTS OF
NONPOINT POLLUTION
The costs of nonpoint pollution are very difficult to assess
accurately. The first problem in assessing the damages is
that the mechanisms of causality are not well known. Con-
sider the following example that illustrates the complex-
ities involved in such measurements. Imagine a farmer
who conventionally plows his field and sprays the field
with herbicides and pesticides. A certain probability exists
that a rainstorm will occur at such a time that it will wash
sediment and chemicals from the field. An additional prob-
ability exists that some of the sediment and chemicals
eventually will reach a nearby waterbody.
The first problem in measuring costs, then, is to identify
the linkages between the farmers' practices and the
amount of chemicals and sediment that appear in the wa-
ter. This is complicated by random and unique events
such as storms and different responses of soil types, wa-
tersheds, timing of chemical applications, and seasons of
the year.
Once the chemicals and sediment reach the water, they
may change some of the water's properties—such as tur-
bidity, carrying capacity of the channel, heat, or chemical
composition of the water. Synergistic effects between the
chemicals may occur. These changes, in turn, may affect
fish and wildlife populations, the probability and severity of
floods, drinking water quality, and the life of reservoirs.
Thus, the second problem in measurement is to accu-
rately identify the relationship between the amount of
chemicals and sediment in the water and its effect on
various uses. This is a complex task. For example, sedi-
ment in the flood channel can increase the probability and
severity of floods. However, stream channels always carry
sediment. If that sediment is not obtained from soil ero-
sion, the stream will erode its bed and banks. Thus, it is
difficult to know accurately the relationships between soil
erosion, sedimentation of the streams, and flooding.
Finally, the changes in the water quality and attendant
changes on potential uses of the stream may affect socie-
ty's evaluation of the stream. Therefore, the third major
measurement problem is to relate the potential impact of
pollutants on water uses to the actual change in society's
uses. This is complicated by the need to know the demand
for the uses with and without the sedimentation or with
and without the chemicals. To use an extreme example, if
there is no demand for drinking water now or in the fore-
seeable future, then there are no damages to drinking
water from the pollution.
Furthermore, the severity of pollution's consequences
is not well established. Where toxic chemicals are in-
volved, for example, a long latency period from exposure
to chemicals and the onset of a disease is possible. It is
difficult therefore to ascertain what level of exposure con-
stitutes a health hazard.
The change in society's use of the polluted waters will
also be influenced by the availability of substitutes for the J
involved uses. One polluted lake surrounded by clean
lakes will not be considered as great a cost from a societal"'
point of view as if there were only one lake within the
commuting distance of the city
Finally, there is the problem of aggregation of costs
across time, users, and regions. Clearly, a research
agenda associated with the costs of nonpoint source pol-
lution restricts the ability to fully employ a legion of hydrol-
ogists, engineers, economists, and biologists. As a result,
one does not find many studies that measure the costs of
nonpoint pollution. Existing studies usually have large var-
iances associated with their cost estimates.
But defining costs is exactly what the authors of the
Conservation Foundation's book, Eroding Soils: Off-Farm
Impacts of Soil Erosion (Clark et al. 1985), seek to accom-
plish. By their own admission, the authors were only mod-
erately successful in quantifying the off-farm damages
from soil erosion. Table 1 summarizes the Conservation
Foundation's research and findings on damage costs. To-
tal impacts of all the damages they were able to estimate
were approximately $6.1 billion per year. (The authors did
not attempt to estimate ground water damages.)
Table 1.—Summary of damage costs (million 1980 dollars).
Single-
Range of value
type of impact
In-stream effects
Biological impacts
Recreational
Water storage facilities
Navigation
Other in-stream uses
Subtotal: In-stream (rounded)
Off-stream effects
Flood damages
Water conveyance facilities
Water treatment facilities
Other off-stream uses
Subtotal: off-stream (rounded)
Total: all effects (rounded)
estimates
950- 5,600
310- 1,600
420- 800
460- 2,500
2,100-10,000
440- 1,300
140- 300
50- 500
400- 920
1,100- 3,100
3,200-13,000
estimate
no est.
2,000
690
560
900
4,200
770
200
100
800
1,900
6,100
Source: Conserv. Foundation research (Clark et al. 1985).
THE COSTS OF NONPOINT POLLUTION
CONTROL
This figure—$6 billion per year—can be interpreted as an
estimate of the costs of "doing nothing" about nonpoint
pollution in the United States. The other types of costs
associated with nonpoint agricultural pollution are the
costs of "doing something." Doing something involves
changing institutions' and individuals' behavior to improve
water quality and to reduce nonpoint pollution problems.
This type of analysis is an example of research that re-
flects the broad definition of economics mentioned earlier.
Costs as difficult to define as those associated with non-
point source pollution are also difficult to control. Alterna-
tives frequently discussed include voluntary programs
with or without cost-sharing, regulatory programs, devel-
opment of user taxes to finance pollution control, or giving
more authority and flexibility to State governments to de-
velop either voluntary or regulatory programs.
There are costs of doing something just as there are
costs of doing nothing. First, the funds used for nonpoint
230
-------
ECONOMICS OF NONPOINT SOURCE POLLUTION
source pollution control obviously could have been used
for other purposes. Economists have a saying: "Anything
worth doing is not necessarily worth doing well." That is,
every dollar spent on water quality is a dollar not spent on
better schools and better highways. Every dollar spent in
ene area of water quality is a dollar not spent in another
area of water quality. These foregone benefits are one
type of costs associated with pollution policy.
Information, enforcement, and administration also cost.
For example, one possible policy is regulation. Regulatory
policies require substantial knowledge of probabilities,
costs, and benefits of pollution control if there is to be a
consensus on the appropriate definition of safety thresh-
olds of pollutants. The regulatory approach has high en-
forcement, monitoring, and administrative costs as well.
Furthermore, regulatory decisions tend to be litigated.
The costs of appealing to courts for solutions can be ex-
tremely high.
Voluntary approaches to pollution control also have high
costs. For example, $13 million has been allocated for
developing the Maryland portion of the Chesapeake Bay
Program. Unfortunately, the history of voluntary programs
in nonpoint pollution suggest low effectiveness in return
for such expenditures. Voluntary approaches have tended
to work best when there has been (1) agreement on objec-
tives to be reached, (2) easily observable noncompliance,
(3) private costs proportional to the private benefits re-
ceived, and (4) a belief that failure to voluntarily comply
will bring government mandatory action (Harrington et al.
1985). As Ken Cook has noted, the voluntary approach
seems to be the "slow boat on waters that remain pol-
luted" (Cook, 1985).
Of course, one can consider other policies such as user
taxes, cost-sharing, subsidies, giving more flexibility to
agencies for individual solutions, trading policies, or "bub-
ble policies." There are not, however, conclusive answers
as to which is the more appropriate policy. It is clear that
controlling off-farm impacts of soil erosion is going to be
very difficult whatever policy is selected.
Other costs are associated with policy development
such as the political costs of changing the present institu-
tions. Who pays and who loses with a policy change can
become quite important. Any policy that attempts to
change the use of the environment is going to change the
distribution of the benefits and costs associated with that
use. It is the distribution more than the magnitude of the
benefits and costs that determines most political deci-
sions.
The challenge of policymakers is to design nonpoint
pollution strategies that minimize these costs of achieving
desired levels of environmental improvement. The chal-
lenge is all the more imposing because such policy design
will necessarily proceed with fragmentary information and
high levels of uncertainty.
The acceptance of any policy designed with limited
knowledge requires public perceptions that something
needs to be done—that maintaining the status quo is
costly—even if the benefits of such actions are not com-
pletely delineated. This is the payoff associated with defin-
ing costs associated with nonpoint source pollution. The
payoff is not obtaining totally accurate estimates; the pay-
off is changing of perceptions. Changing perceptions can
improve research and stimulate the public's desire to do
something about nonpoint pollution, by placing water
quality problems higher in priority and encouraging gov-
ernments to reflect these priorities in policies. I already
see growing evidence that the public is perceiving non-
point pollution as a serious problem and is ready to make
a commitment. (I also agree with those who feel that much
of this commitment is taking place outside traditional agri-
cultural circles.) Research such as the Conservation
Foundation's which estimates the costs of nonpoint
source pollution can aid in creating a climate from which
reasonable nonpoint pollution control policies might ema-
nate.
REFERENCES
Clark, E.H., II, J.A. Haverkamp, and W. Chapman. 1985. Erod-
ing Soils: The Off-Farm Impacts. Conserv. Foundation, Wash-
ington, DC.
Cook, K. 1985. Commentary: agricultural pollution control: a
time for sticks? J. Soil Water Conserv. 40(1): 105-6.
Harrington, W.A., J. Krupnick, and H.M. Peskin. 1985. Policies
for nonpoint source water pollution control. J. Soil Water Con-
serv. 40(1): 27-32.
Sharefkin, M.F., M. Schecheter, and A.V. Kneese. 1983. Im-
pacts, Costs, and Techniques for Mitigation of Contaminated
Groundwater. Reprint 210. Resour. for the Future, Washing-
ton, DC.
231
-------
ECONOMICS OF NONPOINT SOURCE POLLUTION CONTROL: LAKE
TAHOE, CALIFORNIA/NEVADA
DAVID S. ZIEGLER
Long Range Planning Division
Tahoe Regional Planning Agency
South Lake Tahoe, California
INTRODUCTION
Controlling water pollution from nonpoint sources has
been a major concern at Lake Tahoe for over a decade.
Recent amendments to the Tahoe Regional Planning
Agency's Regional Plan require five different nonpoint
source control programs to reduce pollutant loads to the
Lake.
The Agency's nonpoint source control policies may
have a positive long-term impact on the region's economy
and impacts of various types on local government, utility
districts, and individual property owners.
DISCUSSION
The Water Quality Problem. Lake Tahoe is a large ul-
traoligotrophic lake with astonishing clarity and low algal
productivity. Because of the Lake's unique recreational,
scenic, and environmental values, the States of California
and Nevada decided to require export of all sewage from
the Lake Tahoe Basin in the early 1970's. Despite the
sewage export, however, the lake's water quality has de-
clined steadily since scientists started keeping records in
the 1960's.
According to recent Agency studies, increasing algal
productivity in Lake Tahoe is the result of increasing stor-
age of nutrients, particularly dissolved nitrogen. The ele-
vated nutrient loads come from nonpoint sources related
to development of the Tahoe region and surrounding ar-
eas. Development increases nutrient loads in five ways:
1. Soil disturbance liberates nutrients stored in the
soils, allowing them to migrate to the Lake;
2. Displacement of vegetation removes natural filtra-
tion, increasing the concentrations of pollutants in runoff
and tributary flows;
3. Altered hydrology, especially the addition of imper-
vious surfaces, causes flashy runoff, which is more de-
structive than natural runoff;
4. Artificial Inputs, particularly fertilizer, add to nutri-
ent loads; and
5. Air pollution from automobiles contributes nitrogen
to Lake Tahoe through the process of atmospheric deposi-
tion.
The Tahoe region is a resort area with a small perma-
nent population (approximately 45,000). But on a peak
summer day, the population of the region can swell to
300,000 and daily vehicle-miles-of-travel reach 1.7 million.
Most development in the region took place in the 1960's
and 1970's without the benefit of best management prac-
tices (BMP's) and other controls to minimize the environ-
mental impact. Many subdivision roads, for example, lack
adequate drainage and slope stabilization. Storm drain-
age is inadequate. And the problem is aggravated by the
fragile soils, high annual precipitation, and a short grow-
ing season.
Water Quality Control Measures. The Tahoe Regional
Planning Compact (PL. 96-551) gives the Agency author-
ity to regulate land use by adopting a set of environmental
threshold standards, amending the Regional Plan to attain
and maintain the thresholds, and adopting and enforcing
ordinances to carry out the Plan. Violators of Agency ordi-
nances are subject to fines up to $5,000/day. Virtually all
development in the region requires some form of Agency
review, usually including a permit.
In August 1982, the Agency adopted comprehensive
threshold standards covering water quality, soils, air qual-
ity, vegetation, wildlife, fisheries, noise, recreation, and
scenery. They include requirements to reduce annual
loads of dissolved inorganic nitrogen (DIN) to the lake by
25 percent and to restore 480 ha (1,200 acres) of dis-
turbed, developed, or subdivided stream environment
zones.
The Agency has adopted the following control mea-
sures to implement these standards:
1. Application of BMP's on all property to prevent ero-
sion and runoff problems (basically, infiltration, slope stabi-
lization, drainage stabilization, and revegetation);
2. Growth management programs, covering the rate,
location, and type of new development;
3. Implementation of controls on new impervious cov-
erage, based on the capability of the site;
4. A public sector capital improvements program for
erosion and runoff controls, estimated to cost $125 million
($ 1982), including a stream zone restoration program for
disturbed wetlands; and
5. A transportation capital improvements program in-
cluding transit and traffic flow improvements, designed to
reduce air pollution at a cost of about $125 million ($
1982).
Previously, the Agency generally required BMP's only
on new construction and other activities requiring a per-
mit. However, the amended Regional Plan specifically
states, "All persons who own land, and all public agencies
that manage public lands in the Lake Tahoe region shall
put best management practices (BMP's) in place; main-
tain their BMP's; protect vegetation on their land from
damage; and restore the disturbed soils on their land."
Economic Implications. Although it is difficult to isolate
and quantify the economic impacts of these nonpoint
source control measures, the Agency has made qualita-
tive assessments and used computer models to investi-
gate the impacts on the regional economy, local govern-
ment, utility districts, and individual property owners.
Three factors, not unique to the Tahoe region, compli-
cate the assessment of economic impacts. First, the re-
gional economy depends heavily on unrelated factors in-
side and outside the region (for example, interest rates).
Second, no database of key economic indicators is availa-
ble. Third, available data were collected by different au-
thors, and are not necessarily comparable.
In response to these deficiencies, the Agency plans to
monitor economic indicators and identify trends that might
result from nonpoint source controls. Although the causal
relationships are always difficult to understand, trend
monitoring will be useful.
232
-------
ECONOMICS OF NONPOINT SOURCE POLLUTION
The Lake Tahoe region provides recreation opportuni-
ties for tourists from not only California and Nevada but
Ihe entire world. A1976 study for the Agency said that the
Tahoe region had reached a phase of market maturity and
saturation where growth is slow. Thus, with or without
•controls on nonpoint sources, the regional economy will
not reflect vigorous growth.
In January 1985, an Urban Land Institute study panel
essentially confirmed these findings. The panel said that
the region had an oversupply of commercial property, re-
flected in low rents and a high vacancy rate, that the re-
gion lacked pulling power, and that the region needed a
bigger "bang for the buck" from each visitor. According to
the panel, economic growth will result not from more
stores and motels, but from more productive stores and
motels.
The Agency feels that the threshold standards and pol-
lution control measures will have a positive impact on tour-
ism and the economy because of increased attention to
environmental factors, transportation, redevelopment,
and the quality of life in general. ,
The remedial programs in the Plan also create jobs in
the construction sector. For the 1985 building season, lo-
cal government has already programmed $6 million in
erosion and runoff controls. It is unlikely, however, that
construction employment will return to the peak levels of
the mid-1970's.
Economic Impacts on Local Government. The main
economic impacts of the Regional Plan on local govern-
ment stem from the Agency's growth management poli-
cies and the local government's obligations to provide
capital improvements.
In the area of growth management, tax revenue losses
are of local concern, but research reveals that such im-
pacts will be minor. Only one jurisdiction, the city of South
Lake Tahoe, is entirely within the Tahoe region. The five
county governments belonging to the Agency include
large population centers outside the Basin, which buffer
them from the economic impacts of land use control at
Tahoe. Since the counties can direct growth outside the
Basin, the counties will continue to provide services at the
present level. Also, since the city of South Lake Tahoe has
not financed capital improvements with future growth, the
lower projected revenues do not present a problem. In the
long run, the city anticipates a substantial amount of rede-
velopment and new tax income.
Units of local government are also concerned that they
lack the ability to carry out the water quality and transpor-
tation capital improvement programs. The Regional Plan
calls for local government to provide erosion and runoff
controls on its righls-of-way and, with assistance from
State and regional agencies, provide necessary transit
service. To capitalize the $250 million capital improvement
program will require annual revenues of about $30 million,
far above the existing level of investment in the program.
An Agency study in 1983 concluded that, at most, taxes
and assessments on residents and visitors could raise
only half the necessary funds. Therefore, the Agency
adopted a strategy based on the following assumptions:
1. The ability of local government to finance the capital
improvement program is limited;
2. The program will require regional implementing
agencies with the ability to raise revenues and sell bonds;
3. Local governments should contribute a fair share to
implement the capital improvement program, with the goal
of achieving a fair and equitable burden on the taxpayer
regionwide; and
4. When the region is able to demonstrate its commit-
ment to implementing the program through the measures
above, it will approach the States and the Federal govern-
ment for assistance in reaching environmental goals.
Under this approach, the Agency will be able to imple-
ment the Regional Plan capital improvement program
without placing an unfair burden on local government and
creating associated economic impacts.
Impacts on Public Utilities. The utility districts in the
Tahoe region are also concerned that the Agency's growth
management policies will cause a loss in anticipated reve-
nues with subsequent impacts on rates and financial
health. Limits on new development in the Plan do affect
the utility districts because they have to spread their costs
over fewer users and because they will not receive pro-
jected connection fees, which they had planned to use for
both operations and capital expansion. Thus, the impact
of the Regional Plan on the utility districts is related to how
much they use connection fees to pay for operations.
One utility district in the region supplies water and
sewer service to an area with a high percentage of envi-
ronmentally sensitive land. The Regional Plan may cut
projected connection fees from Tahoe Basin properties in
half, but this impact will be offset by development of al-
most 1,000 units outside the Basin. The district will still
have a positive net income and, although service charges
may increase, the district will have adequate revenues to
meet its costs.
Impacts on Individuals. The main economic impact on
individual property owners from the Agency's nonpoint
source controls is the cost of applying BMP's to existing
development. In 1983 the Agency staff estimated the pri-
vate sector costs of BMP's at $63 million. In most cases,
the cost of applying BMP's retroactively should not ex-
ceed 1-2 percent of the cost of the home or business
being treated. A $125,000 home, for example, should not
require more than $2,500 in BMP's.
For single family homes, the BMP's may actually en-
hance the value of the home and offset the cost of installa-
tion. In the case of commercial property, however, retroac-
tive installation of BMP's may be both expensive and
technically difficult, since commercial sites in the Tahoe
region are typically 95-100 percent covered with impervi-
ous surfaces, making infiltration of runoff difficult. For
commercial property owners, the Agency will investigate
both shared infiltration systems and low cost loans for
small businessmen to minimize economic impacts.
Another economic concern for small businesses is that
the Agency's limits on impervious coverage prevent con-
struction of economically viable commercial sites. Com-
puter models of commercial projects showed that allowa-
ble coverage and profitability are proportional to each
other, but no absolute profitability point exists for a project.
Because of the oversupply of commercial property in the
region, the Regional Plan allows very little commercial
growth in the near future. But the Plan does provide lim-
ited coverage incentives for commercial sites to enhance
their economic viability.
CONCLUSIONS
Lake Tahoe is a natural resource of exceptional quality—
"the jewel of the Sierra." But nonpoint source pollution
control measures are necessary to protect its famous wa-
ter quality These nonpoint source controls will have a
beneficial effect on the regional economy over the long
run. Economic impacts on local government and individ-
ual property owners should be minor, assuming that they
share the burden of capital improvements. Utility districts
that finance their operations with connection fees will be
affected to some degree by growth management policies.
233
-------
CONTROLLING AGRICULTURAL RUNOFF:
GOVERNMENT'S PERSPECTIVE
RICHARD S. MAGLEBY
C. EDWIN YOUNG
Economic Research Service
United States Department of Agriculture
Washington, D.C.
INTRODUCTION
Government is concerned with protecting and improving
the well-being of the people. But many problems exist and
many demands are placed on government to take actions
or to provide resources. Government's perspective on ag-
ricultural nonpoint source pollution has to do with how it
views this problem, in terms of priority for allocating gov-
ernment resources and actions to be taken.
In this presentation, we argue for greater consideration
by government, and others, of economic benefits and
costs in making decisions on NFS pollution control. We do
this by presenting some results of the economic evalua-
tion of the Rural Clean Water Program (RCWP) and using
these to illustrate the points we wish to make.
RURAL CLEAN WATER PROGRAM
The experimental Rural Clean Water Program was initi-
ated in 1980 to demonstrate the effectiveness of an agri-
cultural nonpoint source program. Some $60 million was
allocated to 21 projects. Five of these received additional
allocations to permit comprehensive monitoring and eval-
uation. These projects are: the Idaho Rock Creek Project,
the Illinois Highland Silver Lake Project, the Vermont St.
Albans Bay Project, the Pennsylvania Conestoga Head-
waters Project, and the South Dakota Oakwood Lakes-
Poinsett Project.
The comprehensive monitoring and evaluation studies
include both water quality and economic components.
The economic evaluations for the Idaho, Illinois, and Ver-
mont projects are the most complete to date, so our dis-
cussion of results will concentrate on these three. How-
ever, even these evaluations are preliminary. Thus, we
welcome suggestions for improvement.
Water Quality Problems
and Use Impairments
The water quality problems and use impairments origi-
nally identified in the three initial comprehensive monitor-
ing and evaluation projects are listed in Table 1. In the
Idaho project, high levels of sediment in Rock Creek from
irrigation-caused erosion were identified as impairing rec-
reational fishing and downstream water storage capacity
and power generation in the Snake River. In the Illinois
Project, highly erodible natric soils being carried off sur-
Table 1 .—Water quality problems/impairments.
Water quality Use
Project problem impairment
Idaho
Illinois
Vermont
Turbidity/
sediment
Turbidity
Phosphorus (algae,
aquatic weeds)
Fishing, water storage,
power generation,
ditch capacity
Water supply, water
treatment, fishing
Swimming, boating,
fishing, property
values
rounding farmlands into Highland Silver Lake were identi-
fied as impairing municipal water supply and treatment
and recreational fishing. In the Vermont project, phos-
phorus from animal wastes and sewage treatment plants
had stimulated algae and weed growth in St. Albans Bay,
impairing swimming, boating, and recreational property
values.
Water Quality Improvements
and Resulting Benefits
In each of the three projects, water quality improvements
have been projected as the result of RCWP, or, in the case
of St. Albans Bay, Vermont, the joint result of RCWP and
improved sewage treatment. Table 2 summarizes these
projected water quality improvements and our evaluation
of the resulting offsite economic benefits.
In Idaho, sediment in irrigation return flows and Rock
Creek will be greatly reduced. This will generate $411,000
in benefits to recreational fishing and will reduce ditch
cleaning costs by an estimated $185,000. However, this
improvement in Rock Creek will minimally affect the qual-
ity of water downstream in the Snake River. Because of
the hydrologic features of the Snake River, sediment from
streambanks and the river bottom would be picked up,
largely offsetting any savings from reductions in sediment
entering from Rock Creek. Thus, water storage or power
generation benefits appear negligible. Total estimated wa-
ter quality benefits over 50 years are $596,000. In addi-
tion, the crop residue cover from use of conservation till-
age is projected to improve upland game habitat, with a
hunting benefit estimated at just over $200,000. Total off-
site benefits of RCWP in Idaho would be $802,000.
In the Illinois Project, sediment entering the lake will be
reduced, in turn, returning the turbidity in the lake. Costs
of water treatment to remove sediment will be lowered by
some $225,000. Also, recreational fishing will marginally
improve but, because of limitations on access and on boat
size, only some $24,000 in benefits will be generated.
Water storage benefits appear negligible because much of
the sediment will remain in suspension and pass over the
dam, and because the lake's capacity is large relative to
future demand. Thus, total offsite benefits of $249,000
appear likely over a 50-year period.
In the Vermont Project, greatly reduced phosphorus
loadings from RCWP and better sewage treatment will
improve the water quality in St. Albans Bay over time, to
near that in the larger Lake Champlain. This will produce
swimming and other recreational benefits of nearly $4 mil-
lion, and will increase recreational property values by over
$1 million. Costs of weed treatment removal will be re-
duced by $27,000. Thus, the total offsite benefits over 50
years are estimated at nearly $5 million.
The much higher offsite benefits resulting from the Ver-
mont project, compared with the other two, stem from two
factors: a greater downstream improvement in water qual-
ity and greater number of people affected by the improve-
ment.
234
-------
ECONOMICS OF NONPOINT SOURCE POLLUTION
Table 2.—Estimated water quality Improvements and offsite benefits for three RCWP projects (preliminary).
Item
Rock Creek
Illinois Highland
Silver Lake
St. Albans Bay
Water quality improvements
Major reductions in sediment
in Rock Creek.
Minor improvements in Snake
River
Some reduction in
turbidity
Minor change in lake
sedimentation
Major reductions in algae
& aquatic weeds
water quaiiiy oenems; •
Recreation
Water storage
Property values
Water conveyance
Water treatment
Other
Total water quality
Upland hunting benefits
Total offsite
$411
$0
N.A.
$185
N.A.
N.A.
$596
$206
$802
• • • 91 UUU
$24
$0
N.A.
$0
$225
N.A.
$249
No Estimate
$249
$3,886
N.A.
$1,008
N.A.
N.A.
$27
$27
0
$4,921
'Benefits are estimated over 50 years and discounted to current value at an 8 percent rate.
Onsite Benefits
In two of the three projects, RCWP is generating some
onsite economic benefits from preserving soil productivity
or from reducing farmers' operational costs, which more
than offset their RCWP installation costs.
In Idaho, planned implementation of conservation till-
age and other practices that help keep soil in place on the
fields will reduce long-term soil productivity loss and gen-
erate benefits estimated at $814,000 (Table 3). In this
case, these productivity benefits are as great as the offsite
benefits.
In the Illinois project, conservation tillage is the principal
best management practice, but because the soils are
deep and fertile, long-term productivity benefits are negli-
gible.
In the Vermont project, the installation of improved ani-
mal waste storage facilities reduces manure handling and
fertilizer costs over time by more than the farmers' initial
share of putting in the systems. This negative cost of over
$2 million can be considered an onsite private benefit.
Note that it is about 40 percent as large as the public
benefits.
Table 3.—Estimated benefits compared with costs, three
RCWP projects (preliminary).
Idaho Illinois Vermont
Item project project project
Benefits
Offsite
Onsite (productivity)
Subtotal public
Onsite private
Total Benefits
Costs
Government costs
Private costs
(net before taxes)
Total Costs
B/C Ratios
Total benefits/
total costs
Public benefits/
government costs
Offsite benefits/
government costs
.8
.8
1.6
—
1.6
3.4
3.3
6.7
.2
.5
.2
. Mi Minn .
.2
—
.2
—
.2
1.6
.3
1.9
.1
.2
.2
4.9
—
4.9
2.0
6.9
3.9
—
3.9
1.8
1.3
1.3
Source: ERS Economic Evaluation Reports on each project.
Costs
Each project has two cost components: government cost
and private cost (Table 3). Government cost ranges from
$1.6 million each for the Illinois and Vermont projects to
$3.4 million for the Idaho project. This cost includes gov-
ernment cost-share payment, technical assistance, infor-
mation and education expenditures, and local administra-
tive costs.
Private costs are the net costs the farmer incurs from
paying his share of the BMP installation, plus the net
change in operating costs. Notice that the private costs in
the Idaho project are very high, nearly equal to govern-
ment costs. By comparison, in the Vermont Project net
private costs are zero because the reduction in operating
costs exceeds the installation cost, so the negative cost
gets listed as a private benefit.
Benefits Versus Costs
How do the estimated benefits in the three comprehensive
monitoring and evaluation projects compare with the costs
of implementing the projects to generate the benefits?
The answer to this question is affected by which benefits
we compare with which costs. First, let's compare total,
benefits, including both public and private, with total
costs, again including both government (or public) and
private. The Vermont project with a benefit/cost ratio of 1.8
to one is the only project of the three that is economically
justified (Table 3). For this project, total economic benefits
exceed costs. In the Idaho and Illinois projects, total eco-
nomic benefits are only one-fourth or less as large as total
costs.
If we are interested in just comparing public benefits
with public costs, and include productivity benefits as a
public benefit, the result changes slightly. The Vermont
project is still the only project of the three with benefits
exceeding costs, but its benefit to cost ratio drops to 1.3
while the ratios for the Idaho and Illinois projects improve
slightly, but still remain low.
If we say that these projects were undertaken to im-
prove water quality and produce offsite benefits, and we
are interested in how much we are getting for the govern-
ment buck, we would compare offsite benefits against
government costs. When we do this, the benefit to cost
ratio for the Idaho project drops to 0.2 while the others
remain the same.
LIMITATIONS
Before drawing some implications from these evaluations
of benefits and costs, several limitations need to be
235
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
pointed out. First, these evaluations are preliminary and,
at best, give only ballpark numbers. Second, the RCWP
projects were not selected on the basis of anticipated ben-
efit/cost ratios, but rather to experiment or try out the pro-
grams in different problem and geographic situations. Al-
though the Idaho, Illinois, and some other RCWP projects
may have low benefit/cost ratios, the information they pro-
vide will be valuable for guiding future programs.
A third limitation is that the RCWP projects are not sta-
tistically representative of possible agricultural nonpoint
source projects in general. Thus, the benefit/cost results
should not be used to generalize about the economic effi-
ciency of a future program.
IMPLICATIONS
The preliminary results of the three projects do allow us to
draw some implications regarding priorities for allocating
public resources and determining actions to take on non-
point source pollution control.
1. Levels of pollution per se are not a good Indica-
tion of the economic damages being caused by NFS
pollution, or of the likely benefits to the public from its
control. For example, all three projects had highly pol-
luted water, but the public benefits from controlling the
pollution ranged from under $250,000 to nearly $5 million.
2. Economic benefits of NPS pollution control can
vary dramatically among areas. A key factor appears to
be how many people are being affected, particularly with
regard to recreational opportunities. In Vermont, the likely
recreational benefits were sizeable, while in the other two
projects they were low. Also, some projects, such as in
Idaho, will generate onsite soil productivity benefits, while
others will not. One policy question is whether offsite and
productivity benefits receive the same or differing priori-
ties in allocating resources. It could make a major differ-
ence.
3. Some anticipated benefits may actually be negli-
gible. This appears to be the case in both the Idaho and
Illinois projects with regard to water supply benefits, for
example. Thus, careful preproject assessment is neces-
sary.
4. Costs of controlling NPS pollution can differ
greatly among areas, and also within an area, depend-
ing on the measures used. The Idaho and Vermont proj-
ects are experiencing high costs because of implementing
structural measures, while Illinois is relying mostly on less
costly management practices. In Idaho, experience and
evaluation have shown that conservation tillage and irriga-
tion water management will both reduce costs and gener-
ate benefits, compared with costly irrigation and sedimen-
tation retention structures. This redirection in practices is
now being implemented in the Idaho project.
Another factor in costs is that point as well as nonpoint
pollution control may be necessary to generate benefits.
For example, the threshold for getting changes in im-
paired uses in the Vermont project could not be reached
by controlling one problem and not the other.
5. Benefit/cost ratios, which bring together both the
benefit and cost sides, are the best indication of where
government and the public will get the most for money
spent. For example, in which of the three RCWP projects
is the public getting the most for its money? Clearly, in the
Vermont project. If we have limited funds and we want to
provide as many benefits to society as possible, where do
we target our future efforts? We think you will agree, to
areas where the estimated economic benefits are highest,
compared with costs. If we don't consider benefit/cost ra-
tios, we will most likely find out that we have spent a high
proportion of the costs to get a small proportion of the
benefits.
We hope we have illustrated our contention that both
the government and the public will benefit from giving
greater consideration to economics in controlling agricul-
tural runoff. Information and procedures are becoming
available to make pre-project economic assessment feasi-
ble as an aid to project selection.
236
-------
SOIL EROSION AS A NONPOINT SOURCE—A FARMER'S
PERSPECTIVE
ROBERT WARRICK
Warrick & Son's Farms
Meadow Grove, Nebraska
As a full time farmer and an environmental activist in the
state of Nebraska, I do have a somewhat different per-
spective on soil conservation. As a farmer, I am concerned
about the loss of soil and as an environmentalist I am also
concerned about the damage it does to our rivers, lakes
and oceans, thus it was natural that I should try to make
one of the nation's oldest and most active environmental
organizations become involved in protecting our soil re-
sources. This is the third year that the Sierra Club has
made the protection of agricultural soil resources one of
its top priorities. Having served on local conservation dis-
trict boards and having been active in the formation of
natural resource districts (NRD's) I have a very great inter-
est in controlling and stopping the washing away of our
nation's most basic resource—soil.
In the midst of emergency conditions across the Farm
Belt, Congress faces the task of reauthorizing the coun-
try's basic agricultural policies and programs in the 1985
Farm Bill. While current economic problems will dominate
the debate, the time has also come to address a major
threat to the long-term productivity of our farmlands—ero-
sion of valuable topsoil.
Soil erosion is worse now than in the Oust Bowl days
when wind and water erosion damaged 282 million acres
of American farmland. The U.S. Department of Agricul-
ture (USDA) estimates that soil is eroding at more than
twice the tolerable levels on 96 million acres, or 23 percent
of the nation's cropland—an area the size of California. A
USDA survey warns that at current erosion rates, corn
and soybean yields in the Corn Belt states may drop by as
much as 30 percent in the next 50 years as soil fertility
declines.
In addition to threatening productivity, soil erosion con-
tributes to air and water pollution, and costs farmers and
the public billions of dollars each year. Dust dirties our air
and reduces visibility, while runoff from farms chokes our
rivers and often contains fertilizer and pesticide residues.
Erosion costs farmers at least $625 million a year in re-
duced yields, extra fertilizer and soil conservation mea-
sures, while water quality damage attributed to soil ero-
sion is estimated at $2 to $6 billion annually.
Our present soil conservation programs date back to
the 1930's. But they have done far too little for the billions
of dollars spent, and in fact, have often contributed to soil
erosion problems. This year there is an opportunity to
enact major soil protection legislation as part of the 1985
Farm Bill. The Agriculture Committees of both the House
of Representatives and the Senate have jurisdiction over
the legislation.
The Sierra Club believes as I do that soil erosion control
should be focused on prevention of the problem. Sensible
farmland management using such simple techniques as
contour plowing and crop rotation can dramatically reduce
soil loss. Remedial measures for badly eroded lands also
need to be taken.
The Sierra Club supports a 1985 Farm Bill that:
1. Includes strong "sodbuster" provisions that deny
USDA program benefits on all land a farmer owns or has
an interest in if he plows up highly erodible land;
2. Establishes a 30-million acre conservation reserve
by paying farmers reasonable annual fees to convert se-
verely eroding land to such sustainable uses as wildlife
habitat, hay or pasture;
3. Makes farmers' eligibility for USDA commodity pro-
grams conditional on the implementation of sound soil
conservation practices;
4. Funds research programs in alternative farming
techniques that not only reduce farmers' dependence on
expensive inputs like chemical fertilizers and pesticides,
but also conserve soil and water; and finally
5. Includes a "swampbuster" provision that would
deny all federal farm program benefits to farmers who fill,
drain, or otherwise convert wetlands to croplands.
Right now it looks good that most of these conservation
measures will be included in the 1985 Farm Bill, but they
need your support. Write your Senator and Representa-
tive expressing your support for a Farm Bill that includes
strong soil conservation provisions.
237
-------
Agricultural Issues:
Eastern and Southern
Experience
NONPOINT SOURCE POLLUTION: SCS PERSPECTIVE
ERNEST V. TODD
State Conservationist
Soil Conservation Service
Auburn, Alabama
Nonpoint source pollution is the next major environmental
issue facing this Nation. If we are ever to reach that cher-
ished goal of swimmable, fishable waters throughout this
land, then the problem of pollution from these diffuse
sources must be seriously addressed. The American peo-
ple want clean water, according to many public opinion
polls, and they are willing to pay for it. (Counc. Environ.
Qua), et al. 1980; Engineering News Record, 1982)
In an EPA report to Congress (U.S. Environ. Prot.
Agency, 1984), there is a table listing the response of each
State water pollution control agency to the question, "Do
nonpoint sources cause a problem in your State?" Twenty
States reported it to be a major problem, while the rest
indicated that it was just a problem or potential problem.
Alabama, at the time of the survey in 1982, reported NPS
pollution to be just a problem. I am confident, however,
that it would now be rated a major problem. The director of
the Alabama Department of Environmental Management,
the water quality regulatory agency in the State, recently
stated that the industrial sector has done much to clean up
its act, but industry and others are now pointing their col-
lective finger at agriculture. Moreover, he has seen a rapid
increase in complaints in the past 2 years regarding agri-
cultural nonpoint source pollution.
A major difference between industry and agriculture is
that industry can pass on the cost of pollution control to
the consumer, whereas agriculture cannot. A recent study
in Alabama suggests that small-scale hog operations
would have a difficult time remaining in business if forced
to install traditional, expensive waste management sys-
tems, especially if cost-sharing is not provided. Agriculture
cannot be called upon to pay the total direct cost of pollu-
tion control because the mechanism is not and has never
been in place to readily pass on the added cost to the
consumer.
In those States that provide cost-sharing funds, State
soil and water conservation boards or committees and
districts are handling the administrative procedures. In
many of these States, conservation districts have their
own technical people. In others, the State relies upon the
Soil Conservation Service for technical support. Where
Federal funds are used, the institutions providing financial
and technical assistance are also in place. This includes
the Agricultural Stabilization and Conservation Service,
which administers cost-share funds, and SCS and the Co-
operative Extension Service, which provide technical as-
sistance. In many States, these organizations are working
in cooperation with 208 planning committees.
For its part, the SCS, with its conservationists and tech-
nicians located in nearly every county throughout the
country, is uniquely situated to deal with nonpoint source
pollution problems. In fact, we in SCS have been dealing
with them directly and indirectly for a number of years.
Each year our people are installing hundreds of miles of
terraces, grassed waterways, and other engineering prac-
tices to reduce erosion. And our field people have been
actively promoting conservation tillage throughout the Na-
tion. Although conservation tillage greatly reduces soil
loss and the loss of phosphorus in surface runoff, we are
concerned that the increased use of "burn down" herbi-
cides in conservation tillage could adversely affect water
quality. We have been studying this issue but, unfortu-
nately, all of the answers are not yet known.
SCS has been active in many areas in our efforts to
reduce nonpoint source pollution. For instance, SCS is
providing special training in water quality to our field per-
sonnel. By the end of this year, approximately 80 percent
of our people nationwide will have had at least 15 hours of
training on nonpoint source pollution.
SCS has also been involved in the development of the
CREAMS computer model—CREAMS being an acronym
for chemicals, runoff, and erosion from agricultural man-
agement systems. The model was developed by the Agri-
cultural Research Service with input from SCS specialists.
239
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
We have conducted week-long training sessions for se-
lected State office personnel on CREAMS, and the model
is now being used extensively by SCS field staffs through-
out the Country
SCS, the States' soil and water conservation districts,
and related agencies are working on water quality projects
as never before. We are deeply involved in salinity control
in the West and in the reduction of nutrients into the Great
Lakes in the North. The 21 experimental Rural Clean Wa-
ter Projects and the Model Implementation Program have
required thousands of manhours of technical assistance
from SCS and district personnel throughout the country.
The thrust of these programs has been to reduce nonpoint
source pollution in watersheds having especially critical
problems.
SCS is involved in many other water quality programs,
including our ongoing conservation programs that are
daily reducing nonpoint source pollution: the Chesapeake
Bay cleanup, the Cornell University/SCS workshops on
water quality, and our contracted studies in the Northwest
on bacterial pollution. And the list goes on. The question
now is, where do we go from here?
We do not know yet how the budget cuts will affect
SCS's overall program. If we are cut too severely, we may
be forced to address only a narrow range of soil conserva-
tion programs. To be sure, these programs are important
in reducing nonpoint source pollution, but they will not
provide the level of water quality protection we would like
to provide.
We have an organizational structure that lends itself to
working with agricultural nonpoint source pollution. Our
technical field personnel are working closely with farmers
and ranchers on a daily basis and have established with
them a rapport that is often needed to encourage them to
install pollution control practices. We feel that friendly per-
suasion is a whole lot better than forced legislation wher-
ever possible.
We need more research on best management prac-
tices, especially those for controlling pesticides. We know
more than we used to about pesticide runoff but we still
don't know enough.
Financial incentives for pollution control practices are
essential if we are to see progress in reducing agricultural
NPS pollution. If the practices we recommend do not in-
crease crop yields or the margin of profit for livestock and
poultry producers, we will be fighting a losing battle with-
out financial incentives.
I am proud of what we have already accomplished in
nonpoint source pollution control. And I am excited about
what could lie ahead in this area for SCS and for those
agencies with whom we are closely allied.
If, however, the public is not adequately concerned
about our work, the effort to reduce nonpoint source pollu-
tion will be slow at best. Perhaps this conference will be
the foundation on which a public awareness program is
built. For the sake of our precious water resource, I cer-
tainly hope it will be.
240
-------
NONPOINT SOURCE POLLUTION FROM PLANT NUTRIENTS
O. R ENGELSTAD
K. S. BRADY
Division of Agricultural Development
Tennessee Valley Authority
Muscle Shoals, Alabama
Of the essential nutrients for plant growth, nitrogen and
phosphorus are of greatest concern as pollutants. How-
ever, these two nutrients differ markedly in the way they
act in the soil. Nitrogen is a mobile nutrient, while phos-
phorus is immobile. These nutrients are supplied in fertil-
izer to supplement soil supplies.
Soil nitrogen is released from organic forms. Decompo-
sition of soil organic matter by microbial action at first
produces ammonium. Ammonium ions are either ad-
sorbed at the cation exchange sites of soil clay minerals
or, with good aeration and temperatures conducive for
nitrifying bacteria, are converted to nitrate. In the case of
surface erosion, nitrogen is removed from cropland pri-
marily as paniculate organic matter or as ammonium at-
tached to soil particles. Once nitrate is in the soil solution,
however, it is subject to leaching and can ultimately enter
the ground water. Fertilizer nitrogen is subject to these
same processes.
Phosphorus is another case. Any phosphorus that ap-
pears in the soil solution, either from fertilizer or from the
decomposition of organic matter, can become attached to
soil clay minerals as a phosphate complex or be rapidly
converted to an inorganic form of metal phosphate, the
type depending on soil aeration and pH conditions. All of
these inorganic forms of phosphates in the soil are rela-
tively insoluble and result in low concentration in soil solu-
tion (generally <0.2 mg/L). Hence, phosphorus losses to
surface waters generally result through soil erosion.
Losses of phosphorus and inorganic nitrogen to surface
waters can be reduced through soil management prac-
tices that slow surface runoff. The question remains as to
the losses of nitrate to ground water because of its mobil-
ity.
Plant nutrients from either soil or fertilizer become po-
tential pollutants when they are present in quantities that
exceed either plant requirements or the capacity of the
soil to act as a reservoir for future use in plant uptake. To a
large degree, the soil type and climate, as well as agricul-
tural practice, affect the movement of plant nutrients. For
example, where irrigation is practiced, salts as well as
nitrates can become a ground water problem. In the prai-
rie regions, the residual fertility of these soils can add to a
nitrate ground water problem even without the addition of
fertilizer. The potential for nonpoint source pollution from
plant nutrients is a natural one, and the degree to which
the addition of fertilizer components exacerbates this
problem may well be a question to be answered in the
context of regional background levels based on soil type,
climate, the quality of runoff, surface waters, and ground
water of areas under natural cover. When the magnitude
and variability of the water quality problem is illustrated by
regional studies, a single comprehensive policy may not
be practical.
MAGNITUDE OF POLLUTION FROM
PLANT NUTRIENTS
Studies from such different regions as the Chesapeake
Bay, the Great Lakes, the cropland areas of Kentucky, and
other parts of the South including the Coastal Plains soils,
underscore the potential seriousness of the problem cre-
ated by plant nutrients when they are displaced and be-
come pollutants. The Chesapeake Bay and Lake Erie
studies reveal that excessive runoff causes pollution by
plant nutrients. The Lake Erie Wastewater Management
Study concentrated on phosphorus losses accompanying
soil erosion from cropland. Phosphorus is considered to
be the limiting factor in eutrophication of Lake Erie, and
the Study's objective was to improve water quality by con-
trolling phosphorus. In 1980, the Study revealed that
8,400 MT/yr or 51 percent of the total Lake Erie phos-
phorus loading came from runoff (nonpoint source) from
rural land, principally cropland (U.S. Army Corps Eng.,
1983).
The Chesapeake Bay study indicated that nonpoint
source runoff from cropland is currently responsible for
the greatest amount of nutrient load to the Bay. Nonpoint
sources contribute approximately 67 percent of the nitro-
gen and 39 percent of the phosphorus load to the Bay
(Macknis, 1984).
In the South, the source of nitrates in ground water has
not been well identified nor have fluctuations in concentra-
tion been as sensational as in other parts of the country A
study of nitrogen contents in shallow ground water in the
North Carolina Coastal Plain indicates that nitrate concen-
trations in the upper part of the ground water under fertil-
ized cropland are greater than under adjacent wooded
areas, and that nitrate levels are most likely to be higher
under the better drained soils than under adjacent wet
sites (Gilliam et al. 1974). A greater incidence of high wa-
ter tables and tighter soils may contribute to a greater
incidence of denitrification, especially with respect to
Coastal Plain soils. However, separate studies in Georgia
and North Carolina showed nitrate concentrations in-
creased in surface waters after stream channelization
(Bliven et al. 1980; Heath, 1975). The increases were at-
tributed to the deeper channelization penetrating the shal-
low ground water table.
Many studies of nitrate concentrations in stream waters
are concerned with subsurface flow. In Georgia and else-
where most of the nitrate applied as nitrogen fertilizers is
moved into the soil during the first few minutes of rainfall
(White et al. 1967). Some of it then reappears in surface
waters from tile drainage and subsurface flow. Subsurface
flow of water in soils underlain by plinthite in the upland
Georgia Coastal Plain is responsible for 99 percent of the
total nitrate lost (Hubbard and Sheridan, 1983). Studies of
North Carolina Coastal Plain soils show that nitrate enters
surface streams via tile drainage flow (Gilliam et al. 1978).
In the Lake Erie Basin, monitoring of storm events shows
that most of the nitrate reaches surface water sampling
stations during the falling portions of the hydrograph, in
contrast to sediment and phosphorus.
The contribution of this nitrate to drinking water can be
significant, and has reached major proportions in the San-
dusky River in northwestern Ohio. Nitrate exceeds the
standard limit for drinking water about 4 percent of the
year, and about 16 percent of the time during May, June,
and July (Baker, 1985). In addition to high mineralization
and nitrification of organic N during those months, fertil-
241
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
izer use is most frequent then. Total N losses via stream
flow are equivalent to about 43 percent of the fertilizer N
applied (Baker, 1985). Some of this loss could, of course,
have originated from organic sources; but in any event,
serious implications for the fertilizer industry are evident.
Knowledge about the link between fertilizer nitrogen ap-
plication and the nitrate concentration in drinking water
also comes at a time when increased nitrogen application
rates on agricultural lands are predicted. While a case can
be made that fertilizer nitrate losses need to be weighed
against potential improvements in nitrate utilization by the
crop, the fact remains that nitrogen placed on croplands in
a single application early in the season may be suscepti-
ble to loss to surface and ground waters.
CONTROL PRACTICES
Some steps have been taken toward abating pollution
caused by plant nutrients. The first two steps in abatement
usually involve identifying and inventorying the extent of
the problem and projecting reasonably attainable goals (in
terms of time, cost, and percent reduction).
The Lake Erie Wastewater Management Study can
serve as an example. This investigation of nonpoint
sources of phosphorus loading led to identifying soil ero-
sion from cropland as a major source of the problem. The
second step was to make projections on reducing soil
erosion. This depended on the types of conservation prac-
tices instituted. A land resource information data base
helped calculate potential soil erosion in the Lake Erie
Basin. The conclusion: reduced tillage in suitable areas
would achieve a 40-percent reduction of the erosion po-
tential, and both no-till and reduced tillage could attain a
reduction of 69 percent of the erosion potential in the Lake
Erie Basin. The present goal is to reduce erosion by 48
percent, thereby reducing phosphorus loading of Lake
Erie by 32 percent over a 20-year period, with 90 percent
of the reduction occurring in the first 7 years (U.S. Army
Corps Eng. 1983).
With goals identified, erosion control depends on a set
of methods known as best management practices. BMP's
tailor a management system for a specific site and can
depend on a number of factors ranging from environmen-
tal, land, and economic considerations to effectiveness of
a certain practice. For example, in certain parts of the
Lake Erie Basin where plant-available phosphorus levels
can be high, yearly soil tests are suggested to determine if
phosphorus supplements are even needed during that
growing season. In North Carolina and Virginia, buffer
strips along waterways are employed to entrap phos-
phorus and associated sediment originating from field run-
off, thus reducing nutrient loading and turbidity.
BMP's that control phosphorus are linked to controlling
erosion. However, considerable question remains as to
the control of nitrogen. Studies in North Carolina using
buffer strips for sediment and phosphorus control have
been expanded to consider nitrate control, especially from
tile drainage and natural subsurface flow. Because the
woody shrubs and trees in this riparian ecosystem-buffer
zone can remove nutrients from subsurface flow, attempts
have been made to study and manage this alluvial storage
of nutrients for uptake by riparian vegetation. Careful man-
agement of these zones can prolong the usefulness of this
riparian ecosystem as an energy source and as a nutrient
filter (Lowrance et al. 1985).
The no-till system has received increased interest as
both an energy-saving and erosion control measure in the
South and East. A comparative study in Maryland of nutri-
ent losses from two watersheds, one in conventional corn
production and the other in no-till corn, reports a ninefold
decrease in the amount of runoff with no-till corn; a twenty-
fold decrease in total P runoff with no-till corn; and a cor-
responding tenfold decrease in both NO3-N and total ni-
trogen in runoff with no-till corn (Angle et al. 1984). A
similar decrease in nitrate runoff was seen in no-till soy-
beans on soils formed in northern Mississippi loess (Mc-
Dowell and McGregor, 1980). However, studies of ground
water pollution from nitrate, particularly in the Midwest,
cause concern about this nutrient moving into the ground
water, especially when soils are of medium texture and
well drained.
UNCERTAINTIES IN PLANT NUTRIENT
CONTROL
As the preceding discussion intimates, the nutrient of ma-
jor interest is nitrate. Largely because of its high mobility,
nitrate may be extremely difficult to manage. Studies con-
cerning the mobility of nitrogen in the plant/soil environ-
ment underscore growing concerns about nitrate's role as
a source of pollution. Despite other researchers' concerns
that the potential for leaching of nitrates to the ground
water may be greater with no-till, a study in Kentucky
seems to indicate that the potential for nitrate loss to
ground water is related more to application rates than to
tillage practices (Smith et al. 1984). Furthermore, ques-
tions remain as to the percent N tied up in crop residues
and microbial populations, and the percent N that could
be recovered in the grain a year after application.
Increased interest in residue management for no-till cul-
tivation has also increased concerns about nitrogen leach-
ing. Residue management in no-till requires the subsur-
face application of fertilizer and pesticides in the soil to
reduce their loss from runoff without residue incorporation
(Baker and Laflen, 1985).
Researchers have established the benefits of using and
managing the riparian zone parallel to the Atlantic Coastal
Plain rivers to control nitrate in subsurface flow and sedi-
ment and phosphorus in surface runoff. But this buffer
zone is itself an ecosystem, and is of value only if it is used
and monitored wisely. It cannot be regarded as a sink. The
extent to which this nitrogen can be managed so that it
remains in place until used by the riparian vegetation is a
crucial question.
Also, predicted increases in ground water pollution
based on projections of continued (and growing) high ni-
trogen rates, tillage practices, and other agricultural prac-
tices are questionable because of current efforts to inhibit
N losses by controlling nitrification rates. If nitrification is
controlled in such a way that nitrate is made available
when needed by the plant, the nitrogen application rate
may actually decrease in the future. The same concept
holds for controlling losses of fertilizer nitrogen from vola-
tilization. Reducing volatilization can decrease the rate of
nitrogen application, and perhaps result in cutting pro-
jected losses to surface and ground waters.
NEEDS FOR FUTURE RESEARCH
A discussion of uncertainties necessitates deliberation of
future research needs. The control of phosphorus losses
through control of runoff and erosion is more certain than
control of nitrogen losses. Future research must attempt
to understand more completely the role of fertilizer nitrate
as a nonpoint source pollutant.
Attempts to understand nitrate from a fertilizer point of
view will involve studies in:
1. Use of various forms (ammonium, amide, nitrate) of
N-fertilizers;
2. Use of urease and nitrification inhibitors;
3. Fertilizer application timing and placement system
as it relates to crop requirement;
242
-------
4. More thorough 15N tracer accounting of nitrate path-
ways and losses;
5. Assessment of natural background levels of nitrate
in ground water for different climatic and soil regions;
6. Assessment of BMP's for controlling nitrates given
the crop under production; and
7. Careful monitoring and use of the riparian ecosys-
tem for the eastern Coastal Plain.
Studies show that nitrate can be a problem in some
areas. Is it confined to these identified pockets or is it a
more general problem? We need to know how serious and
pervasive nitrate is as a nonpoint source pollutant in
ground water. Research should be expanded to provide
this information.
REFERENCES
Angle, J.S. et al. 1984. Nutrient losses in runoff from conven-
tional and no-till corn watersheds. J. Environ. Qua). 13:431-5.
Baker, D.B. 1985. Regional water quality impacts of intensive
row crop agriculture: A Lake Erie Basin case study. J. Soil
Water Conserv. 40:125-32.
Baker, J.L., and J.M. Laflen. 1983. Water quality consequences
of conservation tillage. J. Soil Water Conserv. 38:186-93.
Bliven, L.F., F.J. Humenik, F.A. Koehler, and M.R. Overcash.
1980. Dynamics of rural nonpoint source water quality in a
southeastern watershed. Trans. Am. Soc. Agric. Eng. 23:
1450-6.
AGRICULTURAL ISSUES: EASTERN AND SOUTHERN EXPERIENCE
Gilliam, J.W., R.B. Daniels, and J.F. Lutz. 1974. Nitrogen con-
tents of shallow ground water in the North Carolina Coastal
Plain. J. Environ. Qual. 3:147-51.
Gilliam, J.W., R.W. Skaggs, and S.B. Weed. 1978. An evaluation
of the potential for using drainage control to reduce nitrate
loss from agricultural fields to surface waters. Rep. No. 128.
Water Resour. Res. Inst. Univ. North Carolina.
Heath, R. 1975. Hydrology of the Albemarle-Pamlico Region of
North Carolina. U.S. Geolog. Surv., Raleigh, NC.
Hub-bard, R.K., and J.M. Sheridan. 1983. Water and nitrate-
nitrogen losses from a small, upland coastal plain watershed.
J. Environ. Qual. 12: 291-5.
Lowrance, R., R. Leonard, and J. Sheridan. 1985. Managing
riparian ecosystems to control nonpoint pollution. J. Soil Wa-
ter Conserv. 40: 87-91.
Macknis, J. 1984. Findings and recommendations of the Chesa-
peake Bay study—nutrient aspects. Speech given to The Fer-
tilizer Inst. at Resour. Use Subcomm. Meet. Jan. 26. U.S.
Environ. Prot. Agency, Washington, DC.
McDowell, L.L., and K.C. McGregor. 1980. Nitrogen and phos-
phorus losses in runoff from no-till soybeans. Trnas. Am. Soc.
Agric. Eng. 23: 643-8.
Smith, S. et al. 1984. The fate of nitrogen fertilizer applied to no-
till corn. Agronomy Res. Rep. Progr. Rep. 281. Univ. Ken-
tucky, Lexington.
U.S. Army Corps of Engineers. 1983. Summary report of the
Lake Erie Wastewater Management Study. Buffalo, NY.
White, A.W., A.P. Barnett, W.A. Jackson, and V.J. Kilmer. 1967.
Nitrogen fertilizer loss in runoff from cropland tested. Crops
Soils Manage. 19(4): 28.
243
-------
NONPOINT SOURCE POLLUTION: MANAGING NUTRIENTS
A KEY TO CONTROL
GEORGE B. WOLFF
Pennsylvania State Conservation Commission
Harrisburg, Pennsylvania
The problem of nonpoint source pollution came to light
with the report from the Chesapeake Bay Study Commit-
tee and was confirmed by the Lake Erie Waste Water
Management Study and reconfirmed by the Lake Wallen-
paupack Study. While different agencies performed each
study, the findings were almost an exact schematic. All of
them found the major problem was excess nutrients: nitro-
gen and phosphorus. Nitrogen is very unstable, and can
be leached from the soil. Phosphorus, on the other hand,
very rapidly attaches itself to soil particles, and if it leaves
the soil, it leaves through erosion.
We have discovered that we know quite a lot about
controlling erosion, and by applying the best conservation
practices we can, to a great degree, eliminate erosion,
thereby reducing the phosphorus loss. But, while we're
doing this we're slowing down the flow of water. More of it
percolates into the ground carrying with it nitrogen, leav-
ing us with a "Catch 22" situation.
A conservation program has to be matched with an
equally efficient nutrient management program. However,
we know very little about nutrient management. A reliable
test for nitrogen in the soil does not exist. Also, the tests
for manure nutrients are so slow in returning from the labs
that they are virtually ineffective. By the time you have the
tests back, you have already spread the manure.
We found almost no research on optimum levels of nu-
trients. Much research shows that you could increase poor
soil yield by applying nutrients, but almost nothing demon-
strates where you begin to have negative yield responses
from too many nutrients. Yet, we are finding farmers who,
from their own operating experience, indicate they have
fields that will give them reduced yields when they apply
more manure. All of this convinces our agricultural people
looking into the nutrient situation that:
1. We must have fast, accurate soil tests for every field;
2. We must have fast, accurate tests for the manure
stored in our waste management facilities;
3. We must develop ways to transport manure from
those farms with a surplus to agricultural lands that need
the excess nutrients; and
4. We must develop new uses for excess nutrients, in-
cluding application to forest land, incorporation and re-
source recovery operations, and methane digestion.
Farmers can see substantial dollar losses (as high as
$90/acre on a wet year) if they don't begin to practice
effective nutrient management.
Private industry has responded to the needs of the
farmer facing net dollar losses. Fertilizer dealers are no
longer just trying to move products; they are requesting
that farmers have soil tests for each field. The soil test
analysis correlates the amount of nutrient currently in the
soil with the nutrient demand of the next crop, thereby
determining the amount of nutrient that should be applied.
The fertilizer dealers have also begun purchasing equip-
ment so that they can split applications of nitrogen on
corn. Farmers responded so positively that the dealers are
now telling them, "We can't service you this year because
our equipment is totally booked up."
Farmers are forming local crop improvement associa-
tions. These associations hire a professional to help ana-
lyze the soil tests, make nutrient application recommenda-
tions, and oversee their herbicide/pesticide programs.
These farmers are paying $4 and $5/acre for this service;
but they are finding that their net savings on nutrients can
total as much as $40/acre and as much as $20/acre in
pesticide applications. These professionals are actually
making moth counts, computing degree days and rainfall,
and helping to space applications so they are done at the
most advantageous time. Many times they are finding that
one or, at times, two whole sprays can be eliminated just
by good management. These are some very positive
steps that agriculture itself has been taking with the gov-
ernment helping only by providing facts on losses incurred
by not adhering to best management practices.
Some areas will be more complicated and will require
governmental assistance. The installation of waste man-
agement facilities is expensive, and responsible cost-shar-
ing between government and agriculture is going to have
to be addressed. Methane digestion is a very effective way
of handling hog and dairy manure; but then you still wind
up with nutrient-rich water, and I think government can
help determine the best ways of handling that resource.
Growing fish and algae and then selling the fish and
recycling the algae through the digester may give almost a
perpetual motion machine—clean water and some profit
in between. Where nitrogen is the problem and phos-
phorus is not, we can spray the effluent on the land, not
work it in, and let the nitrogen aerate off. While this
method is wasteful, it may be the most practical way of
handling the problem. Chicken manure presents an en-
tirely different set of problems. Because of the excess
feathers, this resource may not work in a methane di-
gester.
Therefore, we either have to find extra land for land
application or incinerate and recapture the heat for steam,
using it to make electricity This could very well be a pro-
ductive adjunct to municipal resource recovery plants
since it would allow the plants a more uniform daily BTU
loading. In Pennsylvania we have between 200,000 and
300,000 wood lot owners. The land would make an ideal
area for land application since we would be able to gener-
ate much more wood biomass if we practiced good nutri-
ent management in silviculture.
Obviously, one of the reasons for not applying manure
to woodland is because there has been no practical way to
do it; but today, with the big gun spray irrigation systems
and slurry pumps, this type of application may be not only
practical, but advantageous. Not only do we need a lot
more government help from the experiment and research
side, but we also need people to physically carry the mes-
sage one on one and help in a hands-on way to develop
these many opportunities.
Not long ago, I saw a church bulletin board that said,
"We've all seen many great opportunities brilliantly dis-
guised as impossible situations." I think with the help of all
levels of government, and with the conscious desire of the
farmer's family to save dollars, we will have a very produc-
tive working partnership. I'm equally convinced that if we
educate the people, enough incentive exists—we will not
need laws mandating these practices. Let's convert those
impossible situations into great opportunities.
244
-------
AGRICULTURAL LAND TREATMENT PROJECT PLANNING FOR
OFF-SITE PHOSPHORUS REDUCTION
FRANCIS M. KEELER
U.S. Department of Agriculture, Soil Conservation Service
Winooski, Vermont
Vermont's 1978 Water Quality Plan for Controlling Agricul-
tural Pollution identified excessive phosphorus loading as
the primary nonpoint source of water pollution in eight
major drainages to Lakes Champlain and Memphrema-
gog. The plan also recommends that 21 watersheds within
these drainages be given priority for technical and finan-
cial assistance to treat agricultural nonpoint source pollu-
tion. The Lower Winooski River in northwestern Vermont
is a priority watershed within the central Lake Champlain
basin.
Water quality in central Lake Champlain is deteriorating
under existing phosphorus loads. Algae and weed growth
in the lake plague summer recreationists and lake-shore
residents and cause public beaches to close intermittently.
Water transparency, total phosphorus levels, and the
abundance of diatoms and blue-green algae indicate that
this portion of Lake Champlain has a moderate to high
level of biological activity.
The watershed contains 77 operating dairy farms, aver-
aging 140 ha in size with 100 cows and 50 replacements.
Total manure production is approximately 166,000 metric
tons annually. Corn is grown for silage on the intensively
cultivated cropland and soil erosion rates are as high as
58 metric tons per ha.
Phosphorus is the principal nutrient from nonpoint
source pollution in the 57,000 ha watershed. Watershed
loadings of 14,000 kg annually represent approximately
13 percent of the total basin's point and nonpoint phos-
phorus load (112,000 kg) to central Lake Champlain
(Bogdon, 1978).
Although agriculture was deemed responsible for 62
percent of the watershed's nonpoint phosphorus, a logical
plan of treatment could not be developed without deter-
mining the relative contribution of each farm. Toward this
end, Soil Conservation Service (SCS) specialists used
four computer models to assess the amount of phos-
phorus generated by each source on individual farms. The
five major agricultural sources of phosphorus evaluated
were: soil eroded from cropland, livestock concentration
areas, field-spread manure, milking center wastewater,
and field stacks of manure. After comparing the relative
amount of phosphorus loading by each farm, SCS and
.local authorities designated 52 farms for land treatment
priority. In those priority areas the models were used to
select the most cost effective treatments.
Phosphorus from eroded soil was calculated with the
Phosphorus Reduction Model (PHRED). The basis for this
model's soil- erosion calculations was the Universal Soil
Loss Equation (USLE) (U.S. Dep. Agric. 1978). Soil loss
from sheet and rill erosion was quantified field-by-field us-
ing USLE's five variables: rainfall, soil credibility, slope
length and steepness, cover and management, and exist-
ing treatment practices. A soil's total content of adsorbed
phosphorus was established through laboratory analysis
of each soil. The total amount of phosphorus from erosion
was then determined by the PHRED model as the product
of: soil loss per unit area per year, field size, the soil's total
phosphorus concentration, an enrichment rate to allow for
the finer soil particles in sediment, and a sediment deliv-
ery ratio.
The model measured the effectiveness of Best Manage-
ment Practices (BMP's) through an evaluation of erosion
reduction and the corresponding phosphorus levels.
PHRED also quantified phosphorus from field-spread ma-
nure. For this the model incorporated the following varia-
bles: phosphorus needs of the crop grown, the rate and
season of manure application, and incorporation of the
manure into the soil. The total quantity of manure for a
given farm was determined by the number and breed of
livestock. The amount of manure available for spreading,
however, was based upon a proportioned reduction of this
total to reflect pasturing practices. The model applied the
available manure seasonally to cropland and hayland at
rates identified by the farmer. Phosphorus loss was calcu-
lated using mass values in quantities per unit area derived
from literature sources (U.S. Dep. Agric. 1979).
For field-spread manure, the PHRED model evaluated
treatment by reducing manure application levels to the
recommended agronomic rate (the rate calculated to be
beneficial for plant growth by meeting a particular soil's
fertility requirements.) The model also considers any sea-
sonal variations of application or incorporation due to
proper manure storage and utilization. BMP accomplish-
ments were determined by the extent of changes in quan-
tities of manure applied, season of application, and proper
incorporation.
Livestock concentration area (LCA) runoff was evalu-
ated with the Barnyard Area Runoff Nutrient Yield pro-
gram (BARNY). This program utilized methods described
by Young etal. (1981).
BARNY determined runoff from an LCA by using the
size and hydraulic characteristics of the drainage areas
entering the LCA and the seasonal distribution of rainfall
events of various size. An average manure pack for each
season was also estimated using animal type, amount of
daily use, and scraping interval. Seasonal phosphorus
runoff was the product of runoff, and the phosphorus con-
centration determined by the average manure pack. The
model also predicted phosphorus reduction when runoff
from the LCA passed through a vegetative buffer. The
amount of reduction was established by evaluating the
slope gradient, slope length, and cover condition of the
buffer.
BARNY helped analyze treatment of LCA's with BMP's
that would change the drainage area size, animal use, and
scraping intervals or increase the buffer efficiency.
Phosphorus contributed by a farm's milkhouse waste-
water effluent was predicted with a model similar to that
used for LCAs. Water usage within the milkhouse was
considered as a factor of both fixed and variable needs
(U.S. Dep. Agric. 1975). Those needs considered fixed for
each farm did not vary by size or type of system and
included such items as cleaning the bulk tank. Variable
needs, such as cow preparation and equipment cleaning,
were influenced by the type of milking system and herd
size. Phosphorus output was calculated as the product of
total water usage and an effluent concentration (Regan et
al. 1981). The effect of a vegetative buffer was determined
with the same equations used in the Barnyard Area Run-
off Nutrient Yield Program.
245
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Plan development anticipated milkhouse waste treat-
ment with BMP's that would reduce the amount of efflu-
ent, or change the vegetative buffer factors in the model.
The model used to identify the amount of phosphorus
contributed by manure stacked in the watershed was
based upon information reported by Draper et al. (1979).
Phosphorus from this source was predicted using re-
ported levels of runoff and amounts of manure stacked. A
distance of travel in overland flow was also considered for
linear attenuation of phosphorus as discussed in the
study Treatment in this case assumed complete control
with an approved storage system.
The watershed protection plan developed with these
four models will reduce biologically available phosphorus
from agricultural sources by 51 percent when fully imple-
mented. This reduction will be accomplished by BMP's
installed at a cost of 1.3 million dollars, shared equally
between the Soil Conservation Service and private land-
owners. BMP's planned include conservation cropping
systems, conservation tillage, contour farming, stripcrop-
ping, and animal waste management systems.
Although these four models need further validation,
their use during planning of the Lower Winooski River
watershed provided three major benefits. First, aggrega-
tion of the data by farm enabled planners to specify the
major sources of pollution and to set priorities for treat-
ment. Second, treatment areas could be evaluated and
the cost efficiency and accomplishments of various alter-
natives identified for decision makers. Third, such target-
ing of planned treatment provided for a greater efficiency
in the use of Federal expenditures for treating agricultural
nonpoint source pollution in the Lower Winooski Water-
shed.
REFERENCES
Bogdon, K.G. 1978. Estimates of annual loading of phosphorus
to Lake Champlain. New England River Basin Comm.,
Burlington, VT.
Draper, D.W., J.B. Robinson, and O.R. Coote. 1979. Estimation
and management of the contribution by manure from livestock
in the Ontario Great Lakes Basin to the phosphorus loading of
the Great Lakes. In Best Management Practices For Agricul-
ture and Silviculture, Proc. 1978 Cornell Agric. Waste Manag.
Conf. Ann Arbor Science Publ. Inc.
Regan, R.W., R.R. Wright, and W.R. Detar. 1981. Study and
Evaluation of Water Quality Data Obtained from 19 Animal
Waste Lagoons in Six Northeastern States. Vol. I and II. Inst.
Res. Land Water Resour. Pennsylvania State Univ., State Col-
lege.
United States Department of Agriculture, Soil Conservation
Service. 1975. Agricultural Waste Management Field Manual
[Section 210-ENG SCS Directives System]. Washington, D.C.
United States Department of Agriculture. 1978. Predicting Rain-
fall Erosion Losses. Agriculture Handbook No. 537. Washing-
ton, D.C.
United States Department of Agriculture. 1979. Animal Waste
Utilization on Cropland and Pastureland, A Manual for Evalu-
ating Agronomic and Environmental Effects. USDA Utilization
Res. Rep. No. 6. EPA 600/2-79-059.
Young, R.A., M.A. Otterby, and A. Roos. 1981. An Evaluation
System to Rate Feedlot Pollution Potential. USDA-ARM-NC-
17, Agr. Res. Serv., U.S. Dep. Agric., Washington D.C.
246
-------
Agricultural Issues:
Midwestern Experience
IDENTIFYING CRITICAL NFS CONTRIBUTING WATERSHED AREAS
KENT W. THORNTON
DENNIS E. FORD
FTN Associates
Little Rock, Arkansas
ABSTRACT
By general agreement, nonpoint sources contribute sig-
nificantly to receiving system water quality problems.
Sediment is the greatest pollutant both by weight and
volume. Nutrients, pesticides, and other constituents may
be adsorbed to sediment particles and transported to the
receiving system through erosion. Best management
practices (BMP's) are usually expensive to implement;
however, they do not have to be implemented throughout
the entire watershed to control NPS. In many watersheds,
certain critical areas contribute the majority of the pollu-
tants. Identifying these areas permits the most efficient
implementation of BMP's and the most economical ap-
proach for controlling NPS. Eleven major steps in identi-
fying these critical areas are discussed. These range
from identifying the physiographic characteristics through
cost/benefit analysis and implementation of BMP's. This
approach forms the base for nutrient, organic, or other
mass loading. An example from an rural watershed is
given.
by weight and volume (Vanoni, 1977). Nutrients, pesti-
cides, and other constituents may be adsorbed to sedi-
ment particles and transported to the receiving system
through runoff and erosion (Green et al. 1978; Johnson et
al. 1976; Karickhoff and Brown, 1978). Achieving "fish-
able-swimmable" objectives in aquatic systems, then, re-
quires control and regulation of both point and nonpoint
sources.
Point sources are relatively easy to identify and control.
Treatment technology is available and affordable and efflu-
ent regulations generally have been promulgated. Non-
point sources, by definition, are diffuse and not easily
identified or quantified. The control technologies and best
management practices (BMP) available are generally ex-
pensive to implement. In addition, not all areas of the
watershed contribute equally to the nonpoint source loads
because of the heterogeneities in watershed slopes, soils,
and vegetative cover. Certain critical combinations of
these and other factors result in greater pollutant loadings.
INTRODUCTION
Waste treatment facilities have been significantly up-
graded and improved over the past decade. Reduced or-
ganic and nutrient point source loadings from these waste
treatment facilities have improved water quality in lakes,
streams, and reservoirs but not always as dramatically as
anticipated.
Water quality integrates all sources of pollutants.
Aquatic systems receive and process both point and non-
point source loads from their watersheds. In many aquatic
systems, nonpoint sources may contribute significantly to
receiving system water quality problems. Sediment, for
example, is the greatest pollutant in aquatic systems both
o
o
Spring
Summer
Time
Figure 1.—Temporal variability in stream dissolved oxygen
showing the nonpoint impact In summer and point source
Impact in summer and fall.
247
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
SLOPE
VEGETATION
SOIL EROSION POTENTIAL
CRITICAL CONTRIBUTING AREAS
Critical Areas
l__.l Spring
Summer
Figure 2.—Location of watershed factors that Influence water quality and Intersection of factors potentially contributing the
greatest loads and runoff areas In the spring (Inside dashed line) and summer (Inside solid line).
248
-------
AGRICULTURAL ISSUES: MIDWESTERN EXPERIENCE
Cost-effective nonpoint source pollution control can be
achieved if these critical areas (the areas of significant
• nonpoint source loadings) can be isolated and subjected
to BMP's. This paper discusses steps for a procedure to
identify and evaluate critical watershed areas contributing
. nonpoint source pollution.
ASSUMPTIONS
All techniques and approaches incorporate a set of under-
lying assumptions. These assumptions are important for
proper application and interpretation of any technique.
The major givens underlying this procedure include:
(1) Water quality impairment has occurred. The tech-
nique is also applicable for identifying land resources im-
pairment but this application of the procedure is not dis-
cussed here.
(2) Surface water quality: Ground water quality or the.
coupled surface-ground water system is not considered.
The approach of the following procedure may be applica-
ble to evaluating the impact of nonpoint sources on
ground water recharge areas but this paper considers only
surface water.
(3) Existing information: Data deficiencies or missing
data may be identified through this approach but the pro-
cedure itself uses existing available data in delineating
critical areas.
PROCEDURE
Several different approaches exist to identify and evaluate
critical areas and then select and implement BMP's (Maas
et al. 1985; Monteith et al. 1981). Eleven general steps in
this procedure are:
1. Delineate the watershed and subcatchment
boundaries. This defines the management units. Subcat-
chments may be further subdivided or aggregated de-
pending on objectives and available resources.
2. Document the water quality Impairment. Plotting
the water quality data can indicate the location, temporal
variability, and relative magnitude of the water quality im-
pairment (Fig. 1). Water quality problems that generally
occur during or following storm events imply nonpoint
source contributions, while problems that occur during low
flow conditions imply point source contributions.
3. Determine natural background. The key to this
step is to find a similar but relatively undisturbed area
where natural background concentrations have been
measured. Several states currently use this approach,
sometimes referred to as an ecoregion or physiographic
approach to water quality (Omernik and Hughes, 1983;
Jar man, 1984).
4. Identify point sources. Compared to nonpoint
sources, point sources typically are relatively constant
contributors of pollutants. The location and magnitude of
these contributions must be identified and quantified to
assess the point source impact on receiving system water
quality. The relative point source contribution to the pollu-
tant budget and the natural background must be known to
evaluate nonpoint source contributions. If point sources
account for the major water quality impacts, expenditures
for nonpoint source controls may not be warranted.
5. Compute "back of the envelope" nonpoint
source contributions. Annual runoff estimates, available
from the U.S. Geological Survey, Soil Conservation Serv-
ice, and State or local agencies, can be multiplied by ex-
port coefficients for various constituents to estimate non-
point source constituent loads to the receiving system for
comparison with point source loadings in assessing water
quality impairment (Reckhow and Simpson, 1980; Reck-
how and Chapra, 1983). These calculations can be refined
based on soil series, vegetative species, hydrologic class,
and other considerations. If these preliminary calculations
indicate nonpoint sources may be contributing to water
quality impairment, then the watershed areas contributing
the greatest percentage of the load need to be identified
and BMP's evaluated to reduce these loads.
6. Visit Site. An obvious, and too often ignored, step is
to visit the watershed and inspect the point sources, septic
systems, drainage patterns, vegetative cover, and other
watershed characteristics. Septic system drainage may
not be evident from topographic maps or aerial photo-
graphs but may contribute to water quality impairment.
Interpreting water quality patterns and reaching informed
conclusions on appropriate BMP's can only be achieved
by physically visiting the area and observing the water-
shed conditions and characteristics.
7. Identify critical areas. Considerations include:
a. Controlling factors: Factors known or suspected to
affect erosion, runoff, and water quality—for example,
slope, physiography, geology, soils, vegetative cover, land
use, and cultural resources—might be considered.
b. Class intervals: Class intervals for each controlling
factor should be established to represent similar impacts
or effects on water quality. Slopes between 0-3 percent, for
example, may similarly affect water quality. Forest cover
may represent a generic vegetation class that has minimal
impact on water quality compared with other vegetative
classes such as row crops. Although the specific class
intervals are based on the specific objectives and availa-
ble data, the number of intervals for each factor must be
restricted to three or four.
c. Class interval mapping. The specific watershed ar-
eas exhibiting similar class intervals should be mapped
(Fig. 2). Potential problem areas become apparent as this
exercise continues.
d. Controlling factor overlap: The intersection among
multiple controlling factors indicates potential problem ar-
eas. Transparent overlays can be used to identify the ar-
eas of overlap when a consistent map scale is used for all
class intervals on the watershed. Intersecting areas with
class intervals indicate a high potential for water quality
impacts (Fig. 2d). These areas also can be digitized and
the individual files merged to delineate intersecting areas.
e. Contributing areas: After the potential problem areas
are identified, runoff or transport processes need to be
evaluated to determine the contributing flow areas to the
receiving waters. Variable source contributing area rela-
tions can be determined (Beven and Kirkby, 1979; Beven
and Wood, 1983) or simpler runoff formulations can be
used to determine potential transport to the aquatic sys-
tem (Chow, 1964; Soil Conserv. Serv. 1974). By overlaying
this information on the potential problem areas, the critical
areas contributing pollutants can be defined (Fig. 2d).
Constituent loading from these critical areas can be calcu-
lated using runoff estimates and export coefficients.
8. Evaluate candidate BMP's. Determining the loca-
tion and contribution of critical areas in the watershed
provides the necessary data to evaluate appropriate
BMP's to reduce constituent loads. Slope, soil series, veg-
etative cover, land use, and other factors associated with
each critical area can be used to screen BMP's and deter-
mine candidate BMP's for further evaluation. The percent
reduction in constituent loads should be calculated for
each candidate BMP.
9. Prepare benefit/cost analysis. The costs associ-
ated with various BMP's generally are available from the
Agricultural Stabilization and Conservation Service
(ASCS), State and local agencies, or private contractors.
Associating benefits with various BMP's, however, is more
249
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
difficult since water quality improvements also have intan-
gible benefits. The initial analyses may be performed us-
ing a unit load reduction/dollar cost ratio. This ratio pro-
vides a comparative basis for alternative BMP's.
10. Rank alternatives. Some critical areas may be of
greater priority because of designated stream or lake uses
in that area, desired uses, or other considerations. This
objective or subjective priority can be predicted on factors
such as the ease of implementation, longevity, mainte-
nance, land owner willingness to participate, or budgetary
constraints. Regardless of the criteria, the rationale for
establishing priority order alternatives should be docu-
mented to increase the likelihood of objectivity.
11. Seek funding sources and implement BMP's.
The final step in the procedure is to locate sources of
funds to implement the BMP's. Various cost-sharing pro-
grams are available through both Federal and State agen-
cies for some BMP's. Some local watershed districts also
provide cost-sharing funds.
DISCUSSION
Specific steps in identifying critical watershed areas vary
among investigators (Maas et al. 1985; Monteith, 1981).
However, the perception and acknowledgement of critical
watershed areas is more important than the specific steps
in identification. Funds do, and probably will always, limit
the possibilities so the objective should be to maximize the
benefits derived from each dollar spent. If 10 to 15 percent
of the watershed areas are contributing 80 to 90 percent of
the water quality problem, these areas should receive the
priority for BMP's.
REFERENCES
Beven, K.J., and M.J. Kirkby. 1979. A physically-based variable
contributing area model of basin hydrology. Hydrol. Sci. Bull.
24: 43-69.
Beven, K.J., and E.F. Wood. 1983. Catchment geomorphology
and the dynamics of runoff contributing areas. J. Hydrol. 65:
139-58.
Chow, V.T., ed. 1964. Handbook of Applied Hydrology. McGraw-
Hill: New York.
Green, D.B., T.J. Logan, and N.E. Smeck. 1978. Phosphate ad-
sorption-desorption characteristics of suspended sediments
in the Maumee River Basin of Ohio. J. Environ. Qua). 7: 208-
12.
Jarman, R. 1984. The Development of Aquatic Ecoregions in
Oklahoma. PhD Dissertation. Univ. Oklahoma, Norman.
Johnson, A.H. et al. 1976. Phosphorus loss by stream transport
from a rural watershed: Quantities, processes, and sources. J.
Environ. Qual. 5: 148-57.
Karickhoff, S.W., and D.S. Brown. 1978. Paraquat sorption as a
function of particle size in natural sediments. J. Environ. Qual.
7: 246-52.
Maas, R.P., M.D. Smolen, and S.A. Dressing. 1985. Selecting
critical areas for nonpoint source pollution control. J. Soil Wa-
ter Conserv. 40: 68-71.
Monteith, T.J., R.A.C. Sullivan, T.M. Heidtke, and W.C. Son-
zogni. 1981. Watershed Handbook. EPA-905/9-84-002. Re-
gion V, U.S. Environ. Prot. Agency, Chicago.
Omernik, J.M., and R.M. Hughes. 1983. An approach for defin-
ing regional patterns of aquatic ecosystems and attainable
stream quality in Ohio. Progress rep. to Ohio Environ. Prot.
Agency and Region V, U.S. Environ. Prot. Agency, Corvallis
Environ. Res. Lab., Corvallis, OR.
Reckhow, K.H., and S.C. Chapra. 1983. Engineering ap-
proaches to Lake Management. Vols. 1 & 2, Butterworth
Publ., Boston.
Reckhow, K.H., and J.T. Simpson. 1980. A procedure using
modeling and error analysis for the prediction of lake phos-
phorus concentration from land use information. Can. J. Fish.
Aquat. Sci. 37:1439-48.
Soil Conservation Service. 1972. National Engineering Hand-
book. NEH 4-102. Washington, DC.
Vanoni, V.A. 1977. Sedimentation Engineering. Man. Nq, 54.
Am. Soc. Civil Engr, New York.
250
-------
GROSS EROSION RATES, SEDIMENT YIELDS, AND NUTRIENT
YIELDS FOR LAKE ERIE TRIBUTARIES: IMPLICATIONS FOR
TARGETING
DAVID B. BAKER
KENNETH A. KRIEGER
R. PETER RICHARDS
JACK W. KRAMER
Heidelberg College
Tiffin, Ohio
ABSTRACT
Studies of agricultural nonpoint pollution in selected wa-
tersheds within the Lake Erie Basin have included esti-
mates of gross erosion rates, as well as measurements of
sediment, phosphorus and nitrogen yields. These studies
incidate that under conventional management practices
average gross erosion rates for watersheds are not reli-
able indicators of the export of either soluble or sediment
associated pollutants. Average gross erosion rates are
not even good indicators of sediment yields. Conse-
quently, factors other than gross erosion need to be con-
sidered in watershed level targeting for agricultural pollu-
tion abatement programs.
INTRODUCTION
In the early 1970's, measurements of phosphorus trans-
port during runoff events in the Sandusky River suggested
that phosphorus loading to Lake Erie from agricultural
sources was being underestimated (Baker and Kramer,
1973). Extension of event monitoring to the lake's other
major tributaries confirmed that agricultural sources were
indeed larger than previously estimated. (U.S. Army
Corps Eng. 1975). Phosphorus modeling studies for the
lake indicated that phosphorus reductions from nonpoint
sources would be necessary to achieve desired water
quality in the lake. Consequently, a major focus of the U.S.
Army Corps' Lake Erie Wastewater Management Study
was to evaluate options for reducing phosphorus loading
from agricultural sources (U.S. Army Corps Eng. 1979).
Part of the evaluation established water quality monitor-
ing stations for a number of smaller watersheds, so that
the relationships between watershed characteristics (e.g.
land use, soils, slopes, and gross erosion) and pollutant
export could be incorporated into the management plans.
This paper summarizes some of the relationships between
watershed characteristics and pollutant yields we have
observed during these studies. In a companion paper in
this volume, we have described some of the characteris-
tics of sediment, nutrient, and pesticide transport in area
streams and rivers (Baker et al. 1985). Our sampling meth-
ods and analytical procedures are summarized in that pa-
per.
WATER QUALITY PROBLEMS
ASSOCIATED WITH RURAL NONPOINT
SOURCES IN THE LAKE ERIE BASIN
Phosphorus Loading to Lake Erie
Table 1 shows an estimate of the sources and amounts of
total phosphorus and bioavailable phosphorus loading to
Lake Erie. This estimate draws upon several sources of
information. The rural nonpoint phosphorus loads repre-
sent an average load for the 1970-80 period (Yaksich,
1983; Yaksich et al. 1985). The urban nonpoint load is
based on the Lake Erie study (U.S. Army Corps Eng.
1982).
Point source phosphorus loading to the lake has been
greatly reduced by phosphorus removal programs at mu-
nicipal sewage treatment plants. Thev point source loads
shown on Table 1 represent 1982 loading data. Since by
1982 most municipal dischargers were] meeting the phos-
phorus removal requirements, little further reduction in
municipal phosphorus loading can be expected.
The data for atmospheric inputs and Lake Huron inputs,
as well as the municipal and industrial point source inputs,
are taken from International Joint Commission sources
(1983a, b). Direct point sources empty into river mouth
areas, bays, or the nearshore zone of the lake while indi-
rect point sources empty into streams and rivers in tribu-
tary watersheds. The estimates of bioavailable phos-
phorus loading are based on direct measurements by our
laboratory (Baker, 1983b) and other bioavailability studies
summarized by Sonzogni et al. (1982).
Rural nonpoint sources account for 60 percent of the
total phosphorus loading and 56 percent of the bioavaila-
ble phosphorus loading to the lake. Soluble phosphorus
derived from rural sources is a significant part of the rural
bioavailable load.
The phosphorus target load for Lake Erie is 11,000 met-
ric tons per year (Int. Joint Comm. 1984). The estimated
load of 13,996, based on long-term nonpoint loads and
1982 point source load, exceeds the target load by 3,000
metric tons. Programs to achieve the additional 3,000 met-
ric ton per year reduction are focusing on nonpoint
sources, since the most cost-effective portions of munici-
pal phosphorus removal programs have already been im-
plemented (U.S. Army Corps Eng. 1982).
Other Water Quality Problems
The stream monitoring programs have also measured
sediments, nitrates, and pesticides. Suspended sedi-
ments in area streams and rivers result in the usual prob-
lems associated with erosion (Clark et al. 1985). In the
Sandusky River, nitrate concentrations exceed drinking
water standards about 16 percent of the time in May, June,
and July (Baker, 1985). Rivers are frequently used as pub-
lic water supplies in northwestern Ohio, and nitrates are
not removed in treatment.
Several soluble herbicides are present in the rivers in
May and June and pass directly through conventional wa-
ter treatment plants (Baker, 1983a). Until specific maxi-
mum contaminant levels or health advisory levels are es-
tablished by the U.S. Environmental Protection Agency for
251
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
the currently used pesticides that frequently occur in
drinking water, assessing the significance of these expo-
sures will be difficult.
STUDY WATERSHEDS
From 3 to 10 years of chemical sampling have been com-
pleted for the 15 watersheds listed in Table 2. The table
includes the U.S. Geological Survey stream gauge station
number, the drainage area, the mean discharge and per-
iod of hydrological sampling, the period of chemical sam-
pling, and the number of samples analyzed for sediments
and nutrients. The average annual runoff ranged from
23.2 cm for the Raisin River to 39.8 cm for the Cuyahoga
River. Watershed sizes range from 44 to 16,359 km2.
The land use within each watershed is summarized in
Table 3. Except for the Cuyahoga River basin, cropland
dominates land use for all of the watersheds. The geo-
graphical data system used for developing land use statis-
Table 1.—Total and bioavailable phosphorus loads for Lake Erie, based on 1982 point source loads, long-term nonpolnt
source loads, and recent studies of bioavailable phosphorus loads.
Total Phosphorus
Load
Metric Percent
Tons of Total
I. NONPOINT SOURCES
Rural1
Particulate 80%
Soluble 20%
Subtotal
Urban1
Atmospheric2
Total Nonpoint Source
II. POINT SOURCE
Municipal
Flows >1 MGD
Direct2
Indirect3
Flows <1 MGD
Direct
Indirect
Subtotal
Industrial2
Total Point Sources
III. LAKE HURON INPUT2
IV. TOTAL LOADS
6720
1680
8400
900
660
9960
1388
1061
110
330
2889
67
2956
1080
13,996
48.0
12.0
60.0
6.4
4.7
71.1
9.9
7.6
0.8
2.4
20.6
0.5
21.1
7.7
100
Bioavailable Phosphorus
Fraction Load Percent
Bioavailable Tons of Total
.30
.95
(.43)
.43
.38
.70
.35
.70
.35
.50
.60
2016
1596
3612
387
251
4250
972
371
77
116
1536 •
34
1570
648
6468
31.2
24.7
55.9
6.0
3.9
65:8
15.0
5.7
1.2
1.8
23.7
0.5
10.0
100
'Yaksich, 1983.
2I.J.C. 19833.
3I.J.C. 1983b.
•Sonzogni et al. 1982; Baker 1983b.
Table 2.—Watershed areas, mean annual discharges, period of chemical sampling, and number of samples analyzed for
the Lake Erie tributary monitoring program.
transport Stations
Sandusky River Stations
1. Fremont
2. Mexico
3. Upper Sandusky
4. Bucyrus
Sandusky River Tributaries
5. Wolf Creek, East
6. Wolf Creek, West
7. Honey Creek, Melmore
8. Honey Creek, New Wash.
9. Tymochtee Creek
10. Broken Sword Cr.
Other Lake Erie Tributaries
1 1 . Maumee River
12. Raisin
13. Cuyahoga
14. Portage
15. Huron
U.S. Geological
survey
station number
04198000
04197000
04196500
04196000
04192450
04197300
04197100
04197020
04196800
04196200
04193500
04176500
04208000
04195500
04199000
Drainage
Area
Km2
3,240
2,005
722
230
213
171.5
386
44.0
593
217
16,359
2,699
1,831
1,109
961
mean «
Years of
Record
57
55
57
40
5
5
7
3.908
19
5
58
43
52
51
31
nnuai uisc
nWs
27.75
16.62
6.967
2.461
1.82
1.34
3.908
(0.445)1
4.956
2.45
139.5
19.85
23.14
9.091
8.496
narge
cm
27.0
26.2
28.5
33.8
27.0
24.6
32.0
(32.0) 1
26.3
35.5
26.8
23.2
39.8
25.9
27.9
Chemical
Sampling
Period
1974-84
1974-81
1974-81
1974-81
1976-81
1976-81
1976-84
1979-81;
83-84
1974-81
1976-81
1975-80;
82-84
1982-84
1981-84
1974-78
1974-79
Number of
Samples
Analyzed
45902
2178
2973
2998
2425
2419
45952
2271 2
2471
2512
31 542
80S2
13802
1856
2027
1 Extrapolated from Honey Creek at Melmore.
'Sampling continued on 1985 program.
252
-------
tics for Lake Erie watersheds has been summarized by
Adams et al. (1982).
WATERSHED POLLUTANT EXPORT
The flux-weighted mean concentrations of sediments and
nutrients at each of the transport stations are summarized
in Table 4. These represent the flux-weighted averages for
the entire period of chemical sampling and for the number
of samples indicated in Table 2. For all of the stations,
sampling is conducted throughout the year and conse-
quently the averages reflect the seasonal distribution of
storms that occurred during the sampling period.
One method of estimating mean annual pollutant export
is to multiply the flux-weighted pollutant concentrations by
the mean annual discharge. This procedure will provide
an accurate mean annual load estimate so long as: (1) the
flux-weighted average concentration accurately reflects
current watershed responses to current weather and rain-
fall regimes in the region; and (2) the mean annual dis-
charge for the period of record reflects contemporary wa-
tershed responses to current weather and rainfall
regimes. The latter condition does not appear to be met
since the average discharges for the Maumee, Sandusky,
AGRICULTURAL ISSUES: MIDWESTERN EXPERIENCE
and Cuyahoga rivers during the 1973-83 period are
higher than long-term average discharges (as of 1983), by
17 percent, 26 percent and 30 percent respectively.
Rather than attempt to adjust each watershed to current
"average' discharges, the discharge data in Table 2 have
been adjusted to long-term stations (Baker, 1983c). Other
methods of using these data to estimate mean annual
pollutant loads have been discussed by Baker (1984).
In Table 5, the mean annual loads of sediments and
nutrients are expressed on a unit area yield basis. For
each watershed the mean annual load was divided by the
total watershed area to determine unit area yields. The
gross erosion rates, as calculated in the Lake Erie study
(Logan et al. 1982), are also listed for each watershed.
The unit area total phosphorus yield of 1.26 kg/ha/year for
the Sandusky Basin calculated by this method is signifi-
cantly lower than the 10-year average export of 1.55 kg/
ha/year actually measured at the station (Baker et al.
1985).
The unit area yields shown in Table 5 reflect the inputs
of both point and nonpoint sources. For the parameters
shown in Table 5, point sources constitute significant in-
puts only for phosphorus and for certain watersheds. Ta-
ble 6 illustrates the typical procedure of correcting for
Table 3.—Watershed land use for the Lake Erie tributary nutrient and sediment transport stations.
Cropland Pasture Forest Water
Transport Stations % % % %
Other
Sandusky River Stations
1. Fremont
2. Mexico
3. Upper Sandusky
4. Bucyrus
Sandusky River Tributaries
5. Wolf Creek, East
6. Wolf Creek, West
7. Honey Creek, Melmore
8. Honey Creek, New Wash.
9. Tymochtee Creek
10. Broken Sword Creek
Other Lake Erie Tributaries
11. Maumee River
12. Raisin
13. Cuyahoga
14. Portage
15. Huron
79.9
80.3
78.0
73.3
81.9
83.3
82.6
89.1
84.0
84.7
75.6
67.1
4.2
85.5
75.3
2.3
2.3
3.4
4.9
2.7
1.4
0.6
—
1.2
1.4
3.2
6.8
43.1
3.6
3.5
8.9
8.7
9.1
9.4
6.3
4.7
10.0
7.5
7.6
8.5
8.4
9.0
29.1
5.6
12.5
2.0
2.1
2.0
2.1
2.0
3.1
0.5
—
2.3
1.3
3.5
3.0
3.0
0.9
2.2
6.8
6.6
7.5
10.2
7.0
7.6
6.3
3.4
4.8
4.1
9.4
14.1
20.6
4.3
6.4
Table 4.—Flux weighted mean concentrations of sediments and nutrients at the transport stations for the period of
chemical sampling.
Transport Stations
Suspended
Solids
mg/L
Total
Phosphorus
mg/L
Soluble
Reactive
Phosphorus
mg/L
Nitrate +
Nltrite-N
mg/L
Total
Kjeldahl
Nitrogen
mg/L
Sandusky River Stations
1. Fremont
2. Mexico
3. Upper Sandusky
4. Bucyrus
Sandusky River Tributaries
5. Wolf Creek, East
6. Wolf Creek, West
7. Honey Creek, Melmore
8. Honey Creek, New Wash.
9. Tymochtee Creek
10. Broken Sword Creek
Other Lake Erie Tributaries
11. Maumee River
12. Raisin
13. Cuyahoga
14. Portage
15. Huron
249
230
312
273
246
251
198
254
231
312
218
81.1
188
164
220
0.468
0.409
0.583
0.633
0.479
0.461
0.413
0.458
0.424
0.479
0.474
0.238
0.428
0.402
0.362
0.084
0.061
0.126
0.199
0.109
0.089
0.074
0.088
0.065
0.064
0.090
0.046
0.105
0.119
0.104
4.61
4.32
4.60
3.71
5.32
6.95
4.79
5.05
5.60
5.08
5.13
3.51
1.82
5.89
3.61
1.73
1.79
1.81
1.85
1.23
1.36
253
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
point source inputs. Point source inputs, expressed on a
unit area basis, are subtracted from the total unit area
yield to determine a unit area nonpoint yield. This proce-
dure assumes 100 percent delivery of point source phos-
phorus through the stream system.
The point source inputs for the Maumee River, San-
dusky River-Fremont, Honey Creek-Melmore, and San-
dusky River-Bucyrus reflect inputs from all municipal
sewage treatment plants within the watershed. For the
Raisin River and the Cuyahoga River, only sewage treat-
ment plants with flows greater than 1 million gallons per
day were included. Consequently, the nonpoint yields for
the Raisin River and Cuyahoga River are probably being
overestimated.
In general, the unit area yields of total phosphorus and
total nitrogen (nitrate-N + total Kjeldahl nitrogen) in Lake
Erie tributaries are much larger than the average values
cited for agricultural lands. Based on a thorough review of
the literature on loading studies, Rast and Lee (1983) rec-
ommended the use of 0.5 kg/ha/year and 5 kg/ha/year,
respectively, for estimating total phosphorus and total ni-
trogen inputs into lakes from agricultural watersheds. The
total phosphorus yields from the Maumee and Sandusky
river basins are 2.5 times higher than these national aver-
ages. Total nitrogen yields in northwestern Ohio rivers are
3.5 times higher than the average values. These high
rates of nutrient export occur even though the average
gross erosion rates of 4.1 to 9.8 tons/ha/year (1.9 to 4.4
short tons/acre/year) for these watersheds are lower than
the average for cropland in the United States.
Within the study watersheds, average gross erosion is
not a consistent indicator of pollutant yields. In Table 7 the
ratios of average gross erosion rates and yields of sus-
pended sediments, nonpoint total phosphorus, soluble
phosphorus, and nitrates are shown for three pairs of wa-
tersheds. The gross erosion rate is 2.2 times higher for the
Broken Sword watershed than for the Wolf Creek West
watershed, while the suspended solids and total phos-
phorus yields are 1.79 and 1.50 times higher, respectively.
The gross erosion rate in the Honey Creek watershed is
1.6 times higher than Wolf Creek West, yet the suspended
solids and nonpoint phosphorus yields are essentially the
same.
The Raisin River watershed has the highest gross ero-
sion rate—1.18 times higher than that of the Sandusky
Table 6.—Calculation of nonpoint phosphorus yields at
representative transport stations.
Watershed
Maumee R.
Sandusky R.,
Fremont
Cuyahoga R.
Raisin R.
Honey Cr.
Sandusky R.,
Bucyrus
Total-P
Export
ttg/ha/yr
1.27
1.26
1.71
0.55
1.32
2.14
Point
Source-P
Input
kg/ha/yr
0.20
0.14
1.11
0.11
0.09
1.17
NonPt.
P-Export
kg/ha/yr
1.07
1.14
0.60
0.44
1.23
0.97
Percent
Nonpoint
84
89
35
80
93
45
River watershed. The sediment and nonpoint total phos-
phorus yields for the Raisin are only 0.27 and 0.38 of
those for the Sandusky. In part, the higher sediment and
phosphorus deliveries from the Sandusky Basin result
from finer textured soils.
Land use effects are evident in the data of Tables 5 and
6. The Cuyahoga Basin, which has a small percentage of
cropland, has relatively low nitrate and nonpoint phos-
phorus yields. Stations impacted by large proportions of
point source inputs (e.g. the Sandusky River Bucyrus sta-
tion and the Cuyahoga stations) have high soluble reac-
tive phosphorus yields.
Watershed size has little effect on suspended sediment
and nutrient yields. The Maumee, Sandusky, and Honey
Creek watersheds all have similar unit area exports. This
suggests that instream delivery losses do not increase
with watershed size for this region.
These studies do not support the use of the universal
soil loss equation to select specific watersheds for tar-
geted nonpoint phosphorus control programs. Instead, an
areawide program to support best management practice
(BMP) implementation on critical areas, regardless of sub-
watershed boundaries, would likely result in'the most cost
efficient reductions in sediment and particulate phos-
phorus loads to Lake Erie.
The source areas for high nitrate concentrations in this
region are the tile-drained fields. Fertilizer BMP's need to
be implemented throughout this region. Since much of the
soluble phosphorus export occurs in the winter season
(Baker et al. 1985), the contributing areas are probably
much larger than those for sediment and particulate phos-
phorus.
Table 5.—Gross erosion rates and unit area yields of sediments and nutrients at the transport stations.
Soluble Total
Gross Erosion Suspended Total Reactive Nitrate + Kjeldahl
Rate Solids Phosphorus Phosphorus Nitrate-N Nitrogen
Transport Stations kg/ha/yr kg/ha/yr kg/ha/yr kg/ha/yr kg/ha/yr kg/ha/yr
Sandusky River Stations
1. Fremont
2. Mexico
3. Upper Sandusky
4. Bucyrus
Sandusky River Tributaries
5. Wolf Creek, East
6. Wolf Creek, West
7. Honey Creek, Melmore
8. Honey Creek, New Wash.
9. Tymochtee Creek
10. Broken Sword Creek
Other Lake Erie Tributaries
1 1 . Maumee River
12. Raisin
13. Cuyahoga
14. Portage
15. Huron
8,250
9,370
9,350
7,850
5,110
4,190
6,860
7,060
8,410
9,390
6,840
9,750
—
5,000
7,510
673
601
950
922
663
619
633
811
609
1,110
585
188
749
424
614
1.26
1.07
1.78
2.14
1.29
1.14
1.32
1.46
1.12
1.71
1.27
0.55
1.71
1.04
1.01
0.22
0.16
0.38
0.67
0.29
0.22
0.24
0.28
0.17
0.23
0.24
0.11
0.42
0.31
0.29
12.5
11.3
14.0
12.5
14.3
17.1
15.3
16.1
14.8
18.1
13.8
8.1
7.3
15.2
10.1
4.68
—
—
—
—
—
5.72
5.78
—
—
4.97
2.85
5.42
—
—
254
-------
AGRICULTURAL ISSUES: MIDWESTERN EXPERIENCE
Table 7.—Comparison of erosion rate and nonpoint yields for representative watersheds.
Gross Nonpoint
Watershed
Broken Sword
Wolf, East
Ratio— Broken Sword: Wolf East
Honey Creek
Wolf, East
Ratio— Honey: Wolf East
Raisin River
Sandusky River
Ratio — Raisin R: Sandusky R.
Erosion
kg/ha/yr
9,390
4,190
2.24
6,860
4,190
1.63
9,750
8,250
1.18
SS
kg/ha/yr
1,110
619
1.79
63
619
1.02
188
673
0.27
TP
kg/ha/yr
1.71
1.14
1.50
1.23
1.14
1.08
0.44
1.14
0.38
SRP
kg/ha/yr
0.23 •
0.22
1.04
0.24
0.22
1.09
0.11
0.22
0.50
NO3
kg/ha/yr
18.1
17.1
1.06
15.3
17.1
0.89
8.1
12.5
0.65
The timing of pesticide export during storm events sug-
gests that the contributing areas are those from which
surface runoff water reaches stream systems. This will
vary considerably from year to year depending on rainfall
amounts and intensities.
CONCLUSIONS
1. The nonpoint source nutrient exports from Lake Erie
tributaries are very high relative to average agricultural
export rates even though gross erosion rates in these wa-
tersheds are low.
2. Within this region, average gross erosion rates for
watersheds are not reliable indicators of sediment and
nutrient yields.
3. Contributing areas for nitrate and soluble phos-
phorus probably encompass a larger portion of the land
surface than contributing areas for sediment and pesti-
cides.
REFERENCES
Adams, J.R., et al. 1982. A land resource information system
(LRIS) for water quality management in the Lake Erie Basin. J.
Soil Water Conserv. 37: 45-50.
Baker, D.B. 1983a. Herbicide contamination in municipal water
supplies of northwestern Ohio. Heidelberg College, Tiffin, OH.
. 1983b. Tributary loading of bioavailable phosphorus
into Lakes Erie and Ontario. U.S. Environ. Prot. Agency. Chi-
cago, IL.
_. 1983c. Sandusky Basin sediment and nutrient trans-
port studies. Final Rep. to U.S. Environ. Prot. Agency, Chi-
cago, IL.
1984. Fluvial transport and processing of sediments
in large agricultural river basins. EPA-600/8-83-054. U.S. Envi-
ron. Prot. Agency, Athens, GA.
_. 1985. Regional water quality impacts of intensive
Baker, D.B., and J. W. Kramer. 1973. Phosphorus sources and
transport in an agricultural river basin of Lake Erie. Pages
858-71 in Proc. 16th Conf. Great Lakes Res. Int. Assn. Great
Lakes Res.
Baker, D.B., K.A. Krieger, J.W. Kramer and R.P. Richards. 1985.
Effects of intensive agricultural land use on regional water
quality in northwestern Ohio. In Perspectives on Nonpoint
Source Pollution. Proc. Natl. Conf. May 19-22, Kansas City,
MO. U.S. Environ. Prot. Agency, Washington, DC.
Clark, E.H. II, J.A. Haverkamp, and W. Chapman. 1985. Eroding
soils: the off-farm impacts. Conserv. Foundation, Washington,
DC.
International Joint Commission. 1983a. 1983 report on Great
Lakes water quality. Windsor, Ontario.
1983b. 1983 report on Great Lakes water quality.
Appendix, Great Lakes surveillance. Windsor, Ontario.
_. 1984. Phosphorus load reduction supplement to An-
nex 3 of the 1978 agreement between the United States of
America and Canada on Great Lakes water quality. Windsor,
Ontario.
Logan, T.J., D.R. Urban, J.R. Adams, and S.M. Yaksich. 1982.
Erosion control potential with conservation tillage in the Lake
Erie Basin: Estimates using the univeral soil loss equation and
the Land Resource Information System (LRIS). J. Soil Water
Conserv. 37: 50-5.
Rast, W, and G.F. Lee. 1983. Nutrient loading estimates for
lakes. J. Environ. Eng. 109: 502-17.
Sonzogni, W.C., S.C. Chapra, D.E. Armstrong, and T.J. Logan.
1982. Bioavailability of phosphorus inputs to lakes. J. Environ.
Qual. 11:555-63.
U.S. Army Corps of Engineers. 1975. Lake Erie wastewater
management study: Preliminary feasibility report, Vol. I, main
rep. Buffalo, NY.
1979. Lake Erie wastewater management study
methodology report. Buffalo, NY
_. 1982. Lake Erie wastewater management study. Fi-
row-crop agriculture: a Lake Erie Basin case study. J. Soil
Water Conserv. 40:125-32.
nal rep. Buffalo, NY.
Yaksich, S.M. 1983. Summary report of the Lake Erie wastewa-
ter management study. U.S. Army Corps Eng., Buffalo, NY.
Yaksich, S.M., D.A. Melfi, D.B. Baker and J.W. Kramer. 1985.
Lake Erie nutrient loads, 1970-1980. J. Great Lakes Res. 11:
117-31.
255
-------
LO
TOM DAVENPORT
JOHN LOWREY
U.S. Environmental Protection Agency
Chicago, Illinois
Today every segment of our society is looking for ways to
improve efficiency. Governmental agencies as well as the
business community are increasingly aware that financial
times have changed. No matter how important a cause,
Federal, State and local agencies realize that except for
defense spending, major new Federal funding initiatives
are unlikely in the near future. For agencies to fulfill their
legislative mandates and responsibilities, they must look
for new ways to improve the delivery and efficiency of
existing approaches and programs.
Several nonpoint source control projects—Sec. 108
Great Lakes Demonstration Projects, Clean Lakes Proj-
ects, Sec. 314 Agricultural Conservation Program Proj-
ects, and Rural Clean Water Projects—have been imple-
mented in watersheds critical for agricultural pollution.
Evaluation of these ongoing nonpoint source control proj-
ects is necessary for facilitating future NPS control pro-
grams. Presently in the State of Illinois, two major water-
shed nonpoint source evaluation projects exist.
Recommendations on project selection, development,
and implementation will be discussed based on evaluation
of these projects.
The State of Illinois began seriously to reevaluate its soil
and water conservation programs in 1977, during the de-
velopment of its Water Quality Management (WQM) Plan,
completed in 1979. The WQM Plan identifies the scope
and seriousness of nonpoint source pollutants and as-
signs agency management responsibility. This initial WQM
planning effort documented that agricultural activities are
a major source of pollution and mandated the develop-
ment of plans to control NPS pollution from agriculture.
The most severe problem identified was soil erosion re-
sulting in lake sedimentation. Through the initial WQM
planning process, 11 pirority watersheds were selected in
1979 (Table 1). This study discusses the Lake Pittsfield
(Blue Creek) and Silver Lake (Highland) Watershed proj-
ects.
The Blue Creek Watershed encompasses slightly more
than 2,833 hectares in east central Pike County, Illinois
(Fig. 1). The terrain is hillier than most areas of Illinois, and
the area has a high soil loss potential because of its steep
slopes, fine-textured soils, and agricultural land use prac-
tices. The Blue Creek Watershed drains into Pittsfield City
Lake, which was constructed with PL. 566 assistance
from the Soil Conservation Service in 1961 as a multiple
purpose flood control and water supply reservoir. Average
soil loss was estimated at 18.1 tons (M)/ha/year. Erosion
from livestock operations, primarily hog enterprises, sig-
nificantly contributes to the total basin sediment load. Soil
Erosion and Sediment Transport in the Blue Creek Water-
shed (Davenport, 1983) describes the Blue Creek Water-
shed Project in detail. The Blue Creek project started in
1979 and ended in 1982.
The Highland Silver Lake Watershed encompasses
12,524 hectares in the eastern portion of Madison County,
Illinois (Fig. 1). It drains into Highland Silver Lake through
Silver Creek and numerous tributaries. The City of High-
land built Highland Silver Lake, an artificial impoundment,
in 1962 as a public water supply. Agriculture is the domi-
nant land use with 88 percent of the land devoted to row
crops (Table 2). The terrain is relatively flat, and the soils
have a high detachment potential because of their fine
texture, the influence of sodium, and the agricultural land
use practices. Average soil loss was estimated to be 15.2
tons (M)/ha/years (III. Environ. Prot. Agency, 1979). The
Highland Silver Lake project started in 1980, with signup
lasting to June 1985 and implementation completed in
1990.
Land treatment activities were funded in the Blue Creek
Watershed through the ACP Special Water Quality Pro-
Table 1 .—Priority lakes for agricultural nonpoint source water quality problem abatement in order of priority
(III. Environ. Prot. Agency, 1979).
Ranking
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
Watershed
Lake Pittsfield (Blue Creek)1
Lake Carlinville
Lake Canton
Silver Lake (Highland)1
Spring Lake
Lake Springfield
Lake Taylorville
Lake Lou Yeager
Lake Bloomington
Paris Lake
Lake Paradise
Watershed
size
(ha)
2,838
6,755
3,940
12,399
5,236
66,973
34,029
29,808
17,621
5,184
4,690
Lake size
(ha)
96
68
101
223
112
1,630
465
514
257
89
71
Average
erosion rate
(ton (M)/ha)
18.1
15.7
14.3
15.2
13.9
13.4
13.0
14.6
12.3
11.4
11.2
'Projects evaluated
256
-------
U S. KPARTtCKT OF MfVCULTURC
RURAL CLEAN WATER PROGRAM LAKES
ILLINOIS
so*. CONSERVATION senv
Figure 1.—Rural Clean Water Program Lakes in Illinois.
gram and in Highland Silver Lake Watershed through the
RCWR Both the RCWP and ACP Special Water Quality
Program endorse the concept "that the condition of the
lake is a reflection of the condition and management of its
watershed." Both programs assume that resource man-
AGRICULTURAL ISSUES: MIDWESTERN EXPERIENCE
agement systems in critical water quality areas will im-
prove the quality of the downstream water resources. Both
programs rely on voluntary landowner/operator participa-
tion to address water quality problems.
DISCUSSION
For both the Blue Creek and Highland Silver Lake proj-
ects, the overall goal is to improve water quality by reduc-
ing soil loss. The Blue Creek Project reached its soil re-
duction goal; Highland Silver Lake probably will not
achieve its goal. Two years after most of the Blue Creek
Watershed project was completed, signs of water quality
improvement were noted. In Highland Silver Lake, water
quality has not improved. In both project areas, however,
landowners/operators believe water quality has improved.
Five program elements form the basis for examining the
effectiveness of both projects.
1. Problem Identification/Definition: This is the most
crucial step in the NPS control project implementation
process, because this step determines the type of pro-
gram. The problem identified was excessive sedimenta-
tion in both Pittsfield City and Highland Silver Lakes. In
both cases the cause of the sedimentation was excessive
sheet and rill erosion from agricultural lands. In the case of
Pittsfield City Lake the project sponsors documented the
lake sedimentation problem with a lake sedimentation sur-
vey before applying for funding. In Highland Silver Lake
the project sponsors stated in the application that lake
sedimentation was a problem, although it was not docu-
mented. (Mad. Co. Soil Water Conserv. Dist. 1979). As
part of the Highland Silver Lake monitoring program, a
lake sedimentation survey was completed. It showed that
turbidity, not lake sedimentation, was the water quality
problem (Davenport and Kelly, 1984). In both projects, pro-
grams were implemented to control excessive soil erosion
as the cause of lake sedimentation problems. In the case
of Highland Silver Lake, the program should have ad-
dressed the turbidity problem.
Table 2.—Land use/cover in the Highland Silver Lake (1981) and Blue Creek (1980)' watersheds in hectares and percent.
Highland Silver Lake
Watershed
(III. State Coord.
Comm. 1982)
Blue Creek
Watershed
(Davenport, 1983)
Cropland
(Percent)
Pasture/hay land
(Percent)
Woodland
(Percent)
Urban
(Percent)
Feedlots
(Percent)
Interstate
(Percent)
Wildlife
(Percent)
Farmsteads
(Percent)
Gravel . .
(Percent)
Residential ...
(Percent)
Water
(Percent)
Total
10200 5
(83.3)
... 672 6
(5.4)
505 9
(4.1)
. . 85 0
(0.7)
46 9
(0.4)
198
(0.2)
132 3
(1.1)
250 5
(2.0)
6 1
(0.0)
144 t
(1.2)
3359
(2.7)
1 2 399 6
1 6029
(56.4)
6149
(21.7)
327 8
(11.6)
26 5
(0.9)
126 1
(4-4)
442
(1.6)
—
957
(3.4)
2 838 1
'Year Reported
257
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
2. Critical Area Determinations: Both projects re-
quired that local sponsors determine critical areas. Critical
areas are those areas or sources that contribute most to
impairment of downstream water resources.
In Blue Creek Watershed, a local coordinating commit-
tee defined critical area locations within the watershed
based on three factors. The first factor considered was
whether or not the area was located in the high erosion
hazard area. (High erosion hazard areas were considered
as those with a soil credibility factor of at least 0.37 and a
slope greater than 5 percent.) Second, they considered
the site's close proximity to the lake. The third factor was
whether the site had a feedlot with soil losses exceeding
the allowable limits. These guidelines assumed that ex-
cessive soil erosion was causing the water quality prob-
lem. In Highland Silver Lake Watershed Project the com-
mittee established that the critical area consisted of all
natric soils with 2 percent or greater slope, fine particle
size, and high erodibility, and all non-natric soils with 5
percent or greater slope, high erodibility, and close prox-
imity to the stream system. Since the lake sedimentation
survey documented turbidity rather than sedimentation as
the problem, the critical areas should have been changed.
3. Local Institutional Arrangements: Both projects
used similar working arrangements and methods to en-
sure participation. In both projects three primary meth-
ods—information/education, technical, and financial as-
sistance—were used to encourage adoption of Resource
Management Systems (RMS) in the critical areas. The
"one-to-one" technical assistance and education was a
combined effort of all the involved agencies' personnel.
The major differences were that existing resources fi-
nanced the intensive information/educational effort spear-
headed by the Cooperative Extension Service in the Blue
Creek Watershed, and the effort began 2 years before the
special technical and financial assistance was approved;
but in the Highland Silver Lake Watershed Project, the
National RCWP funded the Cooperative Extension Serv-
ice's activities; thus, Extension's intensive activities
started after Highland Silver Lake had been approved for
special technical and financial assistance.
4. Effectiveness of Practices/Programs: In the Blue
Creek Watershed, practices cost-shared under the ACP
Special Water Quality Project were feasible and accept-
able to landowners/operators and could correct the identi-
fied soil erosion problem. In the Highland Silver Lake Wa-
tershed, 100 percent of the farms had to be treated to
receive cost-sharing, and the high sodium concentration
in the soils made establishing vegetation difficult.
5. Timeframe for Implementation: Both projects had
adequate timeframes for implementation. The Blue Creek
Watershed Project achieved its land treatment goals, but it
is doubtful that the Highland Silver Lake Watershed Proj-
ect can. A major difference is that in the Blue Creek Water-
shed the pre-project information/education efforts ob-
tained early project participation, which paved the way for
the technical assistance personnel. In contrast, the High-
land Silver Lake Watershed technical and educational ef-
forts began simultaneously.
RECOMMENDATIONS
1. Require documentation of the nature and extent of
the water resource impairment. Special physical charac-
teristics that could preclude achieving water quality goals
(whether documented or perceived) must be recognized.
2. The type of water quality problem determines critical
area delineation.
3. The program must be able to correct the identified
water quality problem.
4. The information and education program must pre-
cede the technical assistance and cost-sharing compo-
nents.
5. The use of interagency coordinating committees at
the local, State and Federal levels in RCWP projects has
proven to be effective in providing leadership and man-
agement for water quality projects.
REFERENCES
Davenport, T.E. 1983. Soil Erosion and Sediment Transport Dy-
namics in the Blue Creek Watershed, Pike County, Illinois.
IEPA/WPC/83-004. Plann. Sect. Div. Water Pollut. Control. III.
Environ. Prot. Agency, Springfield.
Davenport, IE., and M.M. Kelly. 1984. Water Resource Data
and Preliminary Trend Analysis for Highland Silver Lake Moni-
toring and Evaluation Project, Phase III. IEPA/WPC/84-U30.
Plann. Sect. Div. Water Pollut. Control. III. Environ. Prot.
Agency, Springfield.
Illinois Environmental Protection Agency. 1979. State of Illinois
Water Quality Management Plan, Vol. III. Plann. Stand. Sect.,
Div. Water Pollut. Control. Springfield.
Illinois State Coordinating Committee. 1982. Highland Silver
Lake Watershed Rural Clean Water Project—Summary Rep.
. 1984. Highland Silver Lake Watershed Rural Clean
Water Project—Summary Rep.
Madison County Soil and Water Conservation District. 1979.
Highland Silver Lake: Application for Rural Clean Water Pro-
gram Funding, Madison County, SWCD, Edwardsville, IL.
258
-------
PRAIRIE ROSE LAKE RURAL CLEAN WATER PROGRAM PROJECT
UBBO AGENA
MONICA WNUK
Iowa Department of Water, Air and Waste Management
Des Moines, Iowa
C. MERLE LAWYER
Soil Conservation Service
U.S. Department of Agriculture
Harlan, Iowa
ABSTRACT
Through the cooperative efforts of private landowners
and a number of Federal, State, and local agencies, the
Prairie Rose Lake Rural Clean Water Program (RCWP)
project substantially reduced sediment and nutrient deliv-
ery to the lake from its agricultural watershed. Landowner
participation has been high, with over 75 percent of the
eligible watershed area included in RCWP contracts.
Best management practices installed to date have re-
duced annual sediment delivery to the lake by 55 percent,
from an estimated 23,670 tonnes (26,300 tons) per year in
1980 to 10,627 tonnes (11,808 tons) per year in 1984.
Similar reductions have been projected for phosphorus.
Lake water quality monitoring has shown improvements
resulting from the project. Water clarity has increased
during most seasons of the year, and lake turbidity no
longer routinely increases following runoff events. How-
ever, algal growths during late summer periods appear to
be increasing because of a combination of high in-lake
nutrient levels and decreased sediment-related turbidity
levels.
BACKGROUND
Prairie Rose Lake is an 86 ha (215 acre) State-owned lake
constructed in 1962 near Harlan in Shelby County, Iowa.
The lake and associated Prairie Rose State Park are used
extensively for camping, picnicking, swimming, boating,
and fishing.
The lake's watershed is 1,844 ha (4,610 acres): 259 ha
(648 acres) is lake and park; 1,459 ha (3,648 acres) is
cropland; 59 ha (148 acres) pasture; and 66 ha (166
acres) farmsteads, roads, and woodland. Prior to the
RCWP project, the watershed had severe erosion prob-
lems, with average annual soil loss of the watershed area
(excluding the park and lake) exceeding 44.8 tonnes per
hectare (20 tons/acre), and erosion on 62 percent of the
nonpark land exceeding 67.3 tonnes per hectare (30 tons/
acre).
This excessive soil erosion was the major source of the
lake's water quality problems. Since its construction in
1962, sediment had significantly reduced both the lake's
surface area and its water volume. Between 1968 and
1977, 4.8 ha (12 acres) of lake surface area were lost to
sediment, and an additional 3.2 ha (8 acres) became inac-
cessible to boats with outboard motors. From 1971 to
1980, lake volume was reduced 19 percent, from 250 ha-
m (2,031 acre-feet) to 203 ha-m (1,650 acre-feet) be-
cause of sediment deposition.
Excessive sediment loads also caused the lake to be-
come extremely turbid following runoff. The high lake tur-
bidity levels reduced the lake's aesthetic acceptability and
its use, prompting numerous public complaints.
Sedimentation and high turbidity levels also affected the
lake's fish populations. Sediment reduced the amount of
aquatic habitat suitable for fish reproduction and growth,
and turbid conditions reduced the ability of sight-feeding
fish to catch their prey. Since these conditions affected
sport fish more than rough fish, rough fish predominated
in the lake. Rough fish predominated so greatly that in
1981 when all fish in the lake were killed to carry out a
restocking program, the rough fish (mainly carp and giz-
zard shad) totalled 689 k per ha (615 pounds per acre);
other fish totalled only 100 k per ha (90 pounds per acre).
Other agricultural pollutants of concern in Prairie Rose
Lake are nutrients and pesticides. Nutrients stimulate al-
gal growths and thus increase lake eutrophication. Pesti-
cides are of concern mainly from a human health perspec-
tive since the lake is the source of drinking water for the
park and a major fishing resource in that area of Iowa.
RURAL CLEAN WATER PROGRAM
Funds to implement agricultural nonpoint pollution control
practices in the lake watershed became available in Au-
gust 1980 when the project was approved for funding un-
der the U.S. Department of Agriculture's (USDA) Experi-
mental Rural Clean Water Program (RCWP).
Under the RCWP, Federal funds pay part of the costs of
installing Best Management Practices (BMP's) on farm-
lands, provided the practices reduce agricultural nonpoint
pollution of a receiving stream or lake. The practices used
may include temporary and permanent soil conservation
practices, animal waste controls, and fertilizer and pesti-
cide management practices.
In RCWP projects, a water quality plan is first developed
for cooperating farms, identifying the practices needed to
control agricultural pollutants and scheduling BMP instal-
lation. The plan serves as the basis for a contract between
the farmer and the USDA; the farmer agrees to install the
needed practices in exchange for RCWP funds. The con-
tracts may cover a 3 to 10 year period.
PRAIRBE ROSE PROJECT
Because the water quality problems in Prairie Rose Lake
were directly related to the large quantities of sediment
entering it, improving the lake's water quality required
substantially reducing that entry. In the RCWP project, this
reduction is being accomplished mainly through soil con-
servation practices on the watershed's agricultural lands.
This approach will also decrease the levels of nutrients
and pesticides reaching the lake since these materials are
frequently attached to eroded soil particles.
The soil conservation practices being used include con-
servation tillage, contour farming, terraces, grassed wa-
terways, grade stabilization structures, and pasture man-
agement. Under the project, RCWP funds pay up to 75
259
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
percent of the installation costs of structural practices
such as terraces and grade stabilization structures, while
a per acre payment is made for management practices
such as conservation tillage. A project goal is controlling
excessive soil erosion on 80 percent of the watershed's
nonpark lands.
The Prairie Rose project also uses nutrient and pesti-
cide management programs. Under the nutrient manage-
ment program, Iowa State University Soil Testing Labora-
tory analyzes soil samples collected from farm fields. The
soil test results are used to recommend fertilizer applica-
tion rates and methods which meet crop nutrient needs
while minimizing potential nutrient runoff into the lake.
The pesticide management program involves scouting
the fields to determine whether weed or insect infestations
are sufficient to justify the use of pesticides. If so, the field
scouting results help determine the pesticides, application
rates, and methods to use. The pesticide use recommen-
dations are designed to assure that weed and insect pests
are adequately controlled while the potential for pesticide
runoff is minimized.
A number of Federal, State, and local agencies are coop-
erating in the Prairie Rose RCWP project. These include
° Agricultural Stabilization and Conservation Service,
USDA
° Soil Conservation Service, USDA
° Shelby County Soil Conservation District
° Cooperative Extension Service, Iowa State University
° Iowa Department of Water, Air and Waste Management
° Iowa Conservation Commission
° University Hygienic Laboratory, University of Iowa
° U.S. Environmental Protection Agency
° Iowa Department of Soil Conservation
The Shelby County Agricultural Stabilization and Con-
servation Service (ASCS) office has major responsibility
for the day-to-day administration of the Prairie Rose proj-
ect. Its duties include entering into contracts with partici-
pating farmers and administering the RCWP cost share
funds. The State ASCS office assists the Shelby County
office as needed and works with other Federal and State
agencies to assure that needed coordination of agency
activities occurs.
Soil Conservation Service (SCS) personnel assigned to
the Shelby County Soil Conservation District develop wa-
ter quality plans for cooperating farms and assist farmers
in selecting the soil conservation practices to be used on
their farms. SCS staff are also responsible for the design
and construction layout of the soil conservation practices
used.
The Cooperative Extension Service (CES) of Iowa State
University, working through the Shelby County extension
office, is responsible for conducting the project's nutrient
and pesticide management programs. In addition, the
CES conducts the project's public information and educa-
tion programs.
RCWP regulations require that a general water quality
monitoring program be carried out, but prohibit the use of
RCWP funds for monitoring. As a result, the Iowa Depart-
ment of Water, Air and Waste Management (DWAWM) has
obtained water quality funds from the U.S. Environmental
Protection Agency to support the monitoring program.
DWAWM also develops the scope of the project's monitor-
ing program and prepares annual water quality monitoring
reports.
The major sample collection activities are conducted by
Iowa Conservation Commission (ICC) field staff located at
Prairie Rose Lake. ICC staff also collect and maintain rec-
ords on lake conditions and lake use.
Under contract with DWAWM, the University Hygienic
Laboratory (UHL) analyzes the water samples and pro-
vides training to ICC staff on sample collection proce-
dures.
The Iowa Department of Soil Conservation (DSC) had a
major role in developing the initial RCWP application for
the Prairie Rose project. DSC also serves on the State
and local project coordination committees.
Prior to the RCWP project, soil conservation practices
were being used on only a small percentage of the Prairie
Rose watershed. Conservation plans had been developed
for 14 of the 47 farms in the watershed, contour farming
was used on about 400 ha (1,000 acres), and 24.2 km
(15.1 miles) of grassed backslope terraces and 2 erosion/
sediment control structures had been constructed (Law-
yer, 1983). Even with these practices, average erosion
rates in the watershed were extremely high, particularly
on the cropland areas (see Table 1).
Since being approved for funding in August 1980, the
Prairie Rose RCWP project has made considerable pro-
gress. The project has been well accepted by farmers in
the lake watershed: as of October 1984, 35 of the 47 land-
owners in the lake watershed had applied for RCWP con-
tracts, 33 had been signed. The 33 signed contracts cover
1,236 ha (3,089 acres), or 79 percent of the eligible area,
and represent a commitment of $351,000 in RCWP funds,
out of a total project cost share allocation of $446,000.
Lands under contract are shown in Figure 1.
Considerable progress has also been made in imple-
menting BMP's in the lake watershed. Table 2 shows the
BMP's implemented as part of the project by December
1984 and includes only those practices using RCWP
funds.
Logond
I I Lotto
ggg Pork Land
Lnnd Undor Controel
Figure 1 .—Land under contract, Prairie Rose Lake Water-
shod (ASCS, 1984).
Table 1.—Cropland erosion rates (Shelby County
ASCS, 1980).
Location
Sidehills
Hilltops
Bottomland
Annual Soil Loss
Tonnes/ (tons/
hectare acre)
67.3 (30)
11.2 ( 5)
11.2 ( 5)
Area
Hectares (acres)
975 (2,438)
197 ( 492)
396 ( 990)
Three landowners not under RCWP contract have also
installed soil conservation practices on their lands. Two of
these landowners established a total of 14.4 ha (36 acres)
of permanent pasture seeding and one constructed 4.5
km (2.8 miles) of terraces.
The RCWP practices and non-cost-shared practices
such as crop residue management and contouring to-
gether are adequately protecting 55 percent of the water-
260
-------
BMP
Table 2.—BMP implementation (ASCS, 1984).
Amount
implemented
Pasture seeding
Terrace system
Waterway systems
Conservation tillage system
Sediment/water control structure
Nutrient management
Pesticide management
12.8 hectares
(32 acres)
70.7 kilometers
(43.9 miles)
4.0 hectares
(10.1 acres)
224 hectares
(560 acres)
8 structures
952 hectares
(2,379 acres)
(27 farms)1
952 hectares
(2,379 acres)
(27 farms)1
AGRICULTURAL ISSUES: MIDWESTERN EXPERIENCE
72.720
Soil Erosion
Lake Sedimentation
'Total contracted as of December 1984—includes some contracted after 1984 grow-
ing season.
shed from excessive soil erosion—substantial progress
toward the project goal of erosion control on 80 percent of
the eligible watershed area.
Soil losses in the lake watershed have been reduced
from a pre-project level of 72,720 tonnes (80,800 tons) per
year to a current level of 33,210 tonnes (36,900 tons) per
year. Assuming a sediment delivery ratio of 32 percent,
the annual sediment delivery to the lake has been re-
duced 55 percent, from a pre-project level of 23,670
tonnes (26,300 tons) to a current level of 10,627 tonnes
(11,808 tons). The reductions in soil erosion and sediment
delivery are shown graphically in Figure 2.
The reduction in sediment delivery has created a paral-
lel reduction in the delivery of sediment-borne nutrients
and pesticides. Implementation of the nutrient and pesti-
cide management program is reducing these pollutant
loads even further. For example, data from participating
farmers indicate that the nutrient management program
has influenced their use of fertilizers, with average appli-
cation rates of phosphorus (as P2O5) declining from 49.3
and 61.7 kg/ha (44 and 55 Ibs/acre) for corn and soy-
beans, respectively, in 1982 to 22.4 and 6.7 kg/ha (20 and
6 Ibs/acre), respectively, in 1984.
Several activities besides the RCWP project have been
Figure 2.—Reduction in soil erosion and lake sedimenta-
tion, 1980-1984.
undertaken to improve the lake. During 1982, the Shelby
County Board of Supervisors reconstructed a road adja-
cent to the lake. As part of the reconstruction, a bridge
spanning the upper arm of the lake was replaced by a
box-inlet culvert that temporarily impounds watershed
runoff above the road, thereby allowing soil particles to
settle out before the runoff enters the lake.
In the fall of 1981 the ICC initiated a complete fish reno-
vation project for the lake. At that time lake levels were
reduced and all fish were killed. The ICC has since con-
ducted a fish restocking program, and has completed sev-
eral projects to improve fish habitat and public access.
WATER QUALITY MONITORING
PROGRAM
A monitoring program tracks both water quality and water-
quality-related data. These data include records of lake
attendance, major use activities, fish population invento-
ries, lake bottom profiles, and lake physical conditions.
Water quality information is obtained from water samples
collected at five in-lake locations from May through Sep-
tember. Fish and sediment analyses are being performed
on an annual basis.
Figure 3.—Sampling locations (Iowa Department of Environmental Quality, 1982).
261
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Figure 3 shows the five in-lake sampling locations (Iowa
Dep. Environ. Quality, 1982):
Maximum Type of
Site Description Depth Sampling
1
2
3
4
5
Upper Arm of Lake
Mid-Lake South of
Swimming Beach
Deepest Part of Lake —
Near Dam
Drinking Water Intake
Swimming Beach
2.4 meters
(8 feet)
3.3 meters
(11 feet)
7.2 meters
(24 feet)
4.5 meters
(15 feet)
3.3 meters
(1 1 feet)
fixed
schedule
fixed
scheduled
fixed
schedule
after
runoff
after
runoff
Samples are collected from sites 1, 2, and 3 biweekly and
analyzed for a variety of water quality parameters. Site 4,
the drinking water intake, is sampled once annually follow-
ing a 5-cm (2-inch) rainfall and analyzed for pesticides and
heavy metals. Site 5, at the swimming beach, is sampled
following all rainfall events greater than 2.5 cm (1 inch)
from June through August and analyzed for fecal coliform.
The DWAWM presents monitoring results in annual wa-
ter quality monitoring reports. These reports discuss both
the current monitoring results and long-term water quality
trends. The first of these reports, Prairie Rose Lake Moni-
toring RCWP Project—Year 1 (1981), also reviewed all
data available from previous monitoring activities. Since
insufficient data existed on pre-1981 lake conditions, it
has been necessary to assume that 1981 monitoring
results reflect preproject conditions.
IN-LAKE WATER QUALITY CHANGES
Since its start in 1981, the monitoring program has yielded
a large amount of water-quality and water-quality-related
data. Data analysis indicates that several significant
changes have occurred in the lake since the beginning of
the RCWP project.
Observations by Prairie Rose Park personnel indicate
that the BMP's installed in the lake watershed have signifi-
cantly reduced the impact of watershed runoff on the
lake's water quality. Their observations indicate that, in
contrast to the preproject situation where lake turbidities
routinely increased immediately following runoff events,
lake turbidity levels now do not change appreciably follow-
ing runoff events.
Monitoring results for turbidity, Secchi depth, and chlo-
rophyll a confirm that lake water quality conditions have
changed appreciably. As compared to 1981 levels, lake
turbidities decreased dramatically in 1982, with surface
and bottom turbidity levels dropping 33 and 50 percent,
respectively. At the same time, lake clarity increased sig-
nificantly, with 1982 Secchi depths being nearly three
times those measured in 1981. Since 1982, turbidity levels
have been increasing, with the 1984 mean surface turbid-
ity values being about the same as those measured in
1981. Table 3 presents the turbidity data for 1981 through
1984.
While the trend of increasing turbidity values in 1983
and 1984 is of some concern, analysis of other monitoring
data indicates these turbidity levels are due mainly to in-
creased algal growths rather than being sediment related.
Chlorophyll a data collected in 1981 showed relatively
low levels of algal growth, indicating that the high turbidi-
ties found at that time were due mainly to the high sedi-
ment loads entering the lake in runoff, as well as by sedi-
ments being stirred up by carp and other bottom feeding
fish. The 1982 monitoring results were somewhat of an
anomaly, since chlorophyll a levels decreased, even
though the low turbidity and high water clarity levels meas-
ured that year would be expected to result in increased
algal growth. Chlorophyll a values did increase signifi-
cantly in 1983 and 1984, with mean surface values at sites
2 and 3 over twice the values measured in 1981. These
increases indicate that lake turbidity is now related primar-
ily to algal growth, rather than sediment loads. Table 4
presents the chlorophyll a values measured for 1981-
1984.
Algal assay results have generally shown phosphorus
to be the limiting nutrient in Prairie Rose Lake, indicating
that further BMP implementation efforts should empha-
size the use of practices effective in reducing phosphorus
delivery to the lake. However, since high levels of phds-
phorus currently exist in the lake and its sediments, reduc-
ing the annual phosphorus load to the lake is unlikely to
reduce algal growth levels in the short term. Other lake
restoration measures, such as precipitation of phosphorus
from the water column or removal of bottom sediments by
dredging, may be required if significant reductions in algal
growths are to be achieved more quickly.
Water samples taken at the swimming beach have gen-
Table 3.—Mean turbidity (NTU's) and ranges (Wnuk, 1984).
Site #1 Site #2
Site #3
Surface
Bottom
Surface
Bottom
Surface
Bottom
1981 Mean
1981 Range
1982 Mean
1982 Range
1983 Mean
1983 Range
1984 Mean
1984 Range
20.7
4.9-75
7.1
2.8-32
12
2.3-28
15.8
3.5-31
31.1
12.0-85
14.7
4.4-44
15
2.7-24
17.5
5.3-32
10.7
5.1-23
3.0
1.5-4.8
8.2
1.6-24
11.5
2.5-20
102
17.2-540
12.2
2.1-22
19
1.9-60
19.2
4.6-50
8.8
1.9-15
2.7
1 .2-4.2
7.5
1.6-20
10.9
2.3-18
84.3
13.0-340
10.3
4.7-16.0
18
2.0-55
14.8
5.5-28
Table 4.—Mean chlorophyll-a (/ig/l) and ranges observed (Wnuk, 1984).
Site #1 Site #2
Site #3
Surface
Bottom
Surface
Bottom
Surface
Bottom
1981 Mean
1981 Range
1982 Mean
1982 Range
1983 Mean
1983 Range
1984 Mean
1984 Range
3.7
16.0-85.0
12
3-29
40
4-98
60
21-116
33
14.0-68.8
14
7-30
41
3-73
54
16-105
21
12.0-38.0
13
3-27
43
3-145
51
16-94
24.4
16.0-38.0
16
3-43
24
3-67
29
8-92
17.3
9.0-33.0
12
4-24
39
3-120
46
7-102
24.1
7.0-87.0
15
4-28
21
3-65
29
6-97
262
-------
AGRICULTURAL ISSUES: MIDWESTERN EXPERIENCE
erally been below the 200 organisms per 100 ml fecal
coliform standard established for primary contact waters
in Iowa's Water Quality Standards, indicating the water is
safe for swimming. Similarly, although several pesticides
were detected in samples collected at the drinking water
intake, the levels found are well below the levels consid-
ered harmful to human health.
ICC records indicate that lake use has fluctuated con-
siderably from year to year, largely because of climatic
factors. Total use levels are generally increasing, however,
with total use in 1984 reported at 129,147 user days. Use
of the swimming beach has increased annually, going
from a 1981 total of 55,279 user days to a 1984 level of
75,500 user days.
Fishing at the lake declined in 1982 and 1983. This
decline was expected, since the fish stocked in the lake
following the 1981 fish renovation had generally not
reached a catchable size. Fishing increased in 1984 and
is expected to continue increasing as the lake's fishery
develops.
OBSERVATIONS ON PROJECT
At this time several general observations can be made
regarding the Prairie Rose RCWP project:
1. Considerable progress has been made toward ac-
complishing the project's goals. Complete achievement of
these goals is expected during the project's lifetime.
2. A major factor in the project's success has been the
willingness of local farmers to participate in it. Their partic-
ipation can be attributed to a number of factors, including:
their awareness of the lake's water quality problems; ex-
tensive efforts by county SCS, ASCS, and CES staffs to
inform farmers about the project and solicit their participa-
tion; and the higher cost share rates available under the
RCWP than under other State or Federal cost share pro-
grams.
3. A large number of local, State, and Federal agencies
are cooperating in the project. This cooperation has not
only helped to ensure the project's success, but has also
resulted in completion of several related projects which
increase the lake's public value. For example, the recon-
struction of the county road through the upper arm of the
lake has provided a secondary sedimentation basin for
runoff entering the lake, and the fish renovation project
will greatly improve the lake's recreational value.
4. The water quality changes observed thus far point
out that implementing BMP's in a lake's watershed will not
necessarily correct all of the lake's water quality problems,
and additional measures may be needed to deal with re-
maining problems. In the case of Prairie Rose Lake, the
reduction of sediment loads appears to have resulted in
increased algal growths. Although these algal growths
may eventually decline as a result of lower annual phos-
phorus loadings entering the lake, the high phosphorus
levels currently found in the lake and its sediments make
short-term reductions unlikely.
5. Although it may appear that the Prairie Rose RCWP
project has simply changed the lake's water quality prob-
lem from one of excessive sedimentation to one of high
algal growth, it is important to recognize that the sedimen-
tation problem was threatening the very existence of the
lake itself. While high algal growths may be aesthetically
objectionable during certain periods of the year, the prob-
lems associated with high algal growth are minor by com-
parison.
REFERENCES
Shelby County Agric. Stabil. Conserv. Serv., USDA. 1980. Proj-
ect plan of work—Prairie Rose Lake Watershed.
Iowa Department of Environment Quality 1982. Prairie Rose
Lake monitoring RCWP project—year 1 (1981). Des Moines.
Lawyer, C.M. 1983. The Prairie Rose Lake rural clean water
project. ASAE: 83-2543. Am. Soc. Agric. Eng., St. Joseph, Ml.
Wnuk, M. 1984. Prairie Rose Lake Monitoring RCWP project-
year 4 (1984). Iowa Dep. Water, Air Waste Manage., Des
Moines.
Agricultural Stabilization and Conservation Service. 1984. An-
nual report—Prairie Rose rural clean water project. Agric. Sta-
bil. Conserv. Serv., U.S. Dep. Agric., Des Moines, IA.
263
-------
AGRICULTURAL SOURCES OF NITRATE CONTAMINATION IN A
SHALLOW SAND AND GRAVEL AQUIFER OF EASTERN SOUTH
DAKOTA
JEANNE GOODMAN
South Dakota Department of Water and Natural Resources
Pierre, South Dakota
ABSTRACT
The South Dakota Department of Water and Natural Re-
sources initiated a water quality study of the Big Sioux
aquifer in eastern South Dakota because of growing con-
cerns over potential water quality degradation in the aqui-
fer. Reports of unacceptable levels of nitrates (> 10 mg/L
No3-N) in 25 percent of available domestic well samples
and in several community water systems were of particu-
lar concern. A random sampling network was established
to characterize the general water quality of the aquifer.
Results of the random network sampling indicated nitrate
contamination was generally confined to numerous local-
ized rather than widespread areas. Specifically, 37 per-
cent of 27 domestic well samples exceeded the United
States Environmental Protection Agency (EPA) drinking
water limit of 10 mg/L No3-N. The major agricultural
sources of nitrates were animal wastes from feedlots and
accidental releases of fertilizers, with the contamination
problems compounded by improper well construction and
poor location. Two special studies were conducted near
rural community water supplies contaminated with exces-
sive concentrations of nitrates. Monitoring in the Egan,
South Dakota, area indicated contamination with concen-
trations reaching 240 mg/L NO3. The source of contami-
nation was determined to be leaking liquid fertilizer tanks
near the city well. Monitoring in Elkton, South Dakota,
also indicated the most likely source of contamination
was a fertilizer distributor. A plume of contamination was
defined with a high concentration of 67 mg/L NO3-N in a
well near the distributor. Because of these findings, a
public education program was initiated as part of the Big
Sioux Aquifer Water Quality Study. This program focused
on nitrate contamination sources, means of contamina-
tion prevention, and recommendations for well construc-
tion and location.
INTRODUCTION
Agriculture is the predominant land use in eastern South
Dakota and is characterized by small diverse farms aver-
aging 175 ha (Census of Agriculture, 1982). Corn, soy-
beans, and small grains are the major cash crops raised
with some feed crops grown for small livestock operations.
Draining the eastern counties, the Big Sioux River drain-
age basin supports the typical agricultural activities of
eastern South Dakota (Fig. 1).
The Big Sioux aquifer is a shallow glacial aquifer associ-
ated with, and underlying the floodplains of the Big Sioux
River and its tributaries. The aquifer, the major water re-
source in the basin, consists of sand and gravel outwash
and alluvial material averaging 9 m (30 feet) in thickness in
most areas. Because these permeable aquifer materials
are at or near the land surface, the aquifer is susceptible
to contamination by land surface activities.
The susceptibility of surficial aquifers to surface con-
tamination is evidenced by the South Dakota Department
of Water and Natural Resources' (DWNR) water quality
data base for shallow wells (less than 50 feet in depth). In
1979, 25 percent of the available samples exceeded the
United States Environmental Protection Agency's (EPA)
Maximum Contaminant Level for public drinking water
supplies of 10 mg/L nitrate-nitrogen (NO3-N). Since the
Big Sioux aquifer is the primary source of drinking water
for approximately one-third of the State's population and
is susceptible to contamination, a possibility that the aqui-
fer was undergoing widespread water quality degradation
prompted the Big Sioux Aquifer Water Quality Study. This
study was initiated to determine the general water quality
of the aquifer and define the areal extent and sources of
nitrate contamination.
A random network of observation and domestic wells
was established and sampled to characterize the general
water quality of the Big Sioux aquifer. Also, several special
studies were conducted in areas of known nitrate contami-
nation. Results of the random network sampling indicated
nitrate contamination was not widespread but consisted of
numerous localized problems. Several sources of nitrates
were identified including landfills, sewage logoons, feed-
lots, septic tanks, and accidental releases of nitrogen-
based fertilizers. The following discussion, however, fo-
cuses on agricultural sources of nitrates causing
contamination of domestic and public drinking water sup-
plies.
DISCUSSION
Domestic Wells
The predominantly rural population in the basin relies pri-
marily on shallow wells constructed into the Big Sioux
aquifer for domestic use and stock watering. Past sam-
pling has revealed that elevated nitrate levels in domestic
wells are common in the aquifer with nearly 25 percent of
136 domestic wells exceeding the EPA drinking water
limit. The severity of the problem was illustrated during the
random sampling conducted for the Big Sioux Aquifer Wa-
ter Quality Study when an alarming 37 percent of the
domestic wells sampled exceeded 10 mg/L NO3-N, with
nitrate concentrations ranging from < 0.1 to 120 mg/L. In
comparison, less than 9 percent of the sampled observa-
tion wells exceeded 10 mg/L NO3-N with a range of <0.1
to 50 mg/L (South Dakota Dep. Water Nat. Resour. 1985).
Several problem areas were identified as contributing to
the contamination of domestic wells. First, the leaching of
nitrogen from animal wastes in feedlots was identified as
the main source of nitrogen. The number of livestock per
farm is usually less than 100. However, over 80 percent of
the farms in the Big Sioux River Basin have some type of
livestock operation (South Dakota Dep. Water Nat. Re-
sour. 1985). Therefore, feedlots are a common source of
nitrogen.
Second, the location of the domestic well was a prob-
lem. Frequently, the contaminated wells were located
close to (and in some cases within) a livestock contain-
ment area and therefore, close to a contamination source.
In addition, if the shallow water table mimics the topog-
raphy, many of the wells sampled were downgradient from
the contamination source. Unfortunately, the wells used
264
-------
AGRICULTURAL ISSUES: MIDWESTERN EXPERIENCE
RED/RAINY
MISSOURI COTEAU BASIN-^ PRAIRIE
/NON-CONTRIBUTING AREA [ \ COTEAU
NON-
CONTRIBUTING
•AREA
ESOTA
'ETSTONE
B/ASIN
N
Fron Stacb, tt.al., 1984
Figure 1.—Location map of Big Sioux Basin.
for stock watering were often also used for domestic sup-
plies, thereby supplying contaminated drinking water for
humans.
Third, improper well construction allowed nitrate con-
tamination of domestic wells. Domestic wells often lacked
a sufficient surface seal to prevent contaminated surface
runoff from entering the well annulus (the area between
the well casing and the sides of the bore hole). Addition-
ally, since 40 percent of the domestic wells constructed
after 1976 were bored, many wells were cased with large-
diameter porous concrete, wood curbing, or similar mate-
rials which allow the infiltration of any surface runoff enter-
ing the well bore into the well. Since many bored wells
were completed to a depth just below the water table they
were likely to have higher nitrate concentrations than if
constructed deeper into the aquifer. The water quality data
from nested observation wells (more than one well at a
site) in the Big Sioux aquifer indicated higher nitrate val-
ues in wells constructed in the upper portion of the aquifer
than in wells constructed into the base of the saturated
material.
Case Study 1. A contaminated domestic well that illus-
trates the three problems discussed above was investi-
gated when a case of infant methemoglobinemia (blue
baby) was reported in 1979 to DWNR. The field investiga-
tion revealed that the well, used for drinking water and
stock water, was a shallow, large diameter, bored well. It
was cased with porous concrete and was located in a
depression down-slope from two hog confinement areas
and from the septic tank drainfield. The feedlots were ac-
tive and had been used for several years. It appeared that
feedlot runoff and leachate as well as septic leachate were
the sources of nitrates. The poor location and improper
well construction compounded and magnified the contam-
ination potential. Well samples had nitrate concentrations
varying from 120 mg/L to 210 mg/L NO3-N (Busch and
Meyer, 1982). This situation was typical of domestic wells
sampled where nitrate concentrations exceeded the EPA
drinking water limit of 10 mg/L NOs-N.
Elkton
IftO
• Data point.
Data point with ammonia.
i Test holt.
t
Aquifer boundary.
J
All numborm orf nltratt
conciMrattOM fn mgSL.
Map location a ucllon 21.
r. IO9 H.. R. 47 W.
| 10 to 29
| \Lnt Ihon 0
From Stick, cl.il., 1984
Figure 2.—Nitrate concentrations in the Elkton, South Da-
kota, area.
Public Supply Wells
In 1979, 33 public supply wells, providing water from the
Big Sioux aquifer, exceeded the EPA drinking water limit
of 10 mg/L NO3-N. As part of the Big Sioux Aquifer Water
Quality Study, several special studies were conducted in
areas near selected contaminated wells. Although there
were several probable sources of nitrogen, mishandling of
commercial fertilizers was the primary agricultural source.
As in the contaminated domestic well situation, the loca-
tion of the public wells close to the fertilizer distributors
magnified the potential for nitrate contamination.
Case Study 1. Elkton, South Dakota—Elkton had con-
sistently exceeded the EPA drinking water limit for nitrates
of 10 mg/L NQrN. The town derived water from two wells
constructed into the Big Sioux aquifer. A special investiga-
tion was conducted in and near Elkton to delineate the
extent of nitrate contamination and to attempt to identify
the source of the contamination.
Several observation wells were constructed from which
water level elevations were measured and water samples
were taken. Nitrate values within the town limits ranged
from 7.3 mg/L to 67 mg/L NOrN (Stach, et al. 1984). As
illustrated in Figure 2, a plume of nitrate contamination
was centered near the north end of Elkton and obviously
affected the public drinking water supply. Field investiga-
tions indicated the location of the highest nitrate concen-
tration was near a commercial fertilizer dealer. Spilled fer-
tilizers or equipment rinsate had reached the ground
water in the area causing a contamination problem for
Elkton.
265
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
66.0
355 57.0
Liquid fertilizer tank
I -20-5
*740.0 e33.0
o City well
20.0
.1R8
16.5
21.0
10.0
20.4
23.0
• Data Point
35.5 Nitrates (N) in ppm
feet
400
Figure 3.—Egan, South Dakota, nitrate study.
R 49 W
R 48 W
1 Mile
- Monitoring Well
2.0 Nitrates (N) in ppm
Figure 4.—Nitrates at Flandreau-Egan, South Dakota.
266
-------
Case Study 2. Egan, South Dakota—Egan also con-
sistently exceeded the EPA drinking water limit of 10 mg/L
NO3-N. The concentration in the public well drastically
increased from 25 mg/L to 70 mg/L NO3-N in 1979. Egan
initiated a study which revealed nitrate contamination of
ground water with a high value of 240 mg/L NO3-N in the
immediate vicinity of several liquid fertilizer storage tanks
(Fig. 3; Stach et al. 1984). The public well was approxi-
mately 1 block from the tanks.
As a special study for the Big Sioux Aquifer Water Qual-
ity Study, an area including Egan and Flandreau, South
Dakota, was investigated in 1982 to determine the extent
and sources of nitrogen contamination. In an attempt to
define the area! extent of the 1979 fertilizer spill, subse-
quent sampling of ground water in the contaminated area
of Egan indicated that even though no further fertilizer
releases had occurred, nitrate concentrations remained
elevated with 77.2 mg/L NO3-N detected near the fertilizer
tanks (Fig. 4; Stach et al. 1984).
CONCLUSIONS
The results of the Big Sioux Aquifer Water Quality Study in
eastern South Dakota indicate that the nitrate contamina-
tion occurs in numerous localized areas rather than being
widespread throughout the aquifer. Several sources of
contamination were identified. Among these were agricul-
tural sources of nitrates causing contamination of domes-
tic wells and public supply wells.
Generally, the major problem areas resulting in high
nitrate concentrations in domestic wells were (1) feedlots
or livestock containment areas as sources of nitrogen, (2)
AGRICULTURAL ISSUES: MIDWESTERN EXPERIENCE
location of the well relative to the source providing con-
taminated drinking water, and (3) improper well construc-
tion allowing contamination. Public supply wells with ni-
trate concentrations exceeding the EPA drinking water
limit had the following common problems: (1) improper
handling or accidental releases of commercial fertilizer as
a source of nitrogen and (2) poor location of the public well
relative to the source providing contaminated drinking wa-
ter.
To disseminate information gained from the study, a
public education program was initiated by the East Dakota
Conservancy Sub-District (a locally elected and State-
funded water district). In addition to making public presen-
tations, the agency prepared three pamphlets presenting
the basic data and information gathered from the Big
Sioux aquifer study stressing the importance of proper
well construction and location. It is the intention of DWNFt
to use the Big Sioux aquifer study to further educate the
general public to prevent contamination of the aquifer and
provide a safe source of drinking water.
REFERENCES
Busch, D. and M. Meyer. 1982. A case of infantile methemoglo
binemia in South Dakota. J. Environ. Health 44(6): 310-11.
Census of Agriculture. 1982. South Dakota Volume, Part 41.
U.S. Dep. Commerce, Washington, D.C.
South Dakota Department of Water and Natural Resources.
1985. The Big Sioux Aquifer Water Quality Study, Draft Re-
port. Pierre.
Stach, R., J. Allen, and S. Chadima. 1984. Big Sioux Aquifer
Study Part 1, Draft Final Report. South Dakota Geol. Surv.
University South Dakota, Vermillion.
267
-------
Agricultural Issues:
Western Experience
AGRICULTURAL ISSUES: THE NEBRASKA PERSPECTIVE
ROGER E. GOLD
University of Nebraska-Lincoln
Lincoln, Nebraska
INTRODUCTION
In the past several years several programs in Nebraska
have addressed the problems associated with nonpoint
sources of pollution. Two specific programs receiving na-
tional attention were the Maple Creek Model Implementa-
tion Project and the Hall County Water Quality Special
Project. Both projects involved interagency, multidiscipli-
nary approaches to the planning, implementation, and
evaluation of best management practices (BMP's) de-
signed to reduce nonpoint pollution. Both projects have
been considered a success, even though each addressed
related, but different problems.
DISCUSSION AND RESULTS
Maple Creek Model Implementation Project. The proj-
ect area was located in the upper reaches of the Maple
Creek drainage in Colfax, Stanton, and Platte counties of
Nebraska. The topography is characterized by steep hills
and irregular slopes covered with deep, silt loam soils.
The project area included 320 landowners and 165 farm
units within 13,440 ha (33,088 acres). An estimated 85
percent of the area was cropland, 11 percent pasture, and
4 percent was used for other purposes.
Conservation land treatments had not been readily ac-
cepted in the area; only 10 percent of the needed prac-
tices had been applied prior to the beginning of the proj-
ect. Soil losses, as calculated by the Soil Conservation
Service (SCS), exceeded 224.2 t/ha (100 tons/acre) on 3
percent of the cropland with 65 percent of the area having
soil losses greater than 11.2 t/ha (5 tons/acre). The mean
soil losses for the project area were 49.3 m/ha (22 tons/
acre) per year with a total annual soil loss of 735,584 t
(724,000 tons). Of that amount, 109,728 t (108,000 tons)
were delivered to the mouth of the watershed.
The project was initiated in 1978 and ran through De-
cember 30, 1982, at which time the final report was pub-
lished. The project objectives were: (1) to demonstrate the
efficiency of certain selected conservation practices
(BMP's) for the abatement of soil losses (nonpoint pollu-
tion) affecting the water quality of Maple Creek, (2) to dem-
onstrate the cost effectiveness of BMP's for controlling
nonpoint pollution, (3) to determine if cost-sharing for non-
point pollution control would enhance adoption and imple-
mentation of BMPs, and (4) to demonstrate the effective-
ness of Natural Resources Districts (NRD's) as areawide
management agencies for nonpoint pollution control.
Agencies participating in the project were Agricultural
Stabilization and Conservation Service (ASCS), Nebraska
Cooperative Extension Service (CES), Nebraska Depart-
ment of Environmental Control (NDEC), U.S. Environmen-
tal Protection Agency (U.S. EPA), Economics Research
Service (ERS), Soil Conservation Service (SCS), Agricul-
ture Research Service (ARS), Nebraska Forest Service
(FS), Nebraska Natural Resources Commission (NNRC),
Nebraska Department of Environmental Control (DEC),
and the Lower Elkhorn Natural Resources District
(LENRD).
BMP's approved for cost-sharing under the project in-
clude vegetative cover, terraces, diversions, windbreaks,
conservation tillage, reduced tillage, impoundments, ero-
sion structures, sod waterways, and animal waste facili-
ties.
Results of the project are summarized in Table 1. During
the 5-year period, 140 of the 280 owner-operators in-
stalled one or more conservation practices resulting in 45
percent of the needed land treatments being completed.
More conservation work was estimated to have been ac-
complished in the project period than in the prior 40 years.
As an example, the installation rate for terrace systems
269
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 1.—Soil and water conservation activities: preproject and project period
comparisons in Maple Creek Model Implementation Project.1
Progress items Preproject
and conservation per
practices Total year 78
Individuals assisted
(No.) -
Technical services
(No.) -
Individuals applying a
practice (No.) —
Conservation plans
(Ac.) 14,800
Land adequately
protected (Ac.)
Terraces (ft) 30,000
Underground outlets
(ft) o
Grassed waterways
(Ac.) 20
Dams (No.) 3
Conservation tillage
(Ac.) 3,500
20 72
100 542
8 58
300 3,841
1,101
5,000 55,751
— 1,040
3 26.0
2 10
— —
79
191
1,048
96
5,978
2,030
37,179
3,230
6.3
11
—
Project period
fiscal year
80 81
157
1,028
112
2,320
2,115
38,674
11,273
8.8
14
—
171
1,030
104
3,060
2,692
67,429
38,062
6.9
3
—
82
64
930
109
1,100
2,880
59,993
38,138
6.6
12
—
Total
655
4,578
479
16,299
10,818
259,026
91,743
54.6
50
6,300
'Information from Table III-2, page 11, of the Maple Creek Model Implementation Project Final Rep., Dec. 30,1982.
increased by eightfold per year. Adoption of conservation
tillage increased by approximately 410 ha/yr (1,000 acres)
with 2,526 ha (6,300 acres) included in 1982 alone. Soil
loss was reduced by at least 20 percent, with gross ero-
sion reduced by approximately 158,500 t (156,000 tons).
Annual reduction of sediment delivery to the mouth of the
watershed was reduced by 12,192 t (12,000 tons). The
tenfold increase in cost-share funds earned by the pro-
ducers in the project area benefited 15 times the acres.
The impact of BMP implementation on water quality
within the watershed is still being quantified. The baseline
studies, completed in 1982, include water analysis, as-
sessment of insect and fisheries resources, and algal
monitoring. The post-treatment monitoring portion of the
project has been tentatively planned for 1987.
Success in terms of adopting and implementing BMP's
by 50 percent of the owners/managers within the project
area was attributed directly to their interest and dedication
to improve water quality by decreasing nonpoint pollution.
Another major factor associated with these successful ef-
forts was the teamwork of the interagency working groups
that met with producers to explain BMP's, arrange for
cost-sharing, and generally educate the owners/managers
on the long-term impacts of nonpoint pollution. The adjust-
ments to cost-share rates which allowed for 90 percent
payment to a $3,500 Agricultural Conservation Program
(ACP) limitation per year was also instrumental. In addi-
tion, the Lower Elkhorn Natural Resources District pro-
vided further financial incentives of up to $16.20/ha ($40/
acre) for approved practices (terraces) completed during
specific time periods. Including all commitments for train-
ing, cost-share, monitoring, and evaluations the total cost
of the Maple Creek Project was estimated at $1.8 million
and involved 26.4 staff years (Maple Creek Model Imple-
men. Proj. Final Rep. 1982).
Hall County Water Quality Special Project. This proj-
ect was located in the central and west center part of Hall
County, Nebraska. The project area was designated at 65
sections (16,848 ha) of which approximately 13,365 ha
(33,000 acres) were irrigated. The soils in the project area
varied but consisted of level to nearly level alluvial topsoil
(sandy loam to silty clay loam) over approximately 15 m of
sand and gravel. Soils were characterized by high water
infiltration rates and low water-holding capacities. Depth
to water table was shallow and ranged from 1 m in the
bottom lands to 20 m on the upland fields. Approximately
90 percent of the area was furrow irrigated from ground
water wells. The long-term average annual precipitation
was 61 cm, with irrigation typically providing an additional
20-75 cm annually. Land use in the area included 85 per-
cent cropland and 15 percent farmland and other uses.
There were 325 farms within the original project area. At
project initiation, an estimated 90 percent of the land
needed treatment based on water quality criteria; 10 per-
cent had been adequately treated.
The Hall County Water Quality Special Project was initi-
ated in 1979 in specific response to the accumulation of
nitrate-nitrogen (NO3-N) in the ground water under
much of the Central Platte River Valley of Nebraska. Stud-
ies showed that nitrate concentrations in water from sev-
eral domestic and irrigation wells exceeded the 10 ppm
permissible standard for public water supplies (Fed. Water
Pollut. Control Admin. 1968; U.S. Environ. Prot. Agency,
1976). Additionally, these and related studies indicated a
steady trend of increasing concentrations through time
(Exner and Spalding, 1976; Olson et al. 1962). Exner
(1984) estimated that the rate of increase of NO3 - N in the
region was approximately 1 ppm/yr.
Specific objectives for the project were to impede the
leaching of nitrates into the aquifer, improve ground water
quality by removing existing nitrates through irrigated corn
production, demonstrate that nitrogen and irrigation could
be managed efficiently and effectively without adversely
affecting crop yields, and develop and demonstrate man-
agement practices applicable in similar situations through-
out Nebraska and the United States.
The methods and procedures used in this project in-
volved introducing and implementing fertilizer and irriga-
tion management. Cost-share monies for these approved
BMP's were provided to participating producers from spe-
cial ACP allocations administered through the ASCS. As
with the Maple Creek Project these efforts were multidis-
ciplinary and multiagency with basically the same repre-
sentation, except that the local Natural Resources District
was the Central Platte NRD.
270
-------
AGRICULTURAL ISSUES: WESTERN EXPERIENCE
Nitrogen management BMP's were implemented with
individual producers based on the NO3 - N content of the
soil (determined through soil sampling and analysis),
NO3-N content of the irrigation water expected to be
applied (based on well sampling), and a realistic yield
goal. Based on the various inputs, a nitrogen fertilizer
recommendation was developed by a project specialist
working one-on-one with the cooperators. Results of the
nitrate management portions of the program were mea-
sured partly through yield checks using a weigh wagon.
The irrigation management program was to demon-
strate the methods and equipment available to manage
irrigation water. Techniques employed to improve water
management included irrigation scheduling and changing
irrigation set sizes, set times, and, where applicable,
length of rows. The use of irrigation water meters, electri-
cal resistance blocks, tensiometers, and soil probes was
stressed. Moisture blocks and tensiometers were installed
in representative sites in the fields. During the irrigation
season field scouts read the blocks and tensiometers and
made irrigation scheduling recommendations. These rec-
ommendations were based on the soil moisture deficit in
the root zone, and calculated based on soil moisture mea-
surements and soil characteristics. The specific goal was
Table 2.—Nitrogen management program participation,
1980-83 Hall County Water Quality Special Project.1
1980 1981 1982 1983
Total irrigated acres in
project area
Producer cooperators
Farms serviced
Acres serviced
33,000 33,000 33,000 33,000
35 48 38 34
52 66 63 58
6,216 8,615 8,378 7,190
Percent of total acres
serviced 19 26 25 22
Average acres per
producer 178 179 220 211
'Information from Table 1, page 10, of the Final Report—Nitrogen and Irrigation
Management, Hall County Water Quality Special Report, February 1, 1984.
to apply the right amount of water at the correct time to
meet the needs of the plant while at the same time reduc-
ing the leaching of nitrogen and other agricultural chemi-
cals through the soil profile.
Table 2 summarizes the participation in the nitrogen
management program associated with the Hall County
Project. Approximately 23 percent of the total irrigated
area within the project area was included in this portion of
the program. Response to the nitrate management recom-
mendations were considered good, since the majority of
the participants followed closely the specialist's recom-
mendations for fertilizer applications. The results for the 4
years of nitrogen management are summarized in Table 3.
Cumulative yield goals, developed for 239 farms, ranged
from a low of 88 q/ha (140 bu/acre) to a high of 126 q/ha
(200 bu/acre). The approximate mean weighted average
of yield goals was 108.6 q/ha (173 bu/acre) with a com-
puted total nitrogen requirement of 1,271 kg/ha (234 Ibs/
acre). The mean weighted average recommended rate of
nitrogen fertilizer corresponding to those yield goals was
842 kg/ha (155 Ibs/acre). Therefore, the estimated nitro-
gen saved by following the recommendations was 429 kg/
ha (79 Ibs/acre) per year over the 4 years of the project.
Results of the irrigation management portions of the
project are summarized in Table 4. Based on the data from
5 production years during which irrigation was managed,
approximately one irrigation was saved per season. The
results of this phase of the project were extremely variable
as expected under field conditions; climatic consider-
ations varied from drought (1980) to extremely wet (1982).
The impact of the nitrate and irrigation management
practices on water quality were indirectly assessed
through irrigation well sampling results (Table 5). In gen-
eral, the NO3 - N concentration in the ground water did
not vary significantly from year to year with section means
ranging from 13.6 ppm to 16.7 ppm. The crop yields asso-
ciated with the nitrate and irrigation management prac-
tices did not vary significantly from yields from compara-
tive farms not participating in the project. In other words,
participating farmers were able to produce the same
yields as their neighbors while using 429 kg/ha (79 Ibs/
Table 3.—Yield goals, nitrogen required and recommended, 1980-1983 Hall County Water Quality special project.1
1980 1981 1982 1983
Range of yield goals— bu/a
Weighted average2 of producers'
yield goals— bu/a
Computed total nitrogen requirement for weighted
average2 yield goal — Ibs/a
Weighted average2 rate of fertilizer nitrogen
recommended — Ibs/a
145-200
177
244
162
140-200
169
234
144
150-190
168
234
156
140-200
167
223
158
Decrease in nitrogen applied (normal practice
nitrogen requirement for yield goal less weighted
average2 recommended rate)—Ibs/a
82
90
78
65
11nformation from Table 2. page 11. of the Final Report—Nitrogen and Irrigation Management, Hall County Water Quality Special Report, February 1. 1984.
'Weighted averages take into account differences in acres among producers for yield goal and nitrogen fertilizer rates.
Table 4.—Irrigation water use and estimated water savings for cooperating producers in Hall County project.1
Water applied per Season
Year
1979
1980
1981
1982
1983
Number
of fields
24
42
62
44
32
irrigation (in.)
Ave.
2.9
3.8
3.7
1.8
3.0
Range
2.5- 6.2
2.5-10.2
1.2- 8.4
0.0- 8.5
1.2- 6.4
irrigation (In.)
Avg.
14.3
21.0
13.3
4.8
21.2
Range
6.9-25.3
10.1-53.3
3.7-34.1
0.0-17.1
7.3-51.0
Irrigation
saved, in.2
2.6
0.7
2.6
1.1
2.3
'Information from Table 8, page 16, of the Final Report—Nitrogen and Irrigation Management, Hall County Water Quality Special Report, February 1,1984.
2Based on measured irrigation amounts and cooperator's estimates of number of irrigations saved because of irrigation scheduling.
271
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
acre) less fertilizer and at least one less irrigation (4.74 cm
of water).
Table 5.—Five-year summary of well sampling for
Nitrate-nitrogen concentrations from the
Hall County project area.1
Year
Number Overall Section Percent of Wells
of Range Average Average2 Greater
Wells ppm ppm ppm than 10 pm
1979
1980
1981
1982
1983
82
139
147
98
123
0.3-33.2
1.0-33.7
0.1-35.4
0.3-33.5
0.1-36.9
15.6
13.6
15.5
16.1
15.1
15.4
13.6
15.3
16.7
15.6
79
67
78
77
77
'Information from Table 9, page 17, of the Final Report—Nitrogen and Irrigation
Management, Hall County Water Quality Special Report, February 1,1984.
2TW3 value was determined by calculating the average concentration for each sec-
tion and then the average of the section averages. Differences were not statistically
significant.
The estimated cost of this project, including cost-shar-
ing to cooperating producers, monitoring, consultations
and educational efforts, was $1.1 million. The exact de-
gree to which the BMP's improved water quality during the
five production seasons is not specifically known other
than the NO3 within the project area did not increase sig-
nificantly. Exner (1984) had estimated increases in NO3-
N in surrounding areas at a rate of approximately 1 ppm/
yr. Five years was generally held to be a minimal time
period in which to detect changes in chemical composition
of an aquifer the size of the project area.
CONCLUSIONS
Even though the two projects discussed in this report ad-
dressed very different problems (soil losses versus ni-
trate-nitrogen leaching) the successes noted were based
on some common principles. Some of these were:
1. Nonpoint pollution problems must be identified and
addressed at the local level with planning, implementa-
tion, and evaluation based on focused programs rather
than areawide or statewide efforts.
2. Education should be emphasized in explaining the
advantages (economic and environmental) of BMP's, es-
pecially when the projects are based on voluntary partici-
pation.
3. Adequate cost-sharing must be provided to offset at
least a significant proportion of the real cost of BMP's,
particularly if the producers do not realize an on-farm
monetary benefit.
4. Projects should be planned to last long enough (with
adequate funding detailed in advance) to enable results to
be quantified and evaluations completed.
5. A communications network needs to be established
between local, State, and Federal agencies to ensure
overall coordination.
The costs associated with the abatement of nonpoint
pollution appear to be high, but the benefits can certainly
be dramatic and the environmental consequences consid-
erable.
REFERENCES
Exner, M. E. 1984. Concentration of nitrate-nitrogen in ground-
water from the Central Platte region, Nebraska. Conserv.
Surv. Div. Inst. Agric. Nat. Resour. Univ. Nebraska-Lincoln.
Exner, M. E., and R. F. Spalding. 1976. Groundwater quality of
the Central Platte region—1974. Resour. Atlas No. 2. Con-
serv. Surv. Div. Inst. Agric. Nat. Resour. Univ. Nebraska-Lin-
coln.
Federal Water Pollution Control Administration. 1968. Water
Quality Criteria. U.S. Gov. Print. Off. Washington, D.C.
Olson, R. A., E. C. Seim, and J. Muir. 1962. Influence of agricul-
tural practices on water quality in Nebraska—a survey of
streams, groundwater, and precipitation. Water Resour. Bull. 9
(2): 301-11.
University of Nebraska-Lincoln. 1984. Final Report—Nitrogen
and Irrigation Management Hall County Water Quality Special
Report. Inst. Agric. Nat. Resour. Coop. Ext. Serv.
U.S. Department of Agriculture: 1982. Maple Creek Model Im-
plementation Project—Final Report. Soil Conserv. Serv. Lin-
coln, NE.
U.S. Environmental Protection Agency. 1976. Quality Criteria for
Water. U.S. Gov. Print. Off. Washington, DC.
272
-------
HYDROPOLITICAL SOLUTIONS TO COMPLEX NONPOINT SALINITY
POLLUTION PROBLEMS IN THE COLORADO RIVER BASIN
JACK A. BARNETT
Colorado River Basin Salinity Control Forum
Bountiful, Utah
Two countries and seven States have worked through a
maze of conflicting interests in the Colorado River so that
water resources may be used, additional water resources
may be developed, and water quality can be protected.
The current success of these long and extended negoti-
ations should be judged as almost a political miracle. Most
recently, cost sharing for the treatment of nonpoint source
agricultural return flows has been agreed to.
By compact and treaty, the waters of the Colorado River
have been divided among the upper and lower basin
States and between Mexico and the United States. Addi-
tional water resource development is anticipated in the
future. Previously, studies had indicated that the use of the
waters of the Colorado over the last several decades has
increased the salinity of the river in the lower reaches. It is
anticipated that additional development will further in-
crease the salinities. The States and the Federal govern-
ment agreed in 1974 on a program that would prevent
salinity increasing above the 1972 levels at three down-
stream mainstem measuring points. It is estimated that
approximately 9 million tons of salt will reach Lake Mead
each year. To maintain the salinity levels in the future, it is
expected that more than approximately 1 million tons of
salt a year must be removed from the river system.
In 1974, Congress gave the Department of Interior the
lead responsibility for studying the salinity problems in the
Colorado River Basin and constructing certain salinity
control units that would remove salts that are being con-
tributed to the river from point sources and also from agri-
cultural nonpoint sources. At that time the States agreed
that they would allow for the repayment of up to 25 percent
of the Federal expenses from the upper and lower basin
power accounts. This repayment was to occur over 50
years with 85 percent of the repayment coming from the
lower basin fund and 15 percent from the upper basin
fund-
After a decade of study, it was determined that the Sa-
linity Control Act of 1974 needed to be modified and pro-
grams established to control nonpoint source pollution,
both from public forest and range lands and from agricul-
tural return flows. The seven basin States had previously
agreed to prepare ongoing plans for salinity control, in-
cluding a program of point discharge control by each. This
point discharge control was accomplished through the
Nonpoint Discharge Elimination System provided for in
Federal water quality laws and regulations. Currently the
basin States are administering approximately 725 NPDES
permits in the basin.
Although all basin senators and many congressmen
cosponsored the legislation and the Administration sup-
ported it, the legislation did not pass the 97th Congress
because of cost-sharing issues and some general reserva-
tions by environmental groups. The legislation was reintro-
duced in the 98th Congress with similar support. For more
than a year and a half various versions were considered
by both houses of Congress and portions passed by the
House.
In the closing days of Congress the issues that seemed
to be delaying the passage of the legislation appeared to
be similar to those of the 97th Congress. At that point the
basin States decided to determine if a solution could be
negotiated with the environmental groups involved and
with the level of required cost sharing. Very late in the
Congress an agreement was reached with the environ-
mental groups that produced their support and an in-
creased cost sharing was agreed to. The legislation
passed in the closing days of Congress.
With respect to nonpoint source issues, two areas are of
particular interest to this Congress. The first relates to the
salt contribution to the Colorado River from the large
amount of Federal land that is managed by the Bureau of
Land Management and other Federal agencies within the
Colorado River basin. The second relates to controlling
the salts being leached back to the river as a result of
irrigation activities within the basin.
The Federal government is by far the largest landowner
in the Colorado River basin. Most of the lands are man-
aged by the Bureau of Land Management; the impact of
this management, including grazing and other activities, is
not well understood with respect to the contribution of salts
from these lands to the Colorado River. Further, the oppor-
tunities to prevent salt discharges from these lands to the
Colorado River are not well understood. Therefore, the
1984 Act requires that the Bureau of Land Management
prepare a comprehensive report and report back to the
Congress by 1987.
Saline return flows from irrigated land can be controlled
in part by better irrigation practices. The effectiveness of
this principle has been demonstrated over the last few
years in areas where farmers have taken advantage of
some cost sharing provided by the already established
Agricultural Conservation Program and have improved
their irrigation practices. These voluntary programs have
typically resulted in the expenditure of about 30 percent of
the total cost by the farmer and 70 percent by the Federal
government. The new program proposes that unless there
is a reason for a special exception the local farmer be
required to contribute at least 30 percent. The additional
70 percent will be paid by the Federal government in rec-
ognition of the basinwide and international benefits that
result from the onfarm improvement.
The new legislation requires that the Basin states pro-
vide 25 percent of the funds from the upper and lower
basin accounts, either in reimbursement the year after the
expenditure is made or over time with the payment of
interest. The State's share will be divided among the up-
per and lower basin States in the 85 percent/15 percent
split as previously agreed to in the 1974 legislation. This
means that for onfarm practices the local and State share
of the expense will be a minimum 51 percent.
273
-------
ACCUMULATION OF SEDIMENT, NUTRIENTS, AND CESIUM-137 IN
PRAIRIE POTHOLES IN CULTIVATED AND NONCULTIVATED
WATERSHEDS
DAN B. MARTIN
U.S. Fish and Wildlife Service
Yankton, South Dakota
Prairie potholes are water-holding depressions of glacial
origin that occur throughout 780,000 km2 of prairie in the
northcentral United States and southcentral Canada
(Sloan, 1972). Collectively, these potholes provide the
most productive wetland habitat for waterfowl in North
America. Although it constitutes only 10 percent of the
continental waterfowl breeding habitat, this pothole region
produces abut 50 percent of the duck crop in an average
year, and much more in bumper years (Smith et al. 1964).
These wetlands also furnish essential resting and feeding
habitat during spring and fall migrations. Potholes are
used extensively by other wildlife for water and habitat,
and recently their importance in flood control and ground
water recharge has been recognized. Numerous potholes
have been eliminated in recent decades by drainage for
agricultural purposes. If the benefits of wetlands are to be
sustained, conservation of the remaining habitat is essen-
tial.
Sediment is currently recognized as an important prob-
lem affecting prairie pothole wetlands. Agriculture is the
predominant land use throughout the region, and erosion
of farmland by wind and water often results in the deposi-
tion of field soil directly onto wetlands. Because of the
rolling topography, and the position of wetlands within this
terrain, potholes are intimately linked to the watershed,
and would appear to be highly reactive to land use prac-
tices.
Resource managers perceive the impacts of sediment
on wetlands as being twofold. First, sediment can reduce
and eventually eliminate wetland habitat by filling the pot-
hole basin. Numerous instances of this process have
been documented by the U.S. Fish and Wildlife Service in
its program on wetland acquisition and easement. Ease-
ment areas that several years ago contained viable, pro-
ductive wetlands, now are used entirely for crop produc-
tion. The wetlands have been filled through erosion of the
surrounding watershed. Second, sediment may serve as
an agent for the transfer of chemical contaminants from
the watershed to the wetland. This is not difficult to visual-
ize when one considers that, in many cases, farming is
carried out right up to the edge of ^e wetland; but, in fact,
it is an impact that is not well documented.
The purpose of our work on sedimentation in pothole
wetlands is to determine the relationship between specific
agricultural land use practices and sediment deposition,
composition, and contaminant impacts. This will better en-
able our resource managers to identify areas of problem
deposition (and the cause), to understand biological ef-
fects of accelerated sediment yield, and to evaluate feasi-
bility and effectiveness of various mitigative measures to
reduce sedimentation.
We began our study by selecting 12 experimental water-
sheds in eastern South Dakota. Each of the watersheds
contained a pothole wetland of similar type. Five of these
watersheds consisted of native grassland in good condi-
tion. These were the control areas, representing natural
conditions, and served as a standard against which condi-
tions in the disturbed watersheds could be judged. The
other seven watersheds contained cultivated cropland
(primarily corn, soybeans, oats, wheat, and sunflowers). In
some of the cultivated watersheds, 100 percent of the
ground was in cropland, while in others a grassed buffer
zone existed between the cropland and the wetlands. It
should be noted that we did not select worst case situa-
tions, where erosion and sediment yield would be severe.
We were more interested in obtaining information on nor-
mal or representative cases of land use.
Initially we set out to obtain information in three different
areas. First, we needed a method to estimate quantita-
tively the rate of accumulation of sediments in pothole
wetlands. Accumulation rate could then be related to land
use practices and/or other variables within the watershed.
Second, we wanted to know what effect land use had on
the composition of wetland sediments. For this we mea-
sured particle size (clay, silt, sand), two important nutrients
(nitrogen and phosphorus), and organic matter content.
Third, we attempted to find a method of relating land use
to the transport of sediment-borne contaminants from wa-
tershed soil to wetlands. In all of these efforts, our underly-
ing objective was to provide resource managers with bet-
ter information to preserve and maintain public and
private wetlands. Some additional benefits may be de-
rived from our study. Since potholes seldom (if ever) over-
flow, they function as continuous, long-term, sediment
traps. They should, therefore, provide basic information
on the nature of erosion and sediment yield from small
watersheds. This information may be of value to workers
concerned with nonpoint source pollution in other types of
aquatic environments.
In our particular study, there were significant differences
in watershed soils. Grassed watersheds were lower in clay
and silt, and higher in sand than cultivated watersheds. At
the outset, it is not possible to determine if these differ-
ences result from land use or if this was a bias introduced
by site selection. Grassed watershed soils also contained
significantly higher concentrations of organic matter and
total nitrogen, a factor undoubtedly related to land use.
Total phosphorus was virtually the same in grassed and
cultivated soils.
Wetland sediments varied significantly with respect to
several constituents. Grassland sediments contained a
higher proportion of coarse inorganic particles (silt and
sand), while cultivated sediments were higher in clay.
Grassland sediments also contained more nitrogen and
organic matter. As with soils, no difference was found for
total phosphorus in sediments.
Accumulation rates of the various constituents were cal-
culated from estimates of sediment density, porosity, verti-
cal accretion rate, and concentration in recent sediments.
These estimates showed that total inorganic sediment ac-
cumulated at twice the rate, and clay at five times the rate,
in cultivated watersheds. This selectivity for movement of
clay-sized particles in cultivated watersheds has clear
implications for the transport of chemical contaminants,
since it is generally noted that the latter move predomi-
nantly with the smaller-size particles. The accumulation
rates of nitrogen and phosphorus were also significantly
greater in cultivated sediments.
274
-------
Cesium-137 was used as outlined by Ritchie and
McHenry (1977) to determine recent sedimentation rates
in the wetlands. These same authors later proposed that
movement of this isotope within a watershed should be of
value in predicting the movement of other soil-borne, non-
point source pollutants (Ritchie, and McHenry, 1978). The
average accumulation rate of cesium-137 in the recent
sediments of wetlands was the same in grassed and culti-
vated watersheds. This result would seem to be in direct
conflict with the previously noted results on accumulation
rates of inorganic particles and nutrients. An explanation
of this apparent conflict can be found in a consideration of
the vertical distribution of the contaminant (cesium) in the
watershed soil. In grassed watersheds, most of the ce-
sium is located in the surface layer of soil. In cultivated
watersheds, the cesium is incorporated throughout the
tillage layer. Thus, even though less particulate matter is
moving in the grassed watersheds, this is apparently off-
set by greater concentrations of cesium adsorbed to the
particulate material.
Our study begins to show certain differences that are
occurring in the sedimentation processes of prairie pot-
hole watersheds under differing land use practices. Addi-
tional information will be needed before we can make
AGRICULTURAL ISSUES: WESTERN EXPERIENCE
quantitative predictions. Two points are already suggested
by the present work. One, we would like to see more
conservation tillage practices to reduce the overall sedi-
ment yield and attendant habitat loss. Second, when agri-
cultural pesticides are applied to watersheds, incorpora-
tion by tillage would appear to be helpful in reducing the
amount that is transported to wetland sediments.
REFERENCES
Ritchie, J.C., and J.R. McHenry. 1977. A rapid method for deter-
mining recent deposition rates of freshwater sediments.
Pages 203-7 in H.L. Golterman, ed., Interactions between
Sediments and Fresh Water. Dr. W. Junk B.V. Publ., The
Hague.
. 1978. Fallout cesium-137 in cultivated and non-
cultivated North Central United States watersheds. J. Envi-
ron. Qua). 7: 40-4.
Sloan, C.E. 1972. Ground-water hydrology of prairie potholes in
North Dakota. Prof. Pap. 585-C. U.S. Geolog. Surv., Washing-
ton, DC.
Smith, A.G., J.R. Stoudt, and J.B. Gollop. 1964. Prairie pothole
and marshes. Pages 39-50 in J. P. Linduska, ed., Waterfowl
Tomorrow. U.S. Bur. Sport Fish. Wild!., Washington, DC.
275
-------
IRRIGATED AGRICULTURE AND NONPOINT SOURCE POLLUTION
IN THE SAN JOAQUIN VALLEY OF CALIFORNIA
WILLIAM R. JOHNSTON
Westlands Water District
Fresno, California
INTRODUCTION
Nonpoint source pollution related to Irrigated agriculture is
becoming a significant problem in regard to maintaining
high water quality in California, and particularly in the San
Joaquin River Basin where large quantities of saline sub-
surface agricultural drainage water are produced (Johns,
1984; San Joaquin Valley Interagency Drainage Program,
1979). The problem will occur eventually in any irrigated
area where subsurface drainage problems develop and
where no readily available salt sink exists to dispose of the
saline subsurface drainage water. Four major types of
nonpoint source pollution generally emanate from irri-
gated agriculture. They are: (1) sediments, (2) pesticides,
(3) nitrates, and (4) salinity, including minor elements.
Frequently, on-farm Best Management Practices
(BMPs) can control nonpoint source pollution problems
without causing serious impacts on farm production or the
environment. However, saline subsurface agricultural
drainage water may contain little, if any, sediments or pes-
ticides, but significant amounts of salts, nitrates, and
heavy metals or other toxic elements. In such instances,
otherwise feasible BMPs are difficult and economically
and environmentally impractical to implement on-farm.
Such is the situation in the San Joaquin Valley of Califor-
nia.
In general, farmers employ good soil and water conser-
vation practices including erosion control, integrated pest
management, proper ground water well sealing, tailwater
recovery systems, and efficient irrigation water manage-
ment. These practices minimize the amount of subsurface
drainage water produced.
However, when approximately 15.24 ha-cm/ha (0.5
acre-ft/acre) of saline drainage water is produced on a
365-day per year production cycle, then each farmer must
be prepared to manage abut 616.8 m3 (21,780 ft3) of drain
water for each ha (acre) of farmland drained. The water
could cause substantial environmental problems, if not
properly managed. In some cases, mismanagement of the
water would threaten wildlife or human health.
RULES AND REGULATIONS
Under the Federal government's regulations, irrigated ag-
ricultural nonpoint source pollution is controlled by BMP's
under Section 208 of the 1972 amendments to the Federal
Water Pollution Control Act. Point source pollution is man-
aged by a permit process under Section 303 of the same
act. Under California law, nonpoint source pollution is not
regulated by permit. However, California recently has
taken steps to regulate drainage discharges from irrigated
agriculture under the 1970 Porter-Cologne Water Quality
Control Act, a law designed to strengthen enforcement,
planning, and water quality efforts in the State. The U.S.
Environmental Protection Agency (EPA) also has autho-
rized the California State Water Resources Control Board
to act as its agent for issuing National Pollutant Discharge
Elimination System (NPDES) permits in California. Nine
Regional Quality Control Boards conduct water quality
planning and must issue and enforce water discharge per-
mits throughout the State. The State Board oversees
statewide policies, issues water rights permits, and re-
views Regional Board decisions when petitioned to do so.
Agricultural drainage discharges require a discharge
permit or an exemption issued by one of the nine regional
boards, even though 1977 amendments to the Federal
Clean Water Act effectively preclude EPA from adopting
standards for agricultural drain or nonpoint source dis-
charges.
PERMIT CONSIDERATIONS
Two types of effluent limitations are available to protect the
beneficial uses of water. They are technology-based limi-
tations and water-quality-based limitations.
Technology-based effluent limitations, minimizing pollu-
tant loads to receiving waters, are based on the concept
that dischargers must take all reasonable steps to reduce
pollutant loads, regardless of whether such reductions are
needed to protect specific beneficial uses of the receiving
water. These limitations are established by EPA for major
discharge categories under Section 301 of the Federal
Clean Water Act.
Water-quality-based effluent limitations are widely used
in California and by EPA wherever technology-based limi-
tations are not sufficient to protect the beneficial uses of a
water body. A standard provision in waste discharge per-
mits issued in California requires that the discharge not
degrade the environment. Generally, this requirement is
sufficient to protect the receiving waters from harmful ef-
fects of the discharge. However, effluent limitations for
specific substances are often included in permits to en-
sure that harmful effects do not occur. Numeric effluent
limitations are based upon safe levels to be maintained in
the receiving waters after initial dilutions, in receiving wa-
ter quality prior to the discharge, and in the amount of
initial dilution.
SALT MANAGEMENT
The nonpoint pollution source attracting national attention
recently is the saline agricultural drainage problem in the
western San Joaquin Valley. The western part of the Valley
has all the factors necessary to cause salt management
problems: tight impermeable clay layers, natural salt and
mineral water, and a limited-to-nonexistent outlet for saline
drainage water and salts.
Early in the development of irrigated agriculture on the
west side, drainage was recognized as a problem (Nelson
and Johnston, 1984). It became apparent by the 1920's
that maintaining productive irrigated agriculture on the
west side depended on good disposal of the salt-laden
drainage water. Since the lowest areas of natural drainage
were the streams and sloughs flowing into the San
Joaquin River from the west, they became natural drains
to remedy the area's high water table. Extensive dredging
of these channels was started.
Drainage problems have been increasing in extent and
severity in the Valley. Ultimately, 200,000 ha (500,000
acres) will be affected by salt problems unless drainage
276
-------
AGRICULTURAL ISSUES: WESTERN EXPERIENCE
facilities are installed and some means of managing the
saline drainage water is provided. Without drainage, pro-
ductivity will be lost. The final result may be seen in parts
of the Middle East where lands that once supported pros-
perous early civilizations are now salt desert and virtually
useless. A similar occurrence in the San Joaquin Valley
would be a tragedy, since the San Joaquin Valley is one of
the world's most productive farming areas.
Today, approximately 31,200 ha (77,000 acres) drain
through Valley wetlands either directly or indirectly into the
San Joaquin River. Saline subsurface drainage water from
about another 3200 ha (8,000 acres) is draining exclu-
sively into the Kesterson Reservoir. Also, agricultural land
is drained with saline water flowing into other streams,
large regional evaporation ponds, and in some cases, on-
farm evaporation ponds.
THE SAN LUIS DRAIN AND KESTERSON
RESERVOIR
The San Luis Unit of the Federal Central Valley Project
was authorized by the U.S. Congress some 25 years ago
(RL. 86-188, 74 Stat. 156,1960). The authorization statute
required the Secretary of the Interior to make provision for
constructing the San Luis Drain to the Delta to meet the
drainage requirements of the Unit as generally outlined in
an earlier report (Department of the Interior, 1956). The
302-km (188-mi), concrete-lined drainage canal was de-
signed to convey saline drainage water from the irrigated
lands in the San Luis service area and discharge it into the
western Delta, where it would be carried by tidal action
through the San Francisco Bay into the Pacific Ocean.
Several studies of various drainage alternatives con-
firmed this disposal method as the most desirable long-
term solution to the Valley's drainage and salt manage-
ment problems (Bureau of Reclamation, 1964; California
Dep. Water Resour. 1965; California State Senate Fact-
Finding Committee, 1965; San Joaquin Valley Interagency
Drainage Program, 1979). To regulate the discharge of
subsurface drainage water, however, the Kesterson Regu-
lating Reservoir was added to the project. The Bureau of
Reclamation constructed a portion of the San Luis Drain
and Kesterson Reservoir between 1968 and 1975. Con-
struction stopped in 1975 because of major concerns
about the effect of Drain discharge on the environment of
the San Francisco Delta-Bay Estuary, and because of
funding limitations. Strong opposition came from San
Francisco Bay Area Civic leaders and environmental
groups, who still fear damage to Delta and Bay water
quality if a San Joaquin Valley agricultural drain is ever
allowed to discharge there.
As a result, only 132 km (82 mi) of the Drain and only
485 ha (1,200 acres) of the planned 1,860 ha (4,600 acres)
of Kesterson Reservoir were completed. Since no addi-
tional drainage disposal facilities have been made availa-
ble, a limit was placed on the amount of effluent allowed
from the 17,000 ha (42,000 acres) of irrigated land in
Westlands where the 3,240 ha (8,000 acres) draining into
the Kesterson Reservoir are located (Fig. 1). This subsur-
face agricultural drainage water contains salts ranging
from 6,000 to 15,000 mg/l Total Dissolved Salts (TDS)
(parts per million (PPM)), with small amounts having salin-
ity concentrations of up to 100,000 mg/l-TDS (ppmnDS).
The drain water is high in common salts, calcium, magne-
sium and sodium sulfates, and boron.
In addition to various salts, drainage water contains
measurable amounts of trace elements such as selenium,
cadmium, chromium, copper, and zinc. Through irrigation,
these naturally occurring elements are leached out of the
soil along with the salts.
Figure 1.—Westlands Water District area where drainage
collector system is complete.
In 1981, one of these elements, selenium, was found in
substantial concentrations in the subsurface drainage wa-
ter and is believed to be the cause of deformities and
mortalities in waterfowl residing at Kesterson Reservoir.
This problem has attracted national attention. The pres-
ence of selenium and other potentially toxic elements in
the drainage water has caused concern about the possi-
ble threat to the quality of the underground water and has
raised questions about imposing stringent regulations on
drainage water disposal.
REGULATING DRAINAGE WATER
The Central Valley Regional Water Quality Control Board
is concerned over the increasing problem of regulating
saline agricultural drainage water and has an ongoing
planning program consisting of three items: (1) coopera-
tive monitoring and sampling, (2) basin planning, and (3)
regulation by permits or waivers. Monitoring is necessary
because little is known specifically about the quality and
quantity of agricultural return flows through the San
Joaquin River Basin. Basin planning is needed to estab-
lish receiving water objectives for surface and ground wa-
ter, and to establish goals and a basis for the regulation
component. Regulation is through waste discharge re-
quirements or possible conditional waivers, such as
"guidelines" or "performance criteria" as used in some
other water quality control programs in the Region.
However, because of the selenium discovery and the
problem with water fowl hatchlings, wetlands owner Rob-
ert James Claus appeared on April 27, 1984, before the
Central Valley Regional Board protesting the discharge of
drainage water into Valley wetlands and Kesterson Reser-
voir. The Regional Board denied his request to require
277
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
immediate discontinuance of the discharges and Mr.
Glaus filed a petition for review with the State Board. The
State Board agreed to review his petition and after three
evidentiary hearings, it adopted a Cleanup and Abate-
ment Order for Kesterson Reservoir on Febuary 5,1985.
The Order required that Kesterson Reservoir be
cleaned up or closed within 3 years. The U.S. Bureau of
Reclamation, the owner and operator of the Reservoir,
was given 5 months to develop a cleanup plan. The State
Board also declared the drain water to be a hazardous
waste because of the way it is being managed in Kester-
son Reservoir and because of the selenium concentra-
tions found in waterfowl. This caused considerable con-
cern because of a California law requiring the
construction of expensive double-lined ponds to control
hazardous liquid waste. However, the Board left some lee-
way to deal with drain water as a nonhazardous waste in
another location.
The Bureau did not wait long to change its course of
action. After contending throughout the State Board hear-
ings that the discharge of saline subsurface drainage wa-
ter, with 300 to 500 /tg/L selenium, was neither a health
threat nor a threat to ground water, the Secretary of Inte-
rior announced on March 15, 1985, before a Congressio-
nal hearing on drainage that he had ordered the Bureau to
immediately close Kesterson Reservoir and stop deliver-
ing irrigation water to the 17,000 ha (42,000 acres) in Wes-
tlands Water District, where the drainage water originates.
The Secretary stated that the action was necessary to
avoid the prosecution of Bureau of Reclamation employ-
ees for violating the Migratory Bird Treaty Act through in-
advertent poisoning of migratory birds in Kesterson Res-
ervoir.
This announcement caused a stir throughout the Na-
tion. It also caused the farming community and all regula-
tory agencies, including the U.S. EPA, to review policies
regarding nonpoint source pollution.
To maintain the flow of irrigation water to the 17,000 ha
(42,000 acres) of drained land, Westlands Water District
has agreed to assume the responsibility of managing this
drainage water within its boundaries and to stop sending
any water to Kesterson Reservoir by June 30, 1986. The
District is working closely with the Central Valley Regional
Board to obtain the necessary environmental clearances
and discharge permits that will allow it to either discharge
the drainage water to new District-constructed, wildlife-
safe, evaporation ponds or to recycle the drainage water
within the District. The other option is to destroy the drain-
age collector system and discontinue drainage service to
the land. Obviously, the problems regarding nonpoint
source pollution in California and, particularly, in the San
Joaquin Valley are not yet resolved.
REFERENCES
California Department of Water Resources. 1965. San Joaquin
Master Drain. Bulletin No. 127.
California State Senate Fact-Finding Committee. 1965. The San
Joaquin Valley Drain. Report.
Johns, G. E. 1984. San Joaquin Valley Drainage Water Quality
and Disposal. Proc. Specialty Conf. Am. Soc. Civil Eng., Irrig.,
and Drain. Div., Flagstaff, Ariz.
Nelson, D.G. and W.R. Johnston. 1984. San Joaquin Valley
Drainage—Development and Impact. Proc. Specialty Conf.,
Am. Soc. Civil Eng., Irrig. and Drain. Div., Flagstaff, Ariz.
San Joaquin Valley Interagency Drainage Program. 1979. Agri-
cultural Drainage and Salt Management in the San Joaquin
Valley. Final Report. U.S. Bureau of Reclamation, California
Dep. Water Resour. and Calif. State Water Resour. Control
Board.
U.S. Bureau of Reclamation. 1964. Alternative Solutions for
Drainage. Report and Addendum. San Luis Unit. CVP.
U.S. Congress. 1960. San Luis Unit Authorization Act. PL 86-
188, 74Stat. 156.
278
-------
Urban Issues:
Runoff
BELLEVUE EXPERIENCES WITH URBAN RUNOFF QUALITY
CONTROL STRATEGIES
PAM BISSONNETTE
City of Bellevue Storm and Surface Water Utility
Bellevue, Washington
The Bellevue Storm and Surface Water (SSW) Utility was
formed out of the city's and citizen's commitment to pre-
serve its network of streams and lakes. Established in
1974, the SSW Utility's mission is to manage the storm
and surface water system in Bellevue to maintain a hydro-
logic balance, prevent property damage, and protect wa-
ter quality for the health, safety, and enjoyment of citizens
and for the preservation and enhancement of wildlife habi-
tat. The basic concept underlying all policies and pro-
grams concerning storm water management in Bellevue
is to use the natural surface water drainage system to
convey and dispose of stormwater runoff. The SSW Utility
system consists of an integrated network of pipes and
stream channels that form the conveyance system; lakes,
wetlands, ponds, and detention basins serve as storage
facilities for flow equalization and water quality control.
The SSW Utility has five major programs: administra-
tion, development regulation, maintenance and opera-
tions, water quality control, and capital improvement. The
1985 operating budget is $5.6 million. The 1980-1985
capital improvement budget is $13 million.
A utility service charge similar to the water or sewer
utility bill provides the major source of SSW Utility reve-
nue. The rate structure currently is based on contribution
of runoff to the drainage system. For a given storm event,
a relationship exists between the amount of runoff from a
property and land area, particularly impervious land area.
Since runoff cannot practically be directly measured for all
properties, this relationship is used to define the rate
structure. All properties, including undeveloped and pub-
licly owned properties, participate financially under this
structure because they contribute runoff and, conse-
quently, benefit from the SSW Utility programs.
Runoff coefficients are the basis of the five rate classifi-
cations: undeveloped (up to 0.25), light (0.25-0.4), moder-
ate (0.4-0.5), heavy (0.5-0.75) and very heavy (0.75-1.0).
A commensurate rate reduction is granted for properties
practicing methods of runoff control. During the SSW Utili-
ty's initial 5 years the focus was almost exclusively on
runoff volume and velocity, and erosion control. I believe
that in the future water quality issues will overshadow all
other urban runoff problems.
Bellevue was one of the 27 cities nationwide to partici-
pate in the National Urban Runoff Program (NURP). Dur-
ing a 6-yr period, extensive monitoring of urban runoff, its
sources of pollutant contamination, and its effects on re-
ceiving waters were undertaken. Over 200 storm events
and atmospheric and impervious surface pollutant con-
centrations were monitored. These results were compared
to the chemical, physical, and biological health of an ur-
ban receiving stream. Various management practices in-
cluding street sweeping, drainage system maintenance,
and detention were applied and monitored to evaluate
whether these practices had any beneficial water quality
effect.
The results of the Bellevue NURP reported in Bellevue
Urban Runoff Program: Summary Report (Pitt and Bis-
sonnette, 1984) cannot be adequately paraphrased here.
In summary, we found that the beneficial uses of streams
that Bellevue seeks to preserve are seriously impaired by
urban runoff. The nature of the impairment is largely due
to the physical effects of uncontrolled runoff: flooding, ero-
sion, and sedimentation. These effects were great enough
to mask most other effects. Were these macroeffects con-
trolled, however, other pollutant effects would become evi-
dent, particularly those associated with metals and or-
ganic toxicants.
We found that management practices targeting the
street surface and structural drainage system cannot pro-
vide a complete solution. An effective control strategy in
Bellevue must look beyond the typical public works and
utilities design and operations and maintenance practices
to source controls, in-stream controls, and even treat-
ment.
To effect the necessary change in direction, we
changed the City's land-use policies. We have already
279
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
changed many programs in the Utility. In the past year we
have adopted operations and maintenance standards for
the cleaning of street surfaces, parking lot surfaces, and
drainage system components for both public and private
properties. We initiated a private system maintenance in-
spection program to ensure that these standards are met
on all properties. Our drainage design standards are to be
updated this year to provide better water quality treatment
via new facilities. An improved emergency response pro-
gram for accidental spills and illegal dumping of wastes
into the storm system has been initiated, including investi-
gative analysis, and cleanup manpower and equipment to
be on call 24 hours a day, 7 days a week. This program is
coordinated with the Washington State Department of
Ecology response program.
Water quality control costs have been segregated for
accounting purposes to allow for future apportionment to
Utility customers through the rate structure to provide in-
centives for pollutant source control programs. The Utility
will continue its data collection on impervious surface pol-
lutants and runoff characteristics from a complete array of
land use activities, funding permitting. We will research
establishing rate-user classifications based on the pollu-
tant load carried by runoff from various land uses. Contin-
ued monitoring will assess whether these management
practices in fact achieve the Utility's goal to protect
stormwater runoff quality for public health and safety con-
cerns and beneficial uses.
Last September, the final Federal NPDES permit regu-
lations were published (40/CFR, part 122), stating that
existing stormwater dischargers (as later defined in the
section) who did not have an effective permit shall submit
an application by March 26, 1985; the deadline later ex-
tended through the end of 1985. Under the regulations
such a permit will be a General NPDES Permit and the
major sections will likely include
1. water quality standards,
2. effluent limitations,
3. monitoring requirements,
4. performance standards,
5. reporting requirements,
6. enforcement, and
7. public participation.
How these general provisions apply to urban runoff will
be unique. For example, the concentrations of pollutants
in runoff vary widely depending on the size of the storm.
While large storms are typically of concern for flooding
purposes, it is the small, frequent, "dirty" storms that
chronically impact water quality. A monitoring program will
be designed for storm events in addition to background
levels. Effluent standards need to be developed for
stormwater. Performance standards will likely be based
on design, operation, and maintenance standards with ap-
propriate inspection programs at the start to ensure com-
pliance.
In the State of Washington, urban runoff has become
viewed as at least as serious a source of pollution to sur-
face and ground water as municipal wastewater effluent.
Urban runoff pollutant source control programs and runoff
treatment are being included as integral components of
the overall strategy to improve and protect the quality of
the State's sole-source aquifer and Puget Sound. The lack
of technical data evaluating the effectiveness of control
practices and treatment facilities hampers these efforts. It
is difficult to develop political courage to allocate enough
resources to deal with problems of this magnitude, partic-
ularly for local governments. If even a small fraction of the
funds that have been spent on the Construction Grants
program for municipal wastewater treatment had been
dedicated to the development and evaluation of runoff
treatment, perhaps we would have the kind of answers we
need today, such as removal efficiencies, specific load
reductions, and sizing data for individual control practices
or treatment facilities. It is essential that the state-of-the-
art for runoff quality control and treatment programs pro-
gress to the point that we have these answers so that
runoff pollution abatement strategies can be followed with
confidence.
REFERENCES
Pitt, Ft., and P. Bissonnette. 1984. Bellevue Urban Runoff Pro-
gram Summary Report. Special pub. for the Environ. Prot.
Agency, Nat. Urban Runoff Program. City of Bellevue, WA.
280
-------
THE EFFECTS OF CARBONATE GEOLOGY ON URBAN RUNOFF
ROGER P. BETSON
JACK D. MILLIGAN
Tennessee Valley Authority
Norris, Tennessee
ABSTRACT
Nonpoint source pollution is a function of runoff. In ar-
eas underlain by soluble carbonate rock (Karst) much of
the potential streamflow drains into the solution cavity
drainage system. When these areas are urbanized, the
increase in storm runoff can be dramatic as compared
with preurban conditions. Nevertheless, the runoff yield
and consequently the nonpoint source pollution loadings
may be less than from areas underlain by less soluble
rocks. However, the potential for contaminating ground
water is increased and the ground/surface water quality
relationships can become very complex. The hydrology
of six watersheds located in Knoxville, Tennessee, which
were studied as part of the EPA NURP program are de-
scribed in terms of their rainfall-runoff relationships and
the implications to ground-water recharge and quality.
These data along with data from an earlier study illustrate
the effect of karst urban hydrology.
BACKGROUND
In 1978, the Environmental Protection Agency (EPA) es-
tablished a National Urban Runoff Program (NURP) to
assess the significance of urban runoff as a contributor to
receiving water quality problems. Twenty-eight urban ar-
eas across the Nation were selected to provide a broad
range of climatological, geological, and social conditions.
Knoxville, TN, was selected for one of the NURP studies
because of its unique location in the Valley and Ridge
physiographic province, characterized by extensive car-
bonate rock geology, and because an earlier study con-
ducted in Knoxville indicated that the hydrology and water
quality loading of the urban streams were significantly af-
fected by the geology.
Table 1 is adapted from a report (Betson, 1976) describ-
ing the earlier Knoxville study conducted during the 3 wa-
ter years 1972-74. The table shows the yield of runoff to be
highly variable, related to the extent of underlying soluble
carbonate rock and the percent of impervious area. The
runoff yield from the Third Creek watershed can be con-
sidered indicative (for this unusually wet period) of that to
be expected from a watershed not underlain by soluble
carbonate rock. Even at the commercialized Fourth Creek
watershed, the yield of runoff was only half that at Third
Creek. During a year with normal rainfall in Knoxville
(116cm) the runoff yield from all of these watersheds
would be less. For example, during water year 1972 when
rainfall at Third Creek totaled 128 cm, the runoff yield was
0.42, which corresponds well with the long-term annual
runoff yield for the entire Tennessee River above Knoxville
(0.41) where long-term mean rainfall totaled 121 cm.
The data from this earlier study will be used to comple-
ment the data from the NURP study described subse-
quently. This paper will focus only on the hydrology of
these urban watersheds. A companion paper (Milligan
and Betson, this vol.) will discuss the water quality as-
pects. A report by Milligan et al. (1984) documented the
NURP study in detail.
THE STUDY WATERSHED
DESCRIPTIONS, DATA, AND ANALYSES
Figure 1 shows the general location of the six watersheds
instrumented as part of the NURP investigation (along
with the four watersheds included in the earlier Knoxville
study). The six watersheds in the NURP study were lo-
cated in the drainage basin of the Second Creek water-
shed which empties into the Tennessee River. Character-
istics of the six watersheds and hydrologic summaries are
shown in Table 2.
The Second Creek watershed is a series of folded and
faulted rocks that resulted from compressive stresses of
the Appalachian Mountain uplift. Ground water occurs in
fractures, joints, and along bedding plains. The fractures
in the carbonate formations (limestone, dolomite, and
calcareous shale) become enlarged by circulating ground
water that creates solution cavities. In some of the carbon-
ate formations underlying the study areas solution cavities
have developed extensively, resulting in surface manifes-
tations typical of karst areas, i.e., sinkholes, blind drains,
and springs.
The watersheds included in the study were selected to
provide a range of urban conditions. Land use in Residen-
tial 1 (R1) watershed is typical medium density single fam-
ily dwellings. Most of the watershed (92 percent) is
drained by separate storm sewers. The Residential 2 (R2)
subbasin has low density single family dwellings with
storm drainage through roadside ditches to a main chan-
nel (no storm sewers). The Strip Commercial (SC) subba-
sin is a long narrow catchment with storm drainage essen-
tially from business and adjoining homes to a separate
Table 1.—Earlier Knoxville study watersheds (after Betson, 1976).
Watershed
Drainage area (km2)
Percent impervious
Percent underlain by soluble carbonate rock
Mean rainfall, WY 1972-74 (cm)
Mean runoff, WY 1972-74 (cm)
Yield (runoff/rainfall)
Plantation
Hills
(PH)
0.62
23
100
153
9.8
0.064
First
Creek
(1Cr)
1.30
16
50
156
16.4
0.105
Third
Creek
(3Cr)
4.14
28
0
150
84.4
0.563
Fourth
Creek
(4Cr)
2.12
45
100
155
41 .51
0.268
'Adjusted for missing data.
281
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Figure 1.—Knoxville, Tennessee, study watershed location
map.
storm sewer draining the area. The Central Business Dis-
trict (CBD) catchment is a typical high density downtown
metropolitan area. Stormwater drainage is through a sep-
arate storm sewer system.
Two of the study watersheds, Upper Sink (US) and
Lower Sink (LS), shown in Table 2 were instrumented to
provide continuous records in an area underlain by very
soluble carbonate rock. The US and LS sites were located
upstream and downstream, respectively, from a short
reach of stream filled with coarse gravel typical of many
streams in this area. Continuous streamflow was meas-
ured at each site beginning in December 1981 through
September 1982. Land use in both watersheds is low den-
sity single family residential; they shared a common rain
gauge.
The remaining four watersheds shown in Table 2 were
each instrumented with a rain gauge and a Patmer-
Bowlus flume with a recorder. Only storm events that were
sampled became part of the data base. Table 3 summa-
rizes the storm event runoff data collected at the four
watersheds used in subsequent analyses.
A regionalized hydrologic model, TVA-HYSIM (Betson et
al., 1980), that can simulate continuous daily streamflow
and storm event data was used to help interpret the proj-
ect data. This model has provisions for internally predict-
ing model parameters (using regionalized prediction
equations) from readily measured or available watershed
and climatological information. Therefore, the model pa-
rameters were not adjusted based upon the streamflow
data. The model has provisions for accounting for the re-
duced storm flow and base flow that occurs in areas un-
derlain by soluble carbonate rock. The capability of the
model to simulate streamflow at the project watershed
was validated at all six watersheds. Table 3 summarizes
the storm event runoff simulations obtained using the
model. The model was used to simulate the annual runoff
information for these four watersheds that was shown in
Table 2. The model was also validated by comparing the
simulated and observed monthly runoff values at the US
and LS sites. It explained 86 and 87 percent of the vari-
ance in monthly runoff at the US and LS sites, respec-
tively.
Table 4 shows flow allocations as predicted by using the
model with observed rainfall as input: volumes of flow that
occurred annually from the impervious and pervious por-
tions of each watershed and the portion of potential
streamflow the model allocated to the underground drain-
age system, i.e., losses that bypassed the gauging sta-
tion.
DISCUSSION
The three mechanisms of storm runoff in humid rural wa-
tersheds were described by Dunne (1983): (1) Morton
overland flow that occurs when rainfall intensity exceeds
the infiltration capacity of the soil; (2) subsurface flow that
contributes significantly where soil conductivity is high be-
Table 2.—Summary of information from six NURP study watersheds.
Watershed
Drainage area (km2)
Percent impervious
Percent underlain by soluble carbonate
rock
Mean rainfall (cm)
Mean runoff (cm)
Yield (RO/RF)
Upper
sink
(US)
0.67
14.3
100
122
6.4
0.052
Lower
sink
(LS)
0.98
14.6
100
122
5.8
0.048
Residential
Sitel
(R1)
0.28
31.7
75
129
9.71
0.075
Site 2
(R2)
0.36
13.4
90
121
7.91
0.065
Strip
commercial
(SC)
0.75
42.8
100
120
22.61
0.188
Central
business
district
(CBD)
0.044
99
0
113
65.31
0.578
'Simulated using model TVA-HYSIM (Betson et al., 1980).
Table 3.—Summary of storm event hydrologic data.
Number of storms
Avg. storm rainfall (cm)
Avg. observed
storm runoff (cm)
storm yield (RO/RF)
Avg. simulated
storm runoff (cm)
correlation coeff.
nesiut
Sitel
(R1)
14
2.26
0.152
0.067
0.157
0.60
muai
Site 2
(R2)
13
2.26
0.145
0.064
0.147
0.92
Strip
commercial
(SC)
16
1.75
0.305
0.174
0.305
0.90
Central
bus. dist.
(CBD)
17
1.60
1.42
0.89
1.42
0.96
282
-------
URBAN ISSUES: RUNOFF
cause of coarse texture or macropores; and (3) saturation
overland flow that develops where the soil becomes satu-
rated by either the perennial ground water table rising to
the surface or by limited lateral or vertical percolation
above an impeding horizon.
As Pilgrim (1983) observed, in humid areas the Horton-
type runoff occurs only during relatively infrequent large
high intensity storms. Therefore, storm runoff primarily
results from saturated overland flow in valley bottoms and
subsurface flow (i.e., the partial area runoff phenomenon).
In areas underlain by soluble carbonate rock, the solu-
tion cavities can provide such efficient drainage that, pro-
vided there are no impeding soil horizons and the topog-
raphy is not too steep, neither subsurface flow nor
saturation overland flow will occur under typical storm
conditions. Gregory and Walling (1973) use the term dry
valley to describe this condition to embrace a wide variety
of geomorphic types where there is no evidence of flowing
water. Storm runoff will occur in these dry valleys during
unusual conditions such as heavy rainfall that exceeds the
infiltration capacity of the soils or the drainage capacity of
the rock solution-channel system, when the ground is fro-
zen, or when the soil surface is disturbed.
Urbanization will increase the yield of runoff from areas
underlain by carbonate rock as a result of the runoff gen-
erated on impervious surfaces. The increase can be dra-
matic relative to preurban conditions, but until the degree
of urbanization becomes substantial, the annual runoff
yield generally will still remain below that experienced
from areas underlain by noncarbonate rock. This effect is
illustrated in Figure 2 which shows the annual runoff yields
versus the percent imperviousness for those watersheds
included in Tables 1 and 2. Considerable scatter exists
among the data because: (1) the extent of carbonate rock
and the associated solubilities vary among the water-
sheds; (2) the drainage conveyance systems vary from
storm sewers to roadside ditches and open channels; and
(3) the rainfall experienced during the period of record
varies (rainfall was much heavier during the earlier study
shown in Table 1, resulting in higher yields).
Nevertheless, these data are generally indicative of the
effect that carbonate rock can have on urban drainage.
The function shown on Figure 2 was arbitrarily selected to
depict these data (it explains 88 percent of the variance
with 3Cr excluded). If, as indicated in the introduction, the
normal runoff yield from watersheds in this area unaf-
fected by carbonate rock is in the order of 0.41, then the
function shown on Figure 2 indicates that the impervious
area in a watershed underlain by soluble carbonate rock
must approach some 80 percent before a comparable
yield occurs.
Urbanization reduces the volume of potential storm run-
off that infiltrates into the solution cavity drainage system
and is therefore lost in relation to a gauging station. The
impervious area runoff occurs as stormflow, markedly
changing the flow regime. Table 4 indicated that most of
the flow from the study watersheds occurred as impervi-
ous area runoff with, as Table 3 indicated, the yield of
storm runoff roughly proportional to the impervious area.
0.6
0.4
A|H)fQ«. Nofmol Tfrld
(TtntiMM* Rivtr at Knoxvillt)
Yiild=O.OOI6 (PI)1'26
40 SO
PERCENT IMPERVIOUS
100
Figure 2.—Annual runoff yield versus percent impervious-
ness for Knoxville study watersheds.
Therefore, in areas underlain by soluble carbonate rock,
urbanization both increases the yield of runoff and creates
a flashier stream.
The Upper and Lower Sink watersheds shown in Table 2
were instrumented to help quantify runoff dynamics within
a catchment. As the table shows, the runoff yield was
actually less at the downstream measuring station. This
indicates that either storm runoff originating in the upper
watershed was lost to permeable reaches in channel be-
tween the two gauging stations or the yield of storm runoff
was less in those portions of the watershed below the US
gauging station (or both).
The lower portions of this watershed are underlain by
two geologic formations most often associated with caves
in this area (Moore, 1973). Based upon the dip of the beds
of these formations (southeast toward a syncline) and the
plunge of the syncline (to the northeast), it is probable that
some of this watershed lies in the recharge area for a
major spring located in the adjoining First Creek water-
shed.
Although not documented in the project data, runoff
yields in large watersheds located in karst terrane tend to
approach that of similar watershed underlain by noncar-
bonate rock. This is because the bypass losses from the
smaller subareas that drain through the solution cavities
often reappear in springs and many times within the same
basin. To the extent that urbanization increases the yield
of stormflow runoff, recharge to the solution cavity system
will be reduced. Consequently, the downstream flow re-
gime is affected.
CONCLUSIONS
The results obtained from these studies in Knoxville are
indicative of the effects of urbanization in karst areas.
However, the magnitude of the changes elsewhere may
differ considerably because of geological and climatologi-
Table 4.—Simulated average runoff allocations for water years 1981-1982 (all values are centimeters).
Impervious Pervious area Total Bypass
Watershed
US
LS
R1
R2
SC
CBD
area runoff
3.6
3.1
6.9
6.4
22.6
65.3
runoff
2.5
2.5
2.8
1.5
0.0
0.0
runoff
6.1
5.6
9.7
7.9
22.6
65.3
losses1
24.9
25.1
37.6
33.3
19.6
3.3
'Potential streamftow that is lost to solution channel drainage and bypasses a gauging station.
283
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
cal differences. Even allowing for leeway in the validity of
the model simulations, a number of conclusions can be
drawn about the effects of urbanization in karst terrane.
1. The yield of runoff in smaller karst basins is typically
only a small fraction of that in watershed underlain by
noncarbonate rock. This results because infiltration rates
are typically high and subsurface drainage through frac-
tures/solution cavities is typically so efficient that seldom
is the soil saturated enough to lead to baseflow and satu-
rated overland flow during storms.
2. When urbanization occurs, the runoff from impervi-
ous surfaces will increase the yield of storm runoff and
reduce recharge to the underground drainage system.
3. Urbanization may have to become extensive be-
fore the yield of runoff in areas underlain by soluble car-
bonate rock approaches that in other areas.
4. Although the yield of runoff in the urban watershed in
karst may remain below that from a watershed not under-
lain by carbonate rock, the change in flow from preurban
conditions can be dramatic because of the increased fre-
quency and magnitude of storm runoff.
5. Since recharge into and flow through the subsurface
drainage system remains high (until impervious areas be-
come extensive) in an urban area, the opportunity is great
for contaminating ground water and springs.
6. Geologic conditions can lead to the transfer of solu-
tion cavity drainage from one basin to another, which, for
example, can greatly complicate defining recharge areas
to springs.
7. Often drainage through the fracture/solution channel
system in the upper subbasins of a watershed reappears
in springs downstream. Consequently, the yield of runoff
in larger catchments underlain by carbonate rock can ap-
proach that of other watersheds. Therefore, the primary
effect of urbanization on these larger basins is to increase
the frequency and magnitude of storm flow and decrease
baseflow.
REFERENCES
Betson, P.P. 1976. Urban hydrology—a systems study in Knox-
ville, TN. Water Systems Develop. Br, Tenn. Valley Author.,
Norris, TN.
Betson, R.P., J. Bales, and H.E. Pratt. 1980. Users'Guide to TVA-
HYSIM—A Hydrologic Program for Quantifying Land-Use
Change Effects. Tenn. Valley Author. EPA-600/7-80-048. U.S.
Environ. Prot. Agency, Washington, DC.
Dunne, T. 1983. Relation of field studies and modeling in the
prediction of storm runoff. J. Hydrology 65: 25-48.
Gregory, K.J.., and D.E. Walling. 1973. Drainage Basin Form
and Processes: A Geomorphological Approach. Halsted
Press (Div. John Wiley and Sons), New York, NY
Milligan, J.D., and R.P. Betson. This vol. The effects of carbon-
ate geology on urban runoff: Part 2—water quality aspects. In
Proc. Perspectives on Nonpoint Source Pollution, a natl.
conf., Kansas City, MO, May 19-22.
Milligan, J.D., I.E. Wallace, and R.P. Betson. 1984. The relation-
ship of urban runoff to land use and ground- water resources.
TVA/ONRED/AWQ-84-1. Tenn. Valley Author., Chattanooga,
TN.
Moore, H.L. III. 1973. Caves of Knox County, Tennessee. Pages
105-13 in Geology of Knox County, Tennessee. Tenn. Div.
Geol. Bull. 70.
Pilgrim, D.H. 1983. Some problems in transferring hydrological
relationships between small and large drainage basins and
between regions. J. Hydrology 65: 49-72.
284
-------
IMPLEMENTING AN URBAN NONPOINT SOURCE
CONTROL STRATEGY
DAVID F. LAKATOS
ALAN CAVACAS
Walter B. Satterthwaite Associates, Inc.
West Chester, Pennsylvania
ABSTRACT
The nonpoint "problem" is often big (for example, the
Chesapeake Bay pollution situation) while the specific
contributors are small ones (say, a single 50-acre devel-
opment in the contributing watershed). Therefore, direct
links (both technical and economic) are somewhat vague.
An implementation strategy developed to meet the needs
of the new statewide nonpoint source management pro-
gram in Maryland had to meet these guidelines: (1) practi-
cal to implement, (2) management procedures that would
make the least changes possible to existing institutional
system, (3) the overall control concept related to eco-
nomics and, where possible, economic benefits, and (4)
the strategy would share objectives, resources, and ben-
efits with ongoing institutional systems. The strategy de-
veloped coupled clearly defined steps with ongoing
stormwater management and flood control activities. The
rise of watershed-level stormwater management pro-
grams has allowed for the development of innovative non-
point source control programs that can be easily incorpo-
rated within the framework of the comprehensive
watershed management plan. The particular aspects of
such an incorporation of these two important areas of
interest will be presented as a case study.
INTRODUCTION
Many new nonpoint source control programs are being
established, at various levels of the regulatory hierarchy,
in response to growing concern about the significance of
"pollutants washed from the land." Much of this concern
comes from studies indicating that pollutant washoff from
storm events may be as bad as or worse than the effluent
of a primary sewerage treatment plant.
Despite these real concerns, implementing an effective
nonpoint source control program can be severely ham-
pered by at least two major dilemmas plaguing both old
and new programs: the inability of lay persons to see non-
point source control benefits, and the great difficulty defin-
ing and justifying economic benefits from a proposed non-
point source control program.
Nonpoint source control programs are difficult to justify
economically because the problem is big while the con-
tributors are small (for example, the Chesapeake Bay eu-
trophication compared to a single 50-acre development in
the contributing watershed). The direct link (both technical
and economic) is somewhat vague. However, we inher-
ently know that if we do something, even on a small basis,
the "little" benefits will eventually add up to improve the
larger situation.
Despite this inability to clearly define the direct link be-
tween individual small activities, such as land develop-
ment projects, and large-scale nonpoint source pollution
impacts, intense political pressure is being applied in
many areas to act on this problem. Regulatory agencies
are being called on to develop strategies and implementa-
tion programs for nonpoint source management. Various
Federal, State, and local regulatory agencies have re-
cently proposed or implemented many types of stormwa-
ter runoff programs directed toward quantifying, permit-
ting, and limiting urban nonpoint source pollution. These
measures include approaches from land use downzoning
to stormwater discharge permitting.
In the opinion of these authors, the nonpoint source
control problems facing regulatory agencies are very im-
portant, and the course of action selected will no doubt
have significant environmental and economic impacts.
The purpose of this paper is to outline a very practical as
well as economical approach to urban nonpoint source
impacts. The approach does not involve setting up a new
program directed toward nonpoint source control, but
rather shares objectives with existing stormwater and
flood control programs.
SHARED OBJECTIVES IN WATERSHED
MANAGEMENT
One approach for developing and implementing a non-
point source management program that can show clearly
defined and unique results in the short term is to share the
positive results from related ongoing watershed manage-
ment programs, coupling clearly defined nonpoint source
management objectives.
Given that nonpoint source pollution is essentially car-
ried by stormwater runoff, a seemingly logical (both tech-
nicaly and politically) approach is to combine nonpoint
source objectives with ongoing watershed/urban runoff
programs. This can potentially be implemented much
more easily than a new program specifically directed to-
ward nonpoint source control.
Joining a nonpoint source management strategy with a
stormwater management strategy is certainly not a new
concept; it has been the preferred approach for many ex-
isting nonpoint source management programs. However,
the specific elements of such a joint water resources man-
agement system have not been defined enough to be fully
evaluated and used for new and different projects.
The following sections describe a joint system that was
used in a recent project in the State of Maryland. The
objectives for the project were primarily to evaluate and
define a flood control system for the watershed. In addi-
tion, broader-based water resources objectives for man-
aging stormwater runoff impacts (quantitative and qualita-
tive) were required under new State of Maryland laws and
regulations.
The Watershed Management approach responded to
both the technical and institutional/political characteristics
of this particular study area. The project has demon-
strated that the comprehensive watershed management
approach can result in implementable, practical, and cost
effective management strategies for both quantity and
quality control.
285
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
SCALE IN MILS'
Figure 1.—Bald Hill Branch watershed.
NONPOINT SOURCE MANAGEMENT:
CASE STUDY PROJECT
Background
The Bald Hill Branch Watershed in Prince George's
County, Md., is 14.3 km2 (5.7 mi2) in size. The upper half
of the watershed is very urbanized, and significant flood-
ing exists along the main channel (Fig. 1). The lower, rela-
tively undeveloped portion of the watershed is now experi-
encing increased development pressures. Existing and
future land uses contribute sediment, nutrients, and other
nonpoint pollutants via stormwater runoff, as is typical of
the region.
The potential downstream impact areas of water quality
degradation are the Western Branch, Patuxent River, and
the Chesapeake Bay. Although good water resources
management of the Bald Hill Branch has relatively small
potential to reduce the impacts at these receiving waters,
the cumulative efforts of the Washington Suburban Sani-
tary Commission (WSSC) and others under the combined
water quantity and quality objectives of the State of Mary-
land Floodplain and Stormwater Management Programs
can be very significant in the future.
A technical work program, involving watershed model-
ing, field investigation, and the application of state-of-the-
art watershed management technology, developed a prac-
tical and cost effective watershed management plan for
this study area. The water resources management objec-
tives addressed by the plan were:
1. 2-, 10-, and 100-yr present and future condition flood
control;
2. Mitigation of stream erosion; and
3. Nonpoint source pollution washoff.
The results of the technical analysis, as well as the
overall management strategy, are presented in the follow-
ing section.
KEY FOR PERCENTAGE INCREASE IN 2 YEARS
ocaiQN
""•m ^/m I in
-. -:.V.r:.;!' 1 t'&JX-. ?"i'T'1
Figure 2.—Watershed map with 2-year storm percent in-
creases at subarea.
Technical Results
An important detrimental effect of land development in a
watershed is increased storm runoff, resulting in flooding,
stream erosion, and stream water quality degradation. In-
creased storm runoff results from covering pervious land
surfaces, such as forests and fields, with impervious cov-
ers (for example, rooftops and parking lots). The purpose
of stormwater management, therefore, is to minimize or
eliminate the adverse stormwater impacts of urban devel-
opment.
The Penn State Runoff Model (PSRM) was used to
study and quantify the stormwater impacts of future devel-
opment in the Bald Hill Branch Watershed for a variety of
storm frequencies. From these measurements, manage-
ment techniques can be prescribed for significantly reduc-
ing these potential storm runoff problems.
To illustrate the typical results of the PSRM modeling
steps, Figure 2 presents the percentage increase in peak
storm runoff rates for the various subareas selected for
the analysis. The modeling assessment identified loca-
286
-------
URBAN ISSUES: RUNOFF
tions in the watershed having significant increases in po-
tential storm runoff rates (Fig. 2), and potential impacts
from anticipated land development activities. This insight
into possible impacts related to increases in peak storm
runoff rates can serve as a key technical input factor into
the development of a shared benefit nonpoint source
management strategy for a watershed.
Identified Storm Runoff Impacts
As described earlier, the technical analysis was directed
toward evaluating potential storm runoff impacts in three
key categories: flooding, water quality, and stream ero-
sion. The modeling framework used for this study involved
two models:
The Penn State Runoff Model (PSRM): to simulate the
complex urban and rural land use conditions of the water-
shed and to model specific runoff volumes and timing
characteristics throughout the basin;
The U.S. Army Corps of Engineers HEC-II: to calcu-
late the hydraulic conditions predicted to occur at the peak
level of the identified watershed flood flows.
The results of the technical analyses, and how they
were of significant value in identifying potential storm run-
off impacts, are presented in the following three subsec-
tions.
Potential Flooding Impacts. The first and most obvious
impact of urban development is the very significant in-
crease in peak runoff rates. The values for increases in
peak subarea runoff rates are illustrated in Figure 2.
These peak runoff rate increases relate to greater inci-
dences of urban flooding and damage.
The flooding damages resulting from "frequent" flood-
ing (for example, a 2-yr flooding event) can accumulate
over a period of time to significantly exceed the amount of
damage that could occur from one "catastrophic" flood
event (for instance, the 100-yr flood). Many urban water
resources managers, therefore, are very interested in
stormwater management programs that address these
frequent flooding events.
Another critical factor of urban watershed flooding,
which can only be fully evaluated by the PSRM watershed
model, involves analyzing specific flows and peak timing
from individual upstream subareas. This phenomenon of
upstream flow contributions to downstream storm flooding
impacts is one of the most important cause-and-effect re-
lationships in a watershed. That is, if the possible cause
for a downstream problem can be identified, the solution is
much easier and less costly to develop. Therefore, this
phenomenon of timing of subarea flows is a central ele-
ment in the development of a truly responsive flood con-
trol strategy for a watershed.
Potential Water Quality Impacts. This watershed is an
important water resource for the region; efforts are under-
way to protect the quality of surface waters, particularly in
changing environments such as an urbanizing watershed.
As a part of this protection effort, the potential for adverse
water quality impacts from existing and projected future
stormwater runoff was evaluated.
The storm runoff quality is highly dependent upon the
land use in the area on which rain falls. Storm runoff qual-
ity characteristics have been identified from analyses of
many water quality sampling programs. The storm runoff
characteristics from urbanizing areas have, however,
been somewhat difficult to identify or predict. Monitoring
results have shown that the quality of storm runoff from
impervious areas (rooftops and pavement) and industrial/
commercial areas is generally worse than that from pervi-
ous areas such as forests and meadows.
Published information confirms that the amount of im-
pervious area in a watershed (and certainly the change in
the amount of impervious cover expected from future land
development) can indicate storm runoff pollution potential
for the area. Therefore, in identifying a nonpoint source
management strategy for an area, it may be more reason-
able to utilize the information and interpretations that have
already been developed than to rigorously sample and
monitor the watershed for new, actual storm pollution val-
ues, which are extremely difficult to gather, interpret and
quantify. The impervious cover of an area might well be
the key watershed parameter on which to concentrate our
interest.
Information that was developed for this particular water-
shed study identified that the amount of future impervious-
ness increases (the potential water pollution contributors)
in the watershed may possibly be attributed to the same
areas that will be producing increases in peak storm run-
off rates (Fig. 2). That is, the areas where impervious
cover will increase are also those areas where peak storm
runoff rates will increase significantly.
This important observation, though not startling or new,
clearly identified a basic technical consideration for the
development of a stormwater management strategy for
this watershed: the proper control of stormwater pollution
might well be achieved through the control of peak storm
runoff rates and runoff volumes from the new developing
areas.
Potential Stream Erosion Impacts. The third area of con-
cern for this stormwater management analysis was the
potential acceleration of stream erosion from urbaniza-
tion. Potential stream erosion impacts are indicated from
the results of the PSRM/HEC-II modeling for this study.
For example, peak storm runoff rates for some subareas
in the watershed were projected to increase by 300 per-
cent for the 2-yr storm (Fig. 2) under proposed land devel-
opment conditions. These significant increases, in what is
an approximate, bankfull flow condition, illustrate the very
real potential for severe stream channel erosion problems
in the watershed.
The potential stream erosion impacts from the signifi-
cant increases in future peak storm runoff rates comprise
the third key, interrelated element included in a compre-
hensive runoff control strategy for this watershed. The
control of storm runoff peak rates appears to have an
important related benefit in possibly reducing the inci-
dences of harmful stream bank erosion damage in this
watershed.
Stormwater Management Analysis
The results of the Bald Hill Branch study identify the tech-
nical basis for controlling storm runoff from projected or
new land developments, as well as in the existing urban
sections of this watershed. To accomplish this, a wa-
tershedwide, multiobjective stormwater management
analysis was performed.
The stormwater management analysis approach was
directed toward identifying a management strategy, pri-
marily using regional stormwater facilities for the future
flooding and water quality impacts of the 2- and 10-year
events in this watershed. The 100-yr event was a very
important consideration, in that an effective stormwater
management strategy must be coordinated with the flood
control strategy for an area.
To develop a comprehensive watershed-level strategy
for controlling the adverse stormwater impacts within the
watershed, a technical work program was performed in-
volving the following steps:
1. Assessment of impacts and needs by field investiga-
tion and through watershed modeling;
2. Preliminary selection of regional stormwater man-
agement sites;
287
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
3. Assessment of the physical and planning param-
eters of the preliminary sites;
4. Field investigation of the preliminary sites;
5. Selection of the best potential sites;
6. Preliminary design of regional stormwater manage-
ment facilities; and
7. Final watershed modeling to ensure that facilities
meet design objectives.
Final Site Selection and Evaluation
The field investigation and engineering assessment re-
sulted in the selection of two regional sites. In addition, an
existing marsh area was considered as a part of the poten-
tial watershedwide stormwater management system be-
cause of the following potential benefits:
1. Significant water quality improvements can be de-
rived from a natural marsh area's potentially large storage
volumes (therefore high trap efficiencies) and wetland
vegetation for the removal of nutrients and other pollu-
tants.
2. The cost of enhancing this area and improving the
flood control and water quality benefits could be relatively
low.
3. The facility can also provide stream erosion and
flood control benefits at downstream locations.
Benefits of the Proposed Stormwater
Management System
The proposed stormwater management strategy that was
developed for this project is of primary importance be-
cause of the unique emphasis on sharing project objec-
tives and benefits for both water quality and quantity man-
agement activities. The management activities for
reducing postdevelopment peak runoff rates to their pre-
development values have the following shared benefits:
1. Reduction in downstream flooding;
2. Reduction in the amount of nonpoint source pollu-
tants that would be discharged into the drainageways in
the watershed; and
3. Reduction in stream channel erosion.
SUMMARY AND CONCLUSIONS
This paper has outlined a comprehensive watershed man-
agement strategy as proposed for the Bald Hill Watershed
in Prince George's County, Md. The comprehensive ap-
proach was oriented to snare the project objectives of
flood control and water quantity control, with the additional
objectives of controlling nonpoint source pollution and
stream channel erosion. The inclusion of these additional
objectives in the watershed plan was done, in part, to
satisfy the new State stormwater regulations (Maryland,
Article 8, Subtitle 11 A). This shared benefit, multiobjective
approach, as applied to the Bald Hill Watershed, is be-
lieved to be a more implementable concept for meeting
nonpoint source control needs. Many single-objective
nonpoint source programs have been difficult to imple-
ment because the public would not support a program
that lacked clear benefits of improved water quality.
The project's technical framework did not detail water
quality analysis, which can be very expensive and can
involve a large lag time before implementation begins.
Instead, the authors used detailed water quality modeling
results combined with their knowledge of nonpoint source
pollution—causes and effects—to develop a practical
(based on available sites for stormwater management fa-
cilities and at locations that will be most effective on a
watershedwide basis) and cost efficient (i.e., utilizing natu-
ral drainage features within the watershed) comprehen-
sive stormwater management plan. The technical frame-
work outlined in this paper will allow the stormwater plan
to be implemented within a much shorter time to provide
more immediate water quantity and water quality benefits.
Overall, in considering the future direction of nonpoint
source pollution programs, the use of a comprehensive
nutrient management plan, sharing water quantity and
quality benefits, as proposed for the Bald Hill watershed,
can provide significant environmental and economic ad-
vantages. Moreover, the tangible benefits of flood and
stormwater controls will help make the implementation of
the total plan much more likely to succeed.
288
-------
URBAN STORMWATER QUALITY MANAGEMENT:
THE FLORIDA EXPERIENCE
ERIC H. LIVINGSTON
JOHN H. COX
Florida Department of Environmental Regulation
Tallahassee, Florida
ABSTRACT
Nonpoint sources are responsible for the majority of the
pollutant load entering Florida surface and ground wa-
ters. In order to protect and manage the state's invaluable
water resources, the Florida Department of Environmen-
tal Regulation developed a regulatory program for the
control of nonpoint sources, especially urban stormwater.
The Stormwater Rule was developed over a two year
period with extensive participation by the regulated and
environmental communities. The rule established a per-
formance standard for the treatment of stormwater which
is based on two properties of stormwater, annual storm
frequency distribution and the first flush of pollutants.
General permits offer encouragement to applicants to
use appropriate best management practices while dele-
gation to the state's regional water management districts
coordinates water quality and water quantity consider-
ations into one permit process.
INTRODUCTION
Section 208 of the Clean Water Act required States to
control nonpoint sources of pollution generated by agricul-
ture, forestry, mining, urban construction, and other activi-
ties. The Florida Department of Environmental Regulation
and 12 designated local agencies received millions of dol-
lars in section 208 grants to assess the extent of the
State's nonpoint pollution problem and to develop techni-
cal and administrative methods of treatment and control.
In Florida, most of the funds given the designated 208
agencies for nonpoint source assessment were used for
urban stormwater problems, as most of the designated
agencies were located in heavily populated areas. The
Department took responsibility for investigating agricul-
ture and forestry activities statewide, and for developing a
management strategy for controlling nonpoint sources of
pollution from those activities.
Stormwater runoff has been recognized for the past
decade as a cause of water quality degradation. In Flor-
ida, rapid urbanization, with its associated land clearing
and paving of pervious areas, has accelerated the prob-
lem recently. While some amount of runoff from rainfall is
a natural occurrence, the volume and rate of runoff and
the accompanying pollutant loads increase as the amount
of paved, impervious surfaces increases. Stormwater
flowing over roofs, streets, lawns, commercial sites, indus-
trial areas, and other permeable and impermeable sur-
faces transports many pollutants into surface and ground
waters. Rain washes sediments from bare soil, motor vehi-
cles' heavy metals, oils, and greases deposited on streets
and parking lots, nutrients from fertilized lawns and crops,
and coliform bacteria from animal wastes into receiving
waters.
Recognition of such problems, along with the availabil-
ity of Federal funds, led Florida to draft regulations to
control stormwater in the late 1970's. The first official
State regulation specifically addressing stormwater was
adopted in 1979 as part of Chapter 17-4, Florida Adminis-
trative Code. Chapter 17-4.248 was the first attempt to
regulate this source of pollution that, at the time, was not
very well understood. Under Chapter 17-4.248 the De-
partment based its decision to order a permit on a deter-
mination of the "insignificance" or "significance" of the
stormwater discharge. This determination seems reason-
able in concept; however, in practice, such a decision can
be as variable as the personalities involved. What may
appear insignificant to the owner of a shopping center
may actually be a significant pollutant load into an already
overloaded stream.
In adopting Chapter 17-4.248, the Department in-
tended that the rule would be revised when more detailed
information on nonpoint source management becomes
available. About 1 year after adoption, the Department
began reviewing the results of research being conducted
under the 208 program. Such research determined that
stormwater discharges were responsible for over half of
the pollution load entering Florida waters and, in many
watersheds, stormwater discharges accounted for all the
pollutant load. In addition, stormwater-associated pollu-
tion is responsible for
1. 80 to 95 percent of the heavy metals loading to Flor-
ida surface waters;
2. Virtually all of the sediment deposit in State waters;
3. 450 times the suspended solids going to receiving
waters and 9 times the load of BOD5 substances when
compared to loads from secondarily treated sewage efflu-
ent; and
4. Nutrient loads comparable to those in secondarily
treated sewage effluent discharges.
THE RULE MAKING PROCESS
Recognition of the water quality problems caused by non-
point sources, especially urban stormwater, led Florida to
begin revising regulations for their control in 1980. The
Department had several major goals in developing a regu-
latory program. First, the rule should be easily understood
and as unambiguous as possible. Second, the rule should
encourage applicants to use appropriate stormwater man-
agement practices by offering exemptions from permitting
or general permits. Third, the rule should establish a clear
performance standard of the level of treatment for which
to aim. And fourth, the rule should recognize that
stormwater management involves the coordination of wa-
ter quality and water quantity aspects. Therefore, the rule
should provide a mechanism for the Department to dele-
gate stormwater regulatory authority to water manage-
ment districts already regulating the flood prevention as-
pects of stormwater.
To begin with, the Department established a stormwater
task force with membership from all segments of the regu-
lated and environmental communities. The rule was devel-
oped after 2 years, more than 100 meetings between de-
partment staff and the regulatory interests, and the
dissemination of 29 official rule drafts for review and com-
ment. As might be expected, the process involved numer-
289
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
ous compromises in developing the final rule language,
which became effective on Feb. 1, 1982. The rule has
been reviewed regularly to ensure that it remains current
with the rapidly changing stormwater management state-
of-the-art and to correct any problems with its treatment
requirements or administration. The rule was revised in
1983. In more recent rule revisions the Department has
established performance standards for incorporating iso-
lated wetlands into comprehensive stormwater manage-
ment systems. In addition, these revisions make consist-
ent treatment requirements for point and nonpoint source
discharges and promote more innovative and comprehen-
sive stormwater management.
STORMWATER RULE STANDARDS
The overriding standards of the Stormwater Rule, Chapter
17-25, Florida Administrative Code, are the water quality
standards and appropriate regulations established in
other Department rules. Therefore, an applicant for a
stormwater discharge permit must provide reasonable as-
surance that stormwater discharges will not violate State
water quality standards. Because of the potential number
of discharge facilities and the difficulties of determining
the impact of any facility on a waterbody or the latter's
assimilative capacity, the Department decided to establish
a rule based on performance standards.
The performance standard is a technology-based efflu-
ent limitation against which an applicant can measure the
proposed treatment system. If an applicant can demon-
strate treatment equivalent to the description in the per-
formance standard systems, then the applicant should be
able to meet applicable water quality standards. The ac-
tual performance standard is the retention or detention
with filtration of the runoff from the first 2.54 cm (1 in.) of
rainfall or the first 1.27 cm (1/2 in.) of runoff for stormwater
discharge facilities that serve a drainage area of 40 ha
(100 acres) or less. Projects that meet this performance
standard and the rule's other design standards can re-
ceive a general permit allowing construction within 30
days after filing a notice. Over 90 percent of all stormwater
facilities are approved with a general permit thus helping
an applicant avoid possible delays associated with the
review of a full permit application.
TECHNICAL BASIS FOR THE
PERFORMANCE STANDARD
To provide reasonable assurance, the performance stand-
ard is built around two properties of stormwater: the an-
nual storm frequency distribution and the first flush of pol-
lutants. Based on long-term rainfall records, statistical
distribution curves have been established that describe
the variability in intensity and duration of individual storm
events. Table 1 shows that an average of nearly 90 per-
cent of a year's storm events occurring anywhere in Flor-
ida produce a total of 2.54 cm of rainfall or less (Anderson,
1980). Also 75 percent of the total annual volume of rain
will fall in storms of 2.54 cm or less.
The first flush of pollutants refers to the highest concen-
tration of stormwater runoff pollutants that characteristi-
cally occurs during the early part of the storm with concen-
trations decaying as the runoff continues (Fig. 1).
Concentration peaks and decay functions vary from site to
site depending on land use, the pollutants of interest and
the characteristics of the drainage basin.
The two properties of stormwater led to the selection of
the first 1.27 cm (1fe in.) of runoff as the performance
standard for sites less than 40 ha (100 acres) in size. Flor-
ida studies (Wanielista, 1979) indicated that for a variety of
land uses the first 1.27 cm of runoff, when projected to
annual loadings, contained 80-95 percent of the total an-
nual loading of most pollutants. However, first flush effects
generally diminish as the size of the drainage basin in-
creases and as the percent impervious area decreases
because of the unequal distribution of rainfall over the
watershed and the additive phasing of inflows from nu-
merous smaller drainages in the larger watershed. Florida
studies indicated that as the drainage area increases in
size above 40 ha the annual pollutant load contained in
the first 1.27 cm of runoff drops below 80 percent because
of the diminishing effects of the first flush in stormwater
from larger watersheds (Wanielista, 1979). As a conse-
quence, the performance standard for projects larger than
40 ha is the treatment of the runoff from the first 2.54 cm
of rainfall.
Stormwater treatment is generally either retention or
detention with filtration. By definition, retention requires
the diversion of the prescribed amount of stormwater to a
separate treatment area with no subsequent discharge of
the diverted water to surface receiving waters. Pollutants
are removed during retention principally by preventing dis-
charge of polluted water, using percolation through soil
and settling. Therefore, retention results in nearly total
treatment of the diverted water. Detention facilities are
typically on-line systems where all of the stormwater from
a site passes through the treatment pond and is subse-
quently discharged to surface waters. The Stormwater
Rule requires that discharges from detention ponds be
Table 1.—Cumulative probability values (%) for 15 Florida locations.
Probability (as °/o) of volumes
Location
Niceville
Tallahassee
Jacksonville
Appalachicola
Gainesville
Daytona
Inglis
Orlando
Tampa
Vero Beach
Clewiston
West Palm Beach
Fort Myers
Miami
Key West
Florida
0-1/2 in.
68.2
70.3
77.1
75.3
76.9
75.9
71.1
80.1
76.4
77.5
74.3
80.6
70.5
82.7
84.9
76.4
1/2-2 in.
84.5
83.7
91.7
87.9
90.0
89.3
85.1
90.0
89.7
89.9
87.3
90.8
86.4
93.3
94.0
89.0
1-2 in.
93.8
94.2
97.7
97.4
97.0
96.2
96.6
98.0
97.9
98.7
97.0
97.0
95.6
98.5
98.4
97.0
2-3 in.
97.7
98.0
99.1
99.3
98.9
98.7
99.2
99.6
99.5
99.3
98.9
98.7
98.4
99.4
99.3
99.0
3-4 in.
98.4
99.6
99.6
99.7
99.8
99.8
99.8
99.9
99.9
99.5
99.6
99.1
99.6
99.6
99.6
99.6
Source: Anderson, 1980
290
-------
URBAN ISSUES: RUNOFF
Amount
ol
PollutonM
FIRST
FLUSH
Pollutant
Concentration
Rat*
of
OitchorQa
Beginning of Storm
Figure 1.—The "first flush" effect.
passed through a suitable filter material, typically 60 cm of
natural soil or mixtures of sand, soil, and gravel, to remove
pollutants. This treatment removes suspended materials
and that fraction of the dissolved pollutants in stormwater
associated with particulate materials.
An applicant can also use other BMP's, alone or in com-
bination, that provide equivalent treatment. However,
these stormwater management systems must be permit-
ted by other than a general permit. To encourage greater
innovation, recent rule revisions establish a general per-
mit for stormwater management systems that incorporate
several percolation and retention mechanisms into a proj-
ect's landscaping. Other rule revisions allow incorporation
of certain isolated wetlands into the stormwater manage-
ment system to provide part of the treatment of pollutants.
ADMINISTRATION OF THE
STORMWATER RULE
Under the Florida Water Resources Act of 1972, the De-
partment of Environmental Regulation, a water quality
agency, served as the umbrella administering agency del-
egating authority to five regional water management dis-
tricts whose primary functions historically have been re-
lated to water quantity management. Therefore, a second
objective in developing the Stormwater Rule was to coor-
dinate the water quality considerations of the Depart-
ment's stormwater permits with the water quantity as-
pects of the Districts' surface water management permits.
Consequently, the Department has delegated stormwater
quality permitting to the South Florida Water Management
District and to the Southwest Florida Water Management
District. These two districts have the resources to adminis-
ter and have requested delegation of the program.
These two aspects of the stormwater program have al-
lowed the Department to administer it with an extremely
small work force. State revenues fund six engineers in our
district offices around the State to process stormwater
applications and review management plans. However,
they also have responsibility for other permitting programs
and for inspecting stormwater facilities for compliance. A
Section 2050) 9rant from EPA funds the six-person Non-
point Source Management Section located in the Depart-
ment's Tallahassee headquarters. This staff provides
technical assistance to the stormwater engineers and to
local governments in establishing or improving their exist-
ing nonpoint source control programs. The Section is also
responsible for nonpoint source water quality manage-
ment planning activities and public education programs,
coordinating the nonregulatory agriculture and silviculture
nonpoint source control programs, coordinating the on-
site wastewater treatment program, managing nonpoint
research programs, and developing revisions to the
Stormwater Rule.
ACKNOWLEDGEMENTS: The authors express great apprecia-
tion to Mimi Drew and Martin Wanielista for their review and
constructive comments on the draft manuscript. We also wish to
thank Alisa Gregory for typing the draft and final manuscripts.
REFERENCES
Anderson, D.E. 1980. Evaluation of swale design. M.S. Thesis.
College Eng. Univ. Central Florida, Orlando.
Wanielista, M.R 1979. Stormwater Management Quantity and
Quality. Ann Arbor Sci., Ann Arbor, Ml.
291
-------
Urban Issues:
Construction Nonpoint
Source Pollution
HAMPTON ROADS WATER QUALITY AGENCY NONPOINT
SOURCE PROGRAM
PAUL E. FISHER
Hampton Roads Water Quality Agency
Norfolk, Virginia
INTRODUCTION
The Hampton Roads Water Quality Agency has been the
"Section 208" areawide water quality management plan-
ning agency for the Greater Hampton Roads, Virginia,
area since 1974. The area includes 10 cities and 4 coun-
ties with a population of over 1.2 million. Land use ranges
from dense urban to extensive agricultural and silvicul-
tural activities.
Since 1979 the Hampton Roads Water Quality Agency
(HRWQA) has focused its efforts on urban and rural non-
point source control planning. We have remained one of
few successful and active "208" agencies long after the
Federal program emphasis was shifted to the States.
THE PROGRAM
The Nonpoint Source (NFS) Control Program for the
Hampton Roads Area includes the following seven com-
ponents:
Water Quality Problem Definition. As part of our plan-
ning process, an extensive water quality sampling and
modeling program was carried out. That program identi-
fied nonpoint source-related water quality problems in the
major drainage basins of the Hampton Roads area. Re-
lated studies have expanded upon and updated the prob-
lem definition. In addition, population and employment
forecasts for the region have been analyzed to determine
the categories of development most likely to contribute to
water quality problems in each drainage basin. The prob-
lem definition serves as the basis for the specific control
recommendations.
Management Agency Designations. Various institu-
tions have assumed leadership in implementing the NFS
Control Program. Each has specific areas of responsibility.
Construction/O&M/Technical Assistance (including
financing)
Urban All area cities, counties, and
towns.
Agriculture
Planning
Regional Coordination
Regulation
All area Soil and Water
Conservation Districts
Peninsula Planning District
Commission; Southeastern
Virginia Planning District
Commission
Hampton Roads Water Quality
Agency
Virginia State Water Control
Board; Virginia Soil and Water
Conservation Commission; U.S.
Environmental Protection
Agency.
Urban NFS Control Program. The Urban NFS Control
Program incorporates the implementation of erosion and
sediment control Ordinance requirements, and local ur-
ban government housekeeping-type activities and the ur-
ban program also includes an ongoing public education
and information program. This component of our nonpoint
source program also includes local government use of the
HRWQA-developed methodology, A Simplified Technique
for Developing Site Specific Non-point Source Control
Plans, in determining which practices to use on specific
sites.
293
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Agricultural NPS Control Program. This component
has four principal elements: (1) implementation agree-
ments with the area's four Soil and Water Conservation
Districts, (2) initiation and continuing involvement in the
$1.8 million Nansemond/Chuckatuck Rural Clean Water
Project, (3) cooperation and participation in other agricul-
tural nonpoint control programs, and (4) public education
and information programs.
Special NPS Control Program. The HRWQMP identi-
fied a significant number of potentially intensive sources
of urban and rural nonpoint pollution which require special
attention in the application of control techniques. Although
site-specific studies of all of these special nonpoint
sources have not been completed, programs to address
most of them are in place. They include
• Animal wastes—a no discharge certificate program
• Agricultural recycling of wastewater treatment plant
residual solids—a no discharge certificate program
• Landfills—State Health Department solid waste dis-
posal regulations
• Construction activities—erosion and sediment con-
trol program
• Septic tanks—State Health Department sewage han-
dling and disposal regulations
• Outside materials storage—best management prac-
tice handbooks
• Drydocks—studies underway
• Marinas—State and Federal regulations and studies.
Monitoring Program. To be effective, any planning and
. implementation effort requires an ongoing monitoring pro-
gram. The HRWQA continues its leadership role in moni-
toring nonpoint source pollution and the effectiveness of
best management practices. A number of special and rou-
tine water quality studies control implementation, and so-
cioeconomic trend monitoring efforts have been under-
taken. Comprehensive water quality management plans
have been developed for several urban watersheds.
Basin-specific NPS Control Program. The Basin-spe-
cific NPS Control Program describes recommended con-
trol activities which are generally applicable throughout
entire basins. It also includes assignments of implementa-
tion responsibility for each program activity The HRWQA
has concluded that individual control measures are best
determined on a site-specific basis. Thus, the Basin-spe-
cific Program does not include site-specific recommenda-
tions for controls, such as BMP implementation or house-
keeping activities. This program does, however, guide
site-specific controls implementation. The manual, Simpli-
fied Technique for Developing Site Specific Nonpoint
Source Control Plans, assists in the transition from basin
level guidance to site-specific implementation.
As a result of these programs, nonpoint source controls
are considered in the local-government-required erosion
and sediment control programs for construction activities.
The program is receiving the support of the Tidewater
Builders Association, and local site planning engineering
consulting firms are also incorporating nonpoint source
control consideration in their plans.
294
-------
PROBLEMS AND PROGRESS IN URBAN SOIL EROSION AND
SEDIMENTATION CONTROL: A BICOUNTY PERSPECTIVE
PETER G. THUM
Lake County Department of Planning, Zoning and Environmental Quality
Waukegan, Illinois
GERALD A. PAULSON
Operation Topsoil
McHenry County Defenders
Crystal Lake, Illinois
ABSTRACT
Controlling urban-related soil erosion and sedimentation
problems in northeastern Illinois requires involvement by
all water resource user groups in the decisionmaking and
enforcement process. The Fox River-Chain O' Lakes re-
gion in Lake and McHenry Counties, Illinois, suffers from
serious sedimentation problems caused partly by con-
struction site soil erosion within the watershed. Despite
having the authority to control such nonpoint pollution
problems, many municipalities lack truly comprehensive
ordinances or the means to effectively enforce such regu-
lations. Anticipated new development in the watershed
may worsen already existing sedimentation problems.
Both counties recognize their common interest in the as-
sets and problems provided by the Fox River-Chain O'
Lakes. Lake County has worked extensively in nonpoint
pollution since the 1970's. McHenry County, on the other
hand, has not yet experienced the development pressure
associated with close proximity to the Chicago metropoli-
tan area. Currently, McHenry County Defenders, a private
citizens advocate group, is spearheading a public educa-
tion effort to prevent further damage to the Fox River-
Chain O' Lakes by soil erosion and sedimentation from
construction sites. Cooperative work between Lake
County and the Defenders intends to address this non-
point pollution concern across the bicounty jurisdiction.
The Fox River, a major tributary of the Illinois River, has its
headwaters just northwest of Milwaukee, Wisconsin. From
its source, the river flows in a southwesterly direction into
Illinois towards its eventual confluence with the Illinois
River in Ottawa, Illinois. Approximately 2,257 km2 (868
mi2) (33 percent) of the drainage area exists in Wisconsin,
2,301 km2 (885 mi2) (34 percent) in northeastern Illinois,
and 2,202 km2 (847 mi2) (34 percent) south of Kane
County, also in Illinois (Northeast. III. Plann. Comm. 1981).
The region reported on involves the bicounty area of
Lake and McHenry Counties in northeast Illinois. The river
enters Illinois at the northwest corner of Lake County
where it passes through the Chain O' Lakes watershed.
This watershed is a series of natural lakes and wetlands
located approximately 80 km (50 mi) northwest of the Chi-
cago metropolitan area. Upon exiting the Chain O' Lakes,
the river flows southwest into McHenry County and tra-
verses a distance of about 21 km (13 mi) before meander-
ing back eastward into the southwest corner of Lake
County. At this point, the river again swings westward into
McHenry County to exit the bicounty area at the Village of
Algonquin. Figure 1 displays the location of the Fox River
basin within the bicounty study area (Northeast. III. Plann.
Comm. 1979).
Two dams exist on the Fox River within the study area,
hereafter named the Fox River-Chain O' Lakes water-
shed. The water level in the Chain O' Lakes is controlled
by a dam on the Fox River at the city of McHenry. A
second dam lying downstream at Algonquin forms a pool
that extends to the base of the McHenry dam. These
dams have transformed the river into a series of artificial
impoundments used extensively for recreation in the re-
gion.
Major tributaries to the Fox River-Chain O' Lakes wa-
tershed (and their respective drainage areas) include
Boon Creek (60.3 km2 (23.2 mi2)), Sequoit Creek
(35.9 km2 (13.8 mi2)), Spring Creek (67.6 km2 (26.0 mi2)),
Squaw Creek (123.8km2 (47.6 mi2)), and Nippersink
Creek (531 km2 (205 mi2)). These tributaries flow on the
east side of McHenry and west side of Lake County (Fle-
mal, 1983). Most of the watershed is still predominantly
rural, with a series of small towns and unincorporated
villages along the waterways. All or parts of 14 communi-
ties with 1980 populations of 3,000 or more are located
within the Fox River-Chain O' Lakes watershed (Table 1).
A large number of people also live in smaller villages or
unincorporated areas.
TOPOGRAPHY AND SOILS
Topographical characteristics of the watershed are a result
of both the past Wisconsin glacial period and the Fox
River fluvial characteristics. Till soils and morainal de-
posits make up the major characteristics of the area. The
terrain is highly variable north of the Chain O' Lakes, yet
displays fairly flat and rolling characteristics south to Al-
gonquin. The river channel is poorly defined and is con-
fined by low banks and wide floodplains (U.S. Army Corps
Eng. 1976). The river falls only about 2.7 m (9 ft) in this 53-
km (33-mi) reach (U.S. Army Corps Eng. 1976).
The soils along the Fox River-Chain O' Lakes water-
shed are developed primarily from glacial till and outwash,
loess, and alluvium. The soil associations are Marsh-Fox-
Boyer, Zurich-Grays-Wauconda, Morley-Markham-
Houghton, and Napponee-Montgomery (Paschke and Al-
exander, 1970). The Fox-Boyer and Wauconda soils
associated with steep slopes are very easily eroded. A
large portion of the soils on the uplands around the Chain
O' Lakes and along the Fox River falls into this highly
erodible category (Metcalf and Eddy, 1980).
WATER QUALITY AND RECREATION
Water quality in the Fox River varies from fair to good with
quality in the Chain O' Lakes being generally poor (III.
Environ. Prot. Agency, 1976). Problems of poor water
quality from nutrient loading, attributed in part to munici-
pal and industrial wastes, agricultural and urban runoff,
and seepage from uncontrolled septic fields, threaten the
quality of recreation afforded by the basin's resources.
Primary problems are high turbidity, nuisance algal
blooms, and low dissolved oxygen levels (CH2M HILL,
1982; Kothandaraman et al. 1977).
295
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Figure 1.—Fox River-Chain O' Lakes (Lake and McHenry
Counties).
Phosphorus and total suspended solids constitute two
of the greatest pollution problems (Flemal, 1983). Soil and
sediments eroded from cultivated croplands and construc-
tion sites and runoff from urban areas are the primary
nonpoint source pollutants (CH2M HILL, 1982; Flemal,
1983).
Despite poor water quality, the Fox River-Chain O'
Lakes is a major recreational area for thousands of people
living in northeastern Illinois. The river also is a major
source of drinking water for people who live along the Fox
River Valley. Approximately 8 million people use the area
for boating, fishing, sightseeing, swimming, and waterfowl
hunting. As many as 15,000 boats and 60,000 people can
be found in the Chain O' Lakes area on any given Sunday
during the summer (Metcalf and Eddy, 1980). Chain O'
Lakes State Park, one of the few publicly owned recrea-
tional sites in the area, attracted 1,200,000 visitors in 1984
(Illinois Dept. of Conservation, 1985).
The city of Elgin (population 63,798), located down-
stream of the Algonquin dam, withdraws water from the
Fox River for public consumption. Other communities are
considering tapping the river for domestic use (Flemal,
1983).
SEDIMENTATION
Sedimentation is a serious problem in the Chain O' Lakes
and behind the dams at McHenry and Algonquin. A recent
U.S. Geological Survey study concluded that 30,690 met-
ric tons (34,100 tons) of sediment are transported annually
to the Chain O' Lakes from Wisconsin and Illinois sources
(CH2M HILL, 1982). Sediments severely limit boating ac-
tivity, destroy wildlife habitat, and degrade water quality in
the river and lakes (Kothandaraman et al. 1977; Metcalf
and Eddy, 1980; Flemal, 1983; CH2M HILL, 1982). Three
primary sources of the sediments have been identified:
sediment transported by surface waters, that eroded from
shorelines, and organic materials generated within the
lakes and surrounding marshes.
Soil particles carried into streams from construction
sites and croplands are considered a major source of the
sediments (Kothandaraman et al. 1977; CH2M HILL,
1982; Stall and Bhowmik, 1974; Brabets, 1977; Metcalf
and Eddy, 1980). Though the exact proportion of sedi-
ments contributed by construction sites versus croplands
is not known, the relative differences in erosion rates
coupled with the area development suggest great contri-
bution by construction activities.
SOIL EROSION AND FUTURE
DEVELOPMENT
Studies have shown that the amount of soil erosion on
land being converted to urban purposes can be about 10
times greater than on land in cultivated row crops, 200
times greater than on pasture land and 2000 times greater
than on forested land (Northeast III. Soil Erosion and Sedi-
mentation Control Steer. Comm. 1981). Also, though ero-
sion from construction sites and urban areas constitutes a
much smaller percentage of the total annual sediment
yield than agriculture, its immediate impact upon urban
areas and streams in the watershed can be as damaging
because of the much higher sediment delivery and the
combined effects of other urban-related pollution on the
waters (III. Environ. Prot. Agency, 1982a).
Development in the Fox River-Chain O' Lakes water-
shed has historically been a result of the recreational ben-
efits provided in the area. In the earlier half of this century
Table 1.—Status of municipal soil erosion and sedimentation control ordinances for incorporated areas with populations
greater than 300 (1980 Census) in Fox River basin, Lake and McHenry Counties, Illinois.
Has an ordinance
Municipality
Algonquin
Antioch
Barrington'
Barrington Hills1
Gary
Crystal Lake
Fox Lake
Lake in the Hills
Lake Zurich
McHenry
Round Lake Beach
Round Lake Park
Wauconda
Woodstock1
Population
5,834
4,032
9,029
3,631
6,640
18,590
6,891
5,621
8,225
10,908
12,291
4,032
5,668
1 1 ,725
Separate,
comprehensive
X
X
X
Under
building
code
X
X
Under
subdivision
code Combination
X
X
X
X
X
No
ordinance
X
X
X
X
'On basin divide
(Sources: Lake County Plann. Dep., 1984; Walters, 1984).
296
-------
URBAN ISSUES: CONSTRUCTION NONPOINT SOURCE POLLUTION
the region consisted mainly of summer cottages and
homes used by individuals living in the Chicago metropoli-
tan area. However, during the last two decades, perma-
nent residences and businesses have come to the area as
a result of an expanding Chicago urban center and an
improved transportation network. The degradation of wa-
ter quality in the Fox River-Chain O' Lakes area has fol-
lowed this new development, partly from construction-re-
lated sedimentation problems.
In the future, the Fox River basin is expected to become
the next major growth area of the Chicago metropolitan
region, as white collar jobs continue to move out of the
city. Much of this new development will concentrate along
the Fox River, on soils with high erosion potential. Studies
done by the Soil Conservation Service show that approxi-
mately 1,330 and 590 ha (3,331 and 1,489 acres) of land
are annually under development in Lake County and
McHenry County, respectively. This equals $68,000 in sed-
iment-related costs (III. Environ. Prot. Agency, 1982b).
Though these figures represent county totals, most of
McHenry's development is in the Fox River Valley
(McHenry County Planning Dep. 1979).
EFFORTS TO CONTROL
CONSTRUCTION SITE EROSION AND
SEDIMENTATION
Efforts to control construction site erosion and sedimenta-
tion in northeastern Illinois span the past decade. In 1973,
the publication Standards and Specifications for Soil Ero-
sion and Sedimentation Control in Northeastern Illinois,
referred to as the "Blue Book," was compiled by the
Northeastern Illinois Natural Resource Service Center for
the Soil and Water Conservation Districts of Northeastern
Illinois. These districts included the Lake and McHenry
County Soil and Water Conservation Districts.
The purpose of the Blue Book was to help developers,
planners, engineers, and local governments effectively
control soil erosion and sedimentation from development
sites (Northeast Plann. Comm. 1973). Though the publica-
tion was comprehensive at the time, it was technical in
nature and designed more specifically for the developer
and engineer.
In March 1978 the Northeastern Illinois Soil Erosion and
Sedimentation Control Steering Committee was orga-
nized to revise the Blue Book. Needs in resource planning
called for updating and including a uniform vocabulary
and a model soil erosion and sedimentation control plan.
This new publication, Procedures and Standards for Ur-
ban Soil Erosion and Sedimentation Control in Illinois, is
referred to as the "Green Book" (Northeast Soil Erosion
and Sed. Steering Committee, 1981). This book was
widely accepted at the time and is still being extensively
promoted in northeastern Illinois today.
In addition to these resource guides, local soil erosion
and sedimentation control ordinances have been recom-
mended in the Areawide Water Quality Management Plan
for northeastern Illinois. Chapter 19 of the plan, for the Fox
River Basin, called for the six counties and all municipali-
ties in the basin to enact soil erosion and sedimentation
control ordinances to control runoff and sedimentation
from land disturbance activities (Northeast III. Plann.
Comm. 1979). Development of model ordinances by the
Northeast Illinois Planning Commission by July 1, 1979,
was designed to aid the local ordinance adoption process
(Northeast III. Plann. Comm. 1980).
Of the 14 larger municipalities in the Fox River-Chain
O' Lakes drainage basin, 10 have some type of ordinance
that addresses soil erosion and sedimentation from con-
struction sites (Table 1). All of the ordinances fall into one
of four categories: (1) a separate, comprehensive ordi-
nance, (2) part of a subdivision ordinance, (3) part of a
building code, or (4) a combination of the first three. Most
of the erosion controls are incorporated into subdivision
ordinances and many consist only of a descriptive para-
graph that allows the city engineer or building inspector to
require soil erosion controls in development plans. Unfor-
tunately, these types of ordinances often leave the require-
ment and need for a plan up to the discretion and knowl-
edge of the village engineer or inspector. A lack of
knowledge and understanding concerning the processes
involved and methods for control often determine the ordi-
nances' true strength (Zeiler, 1985).
A few communities, including Lake County, have
adopted separate comprehensive soil erosion and sedi-
mentation control ordinances. These regulatory ap-
proaches seem to be the most effective because of their
systematic analysis and mitigation of the problem (Koziel,
1985). Both McHenry and Lake County require erosion
controls in development plans, although McHenry Coun-
ty's ordinance is more permissive than comprehensive
(Layer, 1985).
TWO COUNTIES, ONE PROBLEM
In Lake County, construction site soil erosion and sedi-
mentation control has been aggressively promoted since
the early 1970's. With the publication of the Blue Book in
1973, the Lake County Soil and Water Conservation Dis-
trict and Soil Conservation Service have provided techni-
cal assistance to county and municipal governments inter-
ested in controlling erosion from construction sites (Koziel,
1985). Working closely with the County Building and Zon-
ing Department in 1976, a comprehensive soil erosion and
sedimentation control ordinance was passed by the
county which today represents a model example of effec-
tive enforcement (Zeiler, 1985).
Recently, the Soil and Water Conservation District
(SWCD/SCS) and the Lake County Planning Department
have cooperated to strongly promote ordinance adoption
and enforcement through the District's Memo of Under-
standing and the Planning Department's Comprehensive
Stormwater Management Program. As an intergovern-
mental agreement, the memo provides the municipality
with necessary technical expertise and the Soil and Water
Conservation District with an effective approach toward
construction site soil erosion and sedimentation control.
Table 2 identifies the responsibilities set down for both the
District and municipality in the Memo of Understanding.
McHenry County has paid little attention to urban-re-
lated soil erosion and sedimentation control. Since the
only portion of the county experiencing major develop-
ment is the Fox River-Chain O' Lakes, most efforts have
concentrated on agricultural soil erosion and sedimenta-
tion problems. As stated earlier, the county does require
erosion controls in development plans; however, the ordi-
nance is simply a more permissive paragraph incorpo-
rated in the subdivision ordinance, rather than a compre-
hensive identification and regulation of the problem.
Despite all the official recognition given to the problem
of soil erosion and sedimentation from land undergoing
development, most local governments appear ill-equipped
to cope with the new wave of construction activity. Al-
though most of the towns and villages along the river can
require erosion control in development plans, little aware-
ness of the problem or commitment to implement the ordi-
nances seems to exist.
The McHenry County Defenders, a local citizen-based
organization, has identified erosion and sedimentation
from construction sites as a major threat to the future of
the Fox River and Chain O' Lakes. The Defenders are
297
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
attempting to fill an important role that governmental
agencies cannot. To help deal with the problems of soil
erosion and sedimentation in the Fox River basin, the De-
fenders have undertaken a special project called "Opera-
tion Topsoil."
Operation Topsoil is designed to encourage discussion
and action by those sectors of the public affected by Fed-
eral, State, and local policies and programs that deal with
water quality, agriculture, development, and soil erosion.
The goals of the project are to focus issues and seek
solutions to soil erosion in the Fox River-Chain O' Lakes
watershed.
The first goal of Operation Topsoil is to bring private
citizens into the decisionmaking process. This may not be
easy since soil erosion does not incite citizens to action
the way toxic waste landfill or threatened natural areas do.
An understanding of nonpoint source pollution is difficult
to get across; it does not help that the term nonpoint
source is as diffuse as the problem itself.
However, the Fox River-Chain O' Lakes area has sev-
eral assets Operation Topsoil hopes to build upon. First,
the large number of individuals who use the river and
lakes for recreational purposes can be made aware of how
erosion from construction sites affects their use and enjoy-
ment of the waterways. Second, a high level of awareness
of the river and its problems already exists. Third, the Fox
River-Chain O' Lakes Management Agency, which has
the legal powers to maintain the recreational assets of the
area through pollution and flood control projects, has re-
cently been created. This agency may help bridge the two
counties' jurisdictional gap and cooperative efforts.
Operation Topsoil's experience to date shows that the
public is willing to get involved, if they can be given real
solutions to problems that affect them directly. The ques-.
tion remains, "will public initiative and involvement be re-
cognized at the governmental decisionmaking level?"
Without public concern and involvement little may be
done to control erosion from construction sites. Traditional'
agency approaches to implement soil erosion and sedi-
mentation controls have failed to meaningfully link local
governments with citizens. Public interest groups like the
League of Women Voters and McHenry County Defenders
can play an important role in providing that link.
The Fox River-Chain O' Lakes watershed provides unlim-
ited recreational benefits to Lake and McHenry Counties
as well as the general Midwest. However, these benefits
indirectly bring about increasing water quality impacts
throughout the area. Projected development in the water-
shed for the year 2000 suggests worsening problems with
construction-related soil erosion and sedimentation. Cur-
rently, most municipalities in the watershed do have ordi-
nances to control urban erosion; however, the lack of
awareness of the problem reflects the degree to which
these ordinances are enforced.
Lake County has set an example in promoting urban
soil erosion and sedimentation control through the efforts
of the Soil and Water Conservation District, Soil Conserva-
tion Service, and, recently, the Planning Department.
These efforts, along with the goals of the McHenry County
Defenders Operation Topsoil project, aim to involve public
citizens as well as local officials in controlling construc-
tion-related soil erosion. Encouraging planned develop-
Table 2.—Lake County Soil and Water Conservation
District Memo of Understanding.
City/village responsibilities
1. Adopt ordinances or amendments to existing ordinances which will require control of surface water runoff, soil erosion, and
sedimentation on land development and construction sites.
2. Notify the District of the intent of a land developer or builder to prepare a subdivision plat or construction project proposal, and
request the District to perform a site inspection and submit an evaluation report.
3. Require developers (and builders, as appropriate) to submit a runoff, erosion, and sedimentation control plan along with their
preliminary subdivision plats.
4. Request the District to review such plans and provide written evaluations of their adequacy for runoff, erosion, and sedimenta-
tion control.
5. Request the District to conduct onsite investigations.
6. Refer developers, builders, and contractors to the District for advice and information as needed concerning the design and
installation of recommended control practices.
7. Cooperate with the District in conducting training meetings for developers, builders, contractors, and others, as needed.
8. Seek the advice and assistance of the District's technical staff with regard to runoff, erosion, and sediment control on
development and construction projects conducted by the city/village.
9. Adopt Illinois Procedures and Standards for Urban Soil Erosion and Sedimentation Control as the technical reference for
implementing the soil erosion and sedimentation control ordinance.
District responsibilities
1. Assist in developing ordinances and related administrative procedures for controlling runoff, erosion, and sedimentation.
2. Provide standards and specifications for engineering works, vegetative measures, and other practices for controlling runoff,
erosion, and sedimentation.
3. Conduct on-site investigations of proposed development sites, assess surface water management and erosion and sediment
control hazards, and provide a written report of same, with recommendations, to the city/village.
4. Review runoff, erosion, and sediment control plan elements of proposed subdivision plats, and provide an assessment of the
adequacy of such plans.
5. Conduct onsite investigations during the active construction phase of land development projects as needed, to determine
whether the site is being developed in compliance with the approved plan and ordinance requirements, and after construction is
completed to determine whether control practices are being maintained.
6. Consult with land developers, builders, and contractors (upon request and by appointment), concerning the design criteria and
installation procedures for practices recommended to control runoff, erosion, and sedimentation.
7. Provide training meetings as needed for developers, consulting engineers, builders, contractors, and others, as needed,
concerning methods for controlling runoff, erosion, and sedimentation.
8. Provide technical advice and assistance as requested to officers of the city/village regarding control or runoff, erosion, and
sedimentation on public land development and construction projects.
9. Utilize Illinois Procedures and Standards for Urban Soil Erosion and Sedimentation Control as a standard for soil erosion and
sedimentation control plan review, training, on-site technical assistance, and related technical assistance.
298
-------
ment in the watershed will most assuredly provide a multi-
faceted asset for the region; yet, identifying and
controlling soil erosion and sedimentation problems asso-
ciated with development must also be addressed by all the
water resource user groups.
REFERENCES
Brabets, T.P. 1977. Sediment Transport to the Fox Chain O'
Lakes, Illinois. Open file rep. 77-867. U.S. Geolog. Surv., Ur-
bana, IL.
CH2M HILL. 1982. Investigating sedimentation in the Fox River
Chain O' Lakes and associated channels. Prepared for Chi-
cago Dist., U.S. Army Corps Eng.
Flemal, R.C. 1983. Analysis of water quality: Fox River basin,
Illinois. Dep. Geology, Northern Illinois Univ., DeKalb.
Illinois Dep. of Conserv. 1985. Personal comm.
Illinois Environmental Protection Agency. 1976. Assessment and
classification of Illinois lakes.
1982a. Illinois Water Quality Management Plan.
IEPA/WPC/82-012. Div. Water Pollut. Control, Springfield.
_. 1982b. Development of a state-wide construction site
erosion control program. Unpubl. rep. Div. Water Pollut. Con-
trol Plann. Sec., Springfield.
Kothandaraman, V. et al. 1977. Fox Chain of Lakes Investigation
and Water Quality Management Plan. Coop. Resour. Rep. 5,
IL State Water Surv. IL State Geolog. Surv., Urbana.
Koziel, J. 1985. Personal comm. Lake County Soil Water Con-
serv. Dist., Grayslake, IL.
Lake County Planning Department. 1984. Municipal soil erosion
and sedimentation control inventory. Waukegan, IL.
URBAN ISSUES: CONSTRUCTION NONPOINT SOURCE POLLUTION
Layer, R.W. 1985. Personal comm. Staff engineer, McHenry
County Dep. Plann., Woodstock, IL.
McHenry County. 1979. Year 2000 Comprehensive Land Use
Plan.
Metcalf and Eddy, Inc. 1980. Improvements for flood control and
boating, Fox Chain-of-Lakes and Fox River, Lake, McHenry
and Kane counties, Illinois. State of III. Chain of Lakes-Fox
River Comm.
Northeastern Illinois Planning Commission. 1973. Standards
and specifications for soil erosion and sediment control in
Northeastern Illinois. Prepared by Northeast III. Nat. Resour.
Serv. Center, U.S. Soil Conserv. Sen/., Lisle.
. 1979. Areawide water quality management plan.
Part II, Chap. 19 (Fox River basin). Chicago.
_. 1980. Suggested soil erosion and sedimentation
control ordinance: A guide for local officials. Chicago.
1981. Stream use inventory, Fox River. Staff pap.
Chicago.
Northeast III. Soil Erosion and Sedimentation Control Steering
Committee. 1981. Procedures and Standards for Urban Soil
Erosion and Sedimentation Control. Publ. by Assoc. of IL. Soil
and Water Conservation Districts, Springfield, IL.
Paschke, J.E., and J.D. Alexander. 1970. Soil survey of Lake
County, III. Univ. Illinois Agric. Exp. Sta. Soil Rep. 88.
Stall, J.B., and N.G. Bhowmik. 1974. Estimated sedimentation in
the Fox Chain-of-Lakes and the pools behind the dams on the
Fox River-in Illinois. III. State Water Surv., Urbana.
U.S. Army Corps of Engineers. 1976. Plan of Study: Fox River
and Tributaries, Illinois and Wisconsin. Chicago Dist., IL.
Walters, J. 1984. Survey of municipal soil erosion and sedimen-
tation control ordinances in McHenry County. McHenry
County Defenders, Crystal Lake, IL.
Zeiler, C.L. 1985. Personal comm. Lake County Build. Zoning
Dep.
299
-------
Urban Issues:
Hydrologic Modification
and Septic Tanks
NATIONAL PERSPECTIVE ON ENVIRONMENTAL CONSTRAINTS
TO HYDROELECTRIC DEVELOPMENT
S. G. HILDEBRAND
M. J. SALE
G. F. CADA
J. M. LOAR
Environmental Sciences Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee
INTRODUCTION
The U.S. Department of Energy (DOE) initiated a hydro-
power development program in 1977 to promote small-
scale (<30 MW) hydroelectric projects across the country.
Consistent with DOE's support of research on environ-
mental aspects of energy production, it was recognized
that analysis of potential environmental constraints should
be an integral part of the DOE program. The Environmen-
tal Sciences Division at Oak Ridge National Laboratory
implemented studies on the environmental effects of
hydropower development in 1978 in support of the DOE
effort (Hildebrand and Grimes, 1979). This paper summa-
rizes our analyses of two issues that relate to the general
theme of this conference: problems associated with the
concentration of dissolved oxygen in tailwaters below
dams and instream flow requirements for fisheries. In ad-
dition, the need for and technical challenges related to
assessment of the environmental effects of multiple-proj-
ect development in river basins are discussed. Although
the focus of the DOE program is on small-scale hydroelec-
tric development, the issues discussed here are applica-
ble to large-scale facilities as well.
DISSOLVED-OXYGEN CONCENTRATIONS IN
TAILWATERS OF HYDROELECTRIC DAMS
Water quality problems with the discharges of hydroelec-
tric reservoirs can result from the seasonal warming and
consequent thermal stratification of impounded waters.
Hypolimnetic discharges from hydroelectric generating fa-
cilities that have low concentrations of dissolved oxygen
and elevated levels of iron, manganese, heavy metals,
ammonia, and sulfides may adversely affect downstream
biota and water users.
The U.S. Environmental Protection Agency (1976) de-
termined that a dissolved oxygen concentration of not less
than 5.0 mg/L was necessary to maintain the aesthetic
quality of the water, avoid anaerobic conditions (and at-
tendant problems with dissolved iron, manganese, hydro-
gen sulfide, and methane), and support a well-rounded
population of fish. We used the 5.0 mg/L criterion for dis-
solved oxygen to assess the potential for water quality
problems at small-scale hydroelectric projects (<30 MW)
versus larger-scale projects (>30 MW) in the United
States by examining the WATSTORE data base and the
National Hydropower Study data base developed by the
U.S. Army Corps of Engineers (see Cada et at. 1983 for a
detailed presentation of this work).
Using existing data, we calculated the probabilities of
noncompliance (PNC's), defined as the probabilities that
dissolved oxygen concentrations in discharges of cur-
rently operating hydroelectric dams will drop below 5 mg/
L. The continental United States was divided into eight
regions, based on geographic and climatic similarities.
Most of the regions had higher mean PNC's in summer
than in winter, and summer PNC's were greater for large-
scale than for small-scale hydropower facilities. Cumula-
tive probability distributions of PNC also indicated that low
301
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
dissolved oxygen concentrations in the tailwaters of oper-
ating hydroelectric dams are phenomena largely confined
to sites with large-scale facilities. The PNC's are not a
function of electrical capacity per se, but rather appear to
depend on factors related to capacity, such as reservoir
depth.
Although discharges from some small-scale hydroelec-
tric dams have violated the 5-mg/L dissolved oxygen con-
centration criterion, the PNC is relatively low at most exist-
ing sites, particularly in the northern and Pacific Coast
regions. In certain situations, however, small hydropower
development still has the potential to adversely affect dis-
solved gases in rivers. For example, some low-head, retro-
fitted applications may involve the introduction of new tur-
bines into existing dams that are not currently generating
electricity. If the river in question carries a high waste load,
and reaeration during spillage over an existing dam con-
tributes significantly to the oxygen budget of the river, then
loss of this aeration when the flow passes through tur-
bines can depress the concentration of dissolved oxygen
downstream.
Water quality problems could also arise at high-head
applications of small hydropower facilities where air can
be entrained at the penstock entrance. Although this may
serve to reoxygenate tailwaters, nitrogen supersaturation
and downstream fishkills could also occur (Berg et al.
1984).
0STF8EAM FLOW AMD HYDROPOWER
Instream flow needs refer to the amount of water (stream
flow) that is required within a natural stream channel to
maintain instream resource values at acceptable levels
(Bayha, 1978). The issue is primarily one of water quantity,
as opposed to water quality, and focuses on the conflicts
between out-of-stream and instream water uses. One of
the most controversial uses of water within the natural
stream channel has been for the protection of fishery re-
sources, although attention is turning to the maintenance
of instream flows to preserve riparian vegetation, recrea-
tion, and aesthetic values. Flow requirements that include
the needs of fish as well as the integrity of the aquatic
ecosystem as a whole have been difficult to measure and
are often assigned a low priority when they are perceived
as nonbeneficial uses of water resources. Management
for instream flow needs below dams usually takes the
form of minimum release requirements, which are incor-
porated into operation schedules.
The development of analytical techiques for determin-
ing instream flow needs has taken place gradually over
the last two decades (Stalnaker and Arnette, 1976; Ors-
born and Allman, 1976a, b). Although most of this re-
search has occurred in the western United States, where
competition for scarce water resources has prompted
many conflicts between instream and out-of-stream uses,
the need for an objective means for quantifying minimum
instream flows is a national problem.
Three general approaches for determining minimum
flow requirements have been identified: (1) discharge
methods, which rely solely on the use of historical flow
records for making recommendations; (2) hydraulic rating
methods, which construct tradeoff relationships between
stream flow and hydraulic parameters, such as maximum
depth or wetted perimeter (submerged bottom area); and
(3) habitat rating methods, which analyze in detail both the
quality and quantity of habitat available to target fish spe-
cies under different flow regimes. Loar and Sale (1981)
reviewed available methods and made detailed compari-
sons of the strengths and weaknesses of each.
A misconception that often complicates the instream
flow issue is that a single method can serve the assess-
ment needs of all types of water projects. The use of high-
resolution habitat rating methods at small hydropower
projects that have little or no potential for altering stream
flows is as inappropriate as the use of low resolution meth-
ods at large projects that have major effects on stream
flows. Both extremes fail to promote the protection and
management of instream resources. The selection of ap-
propriate methods for determining minimum flow require-
ments should be directed toward achieving a reasonable
match between potential project impacts and resource re-
quirements.
One final note deserves emphasis. There is a significant
lack of validation of existing methods for instream flow
assessments. We recently completed a study (Loar, 1985)
evaluating the validity of using physical habitat indices to
predict the response of trout populations to changes in
stream flow. Based on our results, the assumption that the
abundance of fish varies in direct response to some ex-
pression of their physical habitat could not be rejected.
Habitat-based assessment methods can be valuable tools
for determining minimum flows in southern Appalachian
streams if proper attention is paid to identifying the critical
life stages of a target species and to quantifying the mini-
mum habitat values over annual cycles under the new flow
regime. Much more emphasis needs to be placed on such
evaluation studies and on testing and refining current as-
sessment methods.
MULTIPLE-PROJECT ASSESSMENT AT
One of the most significant challenges to those interested
in the environmental effects of hydropower development
is how to assess the cumulative impacts of multiple hydro-
power project development on river basins. The issue of
cumulative environmental impacts is critical to small hy-
dropower development, especially in California and the
Pacific Northwest. Public pressure and litigation is such
that the Federal Energy Regulatory Commission (FERC)
is not likely to license projects within 22 river basins until
the cumulative impact issue is resolved. In these basins,
more than 150 small hydropower projects have pending
license applications and many more have license exemp-
tions and preliminary permits.
Furthermore, the cumulative impact issue is likely to be
raised in other river basins in the near future (e.g., the
upper Mississippi River basin and the Northeast). The
FERC has proposed a cluster impact assessment proce-
dure to evaluate the impacts of multiple hydropower proj-
ects on a river basin level (U.S. Fed. Energy Reg. Comm.
1985). This procedure is intended to provide the Commis-
sion with sufficient information for deciding which projects
within a basin can be licensed with minimal risks of signifi-
cant cumulative impacts.
The cluster impact assessment procedure consists of a
four-stage process: (1) geographic scoping, (2) resource
scoping, (3) multiple project assessment, and (4) National
Environmental Policy Act (NEPA) documentation. The first
two steps are intended to define the resource system to be
studied (hydropower projects plus nonpower resources
that would be affected by hydropower development). The
final step would be to issue some type of NEPA document,
such as an environmental impact statement or an environ-
mental assessment, to report the findings. All four stages
are designed to solicit and incorporate input from the pub-
lic and resource management agencies.
The cluster proposal is a first step toward the compre-
hensive river basin planning that the Federal Power Act
requires of FERC. Cumulative impacts have rarely been
302
-------
addressed by Federal agencies (Reed et al. 1984), espe-
cially at a programmatic level, as FERC proposes.
As the scientific community rises to the challenge of
assessing the cumulative impacts of multiple hydroelectric
projects, the following appear to be the needed areas of
development:
• Procedures for identifying projects that may contrib-
ute to significant cumulative impacts and projects that are
independent of other hydropower development (in terms
of environmental impacts);
• Guidelines and/or acquisition of data bases for multi-
ple-project assessments;
• Procedures for quantifying the cumulative impacts of
multiple projects;
• Procedures for using multiple project, multiple re-
source impact data in decisionmaking; and
• Technology transfer of all these to all agencies and
the public involved in multiple project assessments.
ACKNOWLEDGEMENTS: We thank Robert M. Cushman and
Virginia L. Tolbert for their reviews of this manuscript. Oak Ridge
National Laboratory is operated by Martin Marietta Energy Sys-
tems, Inc., under Contract No. DE-AC05-840R21400 with the
U.S. Department of Energy. This research was funded by the
Geothermal and Hydropower Technology Division, Office of
Conservation and Renewable Energy, U.S. Department of En-
ergy. Publ. No. 2502, Environmental Sciences Division, Oak
Ridge National Laboratory.
REFERENCES
Bayha, K. 1978. Instream flow methodologies for regional and
national assessment. Instream Flow Info. Pap. No. 7. FWS/
OBS-78/61. Coop. Instream Flow Serv. Group. U.S. Fish Wild.
Serv., Ft. Collins, CO.
Berg, A. et al. 1984. Supersaturation of dissolved air in the wa-
terways of hydroelectric power plants—causal .relationships,
URBAN ISSUES: HYDROLOGIC MODIFICATION AND SEPTIC TANKS
detrimental effects, and preventive measures. Rep. No.
284075. Norwegian Hydrodynamic Lab., Trondheim, Norway.
Cada, G.F., K.D. Kumar, J.A. Solomon, and S.G. Hildebrand.
1983. An analysis of dissolved oxygen concentrations in
tailwaters of hydroelectric dams and implications for small-
scale hydropower development. Water Resour. Res. 19(4):
1043-8.
Hildebrand, S.G., and G.B. Grimes, Jr. 1979. The Department of
Energy environmental subprogram for small hydroelectric de-
velopment. Pages 682-7 in Proc. Waterpower '79: An Interna-
tional Conference on Small-Scale Hydropower. Oct. 1-3. U.S.
Gov. Print. Off., Washington, DC.
Loar, J.M., ed. 1985. Applications of habitat evaluation models in
Southern Appalachian trout streams. ORNL/TM-9323. Oak
Ridge Nat. Lab., Oak Ridge, TN.
Loar, J.M., and M.J. Sale. 1981. Analysis of environmental is-
sues related to small-scale hydroelectric development, v. in-
stream flow needs for fishery resources. ORNL/TM-7861. Oak
Ridge Nat. Lab., Oak Ridge, TN.
Orsborn, J.F., and C.H. Allman, eds. 1976a. Instream Flow
Needs. Vol. I. Am. Fish. Soc., Bethesda, MD.
. I976b. Instream Flow Needs. Vol. II. Am Fish. Soc.,
Bethesda, MD.
Reed, R.M., J.W. Webb, and G.F. Cada. 1984. Siting energy
projects: the need to consider cumulative impacts. Pages
320-8 in Proc. Facility Siting and Routing '84—Energy and
Environment. Banff, Canada.
Stalnaker, C.B., and J.L. Arnette. 1976. Methodologies for deter-
mining instream flows for fish and other aquatic life. Pages
89-137 in C.B. Stalnaker and J.L. Arnette. Methodologies for
the Determination of Stream Resource Flow Requirements:
An Assessment. FWS/OBS-76/03. Western Energy Land Use
Team, U.S. Fish Wild. Serv., Ft. Collins, CO.
U.S. Environmental Protection Agency. 1976. Quality Criteria for
Water. Washington, DC.
U.S. Federal Energy Regulatory Commission. 1985. Fed. Reg.
50(16): 3385-3403. Jan. 24.
303
-------
PERSPECTIVES ON SEPTIC TANKS AS NONPOINT
SOURCE POLLUTION
B. L. CARLILE
Texas A&M University
Department of Soil and Crop Science
College Station, Texas
A septic tank and soil adsorption system, when properly
designed and installed, can effectively remove contami-
nants from human wastes to produce better quality water
than a municipal waste treatment plant providing second-
ary and tertiary treatment.
Although hundreds of thousands of conventional septic
systems have been installed in rural and suburban loca-
tions during the past century, the systems have seldom
been designed according to quantitative theory. The suc-
cess or failure of the system reflects various empirically
derived criteria with designs produced by observations of
systems "working" in a satisfactory number of cases.
When inexperienced, unknowing, or uncaring persons de-
sign septic systems, the criteria may be relaxed or altered.
The normal definition of a failed septic system is when
the sewage effluent collects on the ground surface or
when toilets and drains no longer evacuate the wastewa-
ter. The pollutional aspects of the subsurface hydrologic
environment are often ignored while the surface pollu-
tional problems may be exaggerated since there is little or
no background data to define "natural" conditions.
CONDITIONS AFFECTING SEPTIC
SYSTEM EFFECTIVENESS
Many of the problems associated with septic systems can
be minimized or eliminated through common sense and
some knowledge. For example, common sense warns
that a soil adsorption trench placed in a zone saturated
with perched water will perform poorly when water tables
are high. Some knowledge is required to predict how high
and for how long the perched water table will occur. When
known, a simple design modification (such as shallow
placement of the soil trench and site landscaping) might
be sufficient to overcome problems.
In reviewing literature on the treatment capabilities of
properly functioning septic tank/soil adsorption systems,
Hansel and Machmeier (1980) indicated little or no pollu-
tion of surface or subsurface waters. The study indicated
reductions in biochemical oxygen demand (BOD) from
270-400 mg/L in raw waste to less than 1 mg/L at a point
30 cm (1 foot) below the drainfield trench. Total suspended
soilds (TSS) were reduced from 300-400 mg/L in raw
wastes to zero below the trench bottom. Fecal coliform
bacteria were reduced from 106 to 10"/100 ml in raw
wastes to less than 1/100 ml at depths of 91 cm (3 feet)
below the drainfield. Ammonium nitrogen, organic nitro-
gen and phosphorus were all reduced to background lev-
els within the same distance. Nitrate nitrogen reduction
was more variable—as high as 40 mg/L below some sys-
tems while others showed no increase below the trench.
The data do not necessarily indicate that conventional
systems are working if effluent does not surface. Treat-
ment of the wastes may be seriously impaired if unsatu-
rated and aerobic conditions are not maintained around
the drainfield trenches (Reneau, 1978). Oxygen is essen-
tial for maximizing biological degradation of wastes by soil
organisms and deactivating anaerobic organisms from the
septic tank effluent, which may include disease-causing
pathogens. Problems occur when there is insufficient dis-
tance between the trench bottom and groundwater, when
there is a restrictive horizon over which water can perch,
or when the trench is located immediately over fractured
rock where little or no treatment occurs.
Studies in high water table areas of North Carolina
(Carlile et al. 1981) and in Virginia (Stewart and Reneau,
1981) show that the separation distance between the
trench bottom and the seasonal high water table is the
most significant factor affecting septic system perform-
ance and local grpundwater quality. Hydraulic overloading
of the trenches reduces the thickness of the unsaturated
zone, possibly making a marginal problem into a serious
one. In conventional septic systems where effluent is not
uniformly distributed, 61 cm (2 foot) separations from the
trench bottom to the water table provided excellent treat-
ment of all waste components except viruses in sandy
soils. Separation distances of up to 1.2 m (4 feet) may be
needed in sandy soils to prevent virus mitigation to the
water table. In most systems, viruses were not present in
the groundwater at a 7.6 m (25 foot) horizontal distance
from the trench.
Nonuniform distribution of septic tank effluent in soil
adsorption trenches is a major cause of poor treatment of
wastewater in conventional trench systems (Stewart and
Reneau, 1981; Cogger and Carlile, 1984). A concentrated
plume of pollutants moves from the drainfield toward a
drainage facility in sites with high water tables. The begin-
ning of this plume is generally where the effluent enters
the soil adsorption trench.
MODIFIED AND ALTERNATIVE SYSTEMS
Any septic system that combines dosing, uniform distribu-
tion and shallow placement of effluent increases the effec-
tiveness of the soil filtration process. Such a system maxi-
mizes the unsaturated zone, improves the environment for
aerobic soil organisms and reduces the survival of enteric
bacteria.
The conventional septic system can be modified in
many ways. Most modifications are designed to serve one
or two functions: (1) reduce the clogging potential and
maintain maximum permeability of the soil trench; and (2)
enhance the treatment capabilities of the soil system be-
fore the wastes enter ground or surface waters. The V-
ditch system and the large-pipe, gravelless system en-
hance effluent distribution in the trench and thus decrease
the progressive clogging phenomena common to conven-
tional systems (Carlile and Messick, 1982; Anderson et al.
1983). These improvements reduce the chances that the
trench will "relieve" itself into the nearest road ditch, de-
grading surface water quality.
Pressure dosing and distribution systems use pumps or
siphons to enhance wastewater treatment and maintain
soil permeability. They are used in soils with high water
tables, impermeable layers, shallow depths, or generally
slow permeability. The Low Pressure Pipe (LPP) system
makes optimal use of the entire area of the drainfield,
eliminating local overloading and progressive failure. It
304
-------
provides dosing and resting cycles to maintain aerobic soil
conditions in and around the trenches (Ubler, 1980).
Numerous studies with LPP systems as an alternative
to conventional gravity flow systems have shown their po-
tential to adequately limit bacteria and phosphorus move-
ment and optimize nitrogen losses by denitrification on
site (Carlile et al. 1981; Cogger and Carlile, 1984; Stewart
and Reneau, 1981; Otis, 1977). Movement of pollutant
indicators is limited in contrast to that of gravity-flow sys-
tems when steeper hydraulic gradient was present (Re-
neau, 1978; Carlile et al. 1981). LPP systems provide en-
hanced treatment whether the water table is deep or
shallow.
All of these conditions are conducive to good treatment
of wastewater and minimum impact on surface and
ground water.
Current technology in home waste systems can mini-
mize or eliminate nonpoint pollution from many sites now
considered marginal or unsuited for septic systems. For
the homeowner and the public to profit from this technol-
ogy, alternative home waste systems must be permitted
and encouraged by State and local regulatory agencies.
REFERENCES
Anderson, J.L., R.E. Machmeier and M.R Gaff on. 1983. Per-
formance of gravelless seepage trenches in Minnesota. Pap.
URBAN ISSUES: HYDROLOGIC MODIFICATION AND SEPTIC TANKS
No. 83-2011. In Prof, of Winter Meeting, Am. Soc. Agric. Eng.
Carlile, B.L., C.G. Cogger, M.D. Sobsey, J. Scandura, and S.L
Steinbeck. 1981. Movement and fate of septic tank effluent in
soils of North Carolina coastal plain. Final report. Dep. Human
Resour., State of North Carolina, Raleigh.
Carlile, B.L., and J.K. Messick. 1982. Comparative performance
of conventional and alternative systems under controlled
waste loadings. In Proc. Southeastern On-Site Sewage Treat-
ment Conf., Raleigh, NC.
Cogger, C.G., and B.L. Carlile. 1984. Field performance of con-
ventional and alternative septic systems in wet soils. J. Envi-
ron. Qua). 13:137-42.
Hansel, M.J., and R.E. Machmeier. 1980. On-site wastewater
treatment on problem soils. J. Water Poll. Cont. Fed. 52: 548-
58.
Otis, R.J., J.C. Converse, B.L. Carlile, and J.E. Witty. 1977.
Effluent distribution. Pages 61-85 in Proc. Natl. Home Sew-
age Disposal Symp., Pub. 5-77, Am. Soc. Agric. Engr., St.
Joseph, Ml.
Reneau, R.B., Jr. 1978. Influence of artificial drainage on pene-
tration of coliform bacteria from septic tank effluents into wet
tile drained soils. J. Environ. Qua). 7:23-30.
Stewart, L.W. and R.B. Reneau, Jr. 1981. Spatial and temporal
variation of fecal coliform movement surrounding septic tank-
soil adsorption systems in two Atlantic coastal plain soils. J.
Environ. Qua). 10: 528-31.
Ubler, R.L. 1980. Demonstration and evaluation of alternative
treatment and disposal methods. Triangle J. Council of Gov-
ernments, Research Triangle Park, Raleigh, NC.
305
-------
A. DAVID McKINNEY
Division of Water Pollution Control
Tennessee Valley Authority
Knoxville, Tennessee
The potential impact of impoundment upon the quality of
resultant tailwaters is well documented. While reservoir-
specific characteristics and behavior must be considered,
thermal fluctuations, dissolved oxygen depletion, alter-
nate substrate inundation and dewatering, and sluicing of
sediments are among the stress factors to which tailwater
stream segments are subjected.
Tailwater stream segments or stream segments directly
impacted by tailwaters from large impoundments present
complex water quality management challenges, often of
immediate concern. Power generation, flood control, agri-
cultural development, water contact recreation, and eco-
nomic development receive consideration during formula-
tion of impoundment strategies. Consideration of water
quality issues has only recently begun to result in reser-
voir management plans integrating waste allocation and
hydrological modeling with specific water quality criteria to
achieve a defined level of protection.
A dilemma facing regulatory agencies, power-generat-
ing utilities, and industries and municipalities that use
tailwater stream segments as points of discharge for
treated wastewater includes the following:
° Consideration of designated stream use classifica-
tions
° Identification of applicable water quality criteria
° Characterization of background water quality result-
ing from impoundment, including seasonal dissolved oxy-
gen and temperature behavior
° Coordination of NPDES permit limitations, standards,
and criteria with minimum guaranteed daily average dis-
charges
° Evaluation of worst-case conditions resulting from
maximum daily discharge of pollutants potentially toxic to
fish and aquatic life during periods of no discharge from
impoundment structures.
The complexity of defining water quality management
strategies for tailwater stream segments is sufficiently
challenging without considering acute and cumulative im-
pact of nonpoint source pollution from both urban and
rural sources. Milligan et al. (1984) report that a metropoli-
tan area with a population of 350,000, with mixed commer-
cial, industrial, and residential development, may gener-
ate pollution loading in the form of urban runoff equivalent
to the annual mass loading of pollutants discharged from
the attendant municipal wastewater treatment facility
While National Pollution Discharge Elimination System
(NPDES) permits for industrial and municipal dischargers
to tailwater stream segments are based on daily average
minimum flows, slugs of pollutants from urban nonpoint
source runoff may, under worst-case conditions, occur
during periods of no discharge from the upstream im-
poundment and daily maximum discharge from NPDES
sources.
With few exceptions, water quality management strate-
gies for tailwater segments fail to consider the effect of the
injection of significant urban runoff. Indeed, sufficient site-
specific data to quantify pollutants from urban runoff is
seldom available. In water-quality-limited stream seg-
ments requiring best available technology for NPDES dis-
charges, the suddent influx of nonpoint source pollution
may instantaneously violate water quality criteria for bac-
teria, metals, and temperature. Where applicable, water
quality regulatory agencies responsible for the NPDES
permit system must consider the influence of urban non-
point source pollution as a major factor in management
strategies for stream segments influenced by tailwater re-
leases.
While the acute impact of agriculturally related nonpoint
source pollutants upon unimpounded stream segments is
somewhat ameliorated by attendant increase in flow and
resultant distribution contaminants, tailwater stream seg-
ments, particularly those subjected to extended periods of
no or little release, are unable to benefit from the assimila-
tive capacity mechanisms available to free-flowing
streams. A worst-case condition of extended no release
from a major impoundment combined with short-duration,
high-intensity storm events followed by a period of clear,
warm weather may result in spectacular algae activity with
attendant pH and dissolved oxygen alterations detrimen-
tal to fish and aquatic life. Optimum conditions for using
the nonpoint source nutrients, such as occur in tailwater
stream segments, may influence the establishment and
success of rooted macrophyte communities and associ-
ated epiphytic algae. The potential for significant taste and
odor problems for downstream domestic water supply in-
takes should be considered by water quality management
agencies.
As understanding of the behavior and impact of non-
point source pollution develops, agencies responsible for
reservoir management programs and those responsible
for water quality management, particularly the NPDES
program, must integrate management strategies to ac-
count for this significant source of pollutants. Stream seg-
ments influenced by tailwater releases should receive pri-
ority during evaluation of regional nonpoint source
pollution.
[RiEFElHiEMCES
Milligan, J.D., I.E. Wallace, and R.R Betson. 1984. The relation-
ship of urban runoff to land use and groundwater resources.
Tenn. Valley Author. TVA/ORNED/AWQ-84/1. Knoxville, TN.
306
-------
Rural Issues:
Coal Mining and
Abandoned Land
Reclamation
ACID MINE DRAINAGE: SURFACE MINE TREATMENT AND
IN-SITU ABATEMENT TECHNOLOGY
FRANK TCARUCCIO
GWENDELYN GEIDEL
Department of Geology
University of South Carolina
Columbia, South Carolina
STATEMENT AND NATURE OF THE
PROBLEM
Large quantities of very thick, low sulfur coals are found at
shallow depths in the western part of the United States.
Many problems associated with these coals are primarily
concerned with the high total dissolved solids (TDS) con-
tent of the drainages, the preservation of the fertile val-
leys, and the impact on the ground water resources. In
contrast, the major problems associated with the strip
mines of the bituminous coal fields of Appalachia and the
Midwest center about the instability of backfilled areas,
the reclamation of backfilled mines, and acidic mine drain-
ages. This paper addresses the last problem and outlines
the technologies available for the reclamation of backfilled
mines and acid water treatment.
Acid mine drainage forms under natural conditions
when certain coal seams are mined and the associated
strata exposed to a new oxidizing environment. During
this process a variety of iron sulfides (FeS-FeS2) are ex-
posed to the atmosphere and oxidize in the presence of
oxygen and water to form soluble hydrous iron sulfates.
These compounds commonly appear as white and yellow
salt crusts on the surface of weathered rock faces. Natural
waters flowing over the weathered surfaces readily dis-
solve these compounds that hydrolyze in water, forming
acidic, high sulfate, and high iron drainages. Ferrous and
ferric oxyhydroxides impart the red and yellow color char-
acteristic of acid mine drainage. Iron hydroxide usually
precipitates and forms the "yellow boy" that is commonly
observed in the streams and drainages of some coal mine
areas (Caruccio and Geidel, 1978).
The general chemical reactions explaining the oxidation
of FeS2 and the production of acidity (H +) are given by the
following equations:
2FeS
2,s)
7O2 + 2H2O = 2Fe2t + 4SO42
(1)
(2)
(3)
Fe2+ + 1/4 O2 + H+ = Fe3+ + 1/2 H2O
Fe3+ + 3H2O = Fe(OH)3(s) + 3H+
+ 14Fe3* + 8H2O = 15Fe2+ + 2SCX,2- +
16H+ (4)
(Barnes and Romberger, 1968; Baker, 1975)
The stoichiometry of Eq. (1) shows that 1 mole of FeS2
will produce 2 moles of H+ (acidity). In turn, the Fe2+
generated by the reaction of Eq. (1) can readily oxidize
into Fe3+ and produce an additional 3 moles of H+ (Eq.
(3)). In the references cited by Baker (1975), it has been
shown that FeS2 can also be oxidized in the presence of
excess Fe3+ in solution with water and further hydrolyze
to form 16 moles of H+ (Eq. (4)).
The oxidation of Fe2+ to Fe3+ usually proceeds slowly
under normal conditions, as in Eq. (2) and (3) (Singer and
Stumm, 1970). However, certain bacteria act as catalysts
and greatly enhance and accelerate the chemical reaction
rate. Several workers investigating the oxidation rates of
sterilized versus inoculated samples showed that certain
iron bacteria do indeed catalyze the oxidation reactions
involved in acid mine drainage formation and effectively
increase the acidity produced (Klinemann et al. 1981;
Walsh and Mitchell, 1972).
These bacteria are indigenous to aqueous environ-
ments with pH values that range from 2.8 to 3.2, and the
pH of the ground water controls the occurrence and distri-
bution of these bacteria (Walsh and Mitchell, 1972).
307
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
TREATMENT MECHANISMS
Large-scale Neutralization. When large flows and high
acid loads are encountered, large-scale, manufactured
treatment facilities may be used to neutralize the acidity,
oxidize the iron, and generate a high-density settleable
sludge. Many of these systems are available on a com-
mercial basis.
Small-scale Chemical Neutralization. Because of the
relatively simple and controlled aspects of dispensing a
caustic medium, most mining operations routinely neutral-
ize acidic mine drainages. Numerous acidic seeps are
diverted into one flow channel which may be weired. The
flow rate is measured by a modified Parschall flume; a
float in tandem with a valve dispenses sodium hydroxide
(caustic) directly into the acidic flow, the dosage rate being
determined empirically. The pressure head is provided by
locating the storage tanks at a higher elevation (and adja-
cent to roads to allow supply trucks easy access to refill).
A narrow diameter (1-2 cm) PVC pipe conveys the caustic
from the storage tank to the dispensing valve. This
scheme does not require any electrical power and is well
suited for treating acidic waters in remote parts of the
mine. The viscosity of the caustic, however, is readily af-
fected by temperature changes and the dosage rates
must be continually monitored during the cold winter
months. Further, during periods of heavy rainfall, runoff is
channeled into the flow monitoring device, which activates
additional caustic dispensing for flows that, in fact, have a
lower acid load.
Combinations of sodium carbonate briquettes, calcium
oxide, potassium hydroxide, and various oxidizers (perox-
ide, sodium-calcium hypochlorite) may also be used as
acid treatment. In all of these treatments a sludge de-
velops which is retained in settling ponds. The ponds are
eventually cleaned and the sludge disposed of at the mine
site.
Wetlands. Several studies have demonstrated the ef-
fectiveness of wetlands in removing acidity, sulfate, and
iron from mildly to moderately acidic mine drainages
(Lang and Wieder, 1982). In these natural environments,
sulfate is reduced to either bisulfide (with an attendant
reduction in acidity) or pyrite (with the removal of iron).
This technique impounds a low-lying area that gently
slopes into a wetland, adding to it excessive amounts of
organic matter (straw, hay, etc.). Following a brief period of
decay, a sphagnum moss culture is introduced to the sys-
tem.
The advantages of the treatment are the low mainte-
nance (including cost) and the creation of an aesthetically
pleasing treatment facility. However, the ecosystem pro-
viding the base of the treatment may occupy only certain
natural niches. This treatment process is restricted to con-
tinuously wet areas at certain altitudes. Some research-
ers, however, have developed and selectively bred several
moss species that may be used across a variety of condi-
tions. Some question exists as to the levels of acidity and
iron that the natural system may accommodate before
being detrimentally affected by the acidic waters.
CONTROL MECHANISMS
This paper earlier demonstrated that the natural acid-pro-
ducing elements are pyrite, oxygen, water (moisture), and
the catalytic effects of bacteria and water geochemistry.
All but the pyrite, which occurs as an intrinsic part of the
rock matrix, may be directly affected through in-situ abate-
ment techniques to reduce the flow rate and acid loads.
Detergents and Bactericidal Additives. Limiting the
oxidation of ferrous to ferric irons effectively reduces the
amount of acidity produced by almost half. Eventually, the
oxidation reaction will take place naturally and additional
acidity will be produced. However, restricting the acid re-
actions to Equations (1) through (3) maintains the acid
conditions which may be accommodated by bicarbonate
alkalinity. Should the reaction expressed by Equation (4)
begin, the rate of acid production will be greatly enhanced
and overwhelm the low solubility limited calcareous-gen-
erated alkalinity. Thus, the addition of a suitable bacteri-
cide limits the rate of acid production and may be used to
control the acid loads (Kleinmann and Enrickson, 1983).
This measure can be applied to active as well as aban-
doned mines. When used to complement other treat-
ments, it is an effective control measure; because the bac-
tericide is water soluble and entrained by the recharge
events, it can penetrate deep within the backfill or mine
portals and affect physically remote sections. Depending
on the flow through rate/time, the application of the bac-
tericide must be periodically repeated in severe situations.
Where the acid levels are moderate, however, and may be
neutralized by bicarbonate alkalinity, a single application
of the bactericide may permanently reduce the acid levels
to neutralizing bicarbonate concentrations. In this situa-
tion calcareous material must be present in the overbur-
den of the mine setting.
Phosphate Complexes. The introduction of a phos-
phate compound (primarily in the form of pulverized apa-
tite) into an acid system complexes the free iron radical
and inhibits the hydrolysis reaction, thereby reducing the
amount of acidity produced (Flyn, 1969). The phosphate is
solubilized by only slightly acidic conditions and conse-
quently must be physically juxtaposed to the pyrite reac-
tion site. Some field tests using phosphate mine slurry
have provided varying degrees of success and some fail-
ures have been linked to a mass transfer problem. Accord-
ingly, the application of phosphate must coincide with the
placement and be admixed with the acid spoil material
(Meek, 1984).
If all reaction sites are affected, this treatment reduces,
and in some cases eliminates, acidity. Because it is a com-
plexing technique, acid sources associated with iron hy-
drolysis will be eliminated. The levels of acidity associated
with the reaction (Eq. (1)), however, will continue to be
produced.
Selective Placement of Toxic Material. Obviously, this
strategy should be incorporated as part of the mine plan
and must be implemented during the reclamation phase.
The West Virginia Acid Mine Drainage Task Force advo-
cates, in part, the placement of toxic spoil on top of a
highly permeable inert material pad about 3-5 m thick. In
turn, the acid material is capped with a clay seal, covered
with 1 m of soil, and the surface is reclaimed. Soil in this
case may be considered as the intensely weathered sap-
rolitic profile above the bedrock. In a field test, a small
(6 ha) hydrologically-isolated hill was reclaimed according
to these recommendations. The clay seal effectively pre-
vented infiltrating waters from contacting the toxic spoil,
and the porous pad was instrumental in keeping the water
table below and out of the influence of the acid material.
Through a combination of bromide and iodice tracers and
oxygen isotope data monitored over a 2-yr period, this
method was demonstrated to be effective in preventing
acid drainage formation (Geidel and Caruccio, 1984).
Plastic Liner. Reducing the flow of water through acid
material effectively reduces the acid load of the system. In
this technique the surface of the acid source is graded
and covered with a 20- or 34-ml continuously sealed plas-
tic liner (not unlike those routinely used in hazardous
waste sites). To protect the plastic from ultraviolet ray ex-
posure, which decomposes the plastic, a 1 m layer of soil
is vegetated to afford mass stability.
308
-------
RURAL ISSUES: COAL MINING AND ABANDONED LAND RECLAMATION
In a field test involving a 16 ha mine site where a 20 ml
plastic liner was used, the net acid load emanating from
two highly acidic seeps was reduced to a level sufficient to
amortize the cost of the plastic liner within a 6-yr period
(Caruccio and Geidel, 1983).
For this technique to be totally effective, the plastic liner
must extend to the base of the spoil. The outslopes of
most spoil banks are too steep to accommodate machin-
ery or maintain slope stability necessary for soil retention,
so that a significant portion of the mine is left exposed
(unlined). Recharge avenues then exist; and under some
climatic conditions, the head developed at the outslope is
enough to drive the recharge-wetling front under the plas-
tic (temporarily reversing the hydraulic gradient). Subse-
quent acid leachates drain from beneath the plastic when
the normal gradient is reestablished.
This technique is labor intensive and lends itself best to
active operations. Although, if the acid material could be
identified, concentrated in one area, and encapsulated in
plastic, this procedure has merit in reclaiming abandoned
mines.
Blanket (Surface) Application of Limestone. The geo-
chemistry of the acid system indicates positive results (in
the form of reduced acid loads) if the infiltrating (recharge)
water source is rendered alkaline. Under these conditions,
the elevated pH displaces the bacteria, and pyrite oxida-
tion is inhibited. One possible and economically feasible
method of changing the water chemistry is the application
of limestone in the form of a blanket veneer on the surface
of the mine. Rainwater or snowmelt, in contact with the
limestone, is rendered mildly alkaline and creates an alka-
line wetting front. But, because of the limited solubility of
calcareous material, the amount of bicarbonate produced
is much lower than the amount of acid generated by pyrite
oxidation. As a result, the initial alkaline wetting front has a
negligible effect on the total acid system, and the mitiga-
tion of acid flow is not recognized until several recharge
events, spanning as much as 2 or 3 years, have occurred.
(Geidel and Caruccio, 1982).
This technique is best suited to highly permeable sur-
faces and mildly acidic drainages. In areas where the
slopes are too steep for normal reclamation methods, heli-
copters may dispense the limestone on inaccessible areas
while four-wheel drive lime spreaders are used throughout
the remainder of the mine.
Induced Alkaline Recharge Zones. The blanket appli-
cation of limestone has two disadvantages that detract
from the treatment's effectiveness. First, the ameliorant is
not concentrated in the major recharge areas. Conse-
quently, unless the surface is highly permeable, the major
flow paths do not intercept a large portion of the lime-
stone. Second, the limited solubility of the limestone pro-
duces minor amounts of alkalinity not immediately affect-
ing the more soluble acidic systems.
The mechanisms designed to overcome these two defi-
ciencies (the hydrologic uncertainty and the low alkalinity
produced by the limestone) are incorporated in the in-
duced alkaline recharge zones technique. Here, major
surface water flows are intercepted by shallow (1-m)
trenches that are lined with a bottom layer of sodium bicar-
bonate briquettes (applied at a rate of 1 kg per 0.5 m2) and
covered with 0.3 m of agriculture limestone reject mate-
rial. Surface runoff is then converted into highly alkaline
(the effect from the briquettes), high flow (hence, large
alkaline loads) that may easily overwhelm the acidic sys-
tems. The catalyzing bacteria are then displaced, the acid
system neutralized, and the pyrite oxidation stabilized.
The strongly alkaline system lasts only until the briquettes
are completely dissolved. Normally, this should take about
6 months, during which time the acidic system is rendered
and maintained alkaline. Once achieved, the limestone
should provide sufficient alkalinity to maintain the non-
acidic state (Caruccio and Geidel, 1984).
This technique may easily be applied to abandoned
mined lands with a minimum amount of earth moving ef-
fort. Also, the alkaline ameliorant is most effectively used.
However, should the major recharge areas intercepting
the acid-producing horizons in the backfill not be affected
by the alkaline flow lines (as might occur if the hydrology is
not well understood and the trenches improperly ori-
ented), large amounts of acidity will be produced and the
problem exacerbated. Further, the trenches must be lo-
cated in permeable zones to preclude trench flooding that
will contaminate adjacent areas with sodium, killing vege-
tation.
SUMMARY
Many treatment and control measures are available to
mitigate and abate acid mine drainage from surface
mines. The technique deemed most appropriate for a par-
ticular mine site depends on the age of the mine site, the
permeability and character of the mine surface, the de-
gree of vegetation, the degree of calcareous material
present in the overburden, the availability of a chosen
ameliorant, accessibility of the site by equipment, and the
source (physical and hydrologic) of the acidic seeps in the
backfill or mine regime. Those are but a few consider-
ations to be examined in choosing an effective treatment
technique. The problem site must be explored by drilling;
the hydrogeochemical system must be understood and
the acid source identified before the treatment technology
is implemented.
REFERENCES
Baker, M. 1975. Inactive and abandoned underground mines—
water pollution prevention and control. EPA-440/9-75-007.
U.S. Environ. Prot. Agency, Washington, DC.
Barnes, H.L., and S.B. Romberger. 1968. Chemical aspects of
acid mine drainage. J. Water Pollut. Control. Fed. 40: 371-84.
Caruccio, FT., and G. Geidel. 1983. The effect of plastic liner on
acid loads: DLM site. Proc. 4th Annu. W.Va. Surf. Mine Drain-
age Task Force Symp. W.Va. Dep. Nat. Resour., Charleston.
Caruccio, FT., and G. Geidel. 1978. Geochemical factors affect- •
ing coal mine drainage quality. Pages 129-47 in Reclamation
of Drastically Disturbed Lands. Am. Soc. Agron., Madison, Wl.
Caruccio, FT, and G. Geidel. 1984. Induced alkaline recharge
zones to mitigate acidic seeps. Pages 43-7 in Surf. Mining,
Hydrology, Sedimentology and Reclamation. Univ. Kentucky,.
Lexington.
Flyn, J.P. 1969. Treatment of earth surface and subsurface for
prevention of acidic drainage from soil. U.S. Patent No.
3,443,882. U.S. Patent Off., Washington, DC.
Geidel, G., and FT. Caruccio. 1982. Acid drainage response to
surface limestone layers. Pages 403-6 in Proc. Symp. on
Surf. Mining, Hydrology, Sedimentology and Reclamation,
Univ. of KY, Lexington, KY, 403-406.
1984. A field evaluation of the selective placement of
acidic material within the backfill of a reclaimed coal mine.
Pages 127-31 in Surface Mining, Hydrology, Sedimentology
and Reclamation. Univ. Kentucky, Lexington.
Kleinmann, R.L., D.A. Crerar and R.F. Pacelli. 1981. Biogeo-
chemistry of acid mine drainage and a method to control acid
formation. Mining Eng. March: 300-5.
Kleinmann, R.L., and P. Erickson. 1983. Control of acid drainage
from coal refuse using anionic surfactants. Bur. Mines. Rep.
Inv. 8847, Dep. Interior.
Lang, G.E., and R.K. Wieder. 1982. The use of wetlands to
modify acid mine drainage. Personal comm. Dep. Biology,
West Virginia Univ., Morgantown.
Meek, FA. 1984. Research into the use of apetite rock for acidic
drainage prevention. Proc. 5th Annu. W.Va. Surf. Mine Drain-
age Task Force Symp. W.Va. Dep. Nat. Resour, Charleston.
309
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Singer, P.C., and W. Stumm. 1970. The rate-determining step in Walsh, F, and R. Mitchell. 1972. A pH dependent succession of
the formation of acidic mine drainage. Science, 167: 1121-3. iron bacteria. Environ. Sci. Techno). 6(9): 809-12.
310
-------
COAL INDUSTRY PERSPECTIVES ON NONPOINT
SOURCE POLLUTION
VIRGINIA R LEFTWICH
Senior Environmental Scientist
Peabody Coal Company
St. Louis, Missouri
The objective of the 1972 Federal Water Pollution Control
Act was to ". . . restore and maintain the chemical, physi-
cal, and biological integrity of the Nation's waters." To
achieve this objective Congress declared as one of its
national goals "... that the discharge of pollutants into
navigable waters be eliminated by 1985." Since pollution
abatement was a high priority objective, much attention
was given to point sources of pollution and to methods of
controlling and regulating them. It is now 1985 and
achievement of the national goal, although not 100 per-
cent satisfied, has realized tremendous gains toward that
end.
Attention has now turned to the more subtle offender,
nonpoint source pollution. Nonpoint source pollutants are
those that enter ground and surface waters from nonspe-
cific or unidentified sources. At least one-third of the pollu-
tants entering U.S. waters is estimated to come from non-
point sources. According to the February 1985 draft
Operating Guidance prepared by the U.S. Environmental
Protection Agency's Office of Water and circulated among
State authorities, EPA has promised that by the end of the
year it will issue a status report on nonpoint sources (U.S.
Environ. Prot. Agency, 1985).
In the overall abatement effort nonpoint source pollu-
tants must be dealt with just as arduously as their more
obvious counterparts. The 1972 Act (section 304(e)) rec-
ognized this need and required the EPA administrator to
provide guidelines that would identify and evaluate non-
point source pollutants and methods of controlling them.
In the effort to satisfy this directive, EPA has discovered
that both nature and man have contributed to the problem
and that nonpoint source pollutants are less amenable to
treatments than are point sources.
Those nonpoint sources that result from man's activities
fall, generally, into four categories: agriculture, forestry,
construction, and mining (Ashton, 1975). Though the coal
mining industry is primarily concerned with the fourth cat-
egory, its exploration and reclamation activites involve it,
to some extent, in all four categories. The common thread
that binds all four activities, of course, is the problem of
sediment pollution. In addition to sediment, nonpoint
source pollution resulting from surface mining may in-
clude acid mine drainage and highly mineralized water.
Discharges from active coal mining sites must be di-
rected through sediment control structures and are con-
sidered, therefore, point sources. Since all active mining
sites are heavily regulated under the Surface Mining Con-
trol and Reclamation Act of 1977, and since NPDES per-
mits are required for all point source discharges, all active
coal mines are virtually free of nonpoint source pollution.
Potential nonpoint sources are quickly converted to point
sources by constructing pipes, channels, ditches, con-
duits, etc. The coal industry's major concern with non-
point sources, then, is with abandoned mines. If the aban-
doned mine is a surface mine, sediment will sometimes
result from erosion; if underground, there may be acid
mine drainage. In fact, over half of all acid problems are
directly or indirectly related to inactive underground mines
(Hill, 1975).
The potential for erosion is created early in the opening
of a surface mine. As soon as roads are constructed to
facilitate prospecting and mining activities, the vegetative
cover is removed and bare soil is exposed to rain and wind
action. The mining process itself creates an even greater
area of vulnerability. However, the mining industry recog-
nizes the impact of its actions and is quick to rectify the
situation. Reclamation is contemporaneous and effective.
Coal mining operates under one of the strictest regula-
tions of any industry, so corrective measures are assured.
This has not always been the case, however. There was a
time when the remuneration for coal was so small that it
was not possible to undertake the expensive reclamation
practices that are now so common.
The coal mining industry is unique in that a mechanism
for amelioration of past practices is in place. Title IV, Aban-
doned Mine Reclamation, of the 1977 Surface Mining
Control and Reclamation Act required that a fee collection
program be established for reclamation of abandoned
mined areas. The objectives of this fund were to protect
public health, safety, general welfare, and property from
extreme danger of adverse effects of coal mining prac-
tices and to restore land and water resources degraded by
the mining process.
All coal that is mined in this country is assessed a fee—
15 cents per ton for underground mined coal, 35 cents per
ton for surface mined coal, and 10 cents per ton for lignite.
These fees, collected quarterly from coal mine operators,
are the primary source of revenue for the Abandoned
Mine Reclamation Fund through which the program is fi-
nanced. After the money is collected, it must be appropri-
ated by Congress. The allocation of monies is made annu-
ally according to priorities established by the agency
(State, Indian Tribe, Federal Government) responsible for
the regulatory program of the State or reservation. The fee
collection system became effective October 1, 1977, and
will continue through 1992. Walter Heine, OSM's first di-
rector, stated that the "Abandoned Mine Land Reclama-
tion program is one of the most important and far-reaching
environmental initiatives ever developed by the Federal
Government" (U.S. Dep. Inter, 1978).
The Nation's first Abandoned Mine Lands Reclamation
Conference was held in Pittsburgh on Sept. 19 and 20,
1978. Its purposes were to assess the current state of the
art in reclaiming orphaned coal lands; to develop reclama-
tion plans, regulatory requirements, and fish and wildlife
programs; and to consider technical aspects of such prob-
lems as subsidence, acid drainage, and reclamation of
refuse and gob piles.
It was obvious from the beginning that a large amount of
money would be involved. Even before a model State plan
had been developed, about $104 million had been col-
lected. There were no guidelines for this new program, so
the Appalachian Regional Commission was charged with
developing a plan to serve this need. OSM issued its final
regulations to implement the abandoned mine reclama-
tion programs and published them in the Federal Register
on Oct. 25, 1978.
The inventory to identify the Nation's many acres of
abandoned coal mined lands started in March 1979. Oak
311
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Ridge National Laboratory carried out this initial phase of
the survey. The second phase involved collecting and stor-
ing spatial data in the computer and selecting study areas
for demonstration and testing purposes. Phases three and
four consisted of analyzing, maintaining, and updating
functions.
Once the abandoned sites were identified, the States
made plans to reclaim them. Upon signing a cooperative
agreement with OSM, funds were advanced to coal-pro-
ducing States to help them plan their abandoned mine
land programs. However, States were not eligible to re-
ceive their allocation, which was half the amount col-
lected, until they had approved reclamation and regula-
tory programs in place. It was mid-1980 before any
sizeable amount was made available to them.
As a followup to the regulations published in 1978, OSM
developed guidelines to help States set their priorities for
work to be performed. Problem areas considered were
toxic mine drainage, active or potential slides, erosion and
sedimentation, vegetation, toxic materials, hydrologic bal-
ance, public safety and health, aesthetic values, fish and
wildlife values, air quality, and site characteristics (U.S.
Dep. Inter., 1979). Since OSM's policy was to address first
those problems which involved "extreme danger," most of
the early projects were concerned with open mine shafts,
subsidence, fires, floods, and landslides.
In addition to OSM's general guidelines, each State im-
posed its own priorities into the system. For example, one
of Illinois' rules was that, initially, no projects would be
done on land presently owned by coal companies. First
consideration would be given to publicly-owned land; sec-
ond, to privately-owned land; and last, to land currently
owned by coal companies (Aband. Mine Land, 1985).
Seven years after the Abandoned Mine Land program
was established, with most of the extreme hazard cases
taken care of, work is finally beginning on nonpoint source
problems. So far, there has been a spirit of cooperation
between most coal companies and AML offices to attack
the problems together. The coal industry looks forward to
a continuation of that good working relationship and ex-
pects to see that the efforts of filling voids, grading, seed-
ing, channeling, and stream restoration result in a marked
improvement of our environment.
REFERENCES
Abandoned Mine Land. 1985. Personal commun. with officers of
IL, KY, MO, OH.
Ashton, P.M. 1975. Preface. Pages 1-2 in Nonpoint Sources of
Water Pollution. Proc. Southern Regional Conf., Blacksburg,
VA. May 1 and 2.
Hill, R.D. 1975. Nonpoint pollution from mining and mineral ex-
traction. Pages 67-81 in Nonpoint Sources of Water Pollution.
Proc. Southern Regional Conf., Blacksburg, VA. May 1 and 2.
Department of Interior. 1978. News release. Aug. 9. Off. Surface
Mining, Washington, DC.
1979. News release. Aug. 1. Off. Surface Mining,
Washington, DC.
U.S. Environmental Protection Agency. 1985. Agency operating
guidance. FY 1986-87. Off. Water. Washington, DC.
312
-------
TRENDS IN POST-MINING LAND USES—ARE WE DOING OUR
CHILDREN JUSTICE?
DON EAGLESTON
U.S. Forest Service
Berea, Kentucky
By the end of June 1977, the mineral industry had dis-
turbed an estimated 5.7 million acres in the United States.
Approximately 1 million of the 2.7 million acres disturbed
by the coal industry needed reclamation (U.S. Dep. Agric.,
1979). Excluding coal operations, the mining industry re-
claimed only 22 percent of the land it used from 1930 to
1980. The coal industry reclaimed 75 percent of the land
used from 1930 to 1980 (Adam, 1982). The coal mining
industry is far ahead of the other mining industries in
terms of mined land reclamation. The major difference in
reclamation efforts stems in large part from Federal con-
trol over coal mining and reclamation activities—including
the abandoned mined land reclamation program.
Projections for new mine development and expansion
from 1981 to 1990 show 161 new surface coal mines.
Surface mines will produce a 3 to 1 proportion of overall
production (Coal Mining Proc., 1982). Therefore, surface
mining for coal is not expected to decrease in its impact on
the land. Since surface mining will continue at the present
level for the next 5 years, we should examine nonpoint
source pollution from mines and the most reliable prac-
tices for its reduction and elimination.
ELEMENTS OF NONPOINT SOURCE
POLLUTION
During the mining process, massive amounts of overbur-
den is moved and broken or pulverized. Many fresh ele-
ments are continually exposed to oxidation and other
weathering processes. Mineralization occurs, freeing
these elements to run off. Unfortunately, some overburden
contains elements that are not desirable for release into
the hydrologic system.
One of the more commonly recognized elements of
nonpoint source pollution from coal mines is acid mine
drainage. Other possible components of nonpoint source
pollution from abandoned and active coal mine sites are
manganese, boron, arsenic, selenium, sodium and mag-
nesium salts, and last but not least, sedimentation.
An important step in recognizing and locating potential
sources of pollution is sampling and testing overburden
prior to the mining operation. Overburden sampling allows
the coal mine operator to plan for proper handling and
isolation of toxic materials during the mining and reclama-
tion processes. Many of the unreclaimed abandoned
lands today are the direct result of improper analysis and
planning for handling potentially harmful elements con-
tained in the overburden. These derelict sites are some of
the major contributors to nonpoint pollution today.
PREVENTION PRACTICES
Thanks to improved technology, active reclamation re-
search programs, and a more conscientious coal industry,
many innovative techniques have been developed to elim-
inate the perils of unwanted elements infiltrating the sur-
face and ground water systems. Techniques used to con-
trol nonpoint source pollution from abandoned mined
lands and active surface mine sites include surfactants
on coal wastes and slurry sites, organic compost or mu-
nicipal/industrial wastes, tailored plant/tree seedling in-
oculants, water treatment techniques, flooding, clay caps,
plastic liners, isolation of toxic materials in reconstructed
spoil, sediment ponds, terracing and diversion, and per-
manent ponds or impoundments.
TRENDS IN POST-MINING LAND USES
Through years of practicing soil and water conservation in
America, one lesson we have learned well is to control the
volume and velocity of water. If these two ingredients are
left uncontrolled, then erosion and sedimentation are sure
to occur.
Contour diversions, riprapped waterways and retention
structures are commonly used to control the velocity and
volume of water on the mine site. Conservation measures
are particularly valuable where final steep or long slopes
occur. Unfortunately, unless these practices are stipulated
in the post- mining land use provisions of the mine permit,
they must be destroyed before bonding may be returned.
In other words, the conservation practices are removed
before the land is returned to surface owner use. In Penn-
sylvania, the soil and water conservation districts have
been very active in assisting land owners design their
post-mining options and in specifying the needed soil and
water conservation measures. Adequately addressing
post-mining land use needs allows watershed protection
to be built into long-term land management.
Another potential source of nonpoint pollution is selec-
tion of improper post-mining land use. In 1982, .13 eastern
States permitted 480,000 acres for surface mining. Of
those 480,000 acres, approximately 280,000 acres of for-
est land will be mined. An estimated 90 percent of the
280,000 acres of mined forest land will be converted to
pastureland. Admittedly, many of the mountain areas need
additional space for housing and grazing. However, such
wholesale conversion to grazing land may mean that
many steep slopes will not be reclaimed to a land use that
will provide long-term watershed protection. Many of the
State Regulatory Authorities need to closely examine their
apparent permission of wholesale conversion of forest
land to pasture. The reality of the situation becomes more
apparent when acre upon acre of pasture is never fenced.
REFERENCES
Adam, B.0.1982. Coal industry shows best reclamation record.
Coal Mining Proc. 19(5): 48-50.
Anonymous. 1982. New mine development and expansion,
1981-1990. Coal Mining Proe., 19(5): 28.
U.S. Department of Agriculture. 1979. The Status of Land Dis-
turbed by Surface Mining in the United States. Basic Statistics
by State and County as of July 1, 1977. USDA-SCS-TP-158.
Washington, DC.
313
-------
FACTORS AND TREATMENT OF ABANDONED ACID MINE LANDS
FOR CONTROLLING NONPOINT SOURCE POLLUTION
V P. (BILL) EVANGELOU
W. O. THOM
Agronomy Department
University of Kentucky
Lexington, Kentucky
INTRODUCTION
Generally, a spoil is a heterogeneous mass of geologic
material composed of sandstone, shale, and siltstone at
different stages of physical and chemical weathering. In
this mass, one may find varying amounts of sand, silt, and
clay, depending on the state of physical weathering, and
from very little to large amounts of topsoil that have been
incorporated into the spoil during the grading process
(Evangelou, 1983).
Chemical classification of spoils has always been used
as a means of determining potential nonpoint source pol-
lution. In the past, the practice has been to chemically
classify spoils according to their pH and acid or neutraliza-
tion potential. Table 1 shows this classification as pro-
posed by the Soil Conservation Service.
Such a classification commonly misrepresents mine
spoil material because it assumes that pH is the main
obstacle to reclamation. The data in Table 2 point out infor-
mation that further classifies spoils, especially acid spoils
that have been abandoned for a number of years.
Summarizing the data in Table 3, it can be said that
abandoned acid mine spoils can be divided into four gen-
eral chemical classifications based on electrical conduc-
tivity and pH of the spoil solution.
The following information will show in some detail: (1)
the factors that cause abandoned mine spoils to become
acid and saline-gypsiferous; (2) the influences that acidity
and salinity have on reclamation; (3) the proper frequency
and chemical tests that need to be run to identify a spoil's
proper classification; and (4) ways to approach the solu-
tion of the abandoned acid mine land reclamation prob-
lem.
Table 1 .—Chemical classification of coal mine spoils
(U.S.D.A., 1980).
1. Alkaline and calcareous
2. Medium acid to neutral
3. Acid
4. Extremely acid
5. Alkaline-acid
6. Alkaline-extremely acid
7. Neutral-extremely acid
pH above 7.3
pH 5.6-7.3
pH 3.5-5.5
pH below 3.6
Complex of 1 & 3 above.
Complex of 1 & 4 above.
Complex of 2 & 4 above.
FACTORS CAUSING ACIDITY AND
SALINITY
Abandoned acid mine spoils have three types of acidity:
pyritic, exchangeable, and solution. Agricultural soils have
mainly exchangeable acidity (solution acidity is consid-
ered negligible because of the fact that soils rarely have a
water pH below 4). Upon pyrite oxidation the spoil solution
becomes enriched with iron (Fe), manganese (Mn), and
aluminum (Al) at levels that are toxic to plants and other
biological systems as pH often drops below 4. Under such
conditions disintegration of clay minerals can enrich the
solution with many other metals, having as a source the
clay structural components. These metals may include
Ca, Mg, Na, and K along with Mn, Cu, Zn, and Al. Then,
advanced or long-term oxidation of pyrite (as in aban-
doned acid mine lands) leads to a buildup of exchange-
able acidity, solution acidity, and sulfate salts, mainly cal-
cium sulfate, magnesium sulfate, iron sulfate, manganese
sulfate, and at times, aluminum sulfate (Evangelou, 1983).
Spoils of this nature are easily identifiable by studying
composition of the solution phase. Table 4 shows the solu-
tion composition of saturation extracts of unreclaimed acid
spoils of western Kentucky and Missouri.
TYPES OF ACIDITY TESTS AND
INTERPRETATION
Techniques for assessing the fertility status of agricultural
soils have also been used to assess coal spoils. Although
the practice is justifiable for neutral spoils, for pyritic spoils
it is questionable. The most important aspects of pyritic
spoils are degree of acidity and pH. The pH measure-
ments are usually made on a 1:1 soihwater suspension.
This measurement will have some interpretive value as
long as the pH is equal to or greater than 4 and electrical
conductance is less than approximately 2 mmhos cm"1.
For spoils with a water pH less than 4, oxidizable pyrite is
considered likely to be present.
In terms of acidity, three types of spoils are identified: (1)
spoils of low water pH but of relatively low acidity; (2)
spoils that may have low water pH (not less than 4) and
very high exchangeable acidity; and (3) spoils that contain
pyrite. Spoils of low water pH but of relatively low acidity
are usually sandy and do not contain pyrite. Often such
spoils are limed on water pH because the SMP buffer
does not show any need for lime, even though the water
pH may be lower than 5.
For spoils with low pH (normally higher than 4) and high
concentrations of exchangeable acidity, the most appro-
priate test is one that can measure solution acidity plus
exchangeable acidity. Such results are available from ei-
ther the Woodruff Buffer Method or the Shoemaker,
McLean and Pratt (SMP) buffer pH method (Table 5).
Finally, for pyritic spoils that have been abandoned for
some time, and may have a water pH below 4, the test
most commonly used to determine acidity is that using
peroxide. The pyritic tests are often unreliable no matter
what procedure is followed. The rates of lime reacting with
the acidity (Roberts et al. 1984) and the byproducts
formed may create conditions in the field that are very
different from the laboratory conditions (Evangelou et al.
1982). For these highly pyritic spoils one should take the
wait and see approach. First, apply the lime as split appli-
cations, and before proceeding with vegetative seeding,
recheck the pH. Secondly, check the soluble salts by the
saturation extract method.
Often, laboratories sum the ammonium acetate extract-
able cations and refer to them as cation exchange capac-
ity. This practice is applicable to agricultural soils but not to
314
-------
RURAL ISSUES: COAL MINING AND ABANDONED LAND RECLAMATION
Table 2.—Chemical composition of interpore water from coal mine spoils at saturation (Evangelou, 1983).
Identification
1 . Kentucky
2. Kentucky
3. Kentucky
4. Kentucky
5. Kentucky
6. Missouri
Classification
Acid Gypsiferous
Neutral
Acid-Saline-Gypsiferous
Saline-Gypsiferous-Neutral
Gypsiferous
Acid
Ca2+
18.9
19.1
24.4
19.3
20.2
20.1
Mg2*
13.6
8.1
93.7
94.7
22.9
4.5
K+ Na+
meq/i
0.2 1.2
0.4 0.5
0.1 0.4
0.1 0.2
— 0.5
PH
3.6
7.1
4.0
7.5
4.6
5.4
EC
2.6
2.1
7.0
7.9
3.3
1.9
Table 3.—Chemical classification of abandoned mine land coal spoils.
Class
Chemical characteristics
EC range
(mm hos cnv1)
pH range
IV
Acid
Acid-gypsiferous
Acid-saline-gypsiferous
Saline-gypsiferous
Below 2.25
2.25-2.50
Greater than 2.5
Greater than 2.5
Below 6
Below 6
Below 6
Above 6
Table 4.—Solution composition of saturation extract of Kentucky and Missouri acid mine spoils versus a Kentucky
productive soil (Evangelou, 1984).
Location
Kentucky
Kentucky
Missouri
Missouri
Average Kentucky productive soil
Ca
550
273
520
550
80
Mg
1,130
5,480
1,540
503
10
Na
• • • ppm • • •
10.4
56.6
5.7
14.0
10.0
Mn
221
910
78
61
0
pH
2.5
3.3
2.6
2.9
6.0
EC
12.4
23.6
13.0
7.0
0.8
spoils that have a saturation extract electrical conduc-
tance value of greater than 2 mmhos cnrr1. The saturation
extract test should not be used in lieu of acidity determina-
tion tests but is supplementary to the pyritic acidity test. If
the electrical conductance is less than 2.2 mmhos cm-1
and the pH of the saturation extract is greater than 3, then
neutralization of exchangeable acidity and pyritic acidity
by adding the laboratory-determined amount of lime will
be adequate.
The information presented in Table 6 summarizes the
course of action after obtaining laboratory test results for
acidity and the saturation extract. For a degree of difficulty
greater than + + use a temporary cover such as barley,
which is highly resistant to salts. If the barley grows suc-
cessfully, than a grasslike fescue may become estab-
lished. Legumes are highly sensitive to low pH and salts,
and their establishment in such spoil is always question-
able.
In general, woodbark and wood chips at the rate of 27-
31.5 metric tons/ha (30-35 tons/acre) give good results
when mixed with the necessary plant nutrients, either as
animal wastes (manure) or chemical fertilizers. Seeding
with hydromulching can adversely affect vegetative estab-
lishment on dry soils. Although seeds may germinate in
the ideal environment of the mulch, the dry environment of
the spoils can result in poor seedling survival.
Be aware that mulches do not have a longlasting effect
on vegetative cover. Use them initially only when condi-
tions in the spoils make it necessary. For example, con-
sider mulching when revegetating spoils with a waterhold-
ing capacity (available water) of less than 10 percent,
especially if seeding later 'in spring or if the site has a
southern aspect. When mixed with manures, mulches re-
duce aluminum toxicities to plants that occur at relatively
low pH's since the organic matter tends to tie up soluble
aluminum.
Generally, the following information should be kept in
mind when revegetating abandoned mine lands com-
posed of pyritic spoils.
1. Use calcitic lime in neutralizing pyritic spoils and ap-
ply lime as quickly as possible after spoil exposure to
minimize dissolution.
2. Select a plant species exhibiting salt tolerance (see
Table 9). When possible, use a winter annual or cool sea-
son perennial that is dormant in summer. This reduces
transpirative demand and upward salt flux during the wa-
ter-deficient summer months.
3. Seed during optimum germination conditions (rela-
tively warm seed bed with maximum possible moisture).
Germination delays allow salt migration into the seed,
which kills sensitive, developing seedlings.
4. Leave the spoil surface rough, or deep till (rip) the
upper part of the spoils profile and create furrows parallel
to the slope. This traps water in depressions and encour-
Table 5.—Limestone rates for spoil buffer pH readings
based on SMP (Barnhisel, 1976).
SMP buffer
pH readings
6.8-6.3
6.3-5.9
5.9-5.3
5.3-5.0
5.0-4.5
4.5-4.0
Agricultural limestone
(metric tons/arcre)
required
to adjust to pH 6.4'
1.8-3.6 (2-4)
3.6-5.4 (4-6)
5.4-7.2 (6-8)
7.2-9.9 (8-11)
9.9-13.5 (11-15)
13.5-22.5 (15-25)
'When lime rates are 25 or more (English) tons/acre, the "neutralization potential"
test should be used to determine the actual amount of limestone needed.
315
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 6.—Interpretation of pyritic abandoned mine land spoil chemical characteristics.
Class of
coal spoils
(Table 5)
Degree of
reclamation
difficulty1
Liming
Mulch
Water saving
practices
I
III
IV
Add according to test
results.
Add according to test
results
Add according to test
results.
No.
Depends on soil texture
and amount of acidity
present.
Depends on soil texture
and amount of acidity
present.
Always add 27-32
metric tons of woodbark
or wood chips and
incorporate.
Always add 27-32
metric tons of woodbark
or wood chips and
incorporate.
Leave spoil rough.
Leave spoil rough; build
furrows 60-75 cm deep
and 90-120 cm wide after
mulching.
Leave spoil rough; build
furrows 60-75 cm deep
and 90-120 cm wide after
mulching.
Leave spoil rough; build
furrows 60-75 cm deep
and 90-120 cm wide after
mulching.
+ Low degree of difficulty
+ + + -n- Highest degree of difficulty
Table 7.—Crop sensitivity to salts based on the saturation
extract test (Ayers, 1977).
Expected yield reduction
0% 10% 25% 50%
Crop (Electrical conductivity in mmhos/cm)
Tall wheat grass
Wheat grass
Bermuda .
Barley
Perennial rye grass
Trefoil, birdsfoot narrow
leaf
Tall fescue
Vetch
Trefoil big
Alfalfa
Lovegrass
Corn (forage)
Orchard grass . .
Clover, alsike, ladino, red,
strawberry
7.5
7.5
6.9
6.0
5.6
5.0
3.9
3.0
2.3
2.0
2.0
1.8
1.5
1.5
9.9
9.0
8.5
7.4
6.9
6.0
5.8
3.9
2.8
3.4
3.2
3.2
3.1
2.3
13.3
11.0
10.8
9.5
8.9
7.5
8.6
5.3
3.6
5.4
5.2
5.2
5.5
3.6
19.4
15.0
14.7
13.0
12.2
10.0
13.3
7.6
4.9
8.8
8.0
8.6
9.6
5.7
ages water movement (hence salt leaching) through the
spoil.
5. Apply surface mulches that increase infiltration and
percolation of rainfall and decrease evaporation rates.
Light colored mulches reflect a greater portion of the sun-
light, reducing surface soil temperature heat load and
evaporative demand. As mulch materials decompose, ad-
ditional organic cation exchange capacity is created, tend-
ing to reduce solution salt levels. Avoid the use of salty
mulches containing poultry manure on salt-affected spoil.
CONCLUSIONS
Abandoned acid mine spoils have a complex chemistry,
making their reclamation difficult. To assure success, one
needs to carry through the following tasks:
1. Be careful to obtain a representative sample.
2. Carry out the proper chemical tests to identify the
problem: is it pH alone or high acidity, pyrite, or salinity.
Fertility tests should be run only after the acid compo-
nent has been identified, quantified, and corrected as dis-
cussed in this paper.
REFERENCES
Ayers, R.S. 1977. Quality of water for irrigation. J. Irrig. Drain.
Div. Am. Soc. Civil Eng. Proc. Pap. 13010. 103(1R2): 135-54.
Barnhisel, R.I. 1976. Lime and fertilizer recommendations for
reclamation of surface-mined spoils. AGR-40. Agron. Dep.,
Univ. Kentucky, Lexington.
Evangelou, V.R 1983. Pyritic coal spoils: Their chemistry and
water interactions. Pages 175-227 in S.S. Augustithis, ed.
Comm. for the Studies of Bauxites, Alumina and Aluminum.
Theophrastis Publ. Athens, Greece.
Evangelou, V.P. 1984. Solution composition of saturation ex-
tracts of mine spoils and soils. Unpubl. Data. Agron. Dep.,
Univ. Kentucky, Lexington.
Evangelou, V.R, R.E. Philips, and J.S. Shepard. 1982. Salt gen-
eration in pyritic coal spoils and its effect on saturated hydrau-
lic conductivity. Soil Sci. Soc. Am. J. 46: 457-60.
Roberts, K., V.P. Evangelou, and W. Szereres. 1984. Kinetic dis-
solution methods for differentiation of siderite, calcite and do-
lomite. Min. Environ. 6:72-6.
U.S. Department of Agriculture, Soil Conservation Service.
1980. Mine-soil classification and use. No. KY-TCP-1. Soil
Conserv. Serv., Washington, DC.
316
-------
Rural Issues:
Impact on
Small Communities
MONITORING THE MANAGERS: A COMMUNITY ENTERPRISE
HARRY MANNING
Sandusky County Health Department
Clyde, Ohio
INTRODUCTION
The Chemical Waste Management Citizens Monitoring
Committee was chartered with the May 22, 1984, filing of
the Consent Decree between the State of Ohio ex. rel
Anthony J. Celebrezee, Jr., Attorney General of Ohio,
plaintiff, and Chemical Waste Management, Inc., defend-
ant.
The Committee consists of the Sandusky County Health
Commissioner, Kenneth W. Kerick, M.RH., who serves as
chairman, two members of the Sandusky County Board of
Health, three local citizens appointed by the chairman,
and an Ohio EPA official.
Once officially constituted, the Committee was called
together by Chairman Kerik to organize on June 28,1984,
at the Chemical Waste Management, Inc. facility in Vick-
ery. That meeting included a review of Section 41 of the
Consent Decree which outlines the general responsibili-
ties of the Committee.
Those responsibilities dictated that Chemical Waste
Management, Inc., shall hold a meeting of the Committee
at the facility no more than once per month to answer
questions and complaints about the operation of the facil-
ity, to provide an update on the activities taken pursuant to
the Decree, and to give the Committee a tour to observe,
firsthand, the activities taken to implement the Decree.
The Committee decided to meet on the third Thursday
of each month and to mix the meeting format. Some meet-
ings would be open, public meetings, while others would
use both closed and open formats.
The Committee agreed to become as familiar and
knowledgeable of the Consent Decree as possible. The
next meeting and subsequent meetings, if needed, would
focus on the provisions of the Consent Decree with both
Chemical Waste Management, Inc., and Ohio EPA per-
spectives.
Secretarial needs for the Committee would be provided
by the Board of Health, reimbursed by Chemical Waste
Management, Inc. Settlement monies as well as reim-
bursement to Committee members were set at $20 per
meeting and 20$ per mile for travel. This provision would
exclude the Chairman and Ohio EPA representative.
COMMITTEE GOALS AND OBJECTIVES
The two primary goals of the Committee include the fol-
lowing:
1. To meet the requirements of the Consent Decree
filed on May 22, 1984, in the Sandusky County Court of
Common Pleas and, more importantly;
2. To provide a focal point in the community for the
interaction of differing community interests with the indus-
try, for the purposes of public information.
The scope of the Committee's function and responsibil-
ity was outlined in the Consent Decree; to expand activi-
ties beyond those mandates would be unauthorized. The
Committee has no regulatory authority under the provi-
sions of Section 41, and the Ohio EPA continues to be the
permitting and regulatory entity.
With the focus on public information, the Committee
agreed on the following objectives to meet its stated goals:
1. To become informed about the facility's activity or
operations;
2. To provide a means of disseminating information to
the public on the operation and status of the facility;
3. To provide a mechanism for citizens' concerns and
complaints to be addressed;
4. To provide community input to Chemical Waste Man-
agement, Inc. personnel;
5. To assess the progress of the facility in meeting the
conditions and dictates of the Consent Decree;
317
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
6. To report perceived discrepancies with the mandates
of the Decree to Ohio EPA.
EVALUATION OF COMMITTEE'S
PROGRESS IN MEETING ITS
OBJECTIVES
To become informed about the facility's activities and
operation: During the term of the Committee's existence,
to date, much time and effort has been expended expand-
ing each Committee member's level of knowledge regard-
ing the Chemical Waste Management, Inc. site. Obviously,
several Committee members had been intimately involved
for several years, while others had only tangentially been
exposed through media coverage or other community in-
formation sources. The meshing of the Committee mem-
bers' knowledge and aptitudes would provide a stable
foundation for the work of the Committee.
Initial meetings have involved an in-depth review of the
facility's operation from the acceptance of waste materi-
als, laboratory sampling, and analytical procedures to ulti-
mate disposal in the deep wells. Meeting agenda items
included a review of deep well technology, current moni-
toring procedures, environmental sampling, and interpre-
tation of sample results. The Committee reviews copies of
consultant reports as well as correspondence and/or other
pertinent documents relative to the current and past oper-
ation of the facility.
To provide a means of disseminating information to
the public on the operation and status of the facility:
The Citizens Committee has used various tools to accom-
plish this objective. The Committee chairman has issued
news releases following each meeting of the Committee.
The chairman has been designated as the official spokes-
man for the Committee, and only he has made official
announcements.
Committee members have responded to many citizens'
calls as well as the news media. Periodic public meetings
have been held and planned. The Committee sponsored
one meeting to review the closure plan prior to the Ohio
EPA public hearing and another is currently planned to
review the Phase I activities of the closure plan with area
residents.
Press conferences following committee meetings and
open meetings to the media have been conducted off of
the Chemical Waste Management, Inc. facility premises.
Monthly on-site meetings at the facility have not been
open to the media or public at the request of Chemical
Waste Management, Inc. officials.
To provide a mechanism for citizens' concerns and
complaints to be addressed: Several citizens' concerns
relating to such problems as truck traffic flow, air pollution
and dust emissions, chemical fixation of the sludges,
cloud emissions, animal and health effects, etc. have
been brought to the Committee for discussion and action.
The Committee has attempted to address all concerns as
presented to them individually or collectively as Commit-
tee members. Probably the most significant concern that
has been brought to the Committee's attention has been
that of possible health effects, both acute and chronic,
that may have resulted from either short- or long-term
exposure to airborne releases from the facility.
To provide community input to Chemical Waste Man-
agement, Inc. personnel: The Committee serves to bring
community concern and input to Chemical Waste Man-
agement, Inc. at each of its on-site meetings with the in-
dustry. These meetings allow for informal interaction with
Chemical Waste Management, Inc. officials on differing
problems. Committee members receive varying input
from residents and bring these concerns to each of the
meetings. The process has worked well in establishing an
ongoing dialogue with the Committee and industry.
To assess the progress of the facility in meeting the
conditions and dictates of the consent decree: At each
regular monthly meeting of the Committee, regular pro-
gress reports are submitted to the Committee for review
and discussion. These reports include activities related to
the provisions of the Consent Decree. In addition, the
Ohio EPA provides an update report from their perspec-
tive on the operation.
Copies of all generated reports, correspondence, etc.
are received by the Committee chairman and dissemi-
nated to Committee members between meetings. Such
reports include reported on-site spills, ground water moni-
toring and testing, air pollution and odor emissions, clo-
sure plan activities, inventory depletion and an operating
report on the deep wells that indicates days of operation
and volume injected. These reports and information en-
able the Committee to assess progress toward meeting
the Consent Decree directives.
To report to the Ohio EPA perceived discrepancies
with the mandates of the Decree: The release of a toxic
cloud from Pond No. 7 was probably the most significant
event since the formation of the Committee. Local resi-
dents called the members of the Committee immediately
after the sighting. Committee members investigated the
release and reported their findings to the Ohio EPA on the
next business day. This release resulted in the September
19, 1984, Findings and Orders by the Ohio EPA Director,
which temporarily closed the facility to any incoming
wastes. A fine of $40,000 was also levied against the in-
dustry for violating several hazardous waste laws and
rules.
SUMMARY
In summary, much of the initial activity of the Citizens
Monitoring Committee has been organizational and di-
rected to increase each member's level of knowledge. To
date, 16 meetings have been held over the past 11-month
period. These meetings have been a mixed format with
regular meetings on-site with Chemical Waste Manage-
ment, Inc., officials and open public meetings scheduled
in between.
The Committee's goals and objectives focus on meeting
the requirements of the Consent Decree as well as provid-
ing an improved mechanism for public information. While
the Committee initially took some public criticism for some
of its activities, specifically the public versus private meet-
ing issue, its credibility has improved and its mission is
clear. The individual members of the Committee have
done considerable work and taken up a most unpopular
challenge.
318
-------
SOUTHEAST MINNESOTA'S KARST TOPOGRAPHY LEADS TO
GROUND WATER POLLUTION FROM NONPOINT SOURCES
LONI KEMP
Ground Water Project Director
The Minnesota Project
Preston, Minnesota
The same hilly topography and unique geology that con-
tribute to southeastern Minnesota's scenic charm also
make it particularly susceptible to ground water contami-
nation.
A 1976 study found that 85 percent of the water wells in
four southeastern counties of the State were already con-
taminated to some extent with nitrates or bacteria (Minn.
Dep. Health, 1976). While surface water contamination is
probably no greater here than elsewhere, in this region
contaminants on the surface are much more likely to enter
ground water supplies.
Three natural characteristics of the region explain the
problem. First, limestone bedrock underlies the region.
Called karst, this rock is riddled with cracks, caves, and
fissures—like a giant sponge. The karst contains great
quantities of ground water that flow quickly and freely
through it. In some places, underground streams and riv-
ers flow through the karst. Niagara Cave near Harmony
even has an underground waterfall.
A second natural feature of the region—shallow soils—
allows pollutants to enter the karst almost unrestrained.
Elsewhere in Minnesota, contaminated surface water filt-
ers through thick layers of glacial soil and overburden,
usually becoming purified before reaching ground water.
But the glaciers skipped the southeast corner of the State.
Here, the karst is covered by a shallow soil layer, where
sinkholes, springs, and disappearing streams appear on
the land surface.
The combination of shallow soil and fractured bedrock
allows water on the surface to drain quickly into the
ground. If that water is contaminated, the contaminants
spread quickly through the karst.
When the glaciers skipped southeastern Minnesota,
they also left a third feature: hilly terrain. While these hills
contribute to the scenic beauty of the area, they can cause
severe erosion when combined with careless farming
practices. Runoff from farm fields can carry soil, pesti-
cides, herbicides, and fertilizers into streams and ground
water. Infiltration is also a pathway for agricultural chemi-
cals to enter ground water.
Single sources of ground water pollution and sources
involving particularly hazardous substances have re-
ceived much attention from the media over the past few
years. Leakage of hazardous waste from the Ironwood
Landfill and the recent spill of 210,000 gallons of jet fuel
from the pipeline near Owatonna are cases in point.
But most our ground water pollution comes from every-
day activities, from practically every home and farm. In
our area, we are experiencing ground water pollution from
improperly managed animal feedlots, inadequate city
sewage systems, home septic systems, illegally dumped
garbage, and abandoned water wells.
Unfortunately, the situation is going to get worse before
it gets better. Several southeastern Minnesota communi-
ties have had to shut down their city wells in recent years
and drill new, deeper wells at great expense to taxpayers.
In some cases, a high nitrate level in the water has trig-
gered the problems; in other cases, seepage of industrial
solvents into a city's water supply has caused a shutdown
of the system.
Rural private water wells appear to be even harder hit.
Voluntary testing of private water wells has been done by
most counties in Minnesota at a minimal cost, and in sev-
eral counties in the karst area these tests are showing 30-
40 percent of private wells with nitrate and/or bacteria
contamination above drinking water standards.
The Minnesota Project, a nonprofit organization de-
voted to rural community development, was encouraged
by many different people to develop a project on the issue
of ground water contamination. With the support of the
Joyce Foundation and several local foundations, we are
working to educate the public about the nature of our
ground water problems and how to begin changing prac-
tices to reduce the pollution problem. We have developed
a wide variety of educational materials, including bro-
chures, a newspaper column carried by about 25 weekly
newspapers, radio public service announcements,
models for use in schools, and an aggressive public
speaking and outreach program. In all of our messages,
we have tried to emphasize what the average resident can
do to minimize his or her contribution to pollution.
Working with eight counties in southeastern Minnesota
that have voluntarily formed themselves into two ground
water task forces, we developed a model ordinance for
ground water protection. Starting with the theory that the
Federal and State governments were unlikely to take care
of this problem, the counties agreed that local government
should take the lead in protecting ground water. Indeed,
since land use issues were at the heart of most of the
sources of pollution, it fell into the counties' traditional
area of responsibility.
The model ordinance contains sections regulating
dumping in sinkholes, water well construction and aban-
donment, individual sewage disposal systems, livestock
waste, and erosion control. In some cases, Minnesota had
fairly good State laws on the books, but everyone admit-
ted that these laws were not being enforced at the local
level. Therefore, the model ordinance gets the county into
the act of enforcing certain laws. In every case, we were
able to design enforcement processes that start on a very
friendly discussion basis, and move through court orders
and criminal sanctions only if the landowner is uncoopera-
tive. This model ordinance has been endorsed by the
eight southeastern counties and is in the process of un-
dergoing public hearings. Also, we have received re-
quests for the ordinance from about 30 States.
The one area which we readily admit is not covered by
the model ordinance is agricultural chemicals. In fact, evi-
dence is mounting that indicates that this may be our
biggest source of nitrate and chemical pollution. Direct
leaching of agricultural chemicals into ground water is a
subject that has not been well studied, and we would urge
a much greater emphasis on that particular category of
nonpoint source pollution.
Two points must be emphasized. First, nonpoint source
pollution is a major contributor to ground water pollution,
at least in certain areas of the United States, such as
those with karst topography. Nonpoint source pollution is
not simply a surface water issue.
319
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Secondly, rural areas are just as likely, if not more likely, Unfortunately, local governments are often ill equipped
to suffer the consequences of nonpoint source pollution. to deal with these issues. In Minnesota, we have found
When one considers the fact that most rural families have that if local governments will work together in harmony
to depend on their own untreated private water wells, the with State agencies and researchers, much can be ac-
importance of ground water protection becomes even complished at the local level.
clearer.
320
-------
Rural Issues:
Silvicultural Nonpoint
Source Pollution
U.S. DEPARTMENT OF AGRICULTURE'S PERSPECTIVE ON
SILVICULTURAL NONPOINT SOURCE WATER QUALITY
J. LAMAR BEASLEY
WARREN C. HARPER
U.S. Forest Service
Washington, D.C.
U.S. Department of Agriculture interest and responsibility
for protecting water resources on forested land dates back
to 1897, 88 years ago. The U.S. Forest Service has car-
ried out this responsibility by developing management
programs for the National Forest System that include
managing the water resource through watershed research
and by assisting States with watershed management
through State and private forestry programs. The National
Forest System was created by Congress in 1897 when it
withdrew certain lands from the Federal domain. The De-
partment of the Interior, and subsequently the Forest Serv-
ice under the Department of Agriculture, was charged with
managing the National Forests for continuous production
of timber and favorable conditions of waterflow.
In addition to enabling legislation that created the Na-
tional Forest System in 1897, Congress has continued to
provide specific direction for managing public forest
lands. Even though the Forest Service traditionally man-
aged all resources, its responsibilities were officially
broadened in 1960 beyond water and timber to include
recreation, range, and fish and wildlife. Specifically, Con-
gress directed that National Forest System lands were to
be managed under the concept of multiple use. Forest
Service land management planning incorporates the con-
cept of multiple use and recognizes the importance of
managing National Forest lands to protect water quality.
As a required part of these plans, the Service developed a
strategy for evaluating the impact of other management
activities on the water resource. In addition, the develop-
ment of these plans includes extensive review by public
interests and consultation and cooperation with State and
local authorities. The Congress has directed that the De-
partment promote efforts to prevent or eliminate damage
to the environment while allowing use of the natural re-
sources. To accomplish this task, the Department must
evaluate environmental impacts prior to beginning pro-
grams and activities that might have negative environ-
mental consequences. Following congressional direction
for the management of forest lands, the Forest Service
clearly established programs to balance resource uses
among often competing demands under the concept of
multiple use.
In contrast to multiple-use management responsibilities
of the Forest Service, the Environmental Protection
Agency is responsible for water under the Clean Water
Act. While EPA is the implementing agency, the primary
responsibility for program development and implementa-
tion is delegated to individual States upon request, with
EPA retaining responsibility for program approval and
oversight. Programs developed by EPA and State agen-
cies can produce conflicting requirements for land man-
agement agencies because they are single resource-
driven programs requiring that nonpoint sources be
controlled to the extent feasible; whereas the statutes un-
der which USDA programs are developed require consid-
eration of competing resource uses. Because Federal
agencies are responsible for complying with State and
local water quality program requirements, provisions di-
rected only toward water quality protection can create situ-
ations that make it difficult for USDA to meet its statutory
requirements as Congress mandated. Rather than requir-
ing a balance among resource uses, recent court deci-
sions have narrowly interpreted compliance with this re-
quirement.
The potential conflict between managing a single re-
source and managing multiple resources becomes appar-
ent when legislative authorities of the Forest Service are
considered: the Organic Act requires the Forest Service to
maintain a continuous flow of timber in addition to protect-
ing the water resource; the Multiple-Use Sustained-Yield
Act requires a balance among resources: ind the National
Forest Management Act requires managing National For-
321
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
ests to meet multiple-use objectives. In contrast, the Clean
Water Act addresses only the water resource, providing
authority for State and Federal Governments to develop
water quality standards and comprehensive programs for
water pollution control. Along with other provisions, the
Clean Water Act addresses nonpoint sources by establish-
ing a State planning program that requires States to iden-
tify nonpoint sources, and establish management strate-
gies to meet State and Federal water quality goals.
The concept of feasibiity as contained in section 208 of
the Clean Water Act is an important link between various
"multiple-use" Acts and the Clean Water Act. Because it
is to society's benefit that forest resources are available
for public use, feasibility must be defined to include use of
these resources. To prevent potential conflict, Federal
agencies must cooperate closely with individual States to
ensure program compatibility to the extent "feasible." The
Department remains committed to environmental protec-
tion through developing programs that efficiently balance
resource use and ensure that all forest resources will be
available to future generations.
Responsibilities and obligations to protect the water re-
source on forest lands are specifically identified in the
Forest Service Manual. Objectives related to watershed
management identified in the manual include: (1) protect-
ing, maintaining, and utilizing soil and water resources on
National Forest System lands to provide goods and serv-
ices in harmony with environmental protection, (2) making
National Forest and rangelands fully productive without
destroying or degrading soil and water resources or ad-
versely affecting the environment, and (3) cooperating
with appropriate Federal, State, city and county agencies
responsible for soil, water, and environmental manage-
ment, to meet national economic, social, and environmen-
tal goals.
It is clear from statements of authority as specified in
various legislative actions and in the Forest Service Man-
ual that the Forest Service is responsible for managing
water resources on lands under its administrative control
to provide for a balanced use of all resources.
The Forest Service believes that certain principles must
be incorporated into management strategies designed for
addressing nonpoint sources:
1. Point and nonpoint sources require differing man-
agement strategies and a clear distinction must be made.
2. Natural background levels must be considered.
3. Best management practices (BMP's) as determined
by an approved process to identify such practices must be
defined and used as the primary management strategy for
nonpoint sources.
4. Water quality standards developed for nonpoint
sources must not be used as a direct means of control but
may be used for evaluating effectiveness of practices.
5. Antidegradation policy for nonpoint sources must be
applied on a watershed basis over time, rather than requir-
ing no change for individual points on stream segments.
6. Cumulative effects evaluations must be based on a
number of watershed characteristics rather than only on
water quality.
It is unfortunate that the distinction between point and
nonpoint sources has recently been blurred through incor-
poration of sources traditionally treated as nonpoint into
point source permit programs. This has confused both
regulatory agencies and land managers. Because non-
point sources are dispersed and difficult to quantify, permit
programs for them are difficult to design and to enforce.
During early development of water quality control pro-
grams, emphasis was placed on easily identified point
source discharges associated with manufacturing facili-
ties. In the early to mid-1970's, when point sources had
been fairly well identified and control strategies devel-
oped, initial efforts for controlling nonpoint sources at-
tempted to use methodology developed for point source
control. This proved impossible because of the difficulty of
identifying nonpoint sources, the difficulty in specifying
effluent limitations, and the difficulty in developing and
implementing controls. Point source control strategy was
designed to establish an effluent discharge compatible
with the needs of the receiving water and to require con-
formance with specified effluent requirements. When ap-
plied to nonpoint sources, the point source strategy was
found inappropriate primarily because of the lack of
knowledge concerning the amount of material introduced
as a result of a land management activity. Further, while it
is possible to take corrective action, it is not often possible
to alter or stop a nonpoint discharge once a land manage-
ment activity has taken place.
Natural background accounts for a portion of water
quality levels observed for any water body and is the sec-
ond principle which must be observed in program devel-
opment. Because most water quality constituents poten-
tially affected by forest management occur naturally in the
environment, it is important to understand and include a
consideration of natural background levels when setting
water quality goals and standards. While it may be possi-
ble to control natural background levels, primary manage-
ment concern should be directed toward increases above
those levels as a result of land management activities.
Most published material addressing nonpoint source
impacts ignores the large component of observed water
quality changes of natural rather than human-caused ori-
gin. These materials, without adequately accounting for
natural background levels, cite nonpoint sources as the
primary reason for not meeting goals of the Clean Water
Act. Information on natural background is lacking primar-
ily because current technology for estimation and mea-
surement is difficult to apply and is of low accuracy except
for highly instrumented, site-specific studies. As a result,
accuracy of determination or estimation is low for state-
wide or nationwide evaluations.
With establishment of preventive practices as the cen-
tral component of nonpoint source programs, the third
principle that must be observed is that an approved proc-
ess for determining best management practices must be
provided as opposed to a sample list of practices. In the
Clean Water Act, Congress recognized that nonpoint
sources are best controlled through prevention rather than
through treatment and controlled release. The Depart-
ment has recognized for many years that the most practi-
cal management strategy for protecting the soil and water
resource is one based on defining appropriate practices
before conducting an activity. As the BMP concept has
been developed and applied by EPA and the States over
the years, BMP's have been defined as management
practices designed for the protection of water quality that
are practical, technically feasible, and mindful of institu-
tional, social, and economic factors. The requirement for
development of feasible practices is consistent with the
concept of multiple-use in that it does recognize resource
use.
For purposes of managing forested lands, the Forest
Service Watershed Management Program has been de-
veloped based on defining and applying watershed man-
agement practices designed to protect the soil and water
resource. In implementing this program, the Forest Serv-
ice recognizes that management practices must be tai-
lored to site-specific conditions because it is not possible
to develop a set of watershed management prescriptions
appropriate to all situations and all places. Forest Service
direction at the national and regional level is designed to
define the "process" by which a land manager can arrive
at the appropriate practice or set of practices, for a given
322
-------
RURAL ISSUES: SILVICULTURAL NONPOINT SOURCE POLLUTION
project and site. This same direction should be developed
by the EPA at the national level and by States at the State
level to assist private forest land managers in designing
appropriate practices.
The fourth principle recognizes that water quality stand-
ards must reflect nonpoint source conditions and must
only be used for evaluating effectiveness of BMP's, not as
a direct means of control. Existing water quality standards
for most States were developed for control of point
sources of pollution. By understanding the assimilative
capacity of a receiving water, an effluent limitation could
be set for point sources and the discharge monitored for
compliance. Attempts to apply such a strategy to nonpoint
sources, however, were not very successful. Effluent limi-
tations for nonpoint sources are more difficult to define
and monitor because of natural background levels and
variability. Irrespective of human activity, water quality con-
stituents vary with time,, space, and antecedent condi-
tions. The amount of runoff produced often depends more
upon climatic events than upon human activity. For these
and other reasons, it is not now possible to measure water
quality to a level of accuracy and precision sufficient for
direct regulatory control of nonpoint sources. The
"change" resulting from changes in land use is often less
than the natural variability. Because of these difficulties,
most specialists and managers agree that attempts to use
water quality standards as a direct means of regulatory
control will not be successful.
The difficulty in applying water quality standards to non-
point sources has been acknowledged by EPA. Rather
than relying on water quality standards as the sole control
mechanism, existing EPA policy states that conformance
with approved best management practices constitutes
compliance with water quality standards. While definition
and implementation of management practices is the ap-
propriate management strategy, properly designed water
quality standards can, and should, play a role in defining
such practices. Once water quality standards have been
refined to address natural background levels and variabil-
ity, standards can be used to measure the effectiveness of
management practices. Such information can then form
the basis for adjusting practices where necessary.
The fifth principle recognizes that antidegradation pol-
icy must incorporate a consideration of "change" over
time and area, rather than requiring no change at all
points and at all times. Existing EPA direction provides for
control based on conformance with BMP's, but an inter-
pretation of antidegradation based on no change can
place a land manager in violation of antidegradation while
being in compliance with approved BMP's. EPA water
quality standards regulations currently require States to
include an antidegradation provision in their water quality
programs. EPA antidegradation policy direction results
from an interpretation of the goals of the Clean Water Act
mandating that water quality be maintained or improved. If
water quality is to be maintained or improved, it is cer-
tainly a reasonable interpretation that water quality cannot
be lowered. While the Department agrees with this inter-
pretation, it does not necessarily follow that water quality
can always be maintained at every point, on every stream,
all the time. Such a requirement is impossible if any level
of management activity or resource use is to be allowed
on forest lands. It is important to note that changes in
water quality as a result of forest management activity are
most often of short duration, low level, minimal impact on
beneficial use. While water quality cannot be maintained
at all points at all times, it is important for water quality to
remain constant or improved over both time and space.
This concept allows for minor short-term water quality im-
pacts from resource use, while maintaining water quality
for an entire watershed over the long-run.
Finally, the sixth principle recognizes that evaluation of
cumulative effects must consider characteristics other
than water quality constituents. Concern for identifying
and managing cumulative impacts has recently been
raised by regulatory agencies, Congress and the courts.
Cumulative impact is an extremely complex issue and one
that the Department recognizes a responsibility for resolv-
ing. Unfortunately, while nonpoint strategy needs to ad-
dress cumulative effects, the current state of technology
again makes it difficult.
Addressing nonpoint sources on a watershed basis, in
fact, necessitates a consideration of cumulative effects.
Most watersheds involve a large land base, many land-
owners, and many types of land management activities.
As a result an assessment of cumulative effect must ad-
dress both method of operation and scheduling of where
and when a given operation occurs. Unfortunately, sched-
uling involves specifying how landowners can use their
lands. Both Congress and individual States have been
reluctant to issue land control legislation, particularly in
rural environments. To determine scheduling for various
land uses, or for individual activities, regulatory agencies
have attempted to estimate pollution loads for each activ-
ity or group of activities. Unfortunately, the concept of
loading is difficult to apply because technology does not
exist that will provide sufficiently accurate and precise es-
timates to establish defensible restrictions.
The Department has been working toward resolving the
difficulties of addressing cumulative impacts as related to
forested lands, and believes the technology currently un-
der development by the Forest Service holds some prom-
ise. The concept being explored is management strategy
based upon evaluation of watershed condition. Having
been talked about for many years, the concept is not nec-
essarily new, although it is believed that the current inter-
pretation is somewhat new and offers a solution for ad-
dressing degradation and cumulative impacts. The
concept as now envisioned by the Forest Service provides
for an evaluation of watersheds based on soil and hydro-
logic conditions. The principal idea is that watershed
"health" must remain constant through time, and appro-
priate adjustments made periodically when this is not the
case. Properly developed, watershed condition can pro-
vide information on actual water quality impact and re-
lated beneficial uses, and may reduce the extremely
costly requirement of water quality monitoring. While wa-
ter quality monitoring will be an important part of water-
shed condition evaluation, the complex task of water qual-
ity sampling can be reduced through evaluation of other
factors such as erosion, channel stability, and channel
condition. The end result will be directed toward environ-
mental protection that allows resource use while protect-
ing water quality and related beneficial uses.
The Department of Agriculture believes that control of
nonpoint sources of pollution is important to the Nation.
Further, the Department believes that it is possible to
reach the goals of the Clean Water Act while providing for
conservation and use of the renewable resources availa-
ble from forested lands. In reaching these objectives, it will
be necessary to consider the following 10 points in devel-
oping a rational approach to nonpoint source pollution
control:
1. Changes in water quality result from both natural
events and land management activities; these sources
must be separated so that efforts can be concentrated on
situations created by human activity.
2. Sources from human activity can be effectively con-
trolled, while controlling natural sources may not be eco-
nomically feasible nor even desirable.
3. Nonpoint sources must be appropriately addressed
through prevention rather than through treatment of water
323
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
and controlled release; compliance with approved BMP's
satisfies compliance with water quality standards.
4. Best management practices must be based on the
concept of prevention, site-specific conditions, and feasi-
bility, and on a "process" for determining practices rather
than requiring selection from a "list."
5. Nonpoint water quality standards must include a
consideration of natural background and natural variability
over space and time.
6. Standards should not be used as a direct means of
regulatory control but should be used for evaluation of the
effectiveness of recommended or required practices.
7. Antidegradation policy should include a time and
area component, recognizing some change in the short-
term, no change in the long-term, and a "watershed" ba-
sis.
8. Water quality programs and management strategies
must consider beneficial uses when defining water quality
standards and antidegradation policy.
9. Evaluation of cumulative effect should be based on
watershed condition rather than only on water quality con-
stituents.
10. To provide for use and conservation of all forest
resources, water quality must include "feasibility."
While managing nonpoint sources is an extremely com-
plex challenge, we believe that issues can be effectively
and efficiently resolved, provided the concerns outlined
are considered, and that management strategies are de-
veloped that effectively incorporate these concerns.
324
-------
IMPLEMENTING THE PUBLIC/PRIVATE NONPOINT SOURCE
MANAGEMENT PARTNERSHIP: A STATE FORESTRY PERSPECTIVE
ROGER L. DAVIS
ROBERT L. MILLER
Forestry Division
Oklahoma State Department of Agriculture
Oklahoma City, Oklahoma
ABSTRACT
State forestry agencies currently recognize two principal
areas of concern in nonpoint source water quality man-
agement: to work with the private sector to get the man-
agement job done in the face of poor economic condi-
tions, and to establish a more effective framework in law
and policy for doing so. The new national nonpoint source
policy promises to be an important advancement toward
these ends, particularly because of its support for a coop-
erative public/private partnership approach. Opportuni-
ties exist for improvements where economic factors are
not primarily limiting, particularly in public sector actions.
As first priority, the public sector should put its house in
order by applying the public/private partnership provi-
sions of the new national nonpoint source policy. Action
needs include clear signals, recognition of pertinent tech-
nical and social considerations in assessments and per-
formance evaluations, insuring local private sector partic-
ipation in program development, and recognition of the
necessity for developing local management program in-
frastructure and managerial resources.
INTRODUCTION
The new national nonpoint source policy (U.S. Environ.
Prot. Agency, 1984) offers important opportunities for ad-
vancement in nonpoint source management. From the for-
estry viewpoint, particularly good features of the new pol-
icy are its commitment to a public/private partnership
approach, and its provisions for recognizing relevant so-
cial and technical considerations and suiting programs to
local situations.
This development, along with general management ca-
pabilities and experience gained through the 208 Pro-
gram, and improvements in forestry nonpoint source man-
agement (as noted elsewhere in this proceedings by Ice
and by Bethea) provides a solid foundation. Forestry in
general has implementation tasks yet to do. However, in
the face of a problematic economic outlook, critical eco-
nomic limitations make it imperative to maximize the ef-
fectiveness of nonpoint source management programs.
Serious shortcomings in current assessment and imple-
mentation approaches limit program effectiveness. These
limitations stem in substantial degree from a combination
of persistent, simplistic concepts and analyses and the
current dominance of judicial and other rulemaking proc-
esses that operate on management issues from the top
(Miller, 1985). Public program shortcomings and complex
and locally variable technical and social factors are often
ignored. As a consequence, there are tendencies to un-
derestimate progress, to superficially analyze reasons for
limited performance by parts of the private sector, to be
pessimistic about achieving satisfactory nonpoint source
management, and to overemphasize regulatory ap-
proaches.
Given these considerations, we suggest the need for a
timely introspection by those of us who are involved in the
public sector. An attempt at such introspection on specific
limitations in assessments and other evaluations of imple-
mentation program alternatives follows.
IS THE PUBLIC-SECTOR HOUSE IN
ORDER?
National Assessments
The 1984 report to Congress by EPA on nonpoint source
pollution describes nonpoint source pollution as a major
national problem (U.S. Environ. Prot. Agency, 1984). It
concludes that forestry generates major localized prob-
lems, and causes some problems in most States. As
noted in the panel discussion on status of nonpoint runoff
programs in this conference, a national nonpoint source
update is presently being conducted by the Association of
State and Interstate Water Pollution Control Administra-
tors, under contract from EPA.
From the forestry standpoint, the current assessment
update procedure contains a serious shortcoming in that
no appropriate classification exists for many situations
where new access road construction temporarily in-
creases sediment movement. This problem exists be-
cause the criteria for the protection of water quality and
designated beneficial uses fail to recognize particular pro-
duction characteristics, including the necessity for forest
road development and the associated unavoidable sedi-
mentation, the periodicity and related aspects of other for-
estry operations, and forestry-streamflow interactions.
Because of these considerations, water quality standards,
antidegradation policy, and assessments should be ap-
plied in an appropriate management unit, production cy-
cle framework, that is, in terms of the time and space
considerations as provided in the new national policy. The
initial evaluation report on our program in Oklahoma dis-
cussed these considerations (Miller et al. 1980), as did
Harper (1985) and Beasley (this volume).
The current assessment necessarily depends heavily
on judgments because we lack data on water quality and
effects of pollutants on designated beneficial uses for
many waterbodies. In the case of sediment-related pollu-
tion, nonpoint source assessments are particularly limited
because of the universal inability to separate the three
sources of sediment, that is, natural background, lag de-
posits in channels that have resulted from past practices,
and contributions from current land management prac-
tices.
Because of these limitations, judgments can err sub-
stantially. To the degree that identified problems result
from lag deposits from past practices, the current assess-
ments will overestimate problems caused by current man-
agement practices. Bias may also enter because of a per-
ception that the assessment results may be used to
allocate public funds.
The following actions would improve objectivity and ac-
curacy in future assessments: (1) direct participation of
325
-------
PERSPECTIVES ON NONPOINT'SOURCE POLLUTION
State forestry and other land management agencies in
State assessments; (2) additional guidelines for the exer-
cise of judgment on waterbodies where there are no data
(for example, specifying the conditions for acceptable ex-
trapolation); (3) separate classification of waterbodies on
which the assessment is based entirely on judgment, or
extrapolation of data from elsewhere; (4) identification of
waterbodies that are affected by sediment lag deposits
from past practices; (5) required appraisals of accuracy
and descriptions of the degree to which the assessments
in each classification are based on actual data; and (6)
recognition of the time and space aspects noted previ-
ously
Recent broad evaluations have tended to overgeneralize
from the basis of quite limited research and past conser-
vation programs. Such analyses (for example, Braden and
Uchtmann, 1985; Harrington et al. 1985) fail to adequately
consider the effects of uncertain signals and, in some
instances, plain bad examples in government programs,
as well as the necessity to develop local water quality
management infrastructure and suit programs to local
conditions. As a consequence, perceptions and conclu-
sions about historical private sector conservation practice
performance may be in error.
Readily evident examples of lack of performance by
government are: (1) projected goals and time frames that
are divorced from reality, (2) the politicizing of subsidy
programs and consequent general failure to require cross-
compliance with conservation practices, and (3) the incon-
sistencies of programs through time. A specific manage-
ment example, frequent and obvious in many States, is an
apparent lack of concern by government units about con-
trolling erosion on public roads (the authors specifically
exclude National Forest roads from this criticism). Such
failures send a confused signal to producers about the
seriousness of soil erosion and water pollution, and the
depth of government commitments.
In generalized examinations of economic and other so-
cial considerations, analysts tend to accept the assump-
tions that most nonpoint source management practices
are uneconomic for landowners, that factors other than
the purely economic are not significant to landowner deci-
sions, and that regulatory measures are therefore neces-
sary. Four examples are articles by economists in the Jan-
uary, 1985, issue of the Journal of Soil and Water
Conservation (Braden and Uchtmann; Harrington et al.;
Libby; Epp and Shortie). Yet, the same issue of the Jour-
nal reported on a survey in 17 States in which the majority
of the farmers contacted in each State favored mandatory
cross-compliance between conservation practices and
price supports (Soil Conserv. Soc. Am., 1985). Other ex-
amples of extra-economic interest of landowners and will-
ingness to cooperate in nonpoint source programs are the
Ontario program as reported by Puddister (1985), the will-
ingness of landowners to accept package (cross-compli-
ance) programs in Wisconsin (Konrad et al. 1985), and
landowner participation in Maryland, reported by Magette
et al. (1985). Another example is the recent policy state-
ment by the Farm Bureau in favor of prohibiting price
supports for production on highly erodible soils (Farm Bu-
reau, 1985).
In our experience in Oklahoma with a wide range of
ownership conditions, we have found the most critical
problems in water quality management implementation lie
in awareness, understanding of good practices, and con-
trol of properties and operations, rather than in rejection
on economic grounds. Other examples of private concern
in forestry are the environmental achievement award pro-
grams of the American Paper Institute and National Forest
Products Association (1985), and as reported in this vol-
ume by Dr. George Ice.
Another common assumption is that, simply because
practices can be enforced, a regulatory approach will nec-
essarily be more successful. However, to succeed in
terms of maximizing net social benefits in the long term,
which should be the basic goal of any resource manage-
ment program (and which is implied in the new national
nonpoint source policy), one finds that such an assump-
tion is not always valid when relevant considerations are
carefully examined. For example, as previously noted
(Okla. State Dep. Agric. 1984):
A regulatory approach, if it is adequately funded, would
have the advantages of an acceleration of some practice
implementations. It would have relative simplicity and
ease, from the enforcement agency standpoint, of enforc-
ing a set of highly standardized practices, and evaluation
of progress in terms of numbers of inspections, permits
issued, etc. However, as examined in considerable detail
by Miller et al. (1980), a regulatory program has inherent
shortcomings. . .. Among these limitations are the large
investments in enforcement which could otherwise be
made available for use against management obstacles,
and inherent tendencies to overlook important system in-
teractions and trade-off relationships and to standardize
excessively.
Evaluations of programs and progress also often fail to
recognize.the necessity for establishing effective local
leadership, organization, management support, and other
infrastructure. While the initial 208 program had some se-
rious faults, when viewed in the long term it resulted in a
fundamentally important advancement in across-the-
board nonpoint source management capability. Rittal
pointed this out in a Journal of Soil and Water Conserva-
tion panel discussion (Soil Conserv. Soc. Am. 1985a). The
Congressional staff member on the panel said that Con-
gress does not and will not recognize such developments
as relevant to evaluating progress. Nevertheless, it is our
view that developing local management infrastructure is a
fundamental part of nonpoint source management that
must be recognized in any evaluation.
In summary, evaluations of nonpoint source manage-
ment program alternatives and performance have often
failed to adequately consider public sector shortcomings
and local development actions that are essential to effec-
tive implementation. These essentials are outlined in the
following section.
EFFECTIVE
Program needs and approaches should be established
first at local levels, because technical and social factors
that affect management vary locally, the necessary infra-
structure must be designed for local conditions, and pro-
ducer participation and capabilities must be developed on
a local basis.
Idteuuttoffacaftooim anndl Analysis off LocaD
InffHyeiDces on
Implementation program development should begin with
recognition and analysis of technical and social factors
and interactions that affect management, such as in our
initial program analysis (Miller et al. 1980), in our forestry
nonpoint source management strategy (Okla. State Dep.
Agric. 1984) and in the final report on a pilot 208 program
task in implementation and evaluations (Okla. State Dep.
Agric. 1983). This will form the basis for designing appro-
priate management support and other infrastructure for
developing managerial potentials, for prescribing sound
326
-------
ownership-specific management practices, and for evalu-
ating program effectiveness and regulatory vs. nonregula-
tory choices.
Influences and interactions vary locally, according to ec-
onomic, institutional and other social conditions, as well
as technical aspects. Examples include the differences
between industrial and small, private, nonindustrial forest
ownerships in management objectives, production char-
acteristics, and economic and other limitations. If manage-
ment obstacles related to such conditions are disregarded
in a regulatory approach, socially undesirable side effects
may occur, as Libby (1985) pointed out: "Across-the-board
regulations inevitably hurt some people more than others.
It could be that the human impacts and cost of any pro-
gram to ease the social burden could be greater than the
cost of soil erosion on those farms."
Also, as noted in our forestry nonpoint source strategy
(Okla. State Dep. Agric. 1984), with respect to regulatory
programs:
Trade-off relationships with other social benefits and sys-
tem interactions that affect net benefits tend to be ig-
nored, or viewed as somehow not relevant to water qual-
ity management decisions. As a consequence, practice
prescriptions tend to be standardized and applied without
regard to... the need to design alternatives to suit differ-
ences in social factors at the local site. ... Resulting
inequities are likely to be frequent. Furthermore, because
of the dependence on enforcement, there is less incen-
tive for addressing ownership-specific conditions of an
institutional, managerial, financial or other social nature
that limit practice acceptability and may cause adverse
side-effects on other social benefits. Specific and impor-
tant instances of such conditions in Oklahoma are the
widespread free public use of nonindustrial private forest
tracts, and limited control, higher costs and relatively high
risk for absentee landowners in this situation.
The new national nonpoint source policy recognizes the
need to evaluate program outcomes in terms of net social
benefits. Any adverse social effects of a program alterna-
tive are relevant to the net benefits equation.
Broad surveys can provide data on practice needs.
However, additional information on technical and social
factors and interactions that affect management perform-
ance should be developed on a pilot watershed basis.
Efforts concentrated on selected high priority watersheds
will serve as pilot programs to identify management prob-
lems and to design and test management approaches and
infrastructure, as well as to accelerate accomplishments
in specific waterbodies.
Development of Management Program
Infrastructure and Managerial Resources
In this paper the term infrastructure as applied to local
management implementation refers to these necessary
supports: (1) technical expertise; (2) public agency and
private leadership; (3) educational resources (personnel,
materials, sites); (4) management support (technical as-
sistance; protection from fire, theft, and other loss; private
management services); and (5) cooperative organization.
Where it is lacking, the development of State agency
technical expertise must have priority. It is prerequisite to
satisfactory progress in other infrastructure development.
This requires substantial elements of time in training and
experience development. Making equitable and effective
ownership-specific management practice prescriptions,
through consideration of social as well as technical as-
pects, demands experienced, professional judgment.
Lack of expertise can result in regulatory actions of an
undesirable nature. For example, Popovich et al. (1984)
described cases where problems are created by differ-
ences between counties in required practices, prohibition
RURAL ISSUES: SILVICULTURAJgNONROINT SOURCE POLLUTION
of sound forestry practices, and adverse tax base reclassi-
fications. In studying one of these instances, Goodfellow
and Lea (1985) concluded that foresters must become
better informed to fill a vacuum in expertise that exists in
many local situations.
Youell (1984) pointed out unsatisfactory implementation
conditions that occur in the absence of adequate profes-
sional input:
Why is there so much controversy in forestry circles
about the proliferation of regulations? Basically because
the regulations vary from town to town in their provisions,
soundness, administration, and degree of enforcement.
Many are conflicting. Some are written without profes-
sional forestry input, and as a result contain restrictions
which are often impractical and difficult for foresters and
loggers to follow. Some may be illegal, and all are costly
in terms of time and money. As one logger put it, "You
have to have a lawyer in your back pocket" to keep track
of them all.
The infrastructure supports outlined are necessary for
developing the potentials in managerial resources as ad-
vocated by the new national nonpoint source policy. Par-
ticular needs lie in developing the managerial potentials
associated with the large number of small, nonindustrial
forest ownerships. These include loggers, timber buyers,
consultants, and other managers, as well as landowners.
In the advocated partnership approach, producers must
participate in local program development. This participa-
tion along with positive approaches in cooperative exten-
sion should help develop local leadership, favorable
agency-producer relationships, and high maturity levels
(in situation leadership terms, the latter condition cannot
exist with coercion (Mersey et al. 1979)).
Good examples of these implementation essentials are
programs in Wisconsin (Sorenson, 1985), Maryland (Ma-
gette et al. 1985), and Ontario (Puddister, 1985). Our expe-
rience in Oklahoma also serves as an example.
The Oklahoma Experience
In Oklahoma, industry, State government, and interest
groups became intensively involved early in controversy
about regeneration clearcutting and regulatory proposals.
That involvement, along with foresight by our Department,
led to a State forestry code assignment of primary respon-
sibility on forest environmental concerns to our Division.
This mandate was basic to developing our full participa-
tion in the State's water quality management program,
including 208 program funding and planning, assess-
ments, and inputs into water quality standards revisions,
and to integrating water quality management into the
State's forestry program.
Our program began with monitoring of water quality and
practices in late 1976. We prepared an initial program
analysis and have continued with water quality monitoring
and hydrological studies, development of educational and
technical assistance programs, and a pilot implementation
project (Okla. State Dep. Agric. 1984; Miller, 1984). Coop-
erative activities with Oklahoma State University, Okla-
homa Conservation Commission and other State agen-
cies, USDA Forest Service and Agricultural Research
Service, and the forest industry have advanced the pro-
gram. Present efforts concentrate on evaluating road
practices and trends in sedimentation resulting from
roads, and on further developments in technical assist-
ance and education. Important elements include use of
data on ownership identity recently developed by the
Oklahoma State University Extension Service, develop-
ment of demonstration sites, a followup phase of our initial
pilot implementation project, and the additional develop-
ment of low-cost but effective road practices.
327
-------
PERSPECTIVES ON NONPOINTSOURCE POLLUTION
Our experience indicates that the forestry agency must
be authorized to act as the primary forestry nonpoint
source management agency if water quality management
is to be integrated into forestry. It is prerequisite to for-
estry's appropriate participation in State program plan-
ning and development, development of needed coopera-
tive activities, and development of technical capabilities
and other infrastructure necessary for the analytical and
managerial development tasks as outlined. This experi-
ence also points to a systematic agency assessment and
program analysis as a key initial element, that is, an
agency strategy, as recommended by the new national
policy.
Finally, our experience supports the need for an analyti-
cal approach to questions of enforcement. Through our
initial program analysis and subsequent experience we
have found important needs and ample opportunities for
further development of infrastructure, local leadership and
management capabilities, and basic awareness and
know-how about protecting water quality and beneficial
uses. Development of these opportunities is in any case
essential to long-term implementation success. Such de-
velopment is surely prerequisite to establishing any need
for coercive action.
State forestry agencies currently recognize two principal
areas of concern in nonpoint source water quality man-
agement: to work with the private sector to get the man-
agement job done in the face of poor economic condi-
tions, and to establish a more effective framework in law
and policy for doing so. The new national nonpoint source
policy promises to be an important advancement toward
these ends, particularly because of its support for a coop-
erative public/private partnership approach. In view of the
problematic economic outlook, it is critical that public pro-
grams become more effective.
Water quality assessments, implementation programs
and program evaluations can be improved by more care-
ful recognition and analysis of important technical and
social considerations. Shortcomings in this regard in as-
sessments and evaluations have resulted in problem over-
estimates and in unwarranted pessimism about program
success. In implementation program development, such
recognition and analysis of local factors and interactions
will lead to the appropriate design of management support
and other infrastructure that are necessary for successful
implementation and for the development of managerial
potentials as advocated by the new national nonpoint
source policy.
Other essentials for program success are participation
by producers in local program development, and ade-
quate authority for the state forestry agency to act as the
primary nonpoint source management agency. The latter
is necessary for integrating water quality objectives into
state forestry programs and for developing infrastructure
and managerial potentials. The Oklahoma experience
demonstrates ample opportunity for such development. It
is surely prerequisite to establishing any need for coercive
action.
The public/private partnership provision of the new na-
tional policy should be incorporated into Federal and State
programs by two general actions: (1) putting the public
sector house in order, and (2) full commitment to essential
management development actions. The first is surely top
priority. It calls for clear and consistent signals, more ob-
jective and accurate assessments that recognize perti-
nent technical and social considerations, better perform-
ance in public programs and practices, and less reliance
on threats and bad-mouthing. Regarding the latter—given
the facts of past programs and the current economic hard-
ship for which no relief appears in sight—to shake the
stick at the small private producer (e.g., Cook (1985), and
Tripp (this vol.)) is tantamount to cracking the whip at a
hobbled and one-eyed horse, already suspicious and
balky from past blind-sidings.
In summary, to implement a true partnership means to
replace poor signals, shifting support, bad examples, and
threats by cooperative extension, with well-designed man-
agement support and other infrastructure, and direct pri-
vate producer participation in program planning and de-
velopment.
American Paper Institute. 1985. News Release. Natl. Forest
Prod. Assn.
Braden, J.B. and D.L. Uchtmann. 1985. Agricultural pollution
control: an assessment. J. Soil Water Conserv. 40(1): 23-6.
Cook, K. 1985. Commentary: agricultural nonpoint pollution con-
trol: a time for sticks? J. Soil Water Conserv. 40(1): 105-6.
Epp, D.J., and J.S. Shortie. 1985. Commentary: agricultural
nonpoint pollution control: voluntary or mandatory? J. Soil
Water Conserv. 40(1): 111-4.
Farm Bureau. 1985. Farm Bureau policies, 1985. Approved in
Honolulu, Jan. 10. Farm Bureau, Park Ridge, III.
Goodfellow, J.W., and R.V. Lea. 1985. A town and its harvesting
ordinance. J. Forestry 83(3) 159-61.
Harper, W.C. 1985. National forest trends in water quality. Proc.
Forestry and Water Quality: A Mid-south Symp. May 8-9. Lit-
tle Rock, AR
Harrington, W, A.J. Krupnick, and H.M. Peskin. 1985. Policies
for nonpoint source water pollution control. J. Soil Water Con-
serv. 40(1): 27-32.
Hersey, P., K.A. Blanchard, and W.A. Natemeyer. 1979. Situation
leadership, perception and the impact of power. Center Lead-
ership Stud. Escondido, CA.
Soil Conservation Society of America. 1985a. Nonpoint-source
water pollution: a panel discussion. J. Soil Water Conserv.
40(1): 37-47.
. 1985b. In the news: farmers favor conservation pol-
icy. J. Soil Water Conserv. 40(1): 117.
Konrad, J.G., J.S. Baumann, and S.E. Bergquist. 1985. Non-
point pollution control: the Wisconsin experience. J. Soil Wa-
ter Conserv. 40(1): 55-61.
Libby, L.W. 1985. Paying the nonpoint pollution control bill. 1985.
J. Soil Water Conserv. 40(1): 33-6.
Magette, W.L., R.A. Weismuller, and K.C. Gugulis. 1985. Saving
the Chesapeake: Maryland's agricultural educational pro-
gram. J. Soil Water Conserv. 40(1): 79-81.
Miller, R.L., D.C. Christopher, and K. Atkinson. 1980. Water qual-
ity management in Ouachita Highland headwaters of Okla-
homa. Okla. State Dep. Agric. Forestry Div. Resour. Bull 1.
1985. Trends in nonpoint source regulation and pol-
icy: how should forestry respond? Proc. Forestry and Water
Quality: A Mid-south Symp. May 8-9. Little Rock, AR.
_. 1984. Water quality monitoring and other activities to
support and evaluate the forestry NPS water quality manage-
ment program in Oklahoma. Pages 2-6 in Research and Reg-
ulatory Programs Related to Southern Forestry Management
Practices and Water Quality Protection. Nat. Counc. Air
Stream Improv. Tech. Bull 417.
Oklahoma State Department of Agriculture. 1983. Forestry non-
point source water quality management implementation: a re-
port on the Clayton Lake watershed pilot project and recom-
mendations for program action. Forestry Div. 208 Task 710
Final Rep. Oklahoma City.
. 1984. Oklahoma's statewide water quality manage-
ment strategy for forestry. Forestry Div. 208 Task 1104 Final
Rep. Oklahoma City.
Popovich, L. et al. 1984. The view from Washington. Am. Tree
Farmer May-June: 9-18.
Puddister, M. 1985. Belgrade Creek: a successful nonpoint pol-
lution control project in rural Ontario. J. Soil Water Conserv.
40(1): 84-6.
Sorenson, D.D. 1985. Organizing an information program for
nonpoint pollution control. J. Soil Water Conserv. 40(1): 82-3.
328
-------
RURAL ISSUES: SILVICULTURAtrNONPOINT SOURCE POLLUTION
U.S. Environmental Protection Agency. 1984. Nonpoint source source policy. Prepared for the Nonpoint Source Task Force
pollution in the U.S. Washington, DC. by the U.S. Environ. Prot. Agency. Washington, DC.
_. 1985. Final report on the Federal/State/Local Non- Youell, C.E., 1984. Connecticut forests are ready: the citizens
point Source Task Force and recommended national nonpoint aren't! Am. Tree Farmer May-June: 11.
329
-------
DALE J. McGREER
Potlatch Corporation
Lewiston, Idaho
FOREST DMDUSTOY INVOLVEMENT
The forest industry was involved with shaping State silvi-
cultural nonpoint source control programs even prior to
passage of section 208. Early State agency silvicultural
controls helped shape the central concept of 208 policy:
prevention of unacceptable impacts to water and benefi-
cial uses by controlling problems at their source. These
management practices evolved into the best management
practice (BMP) concept. The U.S. Environmental Protec-
tion Agency defines BMP's as ". . practices . . . that [are]
determined to be ... the most effective, practicable (in-
cluding technological, economic, and institutional consid-
erations) means of preventing or reducing the amount of
pollution generated by nonpoint sources to a level com-
patible with water quality goals" (40 CFR S131.2(g) 1976).
Nationally, the American Paper Institute and National
Forest Products Association have maintained an active
208 committee since 1975. I have chaired the committee
this year. We have worked closely with EPA and other
Federal agencies as they have developed 208 policy. Simi-
lar efforts of the forest industry continue locally with the
States.
The 208 program has matured, and we support it here
today, just as we have before Congress through the years.
208 supports our goal of keeping forested watersheds pro-
ductive for both timber and water. We are committed to the
concept that management of both resources is fully com-
patible.
»TIAL ELEMENTS ©F
Section 208 of the Act, subsequent EPA rules, policies
and guidelines, interactions with other sections of the Act,
and State actions have resulted in well-crafted, practical
programs for controlling silvicultural nonpoint sources. Ex-
perience now demonstrates the essential elements of 208
that have allowed it to work so well.
First, with EPA oversight, program development and im-
plementation is the States' responsibility. State control is
essential because of the tremendous variability of our
lands, waters, and their uses. This in turn determines the
types of potential silvicultural nonpoint source problems
and control strategies. EPA provides the States with the
necessary flexibility to design and implement programs
and control strategies. In addition to variability of forest
ecosystems, this policy recognizes social factors such as
State institutional capabilities and forest ownership pat-
terns.
As an example of regional variation, over 70 percent of
all forest lands are held by small woodlot owners in some
southern States. Topography and stream gradients tend to
be gentle, yet erosion can be significant in some circum-
stances. In these States, nonregulatory programs focus-
ing on education and cooperation have proven most suc-
cessful. In contrast, the far-western States have many
major private landowners, steep slopes, and streams
highly valuable and sensitive for coldwater fisheries.
These States have opted for forest practices acts with
mandatory rules and regulations, in addition to educa-
tional and cooperative programs. Both approaches meet
regional needs and have worked well.
A third essential element of 208 is development and
implementation of the programs by local personnel experi-
enced with silviculture, soils, and water.
THE EXTENT AND NATURE ©F
SiLVICULTURAL NONPOINT SOURCES:
TOE REAL PROBLEMS
Five years ago at a similar conference panel discussion,
our panel moderator, Fred Haeussler, said, "The 208 pro-
gram's approach to silviculture has been hampered by a
lack of high quality cause-effect data to quantify the water
quality related impacts of silvicultural activities." The
statement remains true today, yet we certainly have com-
piled some solid knowledge in these last 5 years, and we
have applied it.
The 208 process itself has contributed to our knowl-
edge. State evaluations of road construction and logging
reveal that concerns such as streamside management,
stream debris, shade and water temperature, chemical
applications, and other potential problems are well con-
trolled through application of BMP's. These surveys have
also found that our primary problem—sediment—is some-
times difficult to manage. Here the carefully conceived
and worded concept of BMP's becomes crucial. Again
EPA's definition includes such language as "economically
practicable," and "amount of pollution compatible with
water quality goals." It is not always possible to prevent
some sediment from reaching streams. The BMP concept
allows this necessary balancing of resource uses and val-
ues to society.
Just how serious is silvicultural sediment? EPA esti-
mates that less than 4 percent of the total man-caused
sediment reaching the Nation's waters is caused by silvi-
culture—silviculture practiced in forests covering roughly
one-third of this Nation. The problem is that silvicultural
sediment can be concentrated in space and time.
The Salmon River in Idaho provides an interesting ex-
ample of abusive practices resulting in damage to water
and fish, followed by control and responsible manage-
ment. Incredibly, roads occupied 25 percent of the total
ground area of highly erodible land in areas of the Middle
Fork of the Salmon River watershed. Landslides streaked
the mountainsides. People still use the Salmon as an ex-
ample of how terrible silviculture is, but those activities in
the Salmon ended over 25 years ago. This sad experience
alerted the public, and may have contributed to develop-
ing 208 and Idaho's Forest Practices Act. Today it is incon-
ceivable that similar practices could be applied. Tremen-
dous progress has been made in responsibly managing
similar lands.
It is interesting to examine rates of erosion from silvicul-
ture. National forests in Idaho commonly limit total in-
crease in erosion during the first few years following road
construction and logging to 200 percent of natural: about
50 tons/mile2 compared to 25 tons/mile2 natural rate on
highly erodible granitics. In comparison, annual erosion
averages about 16,000 tons/mile2 from wheatlands lo-
cated with 100 miles of these same forests. Over a period
330
-------
of 50 years, it is reasonable to estimate that about 500
times as much erosion will occur from an acre of wheat-
land as from managed forest land, including its roads.
So why is silviculture a concern? It is because of the
high quality water, fisheries, and recreational value of our
mountain streams. Riparian habitat must be managed
carefully, and sedimentation of valuable fisheries' rearing
and spawning areas must be controlled.
MAKING 208 WORK BETTER
Foresters generally know what the potential silvicultural
nonpoint sources are, and where and how to control them.
However, we do need better data on the cause-effect rela-
tionship of sediment to fisheries. (Here I would caution
that while sediment indisputably affects fish, current rela-
tionships purporting cause and effect between erosion
and decline in fisheries' productivity are based on a near
absence of in-stream data and verification.) While we
need better data, even this may be fine tuning.
As an example, Idaho just completed a study of whether
its 208 BMP's for silviculture adequately protect water
quality and its beneficial uses. A team of eight, made up of
RURAL ISSUES: SILVICULTURAL NONPOINT SOURCE POLLUTION
State agency and conservation group representatives and
myself as a representative of private landowners, in-
spected 25 logging jobs. We concluded that with currently
proposed changes to a few rules Idaho's Forest Practices
rules constitute BMP's as defined in the Clean Water Act
and that if enforced would protect beneficial uses in most
circumstances. The task force did find administration of
the rules to be inadequate. This points out the greatest
need we have within silvicultural nonpoint source pro-
grams—we need to apply our existing knowledge. The
small landowner needs the most assistance. For the entire
State of Idaho, we have had no more than five and as few
as one man available for assistance with and enforcement
of the Forest Practices rules. Funds are very limited. The
point is, the technology is known, but the States and land-
owners need some assistance implementing it.
One other element germane to 208 needs to be dis-
cussed and better understood. If we are to maintain good
water quality, we must understand the relationship be-
tween its beneficial uses, the ways to prevent its degrada-
tion, and the BMP's—all within the context of the Clean
Water Act.
331
-------
CONTROLLING NONPOINT SOURCE POLLUTION FROM SILVICULTURAL
OPERATIONS: WHAT WE KNOW AND DON'T KNOW
GEORGE W. BROWN
Forest Engineering Department
Oregon State University
Portland, Oregon
For the past 50 years, researchers have attempted to de-
fine the impact of silvicultural operations on nonpoint
source pollution in streams that drain forested catch-
ments. Much information and practical experience exist
that will help managers of forest land minimize nonpoint
source pollution. In 1972, Oregon implemented the Na-
tion's first forest practices act to provide a legal basis for
nonpoint source pollution control. Other States quickly fol-
lowed suit. Research, practical field experience, and legal
precedent provide a strong base from which nonpoint
source pollution can be controlled. As with most problems
as complex as nonpoint source pollution on forest lands, a
great deal remains to be learned.
NFS POLLUTION ISSUES ON
FOREST LANDS
Today's nonpoint source pollution issues on forest lands
are both scientific and legalistic. The scientific issues fo-
cus on describing and understanding changes in water
quality caused by silvicultural operations such as harvest-
ing timber, constructing forest roads, and using silvicul-
tural chemicals such as pesticides or fertilizer. The nonsci-
entific issues focus on using information about changes in
water quality to predict impacts on aquatic resources or
human health.
Research has demonstrated that silvicultural operations
can temporarily change several water quality characteris-
tics in streams draining forest land. Sediment concentra-
tions can increase if erosion accelerates. Temperature of
streams can increase if overstory riparian shade is re-
moved. Accumulations of slash in a stream can deplete its
dissolved oxygen. Organic and inorganic chemical con-
centrations can increase because of harvesting or appli-
cation of pesticides and fertilizers. Today's most pressing
scientific issue is how to predict these changes and their
duration in water quality and how to affect the outcome by
altering management practices.
Today's most important legal issues are imbedded in
the prediction problem. Courts have required managers to
predict the impact of their silvicultural operations in space
and time. It is no longer sufficient to predict the impact of
an operation where it occurs. Managers must also predict
how a change produced by their operation will interact
with changes produced by other activities within a basin
(cumulative effects) and the impact of extreme events
(worst case analyses).
Given these demands to predict and control nonpoint
source pollution from silviculture, what do we know and
what do we need to know before we can adequately re-
spond?
RELATIVE MAGNITUDE OF NONPOINT
SOURCE POLLUTION FROM
SILVICULTURE
An important starting point is to recognize the nature and
magnitude of the nonpoint source pollution problem in
forest streams. The hydrologic nature of these streams
adds an extra dimension of complexity to the nonpoint
source pollution problems in silviculture. Except for pesti-
cides, all of the water quality changes mentioned are
changes in natural or background values in small streams,
such as sediment concentration or temperature. This
means that separating natural from man-caused levels
can be difficult. The changes produced are usually short
lived. Sediment concentrations, for example, usually in-
crease during storms but quickly return to very low levels
even on watersheds that are clearcut and burned. Further,
the background values may be highly variable. Even on
undisturbed watersheds, sediment values may vary 100-
fold at any given discharge level. This means that
changes are often compressed in time and difficult to de-
tect unless greatly different from background values.
On a national level, the magnitude of erosion from silvi-
culture is far less than from agricultural sources. Esti-
mates vary, but erosion rates from silviculture are usually
believed to be 10 to 100 times less than from cropland.
Likewise, the application of chemicals on forest land
has been much less than on agricultural land, even before
herbicides were banned on Federal forests. In 1977,
American farmers applied 180,000 tonnes (400 million
pounds) of herbicides and 76,500 tonnes (170 million
pounds) of insecticides to crops. Nationally, less than
810,000 ha (2 million acres) of forest land received chemi-
cal treatment that year. The Forest Service, our Nation's
largest forest owner, applied only 180 tonnes (0.4 million
pounds) of herbicide and 68 tonnes (0.15 million pounds)
of insecticide in 1976.
Locally, impacts of harvesting or road construction on
water quality can be quite high if proper practices are not
used. A landslide in a small headwaters stream, for exam-
ple, can scour a channel or damage property a short dis-
tance downstream even though water quality several
miles downstream may not be adversely affected.
Regardless of the local nature of most silvicultural im-
pacts on nonpoint source pollution, public perceptions of
such changes far outweigh their local impact. Recent
court challenges to applications of herbicides and harvest
of timber in landslide-prone terrain have had national re-
percussions for management of forest land.
WHAT DO WE KNOW?
With this general background about the nature of nonpoint
source pollution in forest streams and some understand-
ing about the relative magnitude of these problems, what
do we know about controlling these problems? In general,
we have a good understanding of the factors that cause
changes in the temperature, dissolved oxygen, and chem-
ical composition of forest streams. We also understand
how to control these changes.
Several States have implemented forest practice regu-
lations to require the use of best management practices.
In general, these regulations have markedly improved wa-
ter quality and management of streamside zones. Local-
ized problems have and will continue to occur, but such
regulations and their enforcement by State inspectors
have forced operators to focus on ways to minimize im-
pacts on soil and water resources.
332
-------
This leads us directly to another fundamental principle
that we know. Regardless of the rules and the quality of
inspection, much of the success or failure of a program to
control nonpoint source pollution from silviculture rests
with the operator and the care with which he performs the
job. Clear specification of objectives, woodsworker train-
ing, and close supervision are essential components of
any nonpoint source pollution control program. Long be-
fore Oregon's Forest Practices Act became law, loggers
were harvesting timber from several municipal water-
sheds around the State without degrading domestic water
quality. Here, water quality was clearly a primary objective,
and close cooperation between loggers and city water de-
partments provided good examples of what could be
done.
WHAT DO WE NEED TO KNOW?
Erosion and the sedimentation it produces are still the
most important sources of pollution in forest streams. We
still lack the knowledge to adequately predict erosion in
steep terrain, especially from landslides, and how harvest-
ing and road construction practices influence erosion
rates. Several predictive models have been developed,
but most are based on limited data and have not been
adequately verified. Further, once eroded soil reaches a
stream, we are unable to accurately route it downstream.
Mandates from courts to predict cumulative effects are
doomed to fail unless sediment transport is better under-
stood.
We also lack a clear understanding of how sublethal
changes in water quality affect fish and their habitat. Often
RURAL ISSUES: SILVICULTURAL NONPOINT SOURCE POLLUTION
lethal and optimal levels can be identified, but not the
relationship at points between. Further, the short duration
of most water quality changes in forest streams and the
high natural variability described earlier, coupled with the
natural resilience of aquatic organisms, add to the com-
plexity of assessing impact or attributing it to silvicultural
operations at all but the highest levels of nonpoint source
pollution. The long-term impacts from most silvicultural
activities are not well documented. Nearly all research has
concentrated on immediate or short-term impacts. We
need more knowledge on the resilience of aquatic ecosys-
tems and the rates of recovery following silvicultural treat-
ments.
The solution to our nonpoint source pollution problems
from silviculture lies in good, objective research designed
to clearly identify both the relationship between silviculture
and water quality changes and between water quality
changes and the benefits derived from our water re-
sources. This means honest assessment of the resilience
of our aquatic ecosystems to withstand change at less
than lethal levels. Ultimately, we need to know how to
place changes in water quality and their impact on use of
water into a managerial framework. We need to know how
to balance alternative uses and impacts so that decision-
makers can optimize the use of all resources in our for-
ested catchments and not just a few.
Once this is accomplished, we need better systems for
transferring this technology from researchers and plan-
ners to those charged with carrying out silvicultural opera-
tions on the ground, including the woodsworkers. If we
can accomplish these tasks, there is no reason why silvi-
cultural operations and acceptable water quality cannot
be compatible.
333
-------
Noncoal Mining and
Abandoned Land
Reclamation
CRUSHED STONE QUARRIES AND LAND RECLAMATION
R A. RENNINGER
National Stone Association
Washington, D.C.
Nonpoint source pollution from aggregate mining opera-
tions is a natural phenomenon. Stone, sand, gravel, and
clay are found beneath every stream and river on earth. It
is the disturbance of the natural landscape to extract
these minerals that creates a problem. The silt from sur-
face runoff within the quarries and pits is carried into the
natural system, decreasing water clarity and oversilting
the receiving waters.
Federal legislative proposals in both the Senate and the
House of Representatives would give the U.S. Environ-
mental Protection Agency authority to require each State
to conduct a nonpoint source pollution management pro-
gram. This legislation defines for EPA what the programs
should do, including requiring that the States set goals
and a time schedule for attaining those goals.
To date, there is little data on noncoal mining stormwa-
ter runoff on which to base such goals. The EPA has an-
nounced that, beginning Dec. 31, it will require mining
operations to file explanatory information, detailing their
runoff situations, possible pollutants that the water may
assimilate, and the conveyances used for directing runoff.
In essence, mining operations will be filing for National
Pollutant Discharge Elimination System (NPDES) permits.
In the past, EPA has considered any conveyance of
stormwater to be a point source. This has caused prob-
lems for some quarries, particularly those where runoff
from adjacent, undisturbed lands encroaches on their op-
erations. Operators have attempted to redirect the runoff,
preventing any possible pollution from the quarry and
avoiding operating problems related to the excess water.
The problem is in the logic.
If the mining operator directs the stormwater from the
adjacent, undisturbed land away from his operation, then
the mining operation itself is not contributing to the pollu-
tion of the water. Yet, the EPA defines the miner's convey-
ance—a nonpolluting entity—to be the point source, and
the operator becomes liable for the quality of the water
handled.
A second problem faced by quarry operators is posed
by the ground water and runoff from the operation itself.
Operating a quarry on a dry site is the ideal situation.
Mining operators divert the ground water and runoff water
to quarry sumps and, depending on its quality, pump the
water to discharge or further treatment.
No system is 100 percent effective. But the systems do
work, and the amount of water pollutants from a noncoal
mining operation can be economically controlled within
reasonable limits.
Abandoned quarries may be a problem. Most aban-
doned quarries—or sand and gravel pits—were either
abandoned during the heyday of the interstate construc-
tion program or are slated for reclamation in the near fu-
ture.
There is a real shift in how these lands are being man-
aged, a shift caused not only by legislative and regulatory
actions, but also by more responsible attitudes on the part
of the operators. Admittedly, most operators were coerced
into their first reclamation programs; but after one or two
experiences where they see the ultimate benefit—not only
to the ecosystem, but also to their pocketbook—many op-
erators are planning for a more profitable end use prior to
opening or expanding a site.
The National Crushed Stone Association (NCSA), now
the National Stone Association, has taken a leading role in
the effort to change operator attitudes toward reclamation
and rehabilitation.
The first program initiated by NCSA helped operators
handle their environmental problems. Committees within
the association developed handbooks, offering solutions
to air, water, and noise problems.
As a complement to that effort, a community relations
awareness program was expanded in 1975 into a compre-
hensive industry awareness program called About Face.
The first priority was to encourage operators to improve
the general appearance of their sites. This was done, and
continues to be done, through a recognition program in
335
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
which operators provide a panel of judges with illustra-
tions of their sites. The entries are evaluated and awarded
in three categories, depending on the level of beautifica-
tion achieved and the success of their efforts to reduce air,
water, and noise pollution problems.
Another program instituted under the About Face um-
brella was a competition for students in landscape archi-
tecture programs, cosponsored by NCSA and the Ameri-
can Society of Landscape Architects. This program
addresses reclamation attitudes. Students work with oper-
ators, evaluating the existing site and the various environ-
mental and community factors that affect the operation.
From there, the students develop plans for the operator to
improve the site's appearance, reduce its environmental
impact, and begin the process of preparing the land for an
end use. The students also provide a proposed end use
that considers community plans and needs.
This cooperative program is widely accepted as having
a direct impact on changing attitudes toward reclamation.
Not only are operators becoming more amenable to recla-
mation, but there is an increasing pool of professionals in
landscape architecture who have experienced and now
understand the unique problems related to quarry recla-
mation. The program has been expanded to include the
National Sand and Gravel Association as a cosponsor,
further broadening its reach into the aggregate industry.
NCSA has also sponsored a number of programs to
educate operators about the techniques and consider-
ations of land reclamation.
Current trends in demographics have opened new
doors to aggregate mining reclamation. The natural ten-
dency toward reclaiming sites for recreation areas is now
only one of a multitude of alternatives that include indus-
trial parks, residential areas, and school campuses.
Water handling remains a primary obstacle to develop-
ment. But preplanning has done much to resolve those
problems. A good example is a closed quarry in New Eng-
land, now being developed as an office/residential com-
plex. Like many old quarries, this operation was in a rural
area near a city when it opened and now is enveloped by a
suburban environment.
More than 10 years prior to closing the operation, the
owners realized that, at their location, land could yield
prime rates—if it had not been a quarry. They hired a team
of engineers to evaluate their site and define for them the
best way to create salable land at closing. Using the plan,
they continued to quarry stone for 10 more years—shap-
ing the land to fit a design—then sold the property. The
property became valuable to the operator and to the de-
veloper. The land shaping allowed the developers to use a
series of ponds and fountains to handle the excess water
on the site, thereby turning a potential liability into an
asset.
Another example of a second land use for quarries that
converts water liabilities into assets is in Fairfax County,
Virginia. There, a mined-out quarry has been designated
for water storage. In addition to clarifying the water before
it goes into the neighboring river, the site stores the sedi-
ment, which reservoirs cannot do, and becomes a source
for additional water, alleviating the once frequent water
shortages in this fast-growing area.
Worked-out quarries have been converted into indus-
trial/commercial sites, lakeside residential communities,
water reservoirs, botanical gardens, recreational areas,
wildlife habitats, and even college football stadiums.
There is a new view of reclamation in the industry today.
It is a view that reclamation can eliminate many negative
effects aggregates industries may have on communities
or environments. Reclamation and reuse of quarried lands
is an exciting and challenging undertaking. Quarrying
need not be viewed as an undesirable, environmentally
detrimental use of land, but rather a transitional use of
land that enables man to build today while at the same
time creating an environmental asset for the future. It is, in
fact, an exercise in multiple land use planning.
336
-------
RURAL ISSUES: NONCOAL MINING AND ABANDONED
LAND RECLAMATION
GARY UEBELHOER
AMAX Chemical Corporation
Lakeland, Florida
Whenever someone from the phosphate mining industry
in Florida attends a national conference, he must first de-
fine the type of mining we perform in Florida and describe
the differences between surface mining of phosphate rock
and surface coal mining. While both minerals are recov-
ered using surface mining techniques, the differences are
significant. Therefore, I would like to describe our mining
and mineral recovery processes in Florida so our perspec-
tive on nonpoint source issues can be appreciated.
In the heart of central Florida, about midway between
Disneyworld and the Gulf Beaches, lies the largest con-
centrated deposit of phosphate rock in the world. It is
called the Bone Valley Formation. This deposit is of ma-
rine origin, formed during the Miocene and Pleistocene
ages some 15 million years ago. It lies within 30.48 m (100
feet) of the land surface and has supplied approximately
85 percent of America's, and approximately 33 percent of
the world's, requirements for phosphate fertilizers since
the deposit was located in 1881. Currently, 23 mines oper-
ating within the Bone Valley area produce 42 to 45 million
tons of phosphate rock per year. Estimates by the industry
as well as by the U.S. Bureau of Mines suggest that this
deposit could continue to be mined for another 100 to 200
years, pending economic, political, and environmental
conditions.
This massive orebody is located at the approximate
midpoint between Disneyworld and the Gulf of Mexico
beaches. Some people equate this to strip mining the Gar-
den of Eden, and with 4 of the 10 fastest growing cities in
America located within an hour's drive of the deposit, one
can begin to understand public expectations of this indus-
try. Although our history is deep and our future geologi-
cally far-reaching, the political pressures placed upon us
on environmental issues are as stringent as those placed
upon any other industry in America. After all, someone
who spent 30 years working in the heart of Cleveland does
not want to move to Florida to live next door to a strip
mine, nor to live 48 km, (30 miles) downstream and see
the nearby river clouded with mining effluent, whether it is
discharged through a point source or comes in the form of
runoff in a nonpoint source incident.
Florida's climate magnifies the need to control nonpoint
sources of pollution because the average annual rainfall is
135 cm (53 in.) per year. In 1984, our mine experienced
180 cm (71 in.) of rainfall. If the precipitation was spread
statistically throughout the calendar year, management of
surface water issues would be much easier; however, av-
erage monthly precipitation can range from 5 cm (2 in.) in
what we term the dry season in April and May to 33 cm
(13 in.) during our summer rainy season in August and
September.
The environmental characteristics of the land prior to
mining include citrus groves, pine/palmetto scrub land,
pastures, small isolated wet/dry marshes, hardwood bot-
tomland, and wetland forested areas. To produce the 40-
plus million tons of phosphate rock necessary to satisfy
agriculture's demand for phosphate fertilizers, our indus-
try mines approximately 2,600 ha (6,500 acres) of this
land per year.
Phosphate rock ore is recovered using a dragline sur-
face mining technique similar in appearance to what you
see in surface coal mining operations. Medium size drag-
lines carrying 15 to 57 m3 (20 to 75 yd3) buckets first strip
the overburden or barren sand/soil off the phosphate ma-
trix, then excavate the phosphate matrix and place it into
slurry pit wells. From there the phosphate is slurried into a
65-percent water and 35-percent solids slurry for hydraulic
pipeline transport to a central beneficiation plant or mill.
During mining, the total depth of the excavation can range
from 6 to 30 m (20 to 100 ft). The average is 9 to 10.5 m
(30 to 35 feet) total depth from land surface to the bottom
of the mined area. Because the water table is encountered
anywhere from land surface to 3 m (10 feet) below land
surface, water management in phosphate mines is a pri-
mary function. Large pumps dewater the mining areas
and transport this pit water to large recirculation holding
ponds wherein the water will be recycled for process water
use. Recirculation of process water provides 95 percent of
the total water requirements to produce a ton of phos-
phate rock.
Recovery of the phosphate rock from the ore, or matrix
as we call it, is conducted in a central beneficiation plant
where two types of product and two types of waste are
produced. The volume of these two waste products typi-
cally exceeds the void created by removal of the ore and
adds a waste disposal step between mining and reclama-
tion. These waste products consist of silica sand sepa-
rated from the phosphatic sands in a flotation process.
The silica sand is traditionally 99.9 percent silica, dewa-
ters very rapidly, is very permeable when compared to the
other waste product, approximates the overburden per-
meability, and does not possess favorable agronomic po-
tential. The other waste product is a mixture of clays usu-
ally consisting of montmorillonite, attapulgite, and some
kaolinite. The clays dewater very slowly, are very imper-
meable, but possess very favorable agronomic character-
istics when compared to the virgin or unmined overburden
or the sand waste product. The ratio of products to waste
is about equal. That is, an average phosphate mine pro-
duces 3 million tons of recoverable product per year; the
mine will have to dispose of about 3 million tons of clay on
a dry weight basis and 3 million tons of sand to recover
that quantity of phosphate rock.
Waste disposal practices within the industry vary, how-
ever; disposal of the sand and clay waste is integrated into
the reclamation process and is planned in advance of the
mining. Disposal of the sand tailings falls into five general
categories. First, the sand can be used as a construction
material for building dams for the clay waste disposal and
water recirculation. It also can be used to cap the consoli-
dated clay waste settling areas, or it can be admixed with
the clay to produce a sand/clay mix fill material. However,
the most typical use for the sand is to pump this material
from the beneficiation plant to mined areas where the
sand is used to backfill between the spoil piles to desired
post-reclamation elevations.
The waste clay produced at phosphate mines has been
used for various things from kitty litter to soil amendments
to increase the nutrient and moisture holding characteris-
tics. Some mines also admix clay to the sand tailings to
improve its dewaterihg characteristics and the pumping
characteristics of the sand tailings. However, the typical
337
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
sand/clay disposal process consists of pumping a 3-per-
cent solids clay slurry to large clay settling areas, allowing
the clays to settle through gravity from 3-percent solids to
approximately 30-percent solids, and then revegetating
the surface after dewatering is complete.
Land reclamation became mandatory in Florida in 1975.
Voluntary reclamation began long before that and, today,
two types of regulatory programs in Florida relate to phos-
phate-mined land reclamation. Lands mined prior to 1975
and not reclaimed either naturally or by man are covered
under the nonmandatory program, which is similar to the
abandoned lands program in coal mining States. Approxi-
mately 6,400 ha (160,000 acres) fall into this broad eligibil-
ity category. Reclamation of these "old lands" is financed
by the severance tax now paid by phosphate producers on
each ton produced. All other lands, those mined since
1975, fall into the mandatory reclamation liability program
also administered by the Florida Department of Natural
Resources. The mandatory program requires timing, post-
mining land use, revegetation, and other health and safety
measures similar to those imposed by coal mining states.
The waste disposal process results in a postmining ele-
vation and lithological profile and reduces the reclamation
process to grading and revegetation. Reclamation pro-
grams consisting of land and lakes involve regrading spoil
piles without adding waste products. The resulting lakes
with undulating shorelines and a variety of depths are
surrounded by shorelines and uplands used for pasture,
silviculture, citrus and intensive agricultural uses, and res-
idential and commercial uses.
Overburden fill reclamation consists of regrading over-
burden without adding sand or clay wastes, resulting in a
land form similar to that which existed prior to mining,
albeit at a different elevation. Because of the amount of
overburden combined with the depth to the water table, no
lakes are developed in these areas. We have already dis-
cussed the sand backfilling technique whereby sand is
pumped into the rows between the spoil piles. This recla-
mation process results in final land uses that require well-
drained soils. The sand/clay admixture backfilled areas
are a newer development yet to be proven in the form of
the final land use. Agricultural crops requiring well-drained
soils will likely not be used in these areas. The nutrient
and moisture-holding capacity increased by the admixing
of clay sand will result in high yield forage and pasture
uses. Lastly, reclamation of clay settling areas is a process
which consists of drying and developing a clay crust on
the reclaimed surface followed by agriculture and in some
case natural systems. Forage yields from these clay soils
are so far superior to virgin land that the use of clay set-
tling areas for these purposes is considered highly desir-
able by the agricultural industry.
Reclamation timing in phosphate mining varies depend-
ing upon whether the acre just mined is scheduled to
receive waste products or not. If wastes are not to be
added to a mined area and reclamation is limited to grad-
ing overburden and revegetation, reclamation is typically
completed, including a growing season to certify success-
ful revegetation, within 2 years of the date of excavation. If
sand tailings are admixed, the 2-year figure increases to
approximately 3 years. Clay disposal sites are typically
used for 10 to 15 years for filling; once filling is completed,
dewatering, drying, crusting, and revegetation can be
completed within 5 years.
With respect to overall progress, reclamation lags be-
hind mining during the first half of a given mine's life as
sites are set aside for waste disposal purposes. Once
waste disposal sites have been set aside, the rate of min-
ing equals the rate of reclamation. Thus, when a given
mine is fully depleted, usually 75 percent of the mined
area has been reclaimed and released from reclamation
liability. Reclamation of the remaining 25 percent usually
is completed within 5 years after the mine-out date.
Because the reclamation of old lands or nonmandatory
lands is limited by severance tax revenues versus the in-
tent of landowners to reclaim the land, the cash flow from
the trust fund dictates the rate of reclamation. At the cur-
rent severance tax production rate and trust fund interest
generation rate, it appears as if all eligible old lands where
the owner of the property wishes to participate will be
reclaimed no later than 1995. This equals a reclamation
rate of approximately 2,000 ha/yr (5,000 acres/yr).
Postreclamation land uses have been mentioned; how-
ever, it might be useful to go through the reclamation re-
vegetation land-use options. The first option is surprising.
Environmental groups and regulators are all beginning to
recognize the need to leave certain lands unreclaimed
because of the success of natural reclamation processes
in some places. As a matter of fact, some of the old lands
have been deemed ineligible for the nonmandatory recla-
mation refund program because their environmental char-
acteristics are so high from wildlife habitat and recrea-
tional perspectives that the Agency refuses to pay to
convert them to houses or citrus groves.
Citrus is a very profitable crop and is, of course, Flori-
da's best known ag crop. Citrus groves are being planted
and cultivated on the overburden fill type reclamation and
on the sand backfill areas. These two waste disposal op-
tions are the only logical sources for citrus groves be-
cause the groves must be very, very well drained, thereby
precluding the use of any sand/clay or clay fill areas. Ex-
tremely rich soil is not necessary for citrus planting. Plant-
ing citrus trees between alternating rows of land and lakes
in the land-and-lakes process is gaining popularity be-
cause the heat released by the lakes during cold winter
nights prevents freezes. Groves on reclaimed lands lying
southeast of lakes created by the mining process have
never frozen.
Pasture land was the first and most logical choice for
much of the reclaimed land because that was the premin-
ing land use. Florida is the largest cattle State east of the
Mississippi and ranks third behind Texas and Colorado in
beef production.
Row crops also present a postmining land use option in
that they produce high land returns and can capitalize on
Florida's climate. The row crops we speak of are lettuce,
carrots, radishes, cucumbers, strawberries.and other gar-
den vegetables. Admixing some waste clay with regraded
overburden or sand tailings produces a superior soil for
these types of vegetables because of the water and nutri-
ent holding capacity of the clay as compared with the
native soil.
Residential and commercial development has occurred
near surrounding communities typically on sand fill and
land-and-lake areas. A number of lakefront properties
have been created by the mining industry and, as most of
you know, lakefront property draws a premium over a
standard building lot.
Finally, during the past 10 years, the mining industry
has made significant progress and great strides in re-
claiming natural systems, both uplands and wetlands. The
industry has agreed to acre-for-acre reclamation of wet-
lands in regulatory proceedings both in terms of total acre-
age as well as "type for type," meaning marsh or herba-
ceous wetlands versus hardwood swamp forest. Two
companies have received the U.S. Fish and Wildlife Serv-
ice award for outstanding citizenship because of wetland
reclamation projects. These projects have moved from the
experimental research stage because reclamation of wet-
lands is no longer disputed. By allowing wetland reclama-
tion to progress, more acres of phosphate should be re-
covered in the future than in the past. Historically, mining
338
-------
NONCOAL MINING AND ABANDONED LAND RECLAMATION
companies have had to mine around wetlands of any sig-
nificance, thereby leaving that ore in the ground forever as
it is not cost efficient to mine a parcel of less than 200 ha
(500 acres). Further, the industry takes great pride in a
survey environmental groups conducted in Florida over
the past decade in which phosphate mine reclamation of
wetlands was considered perhaps the greatest environ-
mental improvement of the decade. In fact, reclamation of
mined lands allows one of the few opportunities to recre-
ate natural systems, whether they are uplands or wet-
lands, of any development option for raw undeveloped
lands. In only a very few places in Florida are developers
creating wetlands, let alone at the rate the Florida phos-
phate industry is.
Obviously, without some steps to mitigate the potential
for water quality degradation, phosphate mining in Florida
could become a significant nonpoint source of pollution.
This would principally be in the form of suspended solids
or turbidity, since the mineral we mine does not dissolve in
water. Data collected from our process water streams indi-
cate that, with the exception of turbidity, a phosphate
mine's process water stream meets Federal drinking wa-
ter standards and can comply with applicable stream wa-
ter quality standards. Therefore, the question that needs
to be addressed with respect to nonpoint sources of pollu-
tion is to describe the steps necessary to control nonpoint
sources and to minimize the potential for such releases.
During the mining process, all runoff within disturbed
areas is controlled through a series of perimeter ditches
constructed with small construction draglines and de-
signed to drain into the central clay settling/water recircu-
lation system. As a result, runoff is captured and allowed
to settle and clarify in the water recirculation system for
reuse as process water or for ultimate discharge through a
permitted NPDES discharge point. A perimeter ditch will
be constructed by placing the spoil outside the collection
system to act as a small .9- to 1.2-m (3- to 4-foot) dike
encircling and containing all runoff, either process water
spills or rainfall. However, when placing spoil outside the
perimeter ditch, the spoil pile berm needs to be grassed
so that erosion from the berm does not defeat the purpose
of its construction.
A second important step is to reclaim land as quickly as
possible and practical. This reduces the amount of proc-
ess water and rainfall runoff diverted to the clay waste
settling/water recirculation system and makes it more
manageable. Reclamation timing is, however, important
because of Florida's unique wet/dry climate cycle. Recla-
mation grading should take place during the fall and
spring dry seasons to minimize erosion during the active
earth-moving phase when barren land surfaces are
present. Furthermore, the ideal time to finish grading on a
reclamation project is around Thanksgiving; such timing
allows revegetation to begin at the point when gentle win-
ter weather increases the survivability of the planted spe-
cies and decreases the need to irrigate and to use tempo-
rary erosion control measures.
Florida's reclamation rules require tree planting at a
rate of 200 trees per acre with understory grassing to
achieve a minimum of 80 percent cover at the end of one
year. This further minimizes the nonpoint sources of re-
claimed lands. In fact, with little difficulty, very lush vegeta-
tive covers return to reclaimed land within a matter of
weeks of the seeding of those lands.
In fact, probably the largest, most effective means for
long-term control of nonpoint sources of pollution is in the
reclamation design of wetlands, lakes, and stream and
creek channels. By properly designing the lakes, green-
belts, wetlands, and stream systems, phosphate mining
creates additional off-stream storage capability, thereby
increasing base flow during the dry season and reducing
runoff during the wet season. As a result, the hydrologic
cycle variation on an annual basis declines. Concurrent
with that is a reduction in the sediment loading caused by
runoff. Further, because the permeability of most re-
claimed land is as high or higher than virgin land, ground
water recharge rates increase, and runoff rates decrease.
Are all of these measures working? Yes they are. And to
that end, the phosphate industry is spending more than $2
million per year researching and developing techniques to
further improve their ability to reclaim land to more desir-
able uses, and to find better ways to minimize problems in
the interim. In addition, the severance tax paid by the
phosphate industry is used for two very important pur-
poses related to nonpoint source pollution. First, the phos-
phate severance tax funds the Conservation and Recrea-
tion Lands Trust Fund, which is used to purchase
environmentally sensitive lands and keep them from being
developed by other sources.
Secondly, the phosphate severance tax pays for the op-
eration of the Florida Institute of Phosphate Research de-
signed to find solutions to public and environmental prob-
lems associated with phosphate mining and processing.
The Institute is researching reclamation and mining, and
to this end has committed over $4.6 million since it was
formed in 1978. As a result of these efforts, combined with
the private efforts of the industry's member companies, I
am confident that our point source pollution problems
have already peaked and probably will only decrease in
the future.
339
-------
NONCOAL MINERAL MINING AND RECLAMATION (CURRENT AND
ABANDONED OPERATIONS) IN THE TENNESSEE RIVER BASIN
JACK A. MUNCY
Reclamation Specialist
Tennessee Valley Authority
Norris, Tennessee
Surface mining is one of the most complex, critical, and
emotionally charged environmental issues of our day.
Throughout the United States, surface mining is used to
recover about 50 different materials essential to our indus-
trial society. In the seven-State region of the Tennessee
Valley, over 25 noncoal minerals, including mica, feldspar,
kaolin, manganese, copper, phosphate, sand, and brown
iron ore are obtained by surface mining.
Most Tennessee Valley States now have laws regulating
surface mining for certain noncoal minerals. Georgia and
Tennessee passed a noncoal minerals surface mining law
in 1968, followed by Alabama and Virginia in 1969, North
Carolina in 1971, and Kentucky in 1975. These laws pri-
marily require that the environmental impacts from mining
be minimized and that mined sites be stabilized through
vegetative measures.
These environmental protection standards have been
received by the mining industry with varying degrees of
success. Tennessee's 1968 reclamation law was consid-
ered ineffective until around 1972, 4 years after it was
enacted. In addition, the portion of the Tennessee law
addressing sand mining was repealed in 1981, and it was
also noted that limestone, dimension stone, marble, and
chert were not regulated by the act.
It is interesting to compare the similarity of the history of
State legislation of coal mining before passage of the Fed-
eral Surface Mining Control and Reclamation Act of 1977
to the current State legislation regarding noncoal mining.
The Federal Act established the first national standards
for surface coal mining and reclamation, including provi-
sions for reclaiming abandoned coal mine lands on a pri-
ority system.
While abandoned noncoal mineral mine lands overall
may not have the same magnitude of problems as coal
mines, a pressing need still exists for their reclamation
since they affect local environments just as severely. In the
Tennessee Valley approximately 9,315 ha (23,000 acres)
of these lands were created before State laws fixed re-
sponsibility for reclamation (Fig. 1). Furthermore, hun-
dreds of hectares of abandoned sand mines are still being
created annually in Tennessee.
These abandoned mine lands are society's legacy, and
we all share in the responsibility for the problems they
cause as well as the efforts necessary to return them to a
productive part of the region's natural resource web.
MANGANESE
162 H (VA)
81 H (TN)
243 H
BROWN IRON ORE
527 H (AL)_
101 H (TN)
628 H
SAND
607 H (TN)
MICA, FELDSPAR, AND CLAY
145 H (NC)
PHOSPHATE
3,240 H
FARMLAND POTENTIAL
(TN)
COPPER
3.483 H
(TN&GA)
PHOSPHATE
810 H
ENVIRONMENTAL PROBLEM LANDS
(TN)
TOTAL 9,156 HECTARES
Figure 1.—Estimated hectares of abandoned noncoal mineral surface mine lands in the Tennessee Valley in need of
reclamation.
340
-------
NONCOAL MINING AND ABANDONED LAND RECLAMATION
Many environmental problems are caused by unregu-
lated mining of minerals. A major problem is soil erosion
and subsequent sedimentation. This can alter the chemi-
cal balance of the water, bury aquatic organisms, change
feeding and spawning habitats, and suffocate fish by coat-
ing their gills. Sediment clogs streams, rivers, and reser-
voirs, reducing the flow-carrying capacity of streams and
the flood detention capacity of reservoirs, which increases
flood damage potential. In addition, sediment damages
hydroelectric units, reduces recreational value of streams
and reservoirs, and adversely affects water supply sys-
tems by increasing treatment costs, causing excessive
wear on equipment through abrasion, and clogging or
covering intake pipes.
While certain noncoal mineral mining (reclamation) is
covered under State laws, few provisions have been made
to pay for corrective reclamation of earlier mining activi-
ties. Only Tennessee and Virginia have budgets for mini-
mal, on-the-ground reclamation efforts. In recent years,
the cooperative efforts of many participants have helped
the Tennessee Valley Authority reclaim selective critical
lands. A listing of these efforts follows:
Mica, feldspar, and kaolin. TVA, in cooperation with
local, State, and Federal agencies, landowners, and the
mining industry, has treated most (239 ha/590 acres) of
the environmental problem mines in western North Caro-
lina. North Carolina passed legislation in 1984 to provide
financial support to complete the project.
Manganese. In 1983, the State of Tennessee, in coop-
eration with Johnson County Soil and Water Conservation
District and the Soil Conservation Service, established a
mine reclamation demonstration. Also in recent years, Vir-
ginia has reclaimed a few sites.
As part of TVA's rehabilitation efforts of the South Fork
of the Holston River basin, the agency joined hands with
the U.S. Forest Service to reclaim up to 41 ha (100 acres)
of critically eroding abandoned mine lands on the Mount
Rogers National Recreation Area. This work is now under-
way.
TVA is formalizing a plan to reclaim the remaining
unvegetated private lands in upcoming years in a cooper-
ative TVA/State/landowner project. The total hectares of
abandoned lands in the Holston River watershed is esti-
mated at 243 (600 acres). Virginia hopes to treat selective
sites in 1986.
Copper Basin. TVA has been involved with the environ-
mental problems of Copper Basin in the past; however, the
mining companies have been the most consistent recla-
mationists. In 1984, TVA initiated new demonstrations in
cooperation with SCS and the Tennessee Chemical Com-
pany. About 284 ha (700 acres) will be treated in 1985 by
pooling resources. Of the original 12,960 ha (32,000
acres) denuded, 3,483 ha (8,600 acres) still need reclama-
tion.
In recent years, the Tennessee and Georgia SCS have
implemented resource conservation and development
plans showing intensive and minimum degrees of recla-
mation. The local soil and water conservation districts
sponsored the projects.
Phosphate. In 1979, the State of Tennessee initiated a
cooperative program with SCS, local soil and water con-
servation districts, and landowners to reclaim abandoned
phosphate mines. The program goal is to restore these
unproductive lands to their premining land use—agricul-
ture. Through 1984, 170 ha (420 acres) have been re-
claimed, but no funding is available to continue this effort.
Of the 4,050 ha (10,000 acres) judged to be in need of
reclamation, about 810 ha (2,000 acres) are causing of-
fsite problems.
As part of the comprehensive cooperative natural re-
source development plan for the upper Duck River basin
in Tennessee, TVA reclaimed 59 ha (146 acres) of aban-
doned phosphate mines causing offsite environmental
damage. TVA, SCS, and several other Federal and State
agencies developed the plan to address several nonpoint
sources of pollution.
Sand. In 1982, Tennessee and SCS established a
small-scale reclamation demonstration in Benton County.
There are about 607 ha (1,500 acres) of abandoned sand
mines in Tennessee counties alone.
Brown iron ore. Abandoned brown iron ore mines exist
in northwest Alabama (527 ha/1,300 acres) and southwest
Tennessee (101 ha/250 acres). As yet no plans have been
made by the States or others to systematically reclaim
these lands.
SUMMARY
Although progress is being made to reclaim these erosive
abandoned mine lands through cooperative efforts, the
circumstances that allow these lands to develop have not
totally disappeared. Legislation regulating mining is a step
in the right direction, but these regulations must be en-
forced to minimize the environmental problems of surface
mining for resources other than coal. Also, mineral mines
not presently covered by State laws should be periodically
re-evaluated to ensure that their related mining activities
do not cause future environmental problems. More fund-
ing sources are needed for action programs to deal with
the abandoned mine lands that cause offsite environmen-
tal degradation.
341
-------
PHOSPHATE AND PEAT MINING IN FLORIDA
CAROL J. FALL
St. Johns River Water Management District
Palatka, Florida
INTRODUCTION
Florida is often envisioned as a land of beaches, tourists,
and orange groves. However, agriculture and mining are
important components of the State's economy. Florida is
first in the nation in production of nonmetallic minerals
and sixth in production of all minerals (State Fla. 1982).
Nearly 75 percent of the phosphate produced in the
United States is mined in Florida. Nationally, Florida ranks
second in peat production (Bond, 1984).
PHOSPHATE MINING
Florida's phosphate deposits are of sedimentary origin.
Shallow deposits, such as central Florida's Bone Valley
formation, are strip mined. Deeper deposits in eastern
Florida have the greatest long-term potential but require
specialized mining techniques, such as the borehold
slurry process (Fla. Dep. Environ. 1984).
The strip-mining process generally consists of (1) site
draining and dewatering, (2) logging of available timber,
(3) land clearing by bulldozer, (4) removal of overburden,
(5) mining of phosphate ore, (6) pumping of phosphate
slurry through pipeline, and (7) washing and flotation in
the beneficiation plant. This last step produces sand tail-
ings and a clay slurry, also called phosphate slimes.
These colloidal clays settle very slowly and are typically
stored in aboveground, diked impoundments. Approxi-
mately 170,000 acres have been strip mined in Florida,
producing 55,000 acres of slime ponds (State Fla. 1982).
Strip mining can substantially alter hydrologic condi-
tions, disrupting wetlands and intersecting small streams.
Ground water conditions are modified as dewatering
lowers the shallow aquifer and mining pits intersect aqui-
fers, sinkholes, or other karst features. Hydrologic altera-
tions are particularly significant since about 15 percent of
Florida's phosphate reserves are located in wetlands.
Strip mining affects surface water quality through dis-
charge of wastewater and land disturbances. Land clear-
ing and dewatering increase erosion and sedimentation of
receiving waterbodies. Discharge of process water ele-
vates levels of sulfate, fluoride, total phosphorus, nitrogen,
and dissolved and suspended solids in streams (Miller et
al. 1978). Processing plants contribute 65 percent of the
phosphorus load to the Alafia River (Wright, 1980). Spills
and dam breaks pose an infrequent threat to surface wa-
ter quality (U.S. Environ. Prot. Agency, 1978). An addi-
tional concern is radiation potentially produced as ura-
nium and its decay products brought to the land surface
by the mining operation (Fla. Def. Environ. 1984).
Ground water quality is affected by seepage into under-
, lying aquifers from slime ponds and ditches (Miller and
Sutcliffe, 1984). Removal of the overburden during strip
mining allows direct contamination of local or regional aq-
uifers in some areas of Florida (Wright, 1980). The initia-
tion of experimental borehole mining in north Florida has
raised additional ground water quality concerns. In this
technique, a series of large boreholes are drilled through
the ore deposit and high pressure water jets cut and slurry
the ore. The phosphate slurry is pumped to surface hold-
ing ponds for processing. When extraction is complete,
the 10m diameter cavity and borehole are sealed, using a
variety of techniques. Ground water contamination may
occur if the borehole penetrates the confining layer of the
artesian aquifer or by vertical seepage along the casing of
an improperly sealed borehole. In initial test results, water
level and water quality changes were observed following
collapse of the cavity roofs (Hampson, 1984).
Reclamation of strip-mined lands has progressed and
is mandatory for lands mined after July 1, 1975. Florida
statute requires reclamation of wetlands on an acre for
acre basis. However, data on wetland reclamation are in-
conclusive (Dames and Moore, 1983), generating much
opposition to proposed phosphate mining leases in the
Osceola National Forest. The U.S. Department of the Inte-
rior (1983) concluded that sufficient technological capabili-
ties do not exist to reclaim wetland hardwoods. Criteria
suggested to evaluate the effectiveness of wetland recla-
mation include vegetation diversity, water quality and
quantity, plant survival, and wildlife use (Dames and
Moore, 1983). Many reclamation schemes use artificially
regulated or augmented water levels, which may be unsat-
isfactory for long-term maintenance.
PEAT MINING
Peat deposits accumulate in a waterlogged environment.
Therefore, peat mining almost always takes place in wet-
lands. Major deposits of peat in southern Florida are lo-
cated in the Everglades and headwaters of the St. Johns
River. In north and central Florida, there are thousands of
small, scattered deposits.
Peat mining requires dewatering the site through drain-
age ditches or pumping. Overlying vegetation is logged or
cleared, and the overburden removed. The peat is exca-
vated and stockpiled for drying. As in phosphate mining,
the major impacts are altered hydrology and loss of habi-
tat. Mining alters flood water runoff response, ground wa-
ter elevations, surface flow patterns, and minimum stream
discharges (Bond, 1984).
Dewatering of peat mines produces discharges which
have low pH and elevated biochemical oxygen demand
(BOD), nutrient, organic, and solids concentrations. Peat
soils accumulate heavy metals, which may be released
during mining operations. A concomitant effect is the re-
duction in water quality improvements provided by the
wetland, such as sedimentation and denitrification.
Reclamation of peat mines is more haphazard than
phosphate mines. When dewatering ceases, the mined pit
fills with water. Within the St. Johns River Water Manage-
ment District, a reclamation plan is required based on the
following guidelines: 20 percent of the site will be recre-
ated as a littoral zone, a peat layer of .3 m or greater will
remain at the bottom of the excavation, and the site will be
mulched with a stockpiled or borrowed overburden with a
viable seed bank.
CONCLUSION
The major impacts of phosphate and peat mining are al-
tered surface and ground water hydrology and loss of
habitat, particularly valuable and irreplaceable wetlands.
Nonpoint source pollution problems result primarily from
land disturbance during mining operations. Surface and
342
-------
ground water quality are affected by phosphate-process-
ing plants. Some environmental impacts can be mitigated
by applying best management practices and reclamation
techniques.
REFERENCES
Bond, P. 1984. An Overview of Peat in Florida and Related Is-
sues. Bur. Geol. Fla. Dep. Nat. Resour., Tallahassee.
Dames and Moore. 1983. A survey of wetland reclamation proj-
ects in the Florida phosphate industry. Sponsored by Florida
Inst. Phosphate Res., Bartow.
Florida Defenders of the Environment. 1984. Phosphate Mining
in Florida: A Source Book. Environ. Serv. Center, Tallahassee.
Hampson, RS. 1984. Effects of hydraulic borehole mining on
ground water at a test site in northeast St. Johns County,
Florida. WRI Rep. 83-4149. U.S. Geo. Surv., Tallahassee.
Miller, J.A. 1978. Impact of potential phosphate mining on the
NONCOAL MINING AND ABANDONED LAND RECLAMATION
hydrology of Osceola National Forest, Florida. WRI Rep. 78-6.
U.S. Geo. Surv., Tallahassee.
Miller, R.L., and H. Sutcliffe, Jr. 1984. Effects of three phosphate
industrial sites on ground water quality in central Florida, 1979
to 1980. WRI Rep. 83-4256. U.S. Geolog. Surv., Tallahassee.
State of Florida. 1982. Testimony by the State of Florida before
the Subcommittee on Public Lands and reserved .water of the
Senate Committee on Energy and Natural Resources on leg-
islation to prohibit issuance of phosphate mining leases in the
Osceola National Forest. Office of the Governor, Tallahassee.
U.S. Department of the Interior. 1983. Environmental assess-
ment on state of reclamation techniques on phosphate mined
lands in Florida and their-application to phosphate mining in
the Osceola National Forest. Bur. Land Manage., Eastern
States Off.
U.S. Environmental Protection Agency. 1978. Final Areawide
Environmental Impact Statement for Central Florida Phos-
phate Industry. Vol. I. Atlanta, GA.
Wright, C.R. 1980. Water quality and mining. State Water Qual-
ity Management Plan. Dep. Environ. Reg., Tallahassee, FL:
343
-------
JOHN FORD
Missouri Department of Natural Resources
Jefferson City, Missouri
Missouri is a major producer of lead. One active and one
inactive mining area are located completely within the
State, and the State shares a second inactive mining area
with Oklahoma and Kansas. Hundreds of small vertical
shaft mines and hundreds of small to moderate-sized tail-
ings piles characterize the last of these areas, the Tri-State
lead-zinc area. The other inactive area, the Old Lead Belt,
has fewer mines and much larger tailings piles. Both ar-
eas adversely affect streams, contributing both heavy sed-
iment loads from eroding tailings piles and dissolved
metals (primarily zinc) from both surface runoff and seep-
age from flooded mines. Occasional tailings dam failures
result in deposition of huge amounts of sand and silt-sized
tailings that smother all existing aquatic life and seriously
degrade aquatic habitat for years. A total of 58 miles of
classified stream is affected.
Table 2.—Quality of runoff from tailings in the Tri-State
lead-zinc area compared to quality of Center Creek above
tailings area (Barks, 1977).
AMD SUG»CE WATEF5
Barks (1977) documented the Tri-State area water quality
problems in 1976. The ground water in flooded mines is
highly mineralized and has high levels of some dissolved
metals, particularly zinc and iron. Mineralization in the
mines has affected shallow wells in the area to some de-
gree but has not affected wells in the deep aquifer (Table
1). Considerable artesian flow and subsurface seepage
from flooded mines enters receiving streams even during
dry weather. The inflow of these ground waters is the pri-
mary source of zinc.
Most of the ground water flow recharges Center Creek,
which maintains a dissolved zinc concentration of about
500 fig/L, a figure five times the State of Missouri standard
for protection of aquatic life. Barks likewise found in-
creased mineralization and dissolved metals concentra-
tions in the surface runoff from the tailings (Table 2). This
runoff not only could elevate instream levels of certain
metals during a local rainfall but also could physically
move metal-rich tailings into stream sediments.
The combined effect of resurfacing ground water flows
and surface runoff has seriously undermined the ability of
portions of Center Creek to support aquatic life. Surveys
of aquatic macroinvertebrate benthos and of fish by the
author and others in the Missouri Department of Natural
Table 1 .—Ground water quality in the Tri-State lead-zinc
area (Barks, 1977).
Constituents1
Dissolved solids (mg/l)
Bicarbonate (mg/l)
Sulfate (mg/l)
Zinc (jig/l)
Iron (/jg/l)
Nickel (Mg/l)
Cadmium (^g/l)
Lead fcg/l)
Mines
1,030
140
580
9,400
5,100-
46
25
10
Shallow
wells
327
214
72
1,100
350
7
4
10
Deep
wells
207
195
28
70
30
0
0
6
Constituents1
Dissolved solids (mg/l)
Bicarbonate (mg/l)
Sulfate (mg/l)
Zinc (/tg/1)
Aluminum (/ig/l)
Lead frig/1)
Copper (jig/1)
Cadmium (^g/l)
Nickel (Mg/l)
Runoff
from
tailings
414
62
230
16,000
600
380
46
26
16
Center
Creek
134
136
8
20
30
4
0
1
2
'dissolved fraction only.
Resources have found that much less density and diver-
sity of aquatic life occurs in Center Creek than in other
streams of similar size and substrate.
The Old Lead Belt and Fredericktown areas are located
in east central Missouri. Fewer mines and fewer and
larger tailings piles characterize these areas. Mineraliza-
tion of water in flooded mines also occurs in this area but
has not been serious enough to impair the water's use.
Some communities withdraw water directly from the
mines rather than incur the expense of drilling a well.
Sulfate values average about 225 mg/L in the mine water
and frequently exceed the secondary (aesthetic) drinking
water standard of 250 mg/l. Sulfate concentrations in
ground waters of the area average 10 to 15 mg/l.
Two independent studies conclude that dissolved
metals have caused water quality problems in both the
Old Lead Belt and Fredericktown areas. A chat pile
(coarse sand-sized particles) in the Old Lead Belt contrib-
uted enough dissolved zinc to cause water quality stand-
ards violations in over 2 miles of Flat River Creek (Wixson,
1976). At Fredericktown, artesian flow from a flooded
mine caused violations of water quality standards for
nickel and cobalt in 2 miles of Saline Creek (Hufham,
1981).
'dissolved fraction only
The erosion of tailings into area streams causes greater
concern. These tailings contribute sand-and-silt-sized par-
ticles to streams at a much greater rate than surrounding
lands. This constant rate of erosion occasionally is in-
creased dramatically by the catastrophic failure of a tail-
ings dam. Three such failures have occurred on dams in
the Old Lead Belt and in Fredericktown. On such occa-
sions sediment completely fills stream channels and ei-
ther buries or displaces all aquatic life. At Fredericktown,
sediment severely affected the aquatic benthos (Duchrow,
1983), and the benthic community remained abnormal for
that locality throughout the 1-year study. Czarnezski
(1981) noted similar effects on 5 miles of Big River with
another 25 miles less severely affected.
Studies of the distribution of mussels on Big River and
Flat River Creek (Buchanan, 1980) clearly show the degra-
344
-------
NONCOAL MINING AND ABANDONED LAND RECLAMATION
Location of
Tailings
11
Jl
River Mile
Figure 1.—Number of mussel species found in Big River,
St. Francois County, MO.
dation in aquatic habitat (Fig. 1). While occasional dis-
solved zinc problems on Flat River Creek may also con-
tribute to the lack of mussels there, no problems with any
dissolved heavy metals exist on Big River. Researchers
believe that deposition of tailings is the sole reason for the
mussels' disappearance.
STRATEGIES FOR IMPROVEMENT
No concerted effort to address the problems presented by
lead-zinc mines and their tailings has been made. How-
ever, the author has noted improvements in water quality
in the Tri-State area. Time trend analysis of dissolved zinc
levels in Center Creek show a statistically significant + (P
< .05) decline in concentrations from 1972 to present.
Two possible reasons suggest themselves. Barks (1981)
has estimated that 80 percent of the tailings in the Tri-
State area have been removed. The mineralization in this
part of the State took place within a zone of very cherty
material, and the tailings have been used both for road
building material and for sand blasting. A second possibil-
ity is that the oxidation rates of the pyritic materials in the
flooded mines are declining over time (Warner, 1977; Ste-
wart, 1980).
By contrast, in the Old Lead Belt neither improvement
nor the expectation of significant improvement exists. The
lead-zinc ores in this part of the State were extracted from
dolomites. Some tailings have been removed for agricul-
tural lime, but the amounts are trivial compared to the
tailings that remain. Recent work (Wixson et al. 1983)
showed that using Old Lead Belt tailings as agricultural
lime would result in greater levels of heavy metals in vege-
tation relative to other agricultural limes, but such levels
would not exceed recommended dietary intake.
State and local officials have made two attempts to turn
these large tailings piles from environmental liabilities into
environmental assets. One county uses part of a large
tailings area as a sanitary landfill. This has the advantage
of not having to use other land as a landfill, but has the
disadvantage of being almost impossible to vegetate.
Thus, wind erosion is constantly uncovering and blowing
litter and garbage from the site. In addition, the lure of
hundreds of discarded tires at the landfill poised so near
the top of a long, steep embankment has encouraged
people to roll large numbers of tires down into Big River
where it borders the tailings dam.
A second tailings area has been converted into a State
park that promotes the use of outdoor recreational vehi-
cles within the park. With a great deal of effort, this tailings
area was the most successfully vegetated area in the Old
Lead Belt prior to its conversion to a State park. Now, the
recreational vehicle use appears to be causing some loss
of vegetation.
CONCLUSION
Although removal or proper on-site use of tailings may yet
be realized as viable alternatives, stabilization of tailings
in place seems the best strategy at present for reducing
nonpoint source problems from tailings in the Old Lead
Belt and Fredericktown areas. Planting vegetation on the
large flat areas of tailings can control wind erosion. Water
erosion control requires good maintenance of the existing
internal drains and protection of the dams which are made
of sand-sized material. Sediment traps between the tail-
ings and the area streams need to be constructed and
periodically cleaned.
The water quality problems from abandoned lead-zinc
mine areas in Missouri are well documented and well
known to Federal, State and local government agencies
and private organizations. No existing source of funding,
however, provides for the implementation of realistic solu-
tions.
REFERENCES
Barks, J.H. 1977. Effects of abandoned lead and zinc mines and
tailings piles on water quality in the Joplin Area, Missouri. U.S.
Geolog. Surv. Water Resour. Invest. 77-75.
Barks, J.H. 1981. Personal communication. U.S. Geolog. Surv.,
Rolla, Mo.
Buchanan, A.C. 1980. Mussels (Naiades) of the Meramec River
Basin. Aquatic Ser. No. 17. MO Dep. Conserv. Columbia.
Czarnezski, J. 1981. Personal communication. MO. Dep. Con-
serv. Columbia.
Duchrow, R.M. 1983. Effects of lead tailings on benthos and
water quality of three Ozark streams. Trans. Mo. Acad. Sci.
17:5-17.
Hufham, J. 1981. A Baseline Study of the Heavy Metal Content
of Open Waters of Fredericktown, Missouri. Univ. Missouri,
Rolla.
Stewart, D.R. 1980. Water resources contamination from aban-
doned zinc-lead mining-milling operations and abatement
activities. Ozark Gateway Counc. Gov., Joplin, MO.
Warner, D.C. 1977. Alternatives for control of drainage from in-
active mines and mine waste sites. Joplin Area, Missouri.
Ozark Gateway Counc. Gov., Joplin, MO.
Wixson, B.G. 1976. Missouri Lead Study. Univ. Missouri, Rolla.
Wixson, B.G., N.L. Gale, and B.E. Davies. 1983. A Study of the
Possible Use of Chat and Tailings from the Old Lead Belt of
Missouri for Agricultural Limestone. Univ. Missouri, Rolla.
345
-------
Salinity: A Nonpoint
Source Problem
MANAGING HEADWATER AREAS FOR CONTROL OF SEDIMENT
AND SALT PRODUCTION FROM WESTERN RANGELANDS
WILLIAM L. JACKSON
ERIC B. JANES
BRUCE R VAN HAVEREN
Bureau of Land Management
Denver Service Center
Lakewood, Colorado
ABSTRACT
Control of nonpoint source water pollutants poses special
challenges on western rangelands. The public range-
lands managed by the Bureau of Land Management are
often characterized by unstable sedimentary geologic
parent material, semiarid climate, and sparse vegetation.
Intense summer thunderstorms produce locally heavy
runoff. Where marine shales are exposed at the surface,
their sediments often contain high concentrations of solu-
ble salts. The immense size of the sediment- and salt-
producing areas poses treatment problems, both from a
technical and economic standpoint. Treatment objectives
include retention of runoff water and stabilization of ac-
tively eroding gullies in headwater areas. Watershed im-
provement projects are designed to provide multiple re-
source benefits, such as water supplies for livestock and
wildlife, improvement of water quality, and retention or
enhancement of site primary productivity. Two represent-
ative watershed improvement projects are described:
Sheep Creek Resource Conservation Area in southern
Utah and Lower Missouri Creek Tributaries Stabilization
Project in northwestern Colorado.
Sediment and salts are major nonpoint source, water qual-
ity constituents on western rangelands. They occur natu-
rally in runoff but may be increased by management activi-
ties and become issues when they affect beneficial uses
of water. Sediment production is highest on lands with
steep slopes, sparse vegetation cover, and erodible
soils—common conditions on western U.S. rangelands
(U.S. Dep. Agric., 1980). Salinity is a problem in the Colo-
rado River Basin where eroded sediments have naturally
high soluble salt contents (Hawkins et al. 1977; U.S. Dep.
Inter. 1978). Public lands in the upper Colorado River
Basin produce about 650,000 metric tons of salt annually,
or about 8 percent of the upper basin salt load from diffuse
overland sources (U.S. Dep. Inter., 1978).
We recently reported on the approach the Bureau of
Land Management (BLM) uses to identify nonpoint
sources of pollution on public lands (Van Haveren et al.
1985). In this paper we describe the specific strategies
and control technologies BLM has employed to reduce
salt and sediment production on western rangelands.
CONTROL PLANS
Developing effective salt and sediment control plans re-
quires: (1) the establishment of resource management ob-
jectives, (2) the identification and quantification of man-
ageable hydrologic processes, (3) the investigation of
cause and effect relationships, (4) the stratification of
treatment areas, and (5) the selection and evaluation of
alternative treatment techniques.
Whichever watershed management techniques are
eventually implemented, multiple resource values may be
affected, including forage production, water supplies for
livestock and wildlife, improved water quality, enhanced
wildlife habitat, reduced soil loss, control of downstream
flooding and channel erosion, and reductions in down-
stream sediment and salt delivery. The overall goal in de-
veloping sediment and salinity control plans is to provide
an optimum mixture of resource benefits consistent with
overall resource management objectives.
Establishment of Objectives
Objectives for controlling salt and sediment should relate
347
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
to both the processes to be influenced and the manage-
ment goals to be achieved. In establishing management
objectives for sediment and salinity control, corresponding
objectives need to be established for related, affected re-
source values. This will enable a meaningful analysis of
tradeoffs associated with alternative treatment tech-
niques. If possible, objectives need to be quantified so that
progress in achieving them can be effectively monitored
and evaluated.
Identification and Quantification of
Manageable Processes
The identification and quantification of manageable proc-
esses and variables is accomplished as part of the water-
shed analysis procedure (Solomon et al. 1982; Gebhardt,
1985). However, more detailed or site-specific quantifica-
tion may be required for project design or for ranking indi-
vidual treatment alternatives. Most sediment and salinity
control projects require information on both long-term and
runoff and sedimentation rates, and single-storm design
values for runoff, peak flows, and sediment yield.
In quantifying manageable salt and sediment proc-
esses, it is useful to distinguish between natural and man-
agement induced problems. Generally BLM prefers to cor-
rect management induced problems rather than control
natural processes.
Investigation of Cause-and-Effect
Relationships
Distinguishing between causes and effects is very impor-
tant when evaluating sediment and salinity problems. For
example, high gully erosion rates may be the result of
local or regional changes in base-level controls, or they
may be caused by runoff in excess of the thresholds, the
reduction of streamside vegetation, or some combination
of causes. Proper identification and quantification of the
causes of a problem will more likely lead to the proper
selection of treatment techniques than will a simple quan-
tification of the problem symptom (such as erosion rates).
Of particular importance in investigating salinity issues is
the relationship between sediment and salt. Where highly
saline soils are eroding, we assume that controlling sedi-
ment will also control salt. However, other salt transport
mechanisms, including interflow and ground water flow,
may not be manageable by controlling runoff and erosion.
Stratification of Treatment Areas
Where large watersheds (>50 km2) are to be treated, we
recommend dividing the area into treatment units. The
stratification is based on topographic considerations, in-
cluding soils and vegetation, salt and sediment source
areas, locations where controlling processes can be man-
aged, and treatment potential. After identification, treat-
ment units are ranked, based on both the sediment or salt
production rating and treatability of the area. The applica-
tion of this concept to the Lower Wolf Creek watershed is
discussed later in this paper.
TREATMENT TECHNIQUES
Controlling salinity in surface runoff from rangelands is
closely related to controlling soil erosion. Vegetation cover
is usually the most important management variable influ-
encing runoff and erosion rates on rangelands.
Therefore, vegetation management, either directly
through vegetation manipulation or indirectly through the
design and implementation of livestock grazing plans, is
an important erosion and salinity control technique. How-
ever, on the most highly saline rangelands, maximum po-
tential cover is usually too low to provide meaningful con-
trol of surface runoff and erosion. In these cases, or in
situations where the watershed's condition is so severely
degraded by past management practices that natural re-
covery will be inefficient, mechanical land treatments and
structural alternatives may be the most effective erosion
and salinity control techniques.
Vegetation Management
Vegetation cover, including canopy cover, ground cover,
and litter, reduces upland soil loss by protecting soil from
direct raindrop impact and by reducing surface runoff ve-
locities. Vegetation also intercepts rainfall and enhances
soil infiltration properties, thus reducing runoff volume and
its erosive capacity, both on hillslopes and in stream chan-
nels.
Livestock grazing affects vegetation cover by influenc-
ing species composition, vigor, production, and forage
use. Most studies have shown that runoff and erosion
increase with grazing intensity (Lusby, 1979a; Gifford and
Hawkins, 1978). Generalized relationships between live-
stock grazing and vegetation cover, however, have not
been forthcoming. Common erosion estimation tech-
niques, such as the Universal Soil Loss Equation (Wis-
chmeier and Smith, 1978), that require information on
vegetation cover are difficult to apply given information
only on livestock grazing or forage use. Thus, it is difficult
to accurately predict the effects of livestock grazing sys-
tems on erosion. Nevertheless, proper land use, including
well-designed grazing systems, is the preferred method of
achieving watershed management objectives (Moore et
al. 1979; Van Haveren et al. 1985).
The most common techniques for direct vegetation ma-
nipulation on rangelands include pinyon-juniper control
and big sagebrush control. Both techniques involve elimi-
nating pinyon-juniper or big sagebrush stands by mechan-
ical or chemical means or burning. Either native grasses
and forbs are permitted to reestablish naturally or grasses
are planted. General conclusions concerning the effec-
tiveness of vegetation conversions in reducing runoff and
soil loss on rangelands are not available. However, the
many discrepancies in the literature suggest that vegeta-
tion manipulations may not be reliable techniques for con-
trolling sediment and salinity. In many cases, vegetation
conversions have resulted in more desirable forage spe-
cies for livestock, but have not significantly reduced runoff
or soil loss (Williams et al. 1972; Gifford et al. 1970; Gif-
ford, 1972; Gifford and Busby, 1974; Blackburn and Skau,
1974). In some cases involving sagebrush conversion to
grass (Lusby, 1979b) runoff and sediment yield have been
reduced significantly.
Mechanical Land Treatments
Mechanical land treatments involve soil tillage techniques
such as contour furrowing, ripping and pitting. Tillage is
generally applied to increase infiltration volumes. This
may be accomplished by increasing infiltration capacities
or depression storage (thus, the time available for infiltra-
tion), or both. When successful, runoff and erosion can be
reduced. Salinity benefits will be proportional to the
amount of salt in the controlled runoff and sediment. If
improved soil moisture regimes improve vegetation cover,
benefits derived from mechanical land treatments may be
sustained indefinitely, given compatible subsequent land
use management. If improved cover is not achieved or
maintained, benefits from mechanical land treatments
may be short-lived.
Contour furrows are usually constructed within a re-
seeding and grazing management program, primarily to
increase depression storage and the time available for
348
-------
infiltration. Furrows are not recommended on slopes
greater than 10 percent, and are most effective in medium
to fine textured soils. Furrows have finite lives (Branson et
al. 1966) that are a function of their storage capacity in
relation to runoff and erosion at the site. When functioning
properly, they eliminate most runoff from a site.
Ripping, unlike furrowing, generally influences depres-
sion storage very slightly; the main benefits must be
achieved by increasing soil infiltration capacities. This is
most effective on severely compacted soils such as on
roads or reclaimed mined lands, or on soils where a shal-
low pan layer restricts downward water movement. In
most rangelands, ripping either has not significantly im-
proved infiltration or cover (Branson et al. 1966; Dor-
tignac, 1963), or has produced very short-lived benefits
(Aldon and Garcia, 1972). However, Griffith et al. (1985)
found ripping to be effective in increasing herbage pro-
duction on shortgrass prairie in southeastern Wyoming.
Land treatment techniques must be carefully tailored to
the site, with topography and soil characteristics dictating
treatment types and design.
Structural Techniques
Common structural techniques used in managing runoff,
sediment, and salt yields include rangeland dikes, reten-
tion plugs, retention and detention reservoirs, and gully
plugs. Retention and detention structures trap runoff and
sediment volumes in accordance with their design capaci-
ties. Generally, total runoff retention is required for a struc-
ture to effectively control salinity. Gully plugs usually have
small retention capacities, but provide salt and sediment
control by reducing erosion in active gully systems.
In addition to effectively controlling downstream im-
pacts associated with runoff, erosion, and salinity proc-
esses, retention/detention structures may provide local-
ized onsite benefits. Reservoirs provide water for livestock
and wildlife. Even after filling with sediment they may pro-
vide a riparian-like habitat. Gully plugs, when properly lo-
cated, can cause overincised channels to aggrade and, if
conditions are adequate, result in the creation or restora-
tion of streamside riparian zones (Heede, 1981). Dikes
and widely spaced furrows (> 5 m) usually do not increase
vegetation production (Branson et al. 1966) unless they
are constructed as part of a water-spreading system
(Miller et al. 1969).
To control salinity, reservoirs must be designed with suf-
ficient storage to trap all incoming runoff. While a retention
structure will cease to function for salinity control after it is
filled with sediments in excess of its design capacity, a
proper spillway will keep the structure from failing and
becoming a future source of salt and sediment. Mainte-
nance of retention structures—either by excavating stored
sediments or by increasing their height—will allow the
structures to function beyond their original design life.
In highly saline areas, retention structures are usually
the only practical management alternative. The feasibility
of constructing these types of structures depends upon
identifying secondary benefits, such as flood control, wa-
ter supply, and wildlife habitat. In less saline areas, onsite
benefits to water supply, vegetation production, and ripar-
ian enhancement associated with retention structures of-
ten will be greater than in highly saline areas, but mechan-
ical treatments and vegetation management also may be
feasible treatment strategies, depending upon the man-
agement objectives.
CASE STUDIES
Two BLM watershed improvement projects, both in the
Colorado River Basin, are described here. Both are exam-
NONCOAL MINING AND ABANDONED LAND RECLAMATION
pies of well-planned, properly designed sediment- and
salt-control projects.
Sheep Creek, Utah
Sheep Creek is a tributary to the Paria River, one of the
highest sediment-producing watersheds in the Colorado
Basin. Chosen in the 1950's for an intergovernmental re-
source conservation project, Sheep Creek is an exem-
plary watershed improvement project because of good
interagency cooperation, primarily at the field level, and
because of a well-planned mix of properly designed water-
shed treatments.
The Sheep Creek project area, 50.1 km2 in size, drains
mid-elevation, pinyon-juniper badlands and sagebrush on
the south boundary of Bryce Canyon National Park in
southern Utah. Land ownership is mixed and includes
public lands managed by the BLM, Forest Service, and
National Park Service, and private lands. Treatments in-
cluded a concrete barrier dam on Sheep Creek at the
downstream end of the project area, detention dams and
dike water spreader systems on the sagebrush flats, pin-
yon-juniper to grass vegetation conversions, and gully
checks and reseedings in the upper end of the watershed.
The barrier dam was constructed in 1961 by the Bureau
of Reclamation to provide base-level control for the project
area. As of April 1984, 43.9 ha-m of sediment had been
trapped behind this structure and 915m of the main
Sheep Creek channel were stabilized.
BLM constructed two earthen detention dams on Sheep
Creek Flat, a large sagebrush flat in the upper Sheep
Creek watershed. These dams have accumulated large
sediment deposits and have also been successful water
control structures because their capacities are large in
relation to their contributing areas.
One of the most successful treatments included a series
of several hundred small gully checks constructed at the
extreme upper end of Sheep Creek. These checks were
installed at a very high density and successfully planted to
western wheatgrass. They trapped large quantities of sed-
iment and stabilized a downstream gully system.
Benefits realized from the Sheep Creek watershed pro-
ejcts include the following: (1) an estimated 125 ha-m of
sediment trapped behind erosion and water control struc-
tures, (2) an estimated 1,000 m of main channel aggrada-
tion, (3) an estimated 6 ha of riparian vegetation estab-
lished behind the Sheep Creek Barrier Dam, increasing
both cover and diversity for wildlife habitat, (4) an esti-
mated 10 km of gullies healed, (5) improved watershed
cover on an estimated 200 ha, (6) reduction of flood
peaks, (7) establishment of perennial flow at the Sheep
Creek Barrier Dam, and (8) improved forage production
(unable to quantify).
In addition, dissolved solids in Sheep Creek may have
decreased in concert with the sediment reductions.
Lower Wolf Creek, Colorado
The Lower Wolf Creek project area covers 319 km2 and
represents 58 percent of the entire Wolf Creek drainage,
which is tributary to the White River in northwestern Colo-
rado. Salinity reduction was one of the management ob-
jectives for Lower Wolf Creek (U.S. Dep. Inter., 1982). Be-
cause of its large size, the Lower Wolf Creek project area
was stratified into treatment units (Table 1). Treatment
techniques were designed to trap and retain runoff and
sediment from saline soils.
The Lower Wolf Creek project is in its third year of im-
plementation. Initial treatments included large reservoir
repair and maintenance, pit reservoirs, gully checks, and
earthen retention dams. These initial treatments have
been applied to the high-priority treatment units. As a step
349
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
in determining the cost effectiveness of the project, bene-
fit/cost ratios were determined for each structural treat-
ment type, using salinity control as the primary benefit
(Table 2). This information was used in the project plan-
ning to ensure that the overall mix of treatments had a
positive benefit/cost ratio.
We do not have any results from the Lower Wolf Creek
project at this time, as it will be several more years before
the project is fully implemented. We feel this project is an
excellent example of how to approach sediment and salt
control in a large watershed.
SUMMARY AND CONCLUSIONS
The development of plans for salt and sediment control on
western rangelands requires: (1) the establishment of re-
source management objectives, (2) the identification and
quantification of manageable hydrologic processes, (3)
the investigation of cause and effect relationships, (4) the
stratification of treatment areas, and (5) the selection and
evaluation of alternative treatment techniques. BLM pre-
fers to incorporate salt and sediment control objectives as
part of management plans for grazing, wildlife manage-
ment, and other resource activities. When objectives can-
not be met this way, techniques including vegetation man-
agement and mechanical and structural treatments may
be used to control salt and sediment problems. Almost all
salt and sediment control techniques influence multiple
resource values. Because of the location of public lands in
the significant sediment- and salt-producing river basins,
BLM concentrates its control efforts in small headwater
streams. Watershed projects at Sheep Creek, Utah, and
Lower Wolf Creek, Colorado, are specific examples of suc-
cessful salt and sediment control programs.
REFERENCES
Aldon, E.F., and G. Garcia. 1982. Vegetation change as a result
of soil ripping on the Rio Puerco in New Mexico. J. Range
Manage. 25(5): 381-3.
Blackburn, W.H., and C.M. Skau. 1974. Infiltration rates and
sediment production of selected plant communities in Ne-
vada. J. Range Manage. 27: 476-80.
Branson, F.A., R.F. Miller, and I.S. McQueen. 1966. Contour
furrowing, pitting, and ripping on rangelands in the western
United States. J. Range Manage. 19(4): 182-90.
Dortignac, E.J. 1963. Surface runoff and erosion as affected by
soil ripping. U.S. Dep. Agric. Misc. Publ. 970: 156-65.
Gebhardt, K.A. 1985. Erosion, productivity, and rangeland wa-
tershed planning. Pages 175-82 in E.B. Jones and T.J. Ward,
eds. Watershed Management in the Eighties. Am. Soc. Civil
Engin. New York.
Gifford, G.F. 1972. Infiltration rate and sediment production
trends on a plowed big sagebrush site. J. Range Manage. 26:
440-3.
Gifford, G.F., and F.E. Busby. 1974. Intensive infiltrometer stud-
ies on an plowed big sagebrush site. J. Hydrology 21: 81-90.
Gifford, G.F., and R.H. Hawkins. 1978. Hydrologic impact of
grazing on infiltration: A critical review. Water Resour. Res.
14(2): 305-13.
Gifford, G.F., G. Williams, and G.B. Coltharp. 1970. Infiltration
and erosion studies on pinyon-juniper conversion sites in
southern Utah. J. Range Manage. 23: 402-6.
Griffith, L.W., G.E. Schuman, F. Rauzi, and R.E. Baumgartner.
1985. Mechanical renovation of shortgrass prairie for in-
creased herbage production. J. Range Manage. 38(1): 7-10.
Hawkins, R.H., G.F. Gifford, and J.H. Jurinak. 1977. Effects of
Land Processes on the Salinity of the Upper Colorado River
Basin. U.S. Dep. Inter. Bur. Land Manage. Final proj. rep.
Contract 52500-CT5-16, Denver, CO.
Heede, B.H. 1981. Rehabilitation of disturbed watersheds
through vegetation treatment and physical structures. Pages
Table 1 .—Lower Wolf Creek watershed treatment units.
treatment unit
%of
Water-
shed
Description
Treatments recommended
Salt
production Priority
Mancos shale uplands 42
Mancos alluvial 24
Mancos shale ridges, gentle to
moderate slopes, sparse
vegetation, and shallow soil
small drainages and dissected
benches and fans at the base of
Mancos shale outcrops and
grassed waterways
gully plugs, contour furrows, High
grassed waterways, pit
reservoirs, vegetation
manipulation, spreader dikes
reservoirs, spreader dikes, high
vegetation manipulations
Gullied alluvium
Sagebrush uplands
Pinyon-juniper woodland
steep slopes
4
7
23
major gullied bottomlands
upland big sagebrush sites on
sandstone around perimeter of
watershed
steep, inaccessible slopes and
shallow, heavy-textured soil
large detention reservoirs and
riparian planting
vegetation manipulation and
small check dams and pit
reservoirs
none
low to
moderate
low
moderate
to high
3
4
5
Table 2.— Benefit/cost data by watershed treatment.
Treatment
Contour furrow
Gully plug
Pit reservoir
Check dam
Retention dam
Detention dam
Cost
$2,350/km2
$1 ,770/km2
$1 ,000 ea.
$1,550ea.
$5,000 ea.
$60,000 ea.
Structures
per km2
10' spacing
865
• 3
3
2
0.1
Life of
project
in years
10
15
25
25
25
50
Sediment
storage
capacity
8,520 tonne/kW
6,050 tonne/km2
.03 ha-m
.01 ha-m
.41 ha-m
5 ha-m
Salt
retention
256 tonne/km2
181 tonne/km2
11.0 tonne
4.4 tonne
147 tonne
1 ,758 tonne
Retention
benefit
$15,972/km2
$11,293/km2
$686
$274
$9,171
$109,682
B/C
ratio
6.80
6.38
.69
.18
1.83
1.83
Assumptions:
Conversion Factors:
One hectare-meter of sediment weighs 11,878 tonne
3% sediment from Mancos Shale equals the weight of sail
1 tonne of salt retained equals $62.39 benefit downstream
350
-------
257-68 in D.M. Baumgartner, ed., Proc. Interior West Water-
shed Manage. Symp., Washington State Univ., Pullman, WA.
Lusby, G.C. 1979a. Effects of grazing on runoff and sediment
yield from desert rangeland at Badger Wash in Western Colo-
rado, 1953-73. U.S. Geolog. Surv. Water-Supply Pap. 1532-1.
Lusby, G.C. 1979b. Effects of converting sagebrush cover to
grass on the hydrology of small watersheds at Boco Mountain,
Colorado. U.S.Geolog. Surv. Water-Supply Pap. 1532-J.
Miller, R.F. et al. 1969. An evaluation of range floodwater spread-
ers. J. Range Manage. 22(4): 246-57.
Moore, E. et al. 1979. Livestock grazing management and water
quality protection. EPA 910/9-79-67. U.S. Environ. Prot.
Agency, Washington, DC.
Solomon, R.M., J.R. Maxwell, and LJ. Schmidt. 1982. Deter-
mining watershed conditions and treatment priorities. In Proc.
1982 Arizona-Nevada Acad. Sci. Hydrology Section (Vol. 12).
SALINITY: A NONPOINT SOURCE PROBLEM
U.S. Department of Agriculture. 1980. Soil and Water Re-
sources Conservation Act: 1980. Resour. Conserv. Act Coor-
dinating Comm. Appraisal. Rev. draft. Part I.
U.S. Department of the Interior. 1978. The effects of surface
disturbance on the salinity of public lands in the Upper Colo-
rado River Basin, (status report). Bur. Land Manage. U.S.
Dep. Inter, Denver, CO.
Van Haveren, B.P., E.B. Janes, and W.L. Jackson. 1985. Non-
point pollution control on public lands. J. Soil Water Conserv.
40(1): 92-4.
Williams, G., G.F. Gifford, and G.B. Coltharp. 1972. Factors in-
fluencing infiltration and erosion on chained pinyon-juniper
sites in Utah. J. Range Manage. 25(3): 201-5.
Wischmeier, W.H. and D.D. Smith. 1978. Predicting rainfall ero-
sion losses. U.S. Dep. Agric. Handbook 537. Washington, DC.
351
-------
AL R. JONEZ
Colorado River Water Quality Office
Bureau of Reclamation
Denver, Colorado
One of the many forms of nonpoint source pollution is
salinity. This pollution is causing millions of dollars in
damages in the Colorado River Basin. This paper will
discuss three specific areas of nonpoint source salinity
pollution in the Basin and the progress we have made
toward controlling it. The three areas are near the Big
Sandy River in Wyoming, in the Price-San Rafael River
Basins, and in the Dirty Devil River Basin in Utah. I will
briefly explore the source areas, plan for controlling salin-
ity inflow, and benefits, cost-effectiveness, and issues re-
garding the plan.
I remember the time when I used to think about the Colo-
rado River as a place for entertainment, for fishing, boat-
ing, and enjoying the wildlife in the area. I still think of the
Colorado in those pleasant terms, but now I also think in
terms of the growing invisible water pollution problem.
One of the many forms of pollution is salinity—salts. Salts
are minerals or dissolved solids—sodium, chlorides, sul-
fates, and others—that are picked up by the river. By the
time the Colorado reaches Hoover Dam, it is carrying
about 8.1 millions tonnes (9 million tons) of salt annually,
and causes millions of dollars in damages in the Lower
Colorado River Basin.
In this paper I will provide a brief background of the
overall problem in the basin and the progress toward con-
trolling it. I will also discuss briefly three specific nonpoint
sources.
About half of the present salt pollution in the Colorado
River comes from natural sources, including mineral
springs and geysers. The salts originate from water seep-
ing through ancient marine deposits and saline soils,
which are washed into streams. The balance comes from
the concentrating effect of man's use of water for irriga-
tion, municipal and industrial use, and reservoir operation.
The Colorado River at its headwaters in the mountains
of Colorado has a salinity of only about 50 mg/L. The sa-
linity concentrations progressively increase downstream
as a result of the use of water and salt contributions. In
1982, the salinity averaged 825 mg/L at Imperial Dam, the
last major diversion point on the Colorado River in the
United States. The salinity in the river does fluctuate, how-
ever. Record high flows in 1983 and 1984 diluted the salin-
ity levels at Imperial to 710 mg/L and 670 mg/L, respec-
tively (see Fig. 1). While these higher flows and diluting
effects will give us additional time to seek more cost-effec-
tive solutions for salinity control, normal flows will increase
the river's salinity levels to the 800-900 mg/L range in 4-5
years. Without control measures, the concentrations are
projected to increase to over 1,000 mg/L by the year 2020.
Salt pollution affects more than 10 million people and
400,000 ha (1 million acres) of irrigated farmland in the
United States. Economic losses associated with municipal
HISTORICAL SALINITY LEVELS
AT IMPERIAL 0AM
Figure 1.—Historical salinity levels at Imperial Dam from
1941 through 1984.
use occur primarily from increased water treatment costs,
accelerated pipe scaling and appliance wear, and in-
creased soap and detergent needs. Some people prefer
bottled water or softened water to the taste of the salty
water. The Environmental Protection Agency recom-
mends that we drink water with no more than 500 mg/L of
total dissolved solids.
For irrigators, higher concentrations decrease crop
yields, alter crop patterns, and result in higher leaching
and drainage requirements, and higher management
costs. Agricultural losses begin when salinity levels of ap-
plied irrigation water reach 700 to 850 mg/L, depending
upon soil conditions and type of crop grown.
Total annual damages to the Lower Colorado River Ba-
sin water users are approximately $561,000 for each rise
of 1 mg/L in salinity concentration at Imperial Dam. Eco-
nomic losses are estimated to be about $90 million annu-
ally. That does not include the undetermined economic
impact on Mexico.
In June 1974, Congress enacted the Colorado River
Basin Salinity Control Act, P.L. 93-320. Title I of the Act
authorized construction of the Yuma desalting complex
and other features to provide better quality water to Mex-
ico in accordance with Minute 242 of the International
Boundary and Water Commission Agreement.
Title II, the Colorado River Water Quality Improvement
Program, directed the Secretary of the Interior to expedite
completion of planning reports on 12 salinity control units
and authorized the construction of four selected units. Un-
der a cost-sharing approach, one-fourth of the construc-
tion costs of the authorized units were to be provided by
Upper and Lower Basin funds, with revenue obtained from
the sale of hydroelectric power.
P.L. 98-569, signed on Oct. 30, 1984, amends P.L. 93-
320. This legislation amends, enhances, and updates the
10-year-old salinity control act. It is the culmination of a
21/2 year effort by the Colorado River Basin States working
in close cooperation with Federal agencies.
The Colorado River Basin Salinity Control Forum, repre-
senting the seven Colorado River Basin States, believes
352
-------
SALINITY: A NONPOINT SOURCE PROBLEM
that the act as now amended provides the authority for the
pursuit of salinity control measures that will put in place
the necessary salinity controls on the river through the
year 2000. It will ensure, if implemented, the compliance
with the numeric criteria (standards set on the lower main
stem) at least through the year 2005.
To meet the salinity standards now set for the basin, up
to 1 million tons of salt per year must be removed from the
river system by the turn of the century. This level of salt
removal will prevent salinity concentrations from exceed-
ing the numeric salinity criteria while the basin States con-
tinue to use and develop their basin water supplies. The
criteria, set in terms of milligrams per liter of total dis-
solved solids, are essentially a nondegradation standard
based upon 1972 historical data. The maximum salinity
concentration level allowed at Imperial Dam is 879 mg/L
(See Table 1.)
The Bureau of Reclamation has been designated to
lead the Federal effort to reduce the salinity in the river
system. The Colorado River Basin Salinity Control Forum
works with Reclamation, the Department of Agriculture,
and other Federal agencies to implement controls to main-
tain the salinity levels in the Colorado River. Both struc-
tural and nonstructural measures are necessary to inter-
cept and control sources of man-caused and natural salt
load.
Under the Colorado River Water Quality Improvement
Program, construction will occur on a priority basis so that
the most cost-effective measures will be implemented to
meet program goals. To avoid the high energy costs of
desalination plants or vast areas of evaporation ponds,
other beneficial use concepts are being considered.
Reclamation has investigated possible beneficial uses
of the saline water. In a September 1981 Special Report,
entitled "Saline Water Use and Disposal Opportunities,"
Reclamation identified other salinity control strategies that
appear both cost effective and environmentally accept-
able. Several States are considering options using saline
water locally in powerplant cooling towers.
Different ways of solving the problem depend partly
upon the way the salt enters the river. More than a third of
the salt enters the river from irrigation sources.
Irrigation source control would reduce salt loading by
improving irrigation practices that currently leach salts
from marine shales and other saline deposits. The Grand
Valley Unit in Colorado is one major example of irrigation
source control where construction by both Reclamation
and Soil Conservation Service has been going on for sev-
eral years. Reclamation estimates that distribution sys-
tems and on-farm practices in Grand Valley, Lower Gunni-
son, and McElmo Creek Units in Colorado and the Uinta
Basin Unit in Utah could be improved to reduce the river's
salt load by up to 1 million tons per year.
Another source of salt loading involves identified point
sources such as mineral springs, abandoned oil wells, and
geysers. Paradox Valley and Meeker Dome Units in Colo-
rado are point sources. Paradox Valley Unit is under con-
struction, and abandoned oil wells in the Meeker Dome
Unit were successfully plugged during verification studies.
Control opportunities from the third source involve dif-
fuse sources of salt, or what we are discussing today as
nonpoint sources. Diffuse source control measures in-
clude watershed management, land treatment, some irri-
Table 1.—Salinity numeric criteria for the Colorado River.
Annual flow-
.Station weighted average
Below Hoover Dam
Below Parker Dam
At Imperial Dam
723 mg/L
747 mg/L
879 mg/L
DIRTY DEVIL RIVER UNIT
Figure 2.—Dirty Devil River Unit location map.
gation improvements, and the collection and disposal of
poor quality streamflows. Investigations of these diffuse
units are examining a combination of irrigation improve-
ments, vegetation and watershed management, and se-
lective withdrawal and disposal of poor quality stream-
flows.
The three diffuse nonpoint sources are the Big Sandy
River in Wyoming, the Price-San Rafael River Basins in
Utah, and the Dirty Devil River Basin, also in Utah. Of
interest here are the source areas, the plans for controlling
the salinity inflow into the Colorado River, the benefits to
be derived as well as the cost effectiveness of each plan,
and what, if any, current issues affect the plans.
DIRTY DEVIL RIVER UNIT
The Dirty Devil River contributes approximately 170,500
tonnes (155,000 tons) of salt annually to the Colorado
River. The preferred plan would reduce the salt load of the
Colorado River by approximately 22,600 tonnes (20,600
tons) per year. (See Fig. 2.)
The Dirty Devil River Unit is located in south central
Utah. The unit area is sparsely populated, with over 75
percent of the land administered by the Federal Govern-
ment. Two geologic formations contribute significant
amounts of salt to the Dirty Devil River drainage—the
Mancos Shale and Carmel Formations. The Mancos
Shale is responsible for the salinity increase in the irri-
gated area near Emery and along the lower reaches of the
Fremont River.
The Carmel Formation is the salt source in Emery
South Salt Wash and Hanksville Salt Wash. Saline springs
feeding the washes result from water percolating from the
underlying Navajo Sandstone aquifer through the salt-
bearing Carmel Formation and emerging through surface
fractures.
After evaluating numerous plans, one strategy em-
erged. This preferred plan consists of collecting saline
spring water in Hanksville Salt Wash and Emery South
Salt Wash and disposing of it by deep-well injection. Sa-
line water would be collected by pumping at the rate of
.0825 cubic meters (2.75 cubic foot) per second from shal-
low wells in the aquifer. This water would be filtered and
chemically stabilized after which it would be injected into a
geologic formation, the Coconino Sandstone, where it
would be stored indefinitely, isolated from any freshwater
aquifer.
353
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
This would decrease the salt load by approximately
20,600 tons annually, reducing salinity at Imperial Dam by
about 1.88 mg/L. The total capital cost of the project is
expected to be approximately $12 million. In terms of total
costs per unit removed, the cost effectiveness is approxi-
mately $97 per ton of salt removed or $1,063,000 per mg/
L at Imperial Dam. This cost effectiveness level can be
compared to other controls now under construction which
range from $70 to $90 per ton.
This preferred plan will be studied as the planning re-
port/environmental statement is prepared. Advance plan-
ning would take approximately 2 years and construction
about 3 years.
Several unresolved issues remain:
1. Verification of the site-specific hydrologic and geo-
logic conditions of the Coconino Sandstone.
2. Environmental mitigation because the unit plan
would dewater a section of Hanksville Salt Wash.
3. The question of beneficial use of saline water. The
State Engineer has reviewed the plan but has reserved-
judgment until water rights have been applied for.
4. Obtaining State and EPA approval for injection into
the Coconino Sandstone. Completion of further site-spe-
cific study, including drilling a test well, will be necessary
to address this issue.
The Price-San Rafael Rivers Unit is in east-central Utah,
192 km (120 miles) southeast of Salt Lake City. The water-
sheds of the two rivers drain into the Colorado River via
the Green River and add about 430,000 tons of salt annu-
ally to the Colorado River system. (See Fig. 3.) Over 70
percent of the land in the basin is administered by the
Federal Government. Two percent of the land is in irri-
gated alfalfa and feed crops.
After evaluating the salt sources in the area, a preferred
plan has been selected: improve winter stockwater prac-
tices by lining stockwater ponds to reduce seepage. Addi-
tional stockwater ponds would be provided and existing
ones would be improved by enlarging and lining. In some
areas, existing domestic water systems could be ex-
panded to provide supplemental supplies to stockwater
users.
The improved winter stockwater practices could reduce
the salinity at Imperial Dam by about 19,800 tonnes
(22,000 tons) or 2.2 mg/L and the capital cost would be
approximately $8 million. The cost effectiveness of this
plan would be approximately $304,000 per mg/L or about
$31 per ton, which is cost effective compared to other unit
plans.
Figure 3.—Price-San Rafael Rivers Unit location map.
BIG SANDY RIVER UNIT
Figure 4.—Big Sandy River Unit location map.
SO© SAMOY KDVER! UMDTS
The Big Sandy River Unit is in southwestern Wyoming
and includes the lower Big Sandy River, a tributary of the
Green River in the Upper Colorado River Basin. This
study is specifically directed toward formulating plans to
reduce salt pickup from seeps and springs along a
41.6 km (26-mile) reach of the Big Sandy River west of
Eden, Wyoming. (See Fig. 4.)
Investigations indicate that saline seeps, totaling some
1.11 cubic meters (37 cubic feet) per second, surface
along the seepage area. The salinity here varies from
1,000 mg/L to 6,000 mg/L along the Big Sandy River with
a total annual contribution of more than 147,000 tonnes
(164,000 tons) of salt to the river, increasing salinity at
Imperial Dam by 16.6 mg/L. Indications are that salt is
picked up by deep percolation from irrigation water con-
tacting the shale of the Green River Formation and even-
tually seeping into the river.
The Soil Conservation Service studied on-farm solu-
tions to the salinity problem. Reclamation focused on off-
farm solutions and identified various ways to intercept the
springs and seeps and transport, treat, and use or dispose
of the saline water.
The recommended Reclamation plan would collect a
total of .498 cubic meters (16.6 cubic feet) per second of
saline wells near the spring and seep area, pumping it
though a pipeline to the proposed Chevron Chemical Co.
fertilizer plant near Rock Springs, Wyoming. However,
Chevron has deferred expansion plans and the recom-
mended plan is no longer viable.
A new plan being developed would collect saline water
from a collector well field and transport it to the Divide
Basin for disposal. This plan, proposed only as an interim
solution, is not a beneficial use of the water.
Wyoming has been an active partner throughout the
study, with Reclamation working with them to develop al-
ternatives. Reclamation has reformulated alternatives and
will concentrate on disposal in the Divide Basin and use of
saline water at the Jim Bridger Powerplant or Chevron
chemical plant. A consultant is currently studying the fea-
sibility of using saline water for cooling at the powerplant,
and the results of that study will be used in evaluating
these alternatives.
The SCS is evaluating a low head pressure on-farm
sprinkler system. The State will evaluate both SCS's and
Reclamation's studies and will ultimately recommend a
course of action to the Federal agencies.
The draft planning report/environmental impact state-
ment is scheduled to be filed with the EPA in June 1986;
354
-------
SALINITY: A NONPOINT SOURCE PROBLEM
however, this will depend on the development and evalua- Program. As technology changes in the development of
tion of the overall recommended plan of action. powerplant cooling or other beneficial use opportunities,
the development potential of specific units will improve. A
SUMMARY long-range program is being developed for the next 20-25
years, one that will implement the most cost-effective units
We have three units under construction, and the Dirty needed to maintain the numeric criteria set for the river
Devil River, Price-San Rafael Rivers, and Big Sandy River system.
are three of the 12 Reclamation units currently under
study in the Colorado River Water Quality Improvement
355
-------
CONTINUOUS SALINITY STATION MONITORING IN THE COLORADO
RIVER BASIN BY THE UTAH BUREAU OF WATER POLLUTION
CONTROL
ROY D. GUNNELL
Environmental Health Scientist
Utah Bureau of Water Pollution Control
Salt Lake City, Utah
ABSTRACT
The objectives of our salinity monitoring program are to
characterize the Colorado River Basin waters by deter-
mining the total dissolved solids loadings entering and
leaving Utah and the relative contributions from major
basins. A network of nine continuous recording salinity
stations collect temperature and specific conductance
data. The continuous recording devices consist of Hydro-
lab Datasondes* (registered trademark of Hydrolab Inc.)
are programmed and standardized in Salt Lake City be-
fore they are distributed to their specific field locations.
Each salinity station is visited monthly to replace the Da-
tasondes* and to collect ambient water quality informa-
tion. The data from the sondes are read and edited into a
computer storage file. The conductivity data, along with
chemistry and flow data are then combined to determine
loadings.
INTRODUCTION
The objectives of the Bureau of Water Pollution Control's
salinity monitoring program are to characterize the Colo-
rado River Basin waters by determining the total dissolved
solids loadings entering and leaving Utah and to deter-
mine the relative contributions from major drainage ba-
sins.
The Colorado River Basin Salinity Control Forum was
created in 1973 to maintain salinity at or below the levels
found in the lower Colorado River mainstem as of April
1972. The Forum consists of water resource and water
quality representatives appointed by the governors of
each basin State. Most Forum members are also mem-
bers of the Colorado River Salinity Control Advisory Coun-
cil created by P.L. 93-320 to advise the Secretary of Inte-
rior, Secretary of Agriculture, and the Administrator of
Environmental Protection Agency on salinity issues.
Salinity standards, including numeric criteria and salin-
ity control implementation plans, were produced by the
Forum in 1975 and revised in 1978, 1981, and 1984. This
plan and revisions have been adopted by each of the
seven basin States as part of their water quality standards
and have been approved by the Environmental Protection
Agency. The Forum reviews the standards, including the
numeric criteria and plan, each year. The plan is brought
up to date as appropriate but at least once every 3 years.
The numeric criteria are revised only when necessary as
agreed by the Forum States.
The Forum plan of implementation is comprised of a
number of Federal and nonfederal measures to maintain
the adopted salinity criteria of 723 mg/L below Hoover
Dam, 747 mg/L below Parker Dam and 879 mg/L at Impe-
rial Dam (Utah, 1982a).
The Utah Bureau of Water Pollution Control maintains a
network of ambient monitoring stations in the Colorado
River Basin. Grab samples and field observations are
taken at designated sampling stations, nine of which are
continuous monitoring salinity stations (Table 1, Fig. 1)
(Utah, 1982b). Continuous monitoring salinity stations are
important because they can record the periodic storm
events and resulting shock loads common to southern
Utah. Salinity stations are strategically located to monitor
salinity entering and leaving Utah and salinity contribu-
tions from major drainage basins. The station on the
Green River at Dinosaur National Monument and that on
the Colorado River at Cisco monitor salinity entering Utah,
while the rest have been located at the bottom of major
drainage basins to record salinity contributions from their
respective basins. Two stations, Virgin River above First
Narrows and Colorado River below Glen Canyon Dam,
were installed in 1984. The Ashley Creek and Dry Gulch
stations were installed in 1981 and the remainder have
been in operation since 1976.
METHODS AND MATERIALS
When the program began in 1976 each salinity station
consisted of a Hydrolab Surveyor unit coupled to a Ball
Brothers recording device housed in a weather- and van-
dalproof shelter. Temperature and conductivity measure-
ments were recorded every 30 minutes onto a cassette
tape. The stations were serviced biweekly by changing
batteries and installing an unused cassette tape. The cur-
rent continuous recording salinity stations collect hourly
temperature and specific conductivity data. These sta-
tions consist of Hydrolab Datasondes (a registered trade-
mark of Hydrolab, Inc.) housed in permanent, protective 6-
inch diameter steel pipes adjacent to U.S. Geological
Survey gauging stations (Hinshaw, 1985). State personnel
determine stream flows where U.S. Geological Survey
flow data are not available.
The sondes are standardized, programmed and cali-
brated in Salt Lake City before they are distributed to their
specific field locations. Sonde servicing includes a thor-
ough examination, cleaning, and installation of fresh bat-
teries. The sondes are calibrated using a known conduc-
tivity solution prepared by the State Health Laboratory.
Each salinity station is visited monthly to replace
sondes and to collect ambient water quality samples. The
samples and sondes are brought back to Salt Lake City for
routine chemical analyses and data retrieval. The data
from the sondes are loaded onto a WANG personal com-
puter floppy disc, edited, and transferred to the mainframe
computer (Judd, 1985).
The editing consists of adjusting the conductivity read-
ings recorded by the sondes; to do this, conductivity data
from the ambient water quality samples are compared
with field readings. Total dissolved solids (TDS) data from
the ambient water quality samples are used to develop a
TDS/conductivity ratio for each salinity station based on its
drainage basin chemical characteristics. The TDS/con-
ductivity ratio is combined with flow data from U.S. Geo-
logical Survey or from Bureau of Water Pollution Control
to determine loadings at each station.
Currently, the Bureau of Water Pollution Control has
limited resources for data analyses by computer technol-
356
-------
SALINITY: A NONPOINT SOURCE PROBLEM
Table 1.—Salinity Stations.
Storet #
493027
493165
493414
493721
493790
495002
495200
495430
495849
Site
San Rafael River at Chaffin Ranch Bridge
Price River at Woodside
Dry Gulch Creek at U-132 road crossing
Ashley Creek above confluence with Green River
Green River at Dinosaur National Monument
Virgin River above First Narrows
Colorado River below Glen Canyon Dam
Dirty Devil River above confluence with Poison Spring Wash
Colorado River at Dewey Bridge crossing near Cisco, Utah
Latitude
38° 45' 32" N
39° 15' 50" N
40° 15' 50" N
40° 20' 30" N
40° 24' 34" N
37° 01 '05" N
36° 56' 12" N
38° 05' 50" N
38° 48' 39" N
Longitude
110° 08' 24" W
1 10° 20' 45" W
109° 51' 31" W
109° 21 '54" W
109° 14' 05" W
1 13° 39' 58" W
111° 29' 00" W
1 10° 24' 27" W
109° 18' 11" W
"IH- "XI™, ^uV " L" Hi- •'••^495200
Figure 1.—Location map, continuous monitoring salinity
stations. Sites are identified by Storet number.
ogy. Previous attempts to analyze the data have been by
hand manipulation, requiring a great deal of time and ef-
fort.
RESULTS AND DISCUSSION
Nine continuous salinity monitoring stations are operating
in Utah. Two stations located on the Virgin River and be-
low Glen Canyon Dam have been installed recently, result-
ing in very little data for analysis. The following discussion
covers the problems, data, and locations of the seven
stations.
The abundance of conductivity data available for analy-
sis would be overwhelming if the daily flow data correla-
tion was done by hand. Because of the limited computer
analysis capability, the grab sample data must serve as
both the quality control benchmark, and the data base to
which generalized flow data may be applied. Unfortu-
nately, this eliminates the ability to document many unu-
sual flow and salt-loading events such as flash flooding
and dry periods. The accuracy of the averaging still re-
flects the general activity of the stations. Insufficient grab
sample data exist to establish any type of yearly trends in
salinity because they appear to be subject to the quantity
of water available during the year.
The Ashley Creek drainage basin includes 168,074 ha.
The salinity contributions to this drainage are from agricul-
tural return flows, energy exploration activities, and the
natural geology.
Ashley Creek is dewatered at the mouth of Dry Creek
Canyon by irrigation diversions, resulting in the down-gra-
dient water monitored by the salinity station being mostly
irrigation return flow. The daily averages are useful for the
hydrologic/mineralogic cycle of the stream. In 1981, only 8
months of data were available, February through Septem-
ber. The average daily flow for water year 1982 (February-
September) was 154 percent of 1981. Comparing the
1981 data (February-September) with 1982 data of the
same months, the 1982 average daily salt load of 500
metric tons/day was 10 percent lower than the 1981 salt
load of 552 metric tons/day (Ellis, 1984). The flows in
Ashley Creek peak during spring runoff. Because the
flows during the 1982 water year were greater than in
1981, the resulting greater dilution reduced salinity.
The station located on the Colorado River near Cisco
records the salinity contributions from the Colorado and
Dolores Rivers entering Utah. This station has been in
operation since 1976 except when flooded for short peri-
ods during some extremely high water years. As a result
of the flooding, the station was moved downstream .5 mile
to the Dewey Bridge. Salinity contributions in the Colorado
River are from agricultural, industrial, and mining activi-
ties. With data available for 1980, 1981, and 1982, Febru-
ary 1981 indicates the least salt load at 175,086 metric
tons while April 1980 reflects the highest load at 485,704
metric tons. Flow and salt load peaks are noticeable only
for the 1980 and 1982 runoff periods. The 1981 runoff
produced flows and salt loads similar to the autumn peak
of that year (Ellis, 1984).
The Dirty Devil River drainage represents a unique area
in the Colorado River system. Most of the salinity contribu-
tions are natural with very little from agricultural or indus-
trial sources. The drainage area includes 3,074,655 ha
with most nonpoint pollution resulting from runoff and ero-
sion from sparsely vegetated Federal (public) lands. The
Dirty Devil River is the least stable of all the continuously
monitored rivers in terms of both salinity and water flow.
No-flow periods are common during the summer. High
flow peaks occur not in the spring, but in the fall after
upstream irrigation use has ceased and as a result of
thunderstorm activity.
The 2 years with a no-flow period showed the high peak
for both flow and dissolved solids loading as occurring
directly after the no-flow period. September of 1980
showed 155,854 metric tons of dissolved solids carried by
the river following a 24 day no-flow period. This load ex-
ceeded the next highest load (August 1982) by 72 percent.
The next no-flow period extended for 51 days. Two weeks
after the flow of the river resumed, the third largest
monthly peak of flow and salinity occurred. Spring peaks
appeared to be rather mild when compared to the late
summer-early autumn peaks.
These peaks can also be viewed from their contribution
to the total amount of dissolved solids carried in the river
during the year. The September peak carried 58 percent
of the 1980 salinity load. The September peak carried a 36
357
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
percent share of the 1981 load, while the August peak
carried 28 percent of the total salinity load for 1982. Aver-
age concentration of total dissolved solids for each year is
as follows (Ellis, 1984):
1980
1981
1982
2,028 mg/L
1,879 mg/L
2,101 mg/L
The Dirty Devil River is located in a semiarid region and
is dry during summer months; however, peak flows occa-
sionally result from sudden storm events. This may ac-
count for the high flows during late summer and early fall,
especially since 1981 through 1983 have had record
breaking precipitation.
Dry Gulch is unique. The stream exists because of irri-
gation return flows. The monitoring station is vital because
the Bureau of Reclamation and the Soil Conservation
Service have targeted the drainage area for salinity con-
trol projects. The success of these projects will be shown
in the data gathered by this station, but the Dry Gulch
station has been difficult to assess. Problems with loca-
tion, equipment, and flow have resulted in many months of
unreliable data. Bureau personnel have solved many of
these problems, resulting in better data gathering and
compilation.
The salinity station on the Green River at Dinosaur Na-
tional Monument monitors the salinity contributions com-
ing into Utah from the Green and Yampa Rivers. Flaming
Gorge Dam controls flows at this station, resulting in
standard runoff peak flows and daily fluctuations from
electrical power demands. This station records the lowest
values of salinity of all the stations. This station has had
few problems, resulting in good data since its installation
in 1976. The average dissolved solids concentration for 3
years of data is as follows (Ellis, 1984):
1980
1981
1982
339 mg/L
393 mg/L
329 mg/L
The Price River drainage area is 486,066 ha. A large
percentage of the land area is federally owned and is used
for livestock grazing. The sparse vegetation contributes to
increased salinity levels from overland flows. The large
size of these range areas and limited ability to sustain
vegetation offer little opportunity for improvement. The
Price River station has limited data available because of
vandalism and equipment breakdown. Problems have re-
sulted because of probes being silted and because acces-
sibility encourages damage from the curious. The availa-
ble data showed an increase of dissolved solids in the
river during October 1981. The spring of that year showed
little or no influence from runoff at the monitoring station.
The Price River experienced increased flow and dissolved
solids loading in the spring of 1982 as well as later in the
summer of the same year. More data are required to make
any real predictions on the Price River.
The San Rafael River drainage area is 622,488 ha. Most
of the area is sparsely vegetated and offers little opportu-
nity for improvement. The San Rafael River area is semi-
arid resulting in no-flow periods during the summer
months. The station has had silting-in and flooding prob-
lems. This station was recently moved from its old location
at U-24 highway crossing to 3 miles above the mouth of
the San Rafael River. The flow and dissolved solids for the
San Rafael River showed primary and secondary peaks in
a pattern unique for the monitored streams of southern
Utah. The primary peaks of dissolved solids for 1980 and
1982 occurred in the summer. The summer of 1981 re-
corded the lowest value for salt loading. Secondary peaks
for 1980 and 1981 occurred in the fall. This is similar to the
pattern of other southern Utah streams where summer
irrigation cessation allows water to remain in the stream.
One more secondary peak occurred during February and
March of 1980. The average total dissolved solids concen-
tration for each year follows (Ellis, 1984):
1980
1981
1982
1,487 mg/L
3,068 mg/L
2,068 mg/L
In summary, from available data, the profile of streams in
the State vary from the northern and the southern por-
tions with respect to seasonal loads of dissolved solids.
Northern streams appear to have a definite spring/sum-
mer runoff peak. The southern streams appear to have
two definite periods when flows and salt loads can peak-
spring and fall. In the case of the Dirty Devil River, only a
fall peak was observed. The reason for this flow pattern is
unclear. It may be the result of 3 years of unusually high
precipitation. More analyses are needed over longer peri-
ods of record to substantiate the findings.
These data should be considered as only rough indica-
tors of actual salt loading. More accurate daily computa-
tions are necessary in monitoring the unstable streams in
the State. Such computations require improvement in the
quality of continuous data recording for such troublesome
streams as Price River and a computer analysis to synthe-
size the daily salinity and flow data. High runoff has de-
creased total dissolved solid concentrations but increased
flows have increased total loadings to the Colorado River
drainage.
The State of Utah shows great potential for establishing
accurate salt-loading information. The Bureau is currently
hampered by budget and personnel constraints, along
with inadequate computer facilities. When these problems
are solved, the State will be able to provide better salinity
data and information to all interested individuals and
agencies. Salinity will always remain a problem in Utah.
The State of Utah will continue to monitor and analyze
salinity information with the resources available.
ACKNOWLEDGEMENTS: Several people helped with this re-
port. I would like to express appreciation to Russell Hinshaw and
Harry Judd for their assistance, and to Mark Ellis for his interpre-
tation of the sampling data at the salinity stations.
Ellis, M.T. 1984. Salinity monitoring report. Utah Bur. Water Pol-
lution Control, Salt Lake City.
Hinshaw, R.N. 1985. Personal communication. Utah Bur. Water
Pollution Control, Salt Lake City.
Judd, H.L. 1985. Personal communication. Utah Bur. Water Pol-
lution Control, Salt Lake City.
State of Utah Bureau of Water Pollution Control. 1982a. Utah
State Strategy for Salinity Control in the Colorado River Basin.
Salt Lake City.
State of Utah Bureau of Water Pollution Control. 1982b. Monitor-
ing Manual. Salt Lake City.
'358
-------
SALINITY CONTROL IN THE GRAND VALLEY OF COLORADO
FRANK R. RIGGLE
LARRY N. KYSAR
Grand Valley Salinity Program
Grand Junction, Colorado
ABSTRACT
Half the salt annually entering the Colorado river system
comes from manmade Influences, and most comes via
irrigation seepage and on-farm percolation. To reduce the
water entering Imperial Dam by 1 mg/L, nearly 11,000
tons need to be prevented from entering the river. A U.S.
Department of Agriculture and Department of Interior pro-
gram attempted to reduce these loads by 50 percent
through lining and piping the irrigation systems and im-
proving on-farm practices. The project, scheduled to run
through the year 2000, is 10-20 percent complete. Projec-
tions indicate salinity in the Imperial Dam should have
already decreased by 4.3 mg/L.
INTRODUCTION
The Grand Valley in west central Colorado contributes
about 522,000 metric tons (580,000 tons) of salt annu-
ally to the Colorado River system. It is estimated that at
least 50 percent of the river's total salt load comes from
manmade influences. In both the United States and Mex-
ico, the increased salinity in the water supply causes sig-
nificant agricultural and urban economic damages to
downstream water users. Salts are brought into the river
system by subsurface return flows percolating through
saline soils and fractured saline shales. Of the total
522,000 metric tons (580,000 tons), about 475,200 metric
tons (528,000 tons) come from irrigation delivery system
seepage and on-farm percolation losses. Surface runoff,
deep percolation from recreation reservoirs, seepage from
utility canals, and other sources contribute the remaining
salt load of 46,800 metric tons (52,000 tons).
To reduce the salinity concentration at Imperial Dam by
1 mg/L, an estimated 9,900 metric tons (11,000 tons) of
salt must be prevented from entering the river. The annual
economic damage to downstream water users in the
Lower Colorado River Basin States is estimated to be
$561,000 per mg/L for concentrations ranging between
875 and 1,225 mg/L at Imperial Dam. Not included are
economic damages to Mexican water users, which would
add significantly to this dollar amount.
In 1979 and 1980, the Federal government, through the
U.S. Department of Agriculture (USDA) and the U.S. De-
partment of the Interior (USDI), implemented programs in
the Grand Valley to reduce the seepage and associated
salt loading. The program assists local water users to line
or pipe delivery systems and improve on-farm irrigation
systems and irrigation water management. Planning and
application responsibilities of the program are divided,
with the USDI lining and piping off-farm irrigation canals
and laterals and the USDA helping landowners improve
on-farm irrigation systems and management.
The program goal is to reduce seepage and associated
salt loading by at least 50 percent. The estimated improve-
ments to irrigation systems consist of lining 72 km (45 mi)
of off-farm delivery canals; piping 704 km (440 mi) of off-
farm delivery laterals; piping or lining 1,056 km (660 mi) of
on-farm ditches; and installing improved irrigation sys-
tems, including irrigation water management, on 21,2QQ
ha (53,000 acres).
PROBLEM
The Colorado River has eroded the Grand Valley into the
Mancos Formation, a sequence of marine shale about
1,200 m (4,000 ft) thick that contains a high percentage of
salts and gypsum. The salt crystals are commonly found
in open joints and fractures. The Mancos Formation is
impervious at depth, but the weathered zone near the
surface transmits water along joints, fractures, and bed-
ding planes.
An aquifer exists within the Mancos Formation between
the Government Highline Canal and the Colorado River.
Recharge of this aquifer system occurs from canal, lateral,
and on-farm ditch seepage, or where irrigation waters per-
colate into the zone through vertical jointing in the shale.
Essentially all irrigated land in the valley is underlain by
Mancos Shale. Salt types present here are mostly calcium
sulfate. Since many of the soils are derived from the Man-
cos, they exhibit chemical properties similar to that of the
shale. Addition of salts to the river system is not the only
cause of increased salinity concentrations. Removal of
water by phreatophytes and field crops increases the sa-
linity concentration of return flows. Also, removing better
quality water in the Upper Colorado River Basin reduces
the dilution effect on the waters of the downstream
reaches.
AUTHORIZATION
The Grand Valley Unit was authorized for construction by
the Colorado River Basin Salinity Control Act of 1974 (RL.
93-320) as part of a basinwide program for enhancing and
protecting the quality of water available in the Colorado
River for use in the United States and the Republic of
Mexico. Title I of the Act was directed toward controlling'
the salinity of river water below Imperial Dam. Title II,
directed toward salinity control in the United States above
Imperial Dam, authorized the construction of the Grand
Valley Unit and three other units.
The USDA on-farm program in the Grand Valley and the
Uinta Basin of Utah was planned and implemented with
existing authorities. The program took effect in October
1979 when Congress allocated $1.7 million to the Grand
Valley from the Agricultural Conservation Program (ACP),
administered by the Agricultural Stabilization and Conser-
vation Service. Funding at that level under that authority
has continued since with ACP rules and regulations con-
trolling administration of the salinity control program. How-
ever, the ACP program could not continue its work nation-
ally and also support the growing salinity control program.
Moreover, the ACP regulations limited the pace at which
salinity control practices could be installed on individual
farms. Therefore, recognizing the need to accelerate the
salinity control program, Congress passed P.O. 98-568,
-359
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
amending RL. 93-320 and authorizing USDA to establish
a new program for salinity control based on the voluntary
participation of private landowners.
USDI Program
The USDI portion of the program was planned by the
Bureau of Reclamation in two stages, primarily to use in-
formation from Stage One in the planning of Stage Two.
Stage One construction began where data could be gath-
ered to assess effects of initial development and where
environmental impacts were not believed to be significant.
Planning on Stage Two took place concurrently. Stage
One construction, which covered about 10 percent of the
Grand Valley, began in October 1980 and was essentially
completed in April 1983; it included concrete-lining
10.9 km (6.8 mi) of the Government Highline Canal and
consolidating 54 km (34 mi) of open laterals into 38 km
(24 mi) of pipe laterals.
Salinity monitoring data indicate that seepage and salt
loading have declined in the Reed Wash study area, a
hydrologically closed basin north of Loma, Colorado.
Stage One results indicate that salt loading has decreased
by about 13,000-22,500 metric tons/yr (20,000-25,000
tons/yr) or 1.8-2.3 mg/L at Imperial Dam. A moss and
debris removal structure installed at the beginning of the
lined portion of the Government Highline Canal consists
of three trash rakes to remove most of the trash, weeds,
and debris. This systemwide removal has solved most of
the problems; however, additional design modifications to
the turnouts and meters are planned for Stage Two, based
on the experience in Stage One.
A problem developed in Stage One that affected plan-
ning on Stage Two concerning cracks in the concrete-
lined canal. The concrete lining of Stage One, completed
during the winter of 1980-1981, is 6.35 cm (21/2 in) thick,
unreinforced, and placed on compacted embankment
with a thickness of at least 60 cm (2 ft). During the first 2
years following construction, very few cracks appeared in
the concrete; however, in the winter of 1983-84, extensive
cracking occurred. A freeze-thaw action of the canal wa-
ter is believed to be the primary cause of the cracks. The
existing cracks will probably be widened and extended by
further freeze-thaw and by hydraulic uplift. The plan for
Stage Two is to membrane-line the Government Highline
Canal and replace existing open earth laterals with pipe.
Based on expected wildlife habitat losses in Stage One
and Stage Two, compensatory measures are to include
acquiring more than 800 ha (2,000 acres) of riverbottom
land along the Colorado River.
USDA PROGRAM
Since the USDA on-farm program began, about 232 km
(145 mi) of on-farm delivery systems have been lined or
piped, and improved irrigation systems have been in-
stalled on 3,880 ha (9,700 acres). The decrease in annual
salt loading as a result of these improvements is about
24,300 metric tons (27,000 tons) or a reduction of 2.5 mg/L
in salinity concentration at Imperial Dam.
The types of irrigation systems being installed include:
underground delivery pipelines with other gated pipe or
concrete-lined ditch distribution systems for surface irriga-
tion, sprinkler systems where adequate gravity pressure
exists, or drip-trickle irrigation on specialty crops. Many of
the surface-irrigated fields are also land leveled to allow a
more uniform water distribution. A number of recently de-
veloped semiautomated irrigation systems have been
used successfully. Cablegation, ported concrete ditches,
and some farmer-developed automated valves have been
readily accepted.
In an attempt to incorprate more management into the
irrigation system, commercially available automatic water-
switching valves were instrumental on a number of early
systems. Mandatory automation in 1979 generally met
with rejection from farmers because of a lack of operator
understanding and poorly developed technology.
Research and development on automating irrigation
systems is progressing; several automated systems are
now being used successfully. Cablegation, an irrigation
system developed by the Agricultural Research Service in
Kimberly Idaho, is being tested on several farms in the
Grand Valley. Farmer acceptance has generally been ex-
cellent, with several landowners working toward convert-
ing their operations entirely to a Cablegation system.
Cablegation owners report approximately 30 percent less
water usage and higher crop yields compared to earlier
irrigation practices on those fields. A skate gate system for
ported concrete ditches, similar to Cablegation, has also
been successful.
Further research and testing of automatic controls is
needed for valves in pipelines and gated pipe to control
water flows. Some of the design and much of the field
testing has been done by local farmers. A local grower has
developed an automatic valve control for gated pipe using
a rechargeable electric drill and timers that automatically
change water sets on his fields. This type of commitment
to progress is necessary for development and application
of field-reliable automation.
The key to a voluntary Federal program on private land
is landowner acceptance. Generally, the irrigation systems
have performed well for the landowners. The installed irri-
gation systems provide an effective tool to better control
irrigation water. In addition to the reduced seepage from
unlined farm ditches, more precise and uniform water dis-
tribution with accurate measurement is possible with new
systems.
The on-farm irrigation improvement is a two-stage proc-
ess. First, the installed irrigation system reduces seepage
from the unlined farm ditches and provides the landowner
with a manageable system to uniformly apply the irrigation
water. Secondly, followup technical assistance helps the
farmer apply the amount of water needed in a timely man-
ner. The expected salinity reduction benefits are divided
almost equally between the improved system and better
water management; gains are needed in both of these
areas for a successful program.
MONITORING & EVALUATION
Aggressive monitoring and evaluation programs assess
and quantify the actual salinity reduction benefits of the
combined cooperative effort between the Federal govern-
ment and the water users. In the Stage One area, Recla-
mation is monitoring ground water levels and canal inflow
and outflow. The quality of canal water and the quality and
quantity of Reed Wash outflow are also being recorded.
The SCS has implemented an on-farm monitoring and
evaluation program. Electronic flow recorders and remote
weather stations gather the field information to assess
seasonal irrigation performance. Irrigation monitoring is
currently ongoing on 16 sites, with four additional sites
scheduled for installation during the summer of 1985.
The effects of Stage One on fish and wildlife resources
were monitored by the Colorado Division of Wildlife be-
tween 1981 and 1984. Since changes to wildlife habitat
were expected, replacement of endangered habitats was
planned for both Stage One and Stage Two. The monitor-
ing by the Division documented that significant changes to
wildlife habitats occurred in the zones along canals and
laterals, but few changes have been noted farther away
from these waterways.
360
-------
On-farm wildlife habitat changes are monitored at repre-
sentative sites throughout the valley, with long-term im-
pacts of the irrigation practices evaluated. The on-farm
program includes incentives for landowners to voluntarily
apply replacement or enhancement practices. The 1984
legislation provides the authority to implement planned
wildlife measures for Stage One, Stage Two, and on-farm
losses. If habitat losses cannot be offset locally, the pro-
gram provides for acquiring and developing as much as
4,000 acres of wildlife habitat.
SUMMARY
The Federal salinity control program assists local water
users with line or pipe delivery systems to improve on-
farm irrigation systems and irrigation water management.
Such technology and practices reduce the seepage and
associated salt loading. Since the beginning, 76 km
(36 mi) of off-farm delivery systems have been lined or
piped, about 232 km (145 mi) of on-farm delivery systems
have been lined or piped, and improved irrigation systems
SALINITY: A NONPOINT SOURCE PROBLEM
have been installed on 3,880 ha (9,700 acres). The pro-
jected net decrease in annual salt loading as a result of
these improvements is about 43,200 metric tons (48,000
tons), or a reduction of 4.3 mg/L in salinity concentration
at Imperial Dam.
Semiautomated irrigation systems using recently devel-
oped technology have been successful. These systems
reduce delivery seepage, help farmers manage water ap-
plications, and generally reduce the amount of water ap-
plied.
The Federal projects are 10-20 percent completed, with
final program completion scheduled for the year 2000.
When this is accomplished, the total projected decrease in
annual salt loading is estimated to be about one-half of the
total 522,000 metric tons. The salinity concentration at
Imperial Dam should decrease by 25 mg/L. Aggressive
monitoring and evaluation programs are assessing and
quantifying the actual salinity reduction benefits of the
combined cooperative effort between the Federal govern-
ment and the water users.
361
-------
Land Use Issues:
Management and
Assessment
PRACTICAL GUIDELINES FOR SELECTING CRITICAL AREAS
FOR CONTROLLING NONPOINT SOURCE
PESTICIDE CONTAMINATION OF AQUATIC SYSTEMS
R. R MAAS
M. D. SMOLEN
S. A. DRESSING
C. A. JAMESON
J. SPOONER
North Carolina State University
Raleigh, North Carolina
INTRODUCTION
The high cost of the financial incentives, technical assist-
ance, educational support, and monitoring programs that
comprise watershed-scale, nonpoint source control proj-
ects has made it increasingly important to target available
funds to critical watersheds and to "critical areas" within
watersheds. Treatment of all potential sources is neither
practical nor necessary for protecting or restoring most
water resources. By definition, critical areas are those ar-
eas or sources where the greatest water quality improve-
ment can be accomplished per dollar spent.
Although many people recognize the importance of the
critical area concept, guidelines for selecting critical areas
within watersheds are generally not available. Motschall et
al. (1984) have presented a procedure for ranking dairy
barnyard sources for phosphorus control. A recent publi-
cation by Maas et al. (1985) proposes criteria for selecting
sediment, nitrogen, phosphorus, and pathogen critical ar-
eas and summarizes selection approaches presently be-
ing used by 32 agricultural nonpoint source control proj-
ects in the United States. A search of the literature has
revealed no guidelines for selecting critical areas in water-
shed-level projects that address pesticide contamination.
This paper offers some practical selection guidelines
and proposes a rating procedure for ranking farm level
units for pesticide control. The criteria and guidelines dis-
cussed should be useful to managers, government field
personnel, and scientists charged with addressing pesti-
cide-related surface and ground water resource impair-
ments.
CRITICAL AREA SELECTION CRITERIA
The criteria for selecting pesticide critical areas can be
grouped into the broad categories of: (1) type of water
resource impairment, (2) source magnitude criteria, (3)
transport-related criteria, and (4) other criteria.
Type and Severity of Water Resource
Impairment
Accurate problem identification that defines the dynamics
of the pesticide-related impairment is a crucial first step in
selecting critical areas. Nearly all documented water re-
source impairments caused by pesticides have involved
either elimination of aquatic flora and fauna (such as fish-
ery impairment) or human health concerns (impairment of
domestic water supply, fishery, or recreation resource).
The impairments in either case are related to the toxic
effects of pesticides and are due more to concentrations
than to total loadings. Thus, the general nature of pesti-
cide impairments suggests that critical areas and BMP's
should be chosen to reduce peak or ambient pesticide
concentrations. It is important to note that BMP's such as
conservation tillage (Baker and Laflen, 1983), terraces,
and contouring (Maas et al. 1984) reduce pesticide sur-
face loadings primarily by reducing surface runoff volume.
Thus, these practices often do not significantly reduce
edge-of-field pesticide concentrations.
The overall effect on receiving water bodies depends on
the percentage of the watershed where the pesticide is
applied as well as various hydrologic factors. For example,
363
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
in the extreme case where a persistent pesticide was ap-
plied on 100 percent of the watershed, BMP's that pro-
duce proportionate reductions in pesticide loading and
runoff volume (no concentration reduction) theoretically
would not reduce receiving water body concentrations.
In reality, surface loading reductions invariably result in
some decrease in receiving body concentrations because
of dilution from nonsource areas and from reemergence of
subsurface runoff as streamflow. The important point is
that runoff-reducing BMP's will generally reduce pesti-
cide loadings more than concentrations. In the case of
ground water, such practices will increase infiltration and
may actually increase ground water pesticide concentra-
tions. This emphasizes the need for selecting BMP's and
critical areas on the basis of the water resource impair-
ment.
As with other agricultural pollutants, the severity of the
impairment markedly affects the extent of areas or
sources that should be designated as critical. For exam-
ple, a 50 percent concentration reduction requires less
inclusive critical area criteria than an 80 percent reduction
would.
Another important consideration involves determining
which pesticide is impairing a water resource use. This
determination is normally made on the basis of monitoring
the water column, sediment, or biota. Often a knowledge
of the temporal dynamics of the pesticide impairment (for
example, fishkills that occur only in late summer) will help
to isolate which pesticides may be responsible. This infor-
mation eliminates areas where this pesticide is not used
from consideration.
The persistence and biomagnification of the organo-
chlorines result in fish tissue concentrations which can
pose human health dangers. Dissolved concentrations
are seldom measurable. Most documented organophos-
phorus insecticide (OP's) impairments have been associ-
ated with accidental spills or overapplications. In areas of
intensive use, documented triazine and anilide impair-
ments have involved subtle and chronic aquatic ecosys-
tem effects.
The most common types of water resource impairments
and the physical/chemical characteristics which contrib-
ute to these impairments are summarized in the following
sections and in Table 1 for five important classes of pesti-
cides.
This will be by far the most
important criterion for selecting critical areas for pesticide
control. Since almost no pesticides occur naturally, only
areas where they are applied are potential sources to
aquatic systems. Also, it is generally assumed that pesti-
cide losses to surface or ground waters are roughly pro-
portionate to application rate. Thus, the selection process
for pesticide critical areas should focus on the usage pat-
terns in the watershed.
A general summary of usage patterns for the major
pesticide classes discussed here and summarized in Ta-
ble 2 will aid in tracing pesticide sources in agricultural
watersheds. The figures presented are at a national level;
usage patterns within a given region may differ considera-
bly. Pesticide use on specialty crops within a region may
be the predominant pesticide source for the region.
With few exceptions (such as toxaphene) organo-
chlorines were phased out of U.S. agricultural use from
1972 to 1976. Heptachlor and mirex are still used some-
what for fire ant control in agricultural settings. In terms of
water use impairments, however, even banned organoch-
lorines are still of concern. They persist in agricultural
soils and remain available for transport and uptake into
the aquatic food web. Cotton received the bulk of or-
ganochlorine applications during the latter 1960's and
1970's, making it the most likely candidate for organoch-
lorine residual. Toxaphene usage has dropped 80 percent
since 1976. This reduction is almost entirely attributable to
a 95 percent use reduction on cotton lands. Toxaphene
use on other crops such as corn and soybeans remains
very low but has actually increased slightly since 1976
(U.S. Dep. Agric. 1983).
Erosion Rate (ER): In general, ER will be an important
critical area criterion for pesticides lost primarily in the
sediment-adsorbed phase of surface runoff.
Organochlorine Insecticides: Organochlorines have
been shown to adsorb strongly to soil particles. For this
reason they are almost entirely lost in surface runoff in the
sediment-adsorbed phase. Hence, ER should be consid-
ered an appropriate criterion for selecting critical areas to
control organochlorine aquatic inputs. As in the case of
other sediment-adsorbed agricultural pollutants such as
phosphorus, the reduction in pesticide loss will be less
than the erosion reduction because of enrichment on the
fine soil fraction.
Organophosphorus Insecticides: Evidence that the OP,
fonofos, is lost in surface runoff primarily in the sediment-
adsorbed phase (Baker et al. 1976) suggests that the in-
clusion of ER as a selection criterion may be appropriate.
In contrast, modeling efforts indicate that methylparathion
runoff losses are 90 percent dissolved (Beyerlein and
Donigian Jr. 1979), implying that ER should not be used as
a selection criterion for this pesticide.
Carbamates, Triazines, and Anilides: Extensive research
has shown that these three classes of pesticides are lost
Table 1.—Characteristics of common pesticides and associated impairments of water resources.
Water resources Characteristics
Pesticide class affected and impairments
Organochlorine insecticides: DDT, Endrin,
Dieldrin, Toxaphene, Heptachlor
Organophosphorus insecticides: Malathion,
Parathion, Methylparathion, Fonofos
Carbamate insecticides: Carbaryl,
Carbofuran, Aldicarb
Triazine herbicides: Atrazine, Simazine,
Cyanizine
Anilide herbicides: Alachlor, Propachlor
surface waters
surface waters
surface waters
and ground water
surface waters
and ground water
surface waters
and ground water
high persistence, biomagnification, high
chronic toxicity to fish and humans,
carcinogenic
high acute toxicity to fish and humans, low
persistence, not biomagnified
moderate acute toxicity to fish and humans,
low persistence, not biomagnified
low toxicity to fish and humans, high chronic
toxicity to algal communities and vegetation,
suspected health effects, not removed by
water treatment1
low toxicity to fish and humans, high acute
effect on algal communities, not removed by
water treatment1
'Baker, D.B., 1983.
364
-------
LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
Table 2.—Pesticide usage on major U.S. crops in 1982'
Pesticide
Organochlorine
Organophosphorus
Carbamates
Triazines2
Anilides3
Crop
cotton
corn
soybeans
cotton
grain sorghum
wheat
corn
soybeans
grain sorghum
peanuts
corn
soybeans
cotton
grain sorghum
corn
soybeans
peanuts
Percent of acres
11
28
3
14
13
2
7
2
7
16
78
33
15
47
34
25
38
1 From USDA (1983).
Triazines are not used on wheat, peanuts, or tobacco.
3Anilides are not used on tobacco or wheat.
predominately in the dissolved phase of surface runoff
(Maas et al. 1984), thus, ER has limited applicability.
Transport Considerations
Distance To Watercourse (DISWC): This criterion is ap-
plicable to all pesticide classes because some relationship
always exists between DISWC and the percentage of ap-
plied pesticide that reaches the impaired water resource.
The dominant transport mechanisms, however, vary
greatly with pesticide class, thus affecting the importance
of DISWC as a critical area selection criterion.
Organochlorines: Organochlorine surface runoff deliv-
ery efficiency decreases greatly with increasing DISWC.
Likewise, an inverse relationship exists between water-
shed size and delivery efficiency in surface waters. The
potential drift of applied pesticide also increases the im-
portance of DISWC as a selection criterion, particularly for
aerial applications. Two other transport mechanisms, how-
ever, reduce the relative importance of DISWC as a critical
area selection criterion. First, from 20 to 50 percent of
applied Organochlorines are lost by volatilization into the
atmosphere (Maas et al. 1984) depending on air and soil
temperature, humidity, and air circulation rates. However,
the impact of the volatilization transport route on aquatic
systems has been difficult to document because of its
diffuse nature. Second, although the actual percentage of
material which is transported by the biotic route is proba-
bly very small, it is this portion which is the most ecologi-
cally significant.
Organophosphorus Insecticides: DISWC is important for
OP's because of the large drift losses often associated
with their application, and because their relatively low per-
sistence (1 to 8 weeks) in the environment means that
longer transport times result in lower delivery efficiency.
For OP's that are lost primarily in the dissolved phase of
runoff, concentrations dissipate less with overland dis-
tance than sediment-adsorbed materials.
Carbamates: Carbamates are lost primarily in the dis-
solved phase of surface runoff. However, transport effi-
ciency decreases with increasing overland distance be-
cause increased soil-pesticide contact results in increased
adsorption or deposition.
Triazines: As a pesticide class the triazines are relatively
mobile. Studies have found that 0.2 to 16 percent of ap-
plied amounts are lost in surface runoff (Wauchope, 1978)
mostly in the dissolved phase. Soil column leaching ex-
periments show that triazines can move fairly readily
through soils, particularly if the clay content is low (Liu and
Cibes-Viade, 1970). Numerous field studies have found
triazines in ground water (for example, Wehtje et al. 1983).
Thus, triazine transport efficiency decreases only slightly
with increasing DISWC.
Anilides: The anilides are lost almost entirely in the dis-
solved phase (Baker et al. 1982). Edge-of-field studies
show that alachlor is lost in surface runoff even more read-
ily than atrazine (Baker et al. 1976). A watershed study,
however, showed that alachlor had considerably lower de-
livery efficiency to streams than triazines (Wu et al. 1983),
implying that, although the anilides are very mobile ini-
tially, their transport decreases greatly with increasing
DISWC.
Distance To Impaired Water Resources (DISIWR).
This refers to the distance between the nearest water-
course from the pesticide source and the site of the actual
water resource impairment.
The importance of DISIWR as a critical area selection
criterion for various pesticide classes can be generally
estimated from the transport information already pre-
sented. For the Organochlorines, dissipation between the
nearest watercourse and the impaired water resource oc-
curs as a function of sediment deposition. Since or-
ganochlorines are also transported to the impaired water
resource through biotic transport and atmospheric redep-
osition, the importance of DISIWR is reduced.
For the triazine and anilide herbicides, dissipation be-
tween upstream watercourse and the site of impairment
occurs primarily by additional adsorption to particulates
and by plant uptake. Their persistence is on the order of
several months, so degradation between watercourse and
impairment site is generally negligible. In the case of
ground water impairments, however, DISIWR may be very
important since concentrations decrease with increasing
distance through soil profiles.
Other Selection Criteria
Present Management and Conservation Status
(PMCS): As with other agricultural water pollutants, PMCS
should be carefully considered in designating critical ar-
eas for pesticide control. As indicated earlier, the most
important parameter is often the amount of pesticide ap-
plied. Numerous studies have shown that for a given set of
management practices the amount of pesticide lost by
each transport route is roughly proportional to application
rate. If application type and rate information are not availa-
ble, surrogate measures such as level of integrated pest
management (IPM) can indicate how current application
rates compare with what can be achieved without exces-
sive economic risk.
Other important PMCS considerations are the method
and timing of pesticide application. Optimal methods em-
ploy proper drop sizes, ground-based equipment, and for-
mulations that minimize losses through runoff, drift, and
volatilization. Timing options involve avoiding application
on windy days or when precipitation is forecast.
The third major component of PMCS involves the cur-
rently used soil and water conservation practices
(SWCP's) and their effects on surface and subsurface
pesticide losses. A summary of the effects of specific
SWCP's on field losses of the various pesticide classes is
beyond the scope of this report; however, a detailed dis-
cussion can be found in the previously cited Pesticide
Best Management Practices document (Maas et al. 1984).
Designated Priority Subbasin: There may be entire
subbasins within a watershed which for hydrologic or
other reasons do not contribute to the water resource im-
pairment. These subbasins should be deleted from critical
area consideration.
365
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
On-Site Evaluation: The on-site evaluation for deter-
mining pesticide critical areas should focus on: (1) pesti-
cide usage patterns (type, rates, frequency, timing,
method, equipment), (2) disposal practices, and (3) pres-
ence of gullies or sinkholes which short-circuit transport.
Pesticide disposal practices, in particular, can be charac-
terized only by an on-site inspection. The importance of
good disposal practices has become increasingly recog-
nized, since the majority of documented fishkills from pes-
ticides have been attributed to accidental spills and im-
proper disposal (U.S. Environ. Prot. Agency, 1975).
Disposal of containers or tank residue directly into water-
courses or dumping in low lying areas of the farm subject
to periodic flooding represent the worst cases.
CRITICAL AREA SELECTION
PROCEDURE
Based on the preceding selection criteria we propose the
following general procedure for selecting pesticide critical
areas:
Stepl
Characterize the nature and extent of the pesticide-re-
lated impairment, preferably as a quantified impairment to
a designated beneficial use.
Step 2
Characterize to the extent possible the hydrology of the
water resource as it relates to the impairment.
StepS
Use the above information to estimate the pesticide re-
duction needed to protect, improve, or restore the im-
paired use. As noted above, for pesticides this will gener-
ally involve a concentration reduction.
Step 4
Determine the largest potential sources of the sus-
pected pesticides. As a first cut this should be limited to
the farm units or crop acreages where these pesticides
are generally used in the watershed.
StepS
Estimate the extent to which various pesticide BMP's
and BMP systems can reduce the pesticide inputs. The
pesticide BMP review by Maas et al. (1984) provides such
estimates for a variety of BMP's, pesticide types, and
crops.
Step 6
Consider the accessibility of the potential sources to the
impaired water resource. Distance to watercourse ap-
pears to be a good first cut with refinements made on the
basis of distance to impaired water resources.
Step?
Make final refinements on the basis of an on-site evalu-
ation. This evaluation should first verify whether the sus-
pected pesticides are actually being applied. Next, the
present management practices should be evaluated (ap-
plication techniques, timing, formulation, level of inte-
grated pest management, presence of soil and water con-
servation practices) to determine how much reduction can
reasonably be accomplished with additional pesticide
BMP's.
FARM-LEVEL CRITICAL AREA RATING
FORM
Any farm-level rating form for pesticide control cannot be
universally applicable because site-specific consider-
ations such as the nature of the pesticide impairment and
the physical-chemical properties of the pesticide responsi-
ble will affect the importance of various critical area rating
factors. Thus, in Table 3 we propose a general farm level
rating procedure. Referring to Table 3, it is clear that a
minimum score of 100 (indicating that at least the pesti-
cide is actually used) would be required before a farm
could be classified as critical. The actual minimum score
used would depend on the extent of pesticide reductions
needed and the financial resources available. The pres-
ence of soil and water conservation practices which would
reduce edge-of-field pesticide losses is not explicitly in-
Generic factor
Table 3.—Proposed farm-level rating form for selecting critical farms in watersheds
with pesticide-related water resource impairments.
Factor refinement
Points
Use of suspected pesticide
Distance to nearest watercourse (DISWC)
Distance to impaired water resource (DISIWR)
Application method
Level of IPM practiced
Pesticide disposal practice
Erosion rate (use only for sediment-adsorbed
pesticides)
Runoff rate (use only for dissolved pesticides
affecting surface water)
Infiltration capacity (use only for dissolved
pesticides affecting groundwater)
At label recommended rate
Excess of recommended rate
Not used
Short distance (e.g. < 0.5km)
Long distance (e.g. > 0.5km)
Short distance (e.g. < 5km)
Long distance (e.g. > 5km)
Low drift (e.g. ground-based with shields,
recirculators, etc.)
Avg. drift (e.g. ground-based with no shields)
High drift (e.g. aerial)
High
Average
Low
Excellent
Average
Poor (e.g. dumping containers into stream)
High
Average
Low
High
Average
Low
High
Average
Low
100
100
+ Excess %
0
15
0
10
0
0
5
15
-10
0
10
0
15
30
20
10
0
20
10
0
20
10
0
366
-------
eluded in the rating procedure; however, these practices
would affect the points assigned for erosion rate, runoff
rate, or infiltration capacity. The rating procedure assumes
that all of the rating factors apply only to the farm acreage
where the suspected pesticide is actually used.
SUMMARY
Numerous criteria are appropriate for selecting critical ar-
eas for addressing pesticide-related water resource im-
pairments. The most important of these is whether the
pesticide causing the impairment is actually used in the
area and at what rate. The next most important criterion is
the type and severity of the water resource impairment.
The relative importance of other source magnitude and
transport-related criteria depend on the physical, chemi-
cal, or biological properties of the pesticide, particularly
whether it is transported in the dissolved or adsorbed
phase. A farm-level critical area rating form can be used to
rank farm units for pesticide control. Public financial re-
sources for pesticide BMP's should be spent on the basis
of the scores received on the rating form for the greatest
cost-efficiency.
REFERENCES
Baker, D.B. Herbicide contamination in municipal water supplies
of northwestern Ohio. Draft final rep. EPA Grant R005727-01.
Heideberg College, Tiffin, OH.
Baker, J. L, and J. M. Laflen. 1983. Water quality consequences
of conservation tillage. J. Soil Water Conserv. 38(3): 186-93.
Baker, J. L, H. P. Johnson, and J. M. Laflen. 1976. Completion
report: effect of tillage systems on runoff losses of pesticides:
LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
a simulated rainfall study. Trans. Am. Soc. Agric. Eng. 21:
886-92.
Baker, J. L., J. M. Laflen, and R. O. Hartwig. 1982. Effects of
corn residue and herbicide placement on herbicide runoff
losses. Trans. Am. Soc. Agric. Eng. 25(2): 340-3.
Beyerlein, D. C., and A. S. Donigian Jr. 1979. Pages 385-473 in:
Effectiveness of Soil and Water Conservation Practices for
Pollution Control. EPA 600/3-79-106. U.S. Environ. Prot.
Agency.
Liu, L. C. and H. R. Cibes-Viade. 1970. Leaching of atrazine,
ametriyne and prometryne in the soil. J. Agric. Univ. Puerto
Rico 54: 5-18.
Maas, R. P., S. A. Dressing, J. Spooner, M. D. Smolen, and F. J.
Humenik. 1984. Best Management Practices for Agricultural
Nonpoint Source Control—IV: Pesticides. Biol. and Agric.
Eng. Dep. North Carolina State Univ., Raleigh.
Maas, R. P., M. D. Smolen, and S. A. Dressing. 1985. Selecting
critical areas for nonpoint source pollution control. J. Soil Wa-
ter Conserv. 40(1): 68-71.
Motschall, R. M., T. C. Daniel and G. D. Bubenzer. 1984. A soil
sampling method to rank critical barnyards. Trans. Am. Soc.
Agric. Eng. 27(3): 744-6.
U.S. Department of Agriculture, Economic Research Service.
1983: Pesticides, supply and use. Inputs Outlook and Situa-
tion: 3-13.
U.S. Environmental Protection Agency. 1978. Fish-kills caused
by pollution, fifteen year summary. 1961-1975.
Wauchope, R. D. 1978. The pesticide content of surface waters
draining from agricultural fields, a review. J. Environ. Qua).
7(4): 459-72.
Wehtje, G. R., R. F. Spalding, C. B. Orvin, S. Lowry, and J. R.
Leavitt. 1983. Weed Science 31: 610-18.
Wu, T. L., D. L. Correll, and H. E. Remenapp. 1983. Herbicide
runoff from experimental watersheds. J. Environ. Qua). 12:
330-6.
367
-------
R. A. YOUNG
C. A. ONSTAD
Agricultural Research Service
U.S. Department of Agriculture
Morris, Minnesota
D. D. BOSCH
University of Minnesota
St. Paul, Minnesota
W. R ANDERSON
Minnesota Pollution Control Agency
St. Paul, Minnesota
A computer simulation model to analyze nonpoint source
pollution from agricultural watersheds in the State of Min-
nesota has been developed as a predictive tool to investi-
gate water quality problems of different watersheds. This
simulation is based on single storm events defined in
terms of frequency and duration and is intended for use
on watersheds ranging in size from 500 to 23,000 acres.
The model uses geographic cells of data units at a resolu-
tion of 40 acres or 10 acres. The model inputs existing (or
proposed) land management conditions and simulates
the transport of sediment, nutrients, and flow from the
headwaters of a watershed to the outlet in a stepwise
manner so that an assessment can be made for land
parcels in the watershed. The nutrients presently exam-
ined by the model are nitrogen, phosphorus, and chemi-
cal oxygen demand. A small scale version of the model
intended for use on watersheds ranging from 1 to 500
acres has also been developed and tested. This model is
designed to run on a handheld programmable calculator
(such as the HP 41CV) so that the amount of time needed
to assemble and input data and obtain results for a 500-
acre watershed does not exceed 3 mandays.
Current interest in water quality and the importance of
runoff from agricultural lands as a potential nonpoint
source pollutant of surface waters has indicated the need
for developing an objective method to analyze the quality
of runoff water from agricultural watersheds. In the past,
inability to analyze pollution problems from different water-
sheds has resulted in inconsistencies in directing public
funds toward alleviating potential pollution problems.
In response to this need, in November 1981, the Minne-
sota Pollution Control Agency began to develop a uniform
method of analyzing the quality of runoff from agricultural
watersheds within the State. The Agency entered into an
agreement with the Minnesota Soil and Water Conserva-
tion Board, the Soil Conservation Service, and the Agricul-
tural Research Service, to develop two models that can
analyze both sediment and nutrient transport within a wa-
tershed. The first of these two models, Agricultural Non-
point Source (AGNPS I), was to be developed for a main-
frame computer system and would analyze large
agricultural watersheds from 200 to 12,000 ha in size. The
second, called AGNPS II, was to be developed for use on
a small, handheld, programmable calculator to analyze
watersheds from 1 to 200 ha in size.
Us® of Models
The intended use of the models is to compare the per-
formance of a watershed during a runoff event to preset
standards of performance or to the performance of other
watersheds experiencing the same type of event. The
event could be a predetermined design storm, such as a
25-year, 24-hour frequency rainfall, or it could be any other
rainfall event for which the response of the agricultural
watershed needed to be analyzed. Predicted outflow of
water, sediment, and chemicals from the watershed could
be compared to established standards to determine
whether the outflow from the watershed posed a potential
pollution problem. In the absence of established stan-
dards, relative comparisons of outflow from several water-
sheds subject to the same design rainstorm could be used
to determine which watershed presented the greatest po-
tential pollution hazard in terms of outflow of sediment and
nutrients.
Once a watershed has been identified as needing reme-
dial measures, the models can be used to assess the
effects of applying alternative management practices to
bring about desired changes. This can be done by chang-
ing selected input parameters corresponding to various
management alternatives and comparing the new output
results with the original output to see if the proposed
changes have achieved the desired results.
AGNPS I and AGNPS II are event-based models intended
to simulate sediment and nutrient transport from primarily
agricultural watersheds. The basic components of the
models are hydrology, erosion, and sediment and nutrient
transport. The models work on a cell basis. Cells are uni-
form square areas subdividing the watershed, making it
possible to analyze any small area within the watershed.
Proceeding from the headwaters to the outlet, potential
pollutants are routed through the cells step by step so the
flow at any point may be examined.
Cell size varies with desired detail and watershed size.
Four ha (10 acres) ceils are recommended for watersheds
up to 800 ha (2,000 acres), and 16 ha (40 acres) cells are
recommended for larger watersheds. All watershed char-
acteristics are expressed and calculated at the cell level.
Figure 1 shows an example of a typical small watershed of
308 ha (760 acres) after it has been divided into 19 16-ha
cells. Arrows in each cell depict the major drainage pat-
tern of the watershed.
368
-------
LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
Figure 1.—A 308 ha (760 acre) sample watershed divided
into 16 ha (40 acre) cells.
Hydrology
In the hydrology portion of the models, runoff volume and
peak flow are calculated. The Soil Conservation Service
curve number (Soil Conserv. Serv. 1972) is used to esti-
mate overland runoff from each of the cells in the water-
shed using the equation
Q =
(P - 0.2S)2
P + 0.8S
where Q is the runoff, P is the rainfall, and S is a retention
parameter. The retention parameter is determined as
S = 1000/CN-10
(2)
where CN is the curve number. This method was chosen
because of its simplicity, wide use by the agencies in-
volved, and because the inputs are readily available.
The peak flow for channelized flow through each cell is
estimated using an equation from CREAMS (Smith and
Williams, 1980):
Q = 3.790 A07 CS0159 (RO/25.4)<0903A°0168> (LW)-°187 (3)
where Qp is the peak runoff in m3/s; A is the upslope
drainage area in km2; CS is the channel slope in m/km;
RO is the runoff volume in mm; and LW is the length-
width ratio, approximated by L2/A, where L is the water-
shed length and A is the drainage area.
This equation was tested for applicability to the North
Central Region including Minnesota. Data from 20 upper
Midwest watersheds were analyzed to compare measured
peak flows with estimates from Equation (3). A regression
produced the equation
Obs = Est -1.0116 (4)
with Obs being the observed values and Est being the
estimated values. The coefficient of determination, r2, was
0.81.
Eros/on
In the erosion component of the models, a modified uni-
versal soil loss equation is used to estimate upland ero-
sion for single storms (Wischmeier and Smith, 1978):
= EI.K.L.S*C-P* SSF
(5)
where A is the soil loss; El is the storm energy-intensity
value; K is the soil credibility; L is the slope-length factor;
S is the slope-steepness factor; C is the cover and man-
agement factor; P is the supporting practice factor; and
SSF is a calculated factor to adjust for average slope
shape within a cell. The slope shape factors (SSF) were
calculated using complex slope factors found in USDA
Handbook No. 537 (Wischmeier and Smith, 1978). Ero-
sion values are calculated for each cell of the watershed.
Sediment entering channels from upland erosion is di-
vided into five particle size classes: clay, silt, small aggre-
gates, large aggregates, and sand.
Sediment Transport
Following the calculation of upland erosion and runoff
from each cell, detached sediment is routed through the
watershed. The method used involves equations for sedi-
ment transport and deposition as described by Foster et
al. (1981) and Lane (1982). The basic routing equation is
derived from the steady state continuity equation:
Qs(x) = Qs(o) + a* x/L - j D(x) w dx (6)
o
where Qs(x) is the sediment discharge at the downstream
end of the channel reach; Qs(o) is the sediment discharge
at the upstream end of the channel reach; Qs, is the lateral
inflow rate for sediment; x is the downstream distance; w
is the channel width; L is the reach length; and D(x) is the
deposition rate. Deposition rate is represented as
. D(x) = V^/qMIqsM-g&O] (7)
where V^ is the particle fall velocity; q(x) is the discharge
per unit width; qs(x) is the sediment load per unit width;
and gs(x) is the effective transport capacity per unit width.
Effective transport capacity is a modification of the
Bagnold stream power equation (Bagnold, 1966)
9s = V 9s =
TV2
(8)
where gs is transport capacity, TJ is an effective transport
factor; k is the transport capacity factor; T is the shear
stress; and v is the average flow velocity determined by
Manning's equation. The sediment load for each of the
five particle size classes leaving a cell is calculated using
the equation
• Q8(o)+Qsl
vss
Qs(x) =
2q(x)
2q(x) + Ax VE
wAx
r_v*L
|_q(o)
q(x)
9«x)
(9)
Equation (9) is the basic routing equation that drives the
model.
Nutrient Transport
The nutrient portion of the model estimates the transport
of nitrogen (N), phosphorus (P), and chemical oxygen de-
mand (COD) through the watershed. N and P were chosen
because they are essential elements for plant growth and
are considered major contributors to eutrophication of sur-
face waters. COD is a measure of the amount of oxygen
required to oxidize organic and oxidizable inorganic com-
pounds in water and, thus, can be used to indicate the
degree of pollution in the outflow. The equations used to
calculate nutrient transport are from CREAMS (Frere et al.
1980), and a feedlot evaluation model (Young et al. 1982a)
with some modifications for the effects of variation in soil
texture.
Nutrient transport calculations are divided into two
parts, one dealing with sediment-attached nutrients and
the second part dealing with soluble nutrients. Nutrient
yield associated with sediment is calculated using total
sediment yield from each cell according to the equation
Nut** = Nut*. SY. ER
ER =A.Q,.*B*T,
(10)
(11)
369
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 1 .—Input data file.
Line 1. Watershed identification (30 character description)
Line 2. Area of each cell / Number of cells / Precipitation/
Energy-Intensity value of the storm
Line 3. to end of file—cell parameters
Column No.
1 Cell number
2 Number of the cell Into which it drains
3 SCS curve number
4 Average land slope (%)
5 Slope Shape factor (uniform, convex, or concave)
6 Average field slope length
7 Average channel slope (%)
8 Average channel side slope (%)
9 Mannings roughness coefficient for the channel
10 Soil credibility factor (K) from USLE
11 Cropping factor (C) from USLE
12 Practice factor (P) from USLE
13 Surface condition constant (factor based on land use)
14 Aspect—(1 of 8 possible directions indicating the
principal drainage direction from the cell)
15 Soil Texture (sand, silt, clay, peat)
16 Fertilization level (zero, low, medium, high)
17 Incorporation factor (percent of fertilizer left in top
centimeter of soil)
18 Point source indicator (indicates existence of a point
source input within a cell)
19 Gully source level (estimate of amount of gully erosion
in a cell)
20 Chemical oxygen demand factor
21 Impoundment factor (a factor indicating presence of an
impoundment terrace system within the cell)
where Nutsed is N or P transported by the sediment; Nuts!
is N or P content in the field; Qs is sediment yield predicted
by the sediment transport equation; ER is an enrichment
ratio for N or P; factors A and B are assumed constant and
equal to 7.4 and -0.2, respectively; and Tf is a correction
factor for soil texture (Young et al. 1985).
The soluble nutrient algorithm considers the effects of
nutrient levels in the rainfall, fertilization, and leaching.
Soluble N in runoff is
N*,, = CRO * Next . Q . 0.01 (12)
where N^, is the concentration of soluble nitrogen in the
runoff; CRO is the mean concentration of soluble N in the
soil surface during runoff; Next is the extraction coefficient
for movement into runoff; and Q is the total runoff. Soluble
P in runoff is
P«>i = Css * PB«t * Q * 0.01
(13)
where PSOI is the concentration of soluble P in the runoff;
Css is the mean concentration of soluble P in the soil
surface during runoff; Pext is the extraction coefficient for
movement into runoff; and Q is the total runoff.
COD in the models is assumed soluble. Calculations of
the amount of soluble COD in the runoff are based on the
runoff volume and the average concentration of COD in
that volume. Various background concentrations of COD
obtained from the literature for runoff waters from various
land uses are used as a basis for predicting the COD
concentration in the runoff from each cell. Soluble COD is
assumed to accumulate only once channelized, without
any allowable losses.
Point Source Inputs
AGNPS I and II treat nutrient contributions from animal
feedlots as point sources and route them with the other
nutrients. Contributions from feedlots are calculated using
a feedlot pollution model developed by Young et al.
(1982b) as a subroutine in the main model. The feedlot
model calculates nutrient concentration and mass at both
the feedlot edge and at the point of input to a body of
water.
Streambank and gully erosion are accounted for by en-
tering estimated values as point sources. Sediment from
gully sources is also considered in the transport phase of
the model.
MODEL INPUT AND OUTPUT
A summary of the inputs for AGNPS I is shown in Table 1.
The parameters may be obtained from published data and
readily available watershed records.
Various output options are available with the models.
Preliminary output, given for all watersheds being exam-
ined, includes the area of the watershed and the cell size,
the storm precipitation and erosivity value, estimates of
runoff volume and peak flow rate at the watershed outlet,
and of the area-weighted erosion, both upland and chan-
nel, sediment delivery ratio, sediment enrichment ratio,
mean sediment concentration, and sediment yield. These
values are given for each of the five particle size classes,
as well as a total for the watershed. A nutrient analysis
which includes the N, P, and COD mass per unit area for
both soluble and sediment adsorbed nutrients, and the N,
P, and COD concentrations in the runoff is also given.
Table 2 shows the output of sediment and nutrient con-
tributions of a 308-ha watershed resulting from a 25-year
24-hour rainstorm. Given the input conditions, this storm
would produce an estimated runoff volume of 7.62 cm
(3 in) at the outlet with a peak flow rate of 35.76 m3 s~1
(1,263 cfs). The total sediment yield is estimated to be
1,802 tonnes metric (1,987 tons). Substantial amounts of
N and P would also be discharged as indicated. Table 2
also includes a cell by cell summary of runoff and sedi-
ment yield for the entire watershed. This portion of the
output is optional.
Optional information is also available for any individual
cell in the watershed. Information given when examining
individual cells, shown in Table 3, includes a runoff analy-
sis, including drainage area, runoff volume, and peak run-
off rate, and a sediment analysis, with estimates for each
of the five particle size classes of upland erosion, sedi-
ment yield, percentage of the yield from within the cell and
from outside sources, and the percent deposition in the
cell. A detailed nutrient analysis for individual cells, also
shown in Table 3, includes estimates of adsorbed and
soluble nutrients in mass per unit area, and the concentra-
tion of the nutrients in the runoff.
TESTING
The AGNPS models were first tested with data from two
32-ha (80 acre) experimental watersheds located near
Treynor, Iowa (Agric. Res. Serv. 1970) and a 195 ha (480
acre) watershed located near Hastings, Nebraska (Agric.
Res. Serv. n.d.). Although these watersheds are smaller
than will be generally used with the model, the runoff and
sediment yield data necessary for testing were readily
available. Sediment yield estimates from the model com-
pared favorably with the measured values from the Trey-
nor watersheds. A statistical comparison of the observed
and predicted sediment yields showed that the model
overpredicted by 2.3 percent, with a coefficient of determi-
nation, r2, of 0.95. Sediment yield was predicted less accu-
rately for the Hastings watershed and resulted in a coeffi-
cient of determination of 0.76. Since insufficient data were
available for nutrient analyses, nutrients have not been
tested. Additional data are being collected from several
large watersheds in Minnesota and South Dakota and will
be used to further test the model.
370
-------
. LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
Table 2.—Watershed output and cell by cell summary-
Agricultural Nonpolnt Source Pollution Model
Watershed Studied: Meadow Brook 760 Contoureds
The area of the watershed is:
The area of each cell is:
The characteristic storm precipitation is:
The storm energy-intensity value is:
30.40 ha
1.60 ha
12.5cm
75
Values at the watershed outlet: Cell Number 18
Runoff volume (in.):
Peak runoff rate (cfs):
Total nitrogen in sediment (Ibs/acre):
Total soluble nitrogen in runoff (Ibs/acre):
Soluble nitrogen concentration in runoff (ppm):
Total phosphorus in sediment (Ibs/acre):
Total soluble phosphorus in runoff (Ibs/acre):
Soluble phosphorus concentration in runoff (ppm):
Total soluble chemical oxygen demand (Ibs/acre):
Soluble chemical oxygen demand concentration in runoff (ppm):
Sediment analysis
3.0
1263
6.83
3.20
5
3.41
1.15
2
82.51
121
Particle
Type
Clay
Silt
Sagg
Lagg
Sand
Toll
Area Weighted
Erosion
Upland Channel
(t/a) (t/a)
.31
.50
3.12
1.93
.37
6.24
.00
0.00
0.00
0.00
0.00
0.00
Delivery
Ratio
(%)
95
84
60
1
1
42
Enrichment
Ratio
2.3
2.0
1.4
.0
.0
1.0
Mean
Concentration
(ppm)
881
1228
5512
66
12
7698
Area
Weighted
Yield
(t/a)
.3
.4
1.9
.0
.0
2.6
Yield
(tons)
227.3
317.0
1422.7
17.0
3.1
1987.1
Values at each cell
Cell
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Drainage
Area
(acres)
40
80
40
40
40
160
240
80
40
40
40
320
520
80
80
120
160
760
40
Runoff
Volume
(in.)
3.4
3.4
4.0
2.4
3.4
3.3
3.4
2.5
2.4
3.1
4.0
3.4
3.0
2.4
3.1
3.4
3.1
3.0
2.4
Generated
Above
(%)
0
50
0
0
0
77
84
47
0
0
0
90
94
50
50
61
81
95
0
Peak
Rate
(cfs)
226
305
153
185
253
471
570
244
185
129
279
689
1021
249
172
286
322
1263
185
Cell
Erosion
(t/a)
2.5
2.5
9.5
8.3
3.7
7.3
2.5
2.9
8.3
9.6
5.9
2.2
8.3
8.3
9.6
7.7
8.3
3.1
8.3
Sediment
Generated
Above Within Yield
(tons) (tons) (tons)
0.0
68.3
0.0
0.0
0.0
216.4
503.9
214.6
0.0
0.0
0.0
647.1
1226.4
214.6
135.5
237.8
376.5
2107.4
0.0
101.3
101.3
380.1
330.1
146.5
291.4
101.3
114.8
330.1
385.1
234.4
86.9
330.1
330.1
385.1
309.7
330.1
125.6
330.1
68.3
117.4
129.5
214.6
99.0
374.3
494.5
221.8
214.6
135.5
152.6
640.5
1380.4
364.1
237.8
376.5
512.5
1987.1
214.6
Deposition
(%)
33
31
66
35
32
26
18
33
35
65
35
13
11
33
54
31
27
11
35
SUMMARY
Two simplified hydrologic models have been developed to
analyze nonpoint source pollution from agricultural water-
sheds in Minnesota. They are based on single storm
events defined in terms of frequency and duration and are
intended for use on watersheds ranging in size from 1 to
12,000 ha. The models use geographic cells of data units
at a resolution of 1 to 16 ha and simulate the transport of
sediment, nutrients, and flow from the headwaters of a
watershed to the outlet in a stepwise manner so that the
flow at any point within a watershed can be examined. The
nutrients presently examined by the model are N, P, and
COO. Preliminary testing of the models has shown that
estimates from the models are reasonably accurate. The
models are simple and easy to use and input for the
models is minimal. The models were developed using cur-
rent, yet easily adaptable information and equations.
Output information at the watershed outlet can be used
to assess the potential pollution hazard posed by a water-
shed as a whole while the output information for each cell
can be examined to locate those local areas within a wa-
tershed that contribute the greatest amount of pollutants
to a waterway. In this way, areas where remedial measures
might be initiated to improve the quality of runoff at the
watershed outlet can be pinpointed. After a watershed has
been identified as needing remedial measures, the
models can be used to assess the effects of applying
alternative management systems.
371
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 3.—Watershed output and detailed summary of selected cells.
Agricultural Nonpoint Source Pollution Model
Watershed Studied: Meadow Brook 760 Contoureds
The area of the watershed is:
The area of each cell is:
The characteristic storm precipitation is:
The storm energy-intensity value is:
Values at the watershed outlet: Cell Number 18
Runoff volume (in.):
Peak runoff rate (cfs):
Total nitrogen in sediment (Ibs/acre):
Total soluble nitrogen in runoff (Ibs/acre):
Soluble nitrogen concentration in runoff (ppm):
Total phosphorus in sediment (Ibs/acre):
Total soluble phosphorus in runoff (Ibs/acre):
Soluble phosphorus concentration in runoff (ppm):
Total soluble chemical oxygen demand (Ibs/acre):
Soluble chemical oxygen demand concentration in runoff (ppm):
Sediment analysis
760 acres
40.0 acres
5.0 inches
75
3.0
1263
6.83
3.20
5
3.41
1.15
2
82.51
121
Particle
Type
Clay
Silt
Sagg
Lagg
Sand
Totl
Area Weighted
Erosion
Delivery
Upland Channel Ratio
(t/a) (t/a) (%)
.31
.50
3.12
1.93
.37
6.24
.00
0.00
0.00
0.00
0.00
0.00
95
84
60
1
1
42
Enrichment
Ratio
2.3
2.0
1.4
.0
.0
1.0
Mean
Concentration
(ppm)
881
1228
5512
66
12
7698
Area
Weighted
Yield
(t/a)
.3
.4
1.9
.0
.0
2.6
Yield
(tons)
227.3
317.0
1422.7
17.0
3.1
1987.1
Values at each cell
Hydrology
Drainage Overland
Cell Area Runoff
Number (acres) (in.)
3 40 4.0
13 520 2.4
Upstream Peak Row Downstream Peak Row
Runoff Upstream Runoff Downstream Particle
(in.) (cfs) (in.) (cfs) Type
CLAY
SILT
0.0 0 4.0 153 SAGG
LAGG
SAND
TOTL
CLAY
SILT
3.1 1034 3.0 1021 SAGG
LAGG
SAND
TOTL
Cell
Erosion
(t/a)
.5
.8
4.8
2.9
.6
9.5
.4
.7
4.1
2.6
.5
8.3
Sediment
Generated
Above
(tons)
0.0
0.0
0.0
0.0
0.0
0.0
124.4
181.0
881.3
33.9
5.8
1226.4
Within
(tons)
19.0
30.4
190.0
117.8
22.8
380.1
16.5
26.4
165.0
102.3
19.8
330.1
Yield
(tons)
15.6
18.4
86.0
8.3
1.2
129.5
140.8
204.7
983.5
44.1
7.3
1380.4
Deposition
(ft)
18
39
55
93
95
66
0
1
6
68
71
11
Nutrient analysis
Water Soluble
Cell
Number
3
13
Drainage
Area
(acres)
40
520
Runoff
Volume
(In.)
4.0
3.0
Sediment
(Ibs/a)
4.85
1.56
Within
Cell
(Ibs/a)
.81
1.30
Cell
Outlet
(Ibs/a)
.81
3.69
(ppm)
1
5
Water Soluble
Sediment
(Ibs/a)
2.43
.78
Within
Cell
(Ibs/a)
.14
.42
Cell
Outlet
(Ibs/a)
.14
1.34
Water Soluble
Within Cell
Cell Outlet
(ppm) (Ibs/a) (Ibs/a)
0 103.73 103.73
2 42.84 89.40
(ppm)
115
130
1 Chemical oxygen demand
REFERENCES
Agricultural Research Service, n.d. The Central Great Plains
Experimental Watershed. A summary report of 30 years of
hydrologic research. USDA Hydrograph Lab. Univ. Nebraska
Statis. Lab., Hastings. U.S. Dep. Agric. Mimeo Rep. (probably
late 1967).
1970. Annual Research Report, North Central Water-
shed Res. Center, Columbia, MO. U.S. Dep. Agric.
Bagnold,'R.A. 1966. An approach to the sediment transport
problem from general physics. U.S. Geol. Survey Prof. Paper
422-J.
Foster, G.R. et al. 1981. Estimating erosion and sediment yield
on field-sized areas. Trans. Am. Soc. Agric. Eng. 24(5): 1253-
62.
Frere, M.H., J.D. Ross, and L.J. Lane. 1980. The nutrient sub-
model. Pages 65-86 in CREAMS, A Field Scale Model for
Chemicals, Runoff, and Erosion from Agricultural Manage-
ment Systems. Conserv. Res. Rep. 26, Vol. 1. U.S. Dep.
Agric., Washington, DC.
Lane, L.J. 1982. Development of a procedure to estimate runoff
and sediment transport in ephemeral streams. Pages 275-82
in Recent Developments with Explanation and Prediction of
Erosion and Sediment Yield. Proc. Exeter Symp. International
372
-------
Association of Hydrological Sciences. No. 137.
Smith, R.E., and J.R. Williams. 1980. Simulation of the surface
water hydrology. Page 15 in CREAMS, A Field Scale Model for
Chemicals, Runoff, and Erosion from Agricultural Manage-
ment Systems. Conserv. Res. Rep. 26, Vol. 1. U.S. Dep.
Agric. Washington, DC.
Soil Conservation Service. 1972. Hydrology. Sec. 4, Chap. 10 in
SCS National Engineering Handbook. U.S. Dep. Agric.,
Washington, DC.
Wischmeier, W.H., and D.D. Smith. 1978. Predicting rainfall ero-
sion losses. Agricultural Handbook 537. U.S. Dep. Agric.,
Washington, DC.
LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
Young, R.A., M.A. Otterby, and A. Roos. 1982a. A technique for
evaluating feedlot pollution potential. J. Soil Water Conserv.
37(1): 21-3.
. I982b. An evaluation system to rate feedlot pollution
potential. USDA ARM-NC-17 (April). U.S. Dep. Agric., Wash-
ington, DC.
Young, R.A., C.A. Onstad, D.D. Bosch, and W.P. Anderson.
1985. Agricultural nonpoint pollution models (AGNPS) I and
II—Model Documentation. Submitted to Minn. Pollution Con-
trol Agency in partial fulfillment of the Trust Fund Cooperative
Agreement No. 12-14-3001-071. February 1985.
373
-------
BROCK TUNNICLIFF
STANLEY K. BRICKLER
School of Renewable Natural Resources
University of Arizona
Tucson, Arizona
Contrary to some popular opinions regarding pristine nat-
ural environments, examination of recreational water
quality in Grand Canyon National Park has revealed two
naturally occurring, and potentially health threatening,
nonpoint sources of fecal loads in Colorado River and
tributary stream waters. Following monsoon type rainfall
events, fecal loading of the river, as indicated by fecal
coliform densities, exceeded the recreational primary
contact standard, 200 FC 100 ml"'. In contrast, during
extensive drought periods, mean FC densities in the river
and tributaries were 10 and 20 FC 100 ml'1, respectively.
Fecal sources in the Grand Canyon watershed include
wildlife, some livestock, and dispersed recreation.
Stream and lake waters in national parks, forests, and
other wildlands are often viewed by the general public as
relatively pristine, that is, completely free of harmful micro-
organisms (Newman, 1981). Based on this common mis-
conception, backcountry visitors often swim in or drink
untreated natural waters without regard for potential
health hazards associated with nonpoint source fecal con-
tamination. In a study examining waters reflective of natu-
ral conditions, fecal indicator bacteria (coliforms, fecal col-
iforms (FC), and enterococci) were isolated from streams
draining two forested mountain watersheds in Montana,
one open and the other closed to public use (Bissonnette
et al. 1970; Stuart et al. 1971; Water and Bottman, 1967).
In Colorado, Salmonella and Arizona have been isolated
from a high mountain stream in the Cache la Poudre River
basin (Fair and Morrison, 1967). Coliforms and fecal strep-
tococci (FS) have also been isolated from selected moun-
tain streams and lakes in Grand Teton National Park, Wyo-
ming (Stuart et al. 1976). Concentrations of total coliforms,
fecal coliforms, and enterococci were found to vary sea-
sonally in another small Wyoming stream draining a natu-
ral area (Skinner et al. 1974). Presumably, the enteric or-
ganisms contaminating these waters were nonpoint in
origin from wildlife and perhaps dispersed human or live-
stock sources.
The preceding studies were not designed to determine
the suitability of water for recreational contact or potability.
The reported data indicate, however, that the waters ex-
amined met full body contact standards (Ariz. Admin.
Comp. 1980; Water Qual. Criteria, 1967), but not stand-
ards for potable waters (PL. 93-523). Results in these
studies were obtained for nonstorm flow periods. Studies
of storm runoff from urban, suburban, and rural areas in
Ohio have shown that total coliform, fecal coliform, and
fecal streptococci densities can be"highly concentrated,
particularly in the more impervious, paved urban setting
as compared to vegetated suburban and rural locations
(Geldreich et al. 1968).
To determine the effects of storm flow on microbiologi-
cal water quality in a pristine, natural environment, we
examined the Colorado River drainage basin within Grand
Canyon National Park, beginning in 1978. This study, in
contrast to those just reviewed, was able to investigate the
impacts of nonpoint fecal loading on water quality under
the highly ephemeral precipitation and stream flow re-
gimes of the semiarid Southwest.
Study Site. The Colorado River drainage basin within
Grand Canyon includes over 20 perennial and thousands
of ephemeral tributaries. As one of the premier wilderness
recreation areas in North America, Grand Canyon is vis-
ited annually by approximately 15,000 river float trip par-
ticipants and tens of thousands of backpackers (U.S. Nat.
Park Serv. 1979). Peak river recreation use occurs during
the summer.
Field work on the Colorado River extended from Lee's
Ferry to Diamond Creek (362 km), including 26 tributaries
within this reach (Fig. 1). Colorado River stream flow is
controlled by hydroelectric releases from Lake Powell
(65,000 ha), 22.5 km upstream of Lee's Ferry. Summer
flows typically resemble a daily tide as releases fluctuate
between 85 and 935 m3/s. In contrast to the river, peren-
nial tributary flows are generally insignificant (< 0.28 m3/
s). Flash floods can increase tributary flows hundreds to
thousands fold (U.S. Geolog. Surv. 1980). Flood events in
most perennial side streams and ephemeral washes are
of short duration, because of the localized nature of thun-
derstorms and limited basin size (3,500 km2 collectively).
The expansive watersheds of the Paria and Little Colorado
Rivers and Kanab Creek (68,000 km2 collectively) can,
however, generate storm flows lasting from days to
months.
To establish a profile of water quality in Grand Canyon,
the river corridor had to be reached through the canyon.
Travel by white water raft was the only practical way to
meet this requirement. Accordingly, two river rafts, each
capable of carrying four to five investigators with complete
research and camping supplies for 2 weeks, were de-
signed. Each boat consisted of two inflatable surplus
bridge pontoons strapped catamaran style to a rigid alumi-
num frame.
Sample Design. The influence of both drought and
rainfall periods on water quality were examined during
four consecutive summers (1978-81). The research
schedule included six, 2-week sample periods in 1978,
two in 1979, and one each in 1980 and 1981. A fixed-site
design was used in 1978 to establish location-specific pro-
files of river and tributary water quality River sites (46)
were located at attraction sites, camping beaches, and at
positions bracketing tributary confluences. Except for se-
lected tributary bracketing sites, fixed sites were elimi-
nated during subsequent years in lieu of time series sam-
ples. Fixed sites were sampled on 26 tributaries in 1978,
13 in 1979, and 12 in both 1980 and 1981.
374
-------
LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
Lee's Ferry
— Mile 16
NAVAJO
INDIAN
— Mile 32
NATIONAL FOREST
TRIBUTARY LEGEND
I. Poria River
2. Vasey's Spring
3. Nankoweep Creek
4. Little Colorado River
5. Clear Creek
6. Bright Angel Creek
7. Garden Creek
8. Monument Creek
9. Hermit Creek
10. Boucher Creek
11. Crystal Creek
12. Shinumo Creek
13. Elves Chasm
14. Stone Creek
15. Tapeats Creeks
16. Deer Creek
17. Kanab Creek
18. Olo Creek
19. Matkatamiba Creek
20. Havasu Creek
2 I. National Creek
22. Fern Glen
23. Mohawk Creek '
24. Pumpkin Spring
25. Three Springs
26. Diamond Creek
26
Mile 225
Figure. 1 .—Colorado River and Tributaries in Grand Canyon.
With an average flow rate of 8.3 km/h, Colorado River
water has a travel time between Lee's Ferry and Diamond
Creek of about 45 hours. This fast, continual renewal of
water in the river channel means that surface water quality
will vary over short time intervals. Accordingly, time series
samples were collected (1978-81) to detect these
changes. Samples were collected daily at 0800,1200, and
1800 hours, at the research raft locations. These times
were selected to monitor water quality while float trips
make principal use of river water. Because of overnight
and sampling stops, the rafts traveled at a rate much
slower than the current. Accordingly, samples collected at
progressive time intervals were from new upstream units
of stream flow.
Water samples were collected in sterile Whirl Pak bags
from the top 15 cm of water and stored on ice. Fecal
coliform, total coliform, and fecal streptococci concentra-
tions in water samples were determined by membrane
filtration within 6 hours of collection (Stand. Methods,
1980). The field membrane filtration system used in Grand
Canyon consisted of a high volume Guzzler 400 boat bilge
pump as a hand-powered vacuum source, a Millipore
three-place manifold, Gelman magnetic filter funnels, and
Gelman type GN-6 0.45 nm presterilized membrane filt-
ers. Membrane filtration equipment was sterilized by 5-
minute exposure to ultraviolet germicidal lamps (2537
mn), housed in a military radio box equipped with a 12-volt
battery. Fresh media were prepared in the field from pre-
weighted lots of m-FC agar, KF-Streptococcus agar, and
M-Endo agar sealed in sterile airtight vials. Samples were
incubated in Millipore aluminum block incubators, housed
individually in radio boxes equipped with 12-volt batteries.
Nonstorm flow turbidities in the Colorado River and trib-
utaries were usually sufficiently low that suspended sedi-
ment did not accumulate on membrane filters. When sedi-
ment loads became excessive during storm flow periods,
filter volumes were split into smaller units, processed on
separate filters, and counted collectively. For example, a
100-ml filter volume became 4 x 25 ml or if necessary 10
x 10 ml (Geldreich, 1975). Turbidity measurements were
made of each surface water sample using a Hach DR-EL
colorimeter.
RESULTS
Nonstorm flow periods. The summers of 1978-80 were
marked by drought. Consequently, the Colorado River and
its tributaries were free of major storm flow during those
years (Fig. 2). Log mean turbidities for the Colorado River
were < 16 NTU through the drought period, and mean
and median fecal coliform densities were < 2.4 FC/100
ml (Table 1). Based on time series samples, fecal coliform
concentrations during this period were < 10 FC/100 ml
and < 3 FC/100 ml 95 percent and 75 percent of the time,
respectively. In .only three of 443 samples did bacterial
densities exceed 100 FC/100 ml. Of these three samples,
two (245 and 1165 FC/100 ml) exceeded the recreational
full-body-contact standard of 200 FC/100 ml. Both of these
observations were associated with temporary, rain-in-
duced turbidities of 100 NTU.
The third peak observation, 120 FC/100 ml, occurred in
conjunction with a turbidity of 99 NTU. Visible beach and
bed scour within the Colorado River appeared to be the
source of this turbidity. Scour is particularly pronounced
375
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
r—A
MILE 61.5 LITTLE COLORADO RIVER INFLOW JULY 2O
TURBIDITY = 37.500 FTU
FC CONCENTRATIONS = 5.5OO/IOOML.
MILE I.O PARIA RIVER INFLOW JULY 17
TURBIDITY = 262,500 FTU
FC CONCENTRATION = I3.OOO/IOOML.
O LOG FC DENSITY LOG MEAN SAMPLE MCL ZOOFC/IOOml A LOG TURBIDITY
IO.OOO
1,000
30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
RIVER MILES DOWNSTREAM FROM LEES FERRY
Figure 2.—Log FC/100 ml log turbidity for 479 Colorado River water samples, 1978-81. FC = fecal coliform.
when river beaches, exposed during hydroelectric ebb
flows, are eroded during peak flow releases. Notable
beach scour occurred during nonstorm flow periods, but
analysis of variance between fecal coliform densities at
ebb and peak flows showed no significant increase (P >
0.05) in fecal coliform numbers during peak releases.
Scour apparently leads only to occasional water quality
problems.
Fecal coliform densities in tributaries were slightly more
variable during 1978-80 than in the river. Mean and me-
dian fecal coliform concentrations in tributaries were simi-
lar to those in the river (Table 1). Individual observations
exceeded 10 FC/100 ml four times more often than in the
river, but were < 20 FC/100 ml 90 percent of the time.
Tributaries generally carry low stream flow volumes during
drought periods. This may subject them to temporary, but
significant, variations in fecal coliform densities. Tributary
impact on river water quality was minimal during nonstorm
flow periods, even when fecal coliform densities in the
hundreds per 100 ml were found in inflow waters. Appar-
ently, the stream flow volume of the Colorado River during
these periods so exceeded that of the tributaries that any
input was diluted beyond detection.
Based on log means, neither the river nor its tributaries
exceeded the full body contact standard during 1978-80
(Table 1). Of the 26 tributaries examined, Hermit Creek,
Elves Chasm, and Havasu Creek most frequently had in-
dividual fecal coliform observations above 100 FC/100 ml.
The Hermit Creek and Elves Chasm watersheds are in
natural states, but the stream courses are intensively used
for water-based recreation. Havasu Creek drains the
Havasupai Indian Reservation and the village of Supai,
and is also used intensively for recreation by backpackers
and river runners.
In addition to recreational contact, Grand Canyon visi-
tors also use the river and tributaries for drinking water.
Total and fecal coliform data for 1978-80 (Table 1) indicate
that river and tributary waters consistently failed to meet
drinking water standards.
Because disposal of human sewage is carefully regu-
lated, livestock grazing excluded, and pack stock use
carefully restricted, wildlife are probably the most impor-
tant sources of fecal contamination during nonstorm flow
periods. This contention is supported by FC:FS ratios con-
sistently below 0.7, indicating animal-dominated fecal
contamination (Table 1) (Geldreich, 1976). These ratios
cannot be considered reliable unless fecal deposition has
occurred within 24 hours of sampling. Hydrologic move-
ment from watershed surfaces to stream channels is com-
pletely absent during prolonged drought. Accordingly, fe-
cal contamination of the river and tributaries must result
from direct deposition in stream channels. Because of fast
flow rates, stream waters have short residencies in Grand
Canyon. Accordingly, the ratios calculated herein are
probably based on recent deposition.
Storm flow periods. In contrast to the nonstorm flow
periods of 1978-80, the summer of 1981 had a well-devel-
oped rainy season, which generated turbid storm flows in
both the Paria and Little Colorado Rivers. Colorado River
turbidity and fecal coliform levels downstream of these
tributaries were markedly higher than in previous years
(Table 1, Fig. 2). The close association between river fecal
coliform densities and turbidities suggests that storm flow
turbidity may be useful to model fecal coliform (Fig. 3).
River storm flow turbidities and fecal coliform densities
were positively correlated, r = 0.54 (P < 0.05).
Individual samples from the Colorado River during the
storm flow period exceeded the 200 FC/100 ml standard
with a frequency suggesting marginal full-body-contact ac-
ceptability (Fig. 3). Storm flow observations of the Paria
and Little Colorado Rivers clearly indicated that fecal con-
tamination levels there also exceeded contact limits.
The highly ephemeral precipitation pattern of the arid
Southwest was the most important factor influencing rec-
reational water quality in Grand Canyon. Based on 1978-
80 data, high quality waters can be expected in the Colo-
rado River and tributaries during drought portions of the
cycle. In contrast, recreational water contact standards
are likely to be exceeded during storm flows.
Nonstorm flow periods. A highly consistent water
quality profile was found for the river and tributaries during
nonstorm flow periods. Based on this profile, concentra-
tions of s 10 FC/100 ml and < 20 FC/100 ml in the river
and tributaries, respectively, could be expected. These
nonstorm flow data established a baseline fecal contami-
nation level for Grand Canyon comparable to those re-
ported for high quality, mountain streams, where fecal coli-
form densities were generally < 20 FC/100 ml (Skinner et
al. 1974; Varness, 1978). Total coliform densities in these
376
-------
LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
Table 1.—Statistical values for selected water quality parameters, Colorado River and tributaries, 1978-1981.
Water Water Water Water Water
Yr. and
stream
1978
CoIR
Trib.
1979
CoIR
Trib
1980
CoIR
Trib
1981
CoIR
Trib
Statistic
mean
median
n
mean
median
n
mean
median
n
mean
median
n
mean
median
n
mean
median
n
mean
median
n
mean
median
n
temp
(°C) pH
12.9 8.1
13 8.2
410 153
21.6 8.3
21 8.4
178 67
11.6
11
73
21.3
21
26
11.6
11
36
21.8
22
12
13.7
14
38
23.3
23
12
turbidity
NTU
16.01
9
360
4.21
4
154
11.01
10
71
7.51
4
26
7.81
8
35
7.91
8
11
589.01
500
37
32.01
16
12
TC/
100ml
402
4
10
782
53
8
4532
445
8
281 2
380
4
FC/
100ml
2.1
1
338
3.1'
1
189
2.41
1
69
7.91
3.5
26
1.41
0
36
4.41
8
11
66.01
81
36
45.01
19
10
FS/
100ml
631
47
85
63'
66
44
651
3
10
1591
200
5
3241
405
8
FC:FS
ratio
0.103
27
0.063
14
0.023
4
0.303
8
'Col R, Colorado River; Trib, tributary.
tabulated values = antilog (sum of log individual observations/n observations).
'Tabulated values = sum of individual observations/n observations.
''Tabulated values = (sum of individual FC valuasj/individual FS values, where FS
density > 100 FS/100 ml.
mountain streams have also been similar to those in the
Canyon during nonstorm flow periods (Skinner et al. 1974;
Stuart et al. 1971; Stuart et al. 1976; Varness, 1978).
Although bacterial data indicate similarities in quality
between Grand Canyon waters and protected mountain
streams, the Canyon represents hydrologic processes
that differ sharply from humid, mountain environments.
During drought, the hydrologically active portions of the
Grand Canyon watershed are strictly limited, for weeks to
months, to the channels of the Colorado River and its few
perennial tributaries (Novotony and Chesters, 1981).
Mountain streams in humid environments are usually
characterized by hydrologically active zones that exceed
the limits of the stream channel. Water frequently flows
from watershed surfaces to stream channels, but vegeta-
tion and soil litter stabilize soils and other debris, including
fecal material, during all but major runoff events (Cum-
mins et al. 1983; Novotony and Chesters, 1981). Accord-
ingly, high quality stream waters may be produced by dis-
similar processes, such as prolonged absence of
hydrologic movement during drought in the arid environ-
ment and stabilization of erodible fecal material during
hydrologic movement in the humid setting.
Storm flow periods. Arid environments such as the
Grand Canyon are highly susceptible to flash floods, be-
cause of intense rainfall over sparsely vegetated water-
sheds, with extremely steep slopes of exposed bedrock
and impermeable soils. With sufficient rainfall, the hydro-
logically active portion of a desert watershed may expand
quickly from near zero to include a major portion of the
basin. Because considerable fecal debris may have accu-
mulated on watersheds between ephemeral storm events,
highly contaminated storm flows may result. In contrast, in
humid environments vegetation, soil litter, and high soil
infiltration rates reduce hydrograph peaks and runoff vol-
ume. Accordingly, rainfall events of equal magnitude may
cause greater peak nonpoint source fecal loading of
streams in arid wildlands than in humid wildland settings.
The impact of storm events on Colorado River water
quality varies considerably depending on the portion of
the watershed generating storm flow, the volume gener-
ated, and the flow volume of the river available to dilute
the storm flow input. Storm flow in 1981 persisted for
weeks as a result of scattered but frequent thunderstorms
over the Paria and Little Colorado River basins. Collec-
tively, these storms maintained continuous discharge into
the Colorado River. By virtue of flow volume, the Little
Colorado River contributed most to the Colorado River
turbidity and fecal conforms below the confluence of these
two rivers (Fig. 3). Historically, the Little Colorado River is
the major source of storm flow in the Colorado River (U.S.
Geolog. Surv. 1980). Annual stream flow in the Little Colo-
rado is 8 times that of the Paria River and 35.4 times the
Kanab Creek's flow. The 1981 storm events produced
slightly below average volume storm flows in the Paria
and Little Colorado Rivers, 8.6 m3/s and 26 m3/s, respec-
tively. Storm flow events up to 456 m3/s for the Paria and
3,400 m3/s for Little Colorado Rivers have occurred. Pre-
sumably, the volume and concentration of fecal contami-
nants in storm flow will increase with runoff volume, espe-
cially in the arid environment.
The Colorado River exceeded full body contact stand-
ards only during storm flows. Previous studies had not
provided data showing that fecal contamination resulting
from storm flows in pristine wildlands would reach the
magnitude observed in Grand Canyon (Skinner et al.
377
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
3.07O -
2.763 -
2.456 -1
2.149 -
—1
S
0
0 1.842 -
x.
t/J
tr
LJ
CD
•g 1.535 -
2
O
"- 1.228 -
O
0.921 -
0.614 -
0.307 -
0 -
C
^ o i unmrLun i unoiui i ico ^
S - MOW-^TflRMFl OW TIlRRiniTIFI ,«... ., "S
0
0
0 °
o o
o
O 0
o
O 0
• 0
• o o
o o o
o o o
• 0 O
• o o
O o
• 0
• o o
• • o
• o o o
•
•
1 • 1
• • • • •
• • • • • • •
• • • 0
• • •*•• t* • •• • • •• •
i •• • • • •!• •
5 4 6
1 | t *fc W f • • •••• • t •• t
• • • m •• ••' •••!••• o
5 99 94 64544
r i i i f i i i i i ^
i A A • • •• A| •^B^AAA^^B ftABMft A ft ftl
9 7 8 9 9 99 99 969 9 574 5 4
1 1 1 1 1 1 \I-A II •..!•( .
4 •• ••••••• •Mtf* •••MtM *|« •{• • •
- 1175
- 575
- 288
.
-J
- 141 5
0
O
\
(/>
- 69 g
00
5
~ 35
u.
UJ
\-
- 17 <
5
O
(£
— 8 °-
^-
- 4
- 2
-oe.
.37 .74 1.12 1.49 1.87 2.24 2.61 2.99 3.36 3.74
LOG TURBIDITY
/t
13 32 7»» 114 417
APPROXIMATE TURBIDITY N.T.U.
977
2344
5495
NON-STORMFLOW OBSERVATIONS
o STORMFLOW OBSERVATIONS
Figure 3.—Fecal coliform (FC) densities and turbidities in the Colorado River, 1981. The impacts of Paria and Little Colorado
River storm flow discharges on the Colorado River are evident at Miles 1 and 61.5.
1974; Stuart et al. 1971; Stuart et al. 1976; Varness,
1978). Although storm flows exceeded contact standards,
the actual health risks cannot be reliably estimated be-
cause of deficiencies associated with the fecal coliform
standards (Cabelli, 1982). Changes in the standards have
been proposed that would directly link water contact with
rates of gastrointestinal disease, using Escherichia coli
and enterococci as indicator bacteria (Cabelli, 1983; Du-
four, 1982).
For Grand Canyon and other wildland settings, the pro-
posed standards may not be an improvement. Assuming
that wildlife and livestock are the principal sources of fecal
contamination, the occurrence of waterborne pathogens
must be related principally to disease or carrier rates
among these animals. Because the proposed standards
are currently based on epidemiological studies of eastern
beaches contaminated with sewage effluents from human
point sources, they may not be appropriate for western
wildland settings where nonpoint source animal contami-
nation predominates. An appropriate approach for these
wildland areas would be to examine relationships between
indicator bacteria and pathogen occurrence in waters,
and disease incidence rates among users of these recrea-
tional waters.
Turbidity. Because access to the Colorado River is diffi-
cult, routine water quality montoring is not easily accom-
plished. The 1981 storm flow pattern suggested that tur-
bidity may be a useful tool to quantitatively model fecal
coliform loading (Fig. 3). Although a positive correlation
(r = 0.54) between turbidity and fecal coliform densities
was found, only about 29 percent of the variation (r2 =
0.29) in fecal coliform can be explained by storm flow
turbidity. The strength of this relationship suggested that a
turbidity model could predict only general levels of fecal
coliform loading. More extensive monitoring, as sug-
gested by Thornton et al. (1980) may improve this model's
potential.
Storm flow volume is a second parameter that may as-
sist in developing a modeling tool for Grand Canyon. Per-
rier et al. (1977) initially found an r2 value of only 0.19
when trying to predict coliform loading as a function of
stream flow on the Caddo River. This relationship im-
proved to 0.42 when only the rising leg of the storm flow
hydrograph was examined. The rising leg more clearly
isolates the impact of the more contaminated first flush
flows (Davis, 1977). By monitoring Paria and Little Colo-
rado Rivers' storm flow volumes, fecal coliform loadings,
and turbidities before they discharge into the Grand Can-
378
-------
yon, the storm flow water quality of the Colorado River
may become predictable.
ACKNOWLEDGEMENTS: This research was supported in part
by the USDA Forest Service through the Eisenhower Consor-
tium for Western Environmental Forestry Research; the Agricul-
ture Experiment Station, College of Agriculture, University of
Arizona; and the U.S. National Park Service, Grand Canyon
National Park.
REFERENCES
Arizona Administration Comp. 1980. R, 9-21-209.
Bissonnette, G.K., D.G. Stuart, T.D. Goodrich, and W.G. Walter.
1970. Preliminary studies of serological types of enterobacte-
ria occurring in a closed mountain watershed. Proc. Mont.
Acad. Sci. 30: 66-76.
Cabelli, V. 1982. Microbial indicator systems for assessing water
quality. Antonnie van Leeuwenhoek. 48: 613-8.
1982. Health effects criteria for marine recreational
waters. EPA-600/1 -80-031. U.S. Environ. Prot. Agency, Wash-
ington, DC.
Cummins, K.W. et al. 1983. Organic matter budgets for stream
ecosystems: Problems in their evaluation. Pages 299-353 in
Stream Ecology: Application and Testing of General Ecologi-
cal Theory. Plenum Press, New York.
Davis, E.M., D.M. Casserly, and J.D. Moore. 1977. Bacterial
relationships in stormwaters. Water Resour. Bull. 13: 895-
905.
Dufour, A.P. 1982. Fresh recreational water quality and swim-
ming-associated illness. Second Nat. Symp. Municipal Waste-
water Disinfection. Orlando, FL.
Fair, J.F., and S.M. Morrison. 1967. Recovery of bacterial patho-
gens from high quality surface water. Water Resour. Res. 3:
799-803.
Federal Water Pollution Control Administration. 1967. Water
Quality Criteria. U.S. Gov. Print. Off., Washington, DC.
Geldreich, E.E. 1975. Handbook for Evaluating Water Bacterio-
logical Laboratories. EPA-670/9-75-006. U.S. Environ. Prot.
Agency, Washington, DC.
LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
1976. Fecal coliform and fecal streptococcus density
relationships in waste discharges and receiving waters. Pages
349-68 in CRC Critical Reviews in Environmental Control.
Geldreich, E.E., L.C. Best, B.A. Kenner, and D.J. Van Donsel.
1968. The bacteriological aspects of stormwater pollution. J.
Water Pollut. Contr. Fed. 40: 1864-72.
Novotny, V., and G. Chesters. 1981. Handbook of Nonpoint Pol-
lution: Sources and Management. Van Nostrand Reinhold,
New York.
Newman, B. 1981. Don't drink the water. Audubon 83: 95-7.
Perrier, E.R., H.E. Werterdahl, and J.F. Nix. 1977. Water quality
loadings during thirteen storms on the Caddo River, Arkan-
sas. Am. Soc. Agric. Eng. Meet. Pap. No. 77-2529.
Skinner, Q.D., J.C. Adams, PA. Rechard, and A.A. Beetle. 1974.
Effect of summer use of a mountain watershed on bacterial
water quality. J. Environ. Qua). 3: 329-35.
Standard Methods for the Examination of Water and Wastewa-
ter. 1980.15th ed. Am. Pub. Health Assn., Washington, DC.
Stuart, D.G., G.K. Bissonnette, T.D. Goodrich, and W.G. Walter.
1971. Effects of multiple use on water quality of high-mountain
watersheds: bacteriological investigations of mountain
streams. Appl. Microbiol. 22: 1048-54.
Stuart, S.A., G.A. McFeters, J.E. Schillinger, and D.G. Stuart.
1976. Aquatic indicator bacteria in the high alpine zone. Appl.
Environ. Microbiol. 31:163-7.
Thornton, K.W., J.F. Nix, and J.D. Bragg. 1980. Conforms and
water quality: use of data in project design and operation.
Water Res. Bull. 16: 86-92.
U.S. Geological Survey. 1980. Water resource data for Arizona.
U.S. Geolog. Surv. Water Data Rep. AZ-80-1.
U.S. National Park Service. 1979. Colorado River management
plan, Grand Canyon National Park. Grand Canyon Nat. Park,
AZ.
Varness, K.J., R.E. Pacha, and R.F. Lapen. 1978. Effects of
dispersed recreational activities on the microbiological quality
of forest surface water. Appl. Environ. Microbiol. 38: 95-104.
Walter, W.G., and R.P. Bottman. 1967. Microbiological and
chemical studies of an open and closed watershed. J. Envi-
ron. Health, 39:157-63.
379
-------
PETER R. WILIENBRING
WILLIAM D. WEIDENBACHER
E. A. Hickok and Associates
Wayzata, Minnesota
This paper examines the role that wetlands play in treat-
ing stormwater runoff. Physical and chemical mecha-
nisms through which wetlands remove pollutants from
nonpoint source runoff are reviewed and specific wetland
characteristics that can increase the pollutant removal
efficiencies are identified. Watershed factors that affect
the quality of the nonpoint source runoff generated from a
given watershed and the ability of a given wetland to treat
this runoff are outlined. The paper also presents a proce-
dure that can be used for planning purposes to quantify
the need for preserving a wetland in a given watershed.
The need for preserving the wetland is based on the wet-
land's ability to mitigate adverse impacts to the surface
water resources in its watershed.
Wetlands have been identified as natural areas that have
the ability to remove nutrients, solids, and other pollutants
from stormwater runoff. Previous studies (U.S. Environ.
Prat. Agency, 1977) have indicated that these nutrients,
solids, and other pollutants are removed through a combi-
nation of physical entrapment, microbial transformation,
and biological utilization. For example, phosphorus pass-
ing through a wetland can be readily immobilized in soils
by adsorption and precipitation reactions with aluminum,
iron, calcium, and clay minerals. The reactions with cal-
cium occur predominantly under alkaline soil conditions
and aluminum and iron reactions occur predominantly un-
der acid or neutral soil conditions (Nichols, 1983). Nitro-
gen can also be removed from water passing through a
wetland, primarily by sedimentation and through the proc-
ess of denitrification, which occurs under anaerobic condi-
tions and is accomplished by facultative bacteria that use
nitrate (NO3) in place of oxygen (O£ to facilitate respira-
tion.
Other nutrient and solids removal processes are facili-
tated by vegetation in the wetland. Vegetation acts to
physically filter water passing through the wetland, reduce
the water's velocity, and allow inorganic and organic par-
ticulate matter and the nutrients associated with it to settle
out. The vegetation also provides a substrate to which
decomposer microorganisms can attach themselves.
These microorganisms assimilate pollutants from
stormwater runoff as they grow and reproduce in a man-
ner similar to that of microorganisms present on the rocks
of a trickling filter at a wastewater treatment plant. Wet-
land vegetation also removes nutrients from the soils and
water in the wetland as it grows and incorporates these
nutrients into its vegetative mass (Wenck, 1981).
Unfortunately, not all of these nutrient removal mecha-
nisms are permanent. Phosphorus that is adsorbed to soil
particles will be desorbed under certain conditions. Nutri-
ents that were incorporated into the wetland's vegetative
mass may re-enter the water column when the vegetation
dies. Previously sedimented particulate matter can be re-
suspended and flushed out of the wetland during periods
of high flow.
In spite of these factors, removal efficiencies in some
wetland treatment systems have exceeded 60 percent tor
some nutrients and 95 percent for suspended solids (Wil-
lenbring, 1984).
Nutrient removal efficiencies for a given wetland may be
the result of a number of different factors linked to the
various removal processes. These factors are generally
either wetland-specific factors or watershed-specific. Wet-
land-specific factors include:
1. Type of vegetation in the wetland.
2. Type of soils in the wetland.
3. Temperature of water and soils in the wetland.
4. Characteristics of water flow through the wetland.
5. Water retention characteristics of the wetland and its
associated outlet structure.
6. Normal depth of water in the wetland.
7. Area of wetland.
Watershed-specific factors include:
1. Drainage area tributary to the wetland.
2. Amount of stormwater storage present in tributary
drainage area.
3. Concentration of nutrients and solids in stormwater
runoff entering wetland.
4. Hydraulic loading.
5. Soils, vegetation, and land use in tributary water-
shed.
Improving the nutrient and solids removal efficiency of a
given wetland generally involves modifying one or more of
these factors to allow the natural removal processes to
occur more completely, at a greater pace, over a longer
time or over a larger area. The nutrient removal efficiency
of an existing wetland can be estimated using these fac-
tors and the results of past wetland monitoring.
The Rice Creek (Minnesota) Watershed District, which
encompasses many of the suburbs north of the Minneapo-
lis-St. Paul metropolitan area, has recognized the bene-
fits of using and preserving wetlands. The District has
adopted a wetland preservation guideline to preserve wet-
lands for the purpose of treating stormwater runoff.
The guideline uses some of the watershed-specific fac-
tors previously listed to estimate the phosphorus loading
from the watershed tributary to the wetland and some
wetland-specific factors to estimate the phosphorus as-
similative capacity of the wetland. The nutrient loadings
and wetland assimilative capacities are then compared to
determine if the assimilative capacity exceeds the nutrient
loading. If this is the case, a portion of the wetland may be
filled.
380
-------
LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
Table 1 .—Nutrient loading according to land use.
Nutrient load
Land Use (pounds/acre/year)
Open Space
Residential:
lots larger than 1 acre
lots 1 acre or less
multiple dwellings
Commercial/industrial
0.30
0.40
1.44
3.46
1.33
WETLAND PRESERVATION GUIDELINE
The formal procedure for calculating the area of a given
wetland that could be filled is as follows:
Step 1: Calculate the area of each proposed land use in
the basin that drains to the wetland.
Step 2: Calculate the nutrient load (Ib/year) generated
by the watershed that drains to the wetland according to
the ultimate land use.
Step 3: Calculate the nutrient assimilative capacity of
the wetland.
Step 4: Balance the lost assimilative capacity per acre
of fill plus the additional loading due to the use of that area
with the surplus nutrient assimilative capacity of the wet-
land. This can be written in the following equation:
Finable
Acres
Nutrient loads and wetland assimilative capacities can
be estimated from available information or whatever data
collection effort seems reasonable. In the absence of non-
point source runoff data, the Rice Creek Watershed Dis-
trict used the nutrient loadings and wetland assimilative
capacities shown in Tables 1 and 2. The values shown in
these tables were derived from a previous study in the
area (Environ. Prot. Agency, 1977).
SUMMARY AND CONCLUSIONS
Wetlands can remove nutrients, solids and other pollu-
tants from stormwater runoff. Wetlands can also be pre-
Total Assimilative Capacity
of Existing Wetland
Assimilative Capacity of
Wetland (per acre)
Total Loading from
Ultimate Runoff
Ultimate Loading Due
to Use (per acre)
Table 2.—Nutrient assimilative capacity of wetlands.
Assimilative
capacity
Wetland Type (pounds/acre/year)
Cattail marsh (continuously wet)
Grassy marsh (alternately wet-dry)
4.0
12.0
served on this basis. Such a wetland preservation regula-
tory program is being administered by the Rice Creek
Watershed District.
The wetland preservation formula presented in this pa-
per can be used in areas for which no water quality moni-
toring information is available, or for areas in which exten-
sive nonpoint source monitoring programs have been
completed. By adding more detailed data on expected
nutrient loadings and wetland assimilative capacities, wet-
land preservation requirements can be made more site
specific. The formula presented here will accommodate
all these levels of effort.
In today's rapidly developing world where preservation
of wetlands is important for many reasons, but where reg-
ulatory denial of a proposed wetland alteration could con-
stitute an unjust taking of property, site-specific ap-
proaches to wetland preservation are needed. The
approach presented in this paper allows the value of the
wetland to be established on its ability to treat nonpoint
source runoff.
REFERENCES
Nichols, D.S. 1983. Capacity of natural wetlands to remove nu-
trients from wastewater. J. Water Pollut. Control Fed. 55(5).
U.S. Environmental Protection Agency. 1977. Urban Runoff
Treatment Methods. Vol. 1. Non-structural wetland treatment.
EPA 600/2-77-217. U.S. Environ. Prot. Agency, Washington,
DC.
Wenck, N.C. 1981. Wetlands and organic soils for the control of
urban stormwater. In Proc. Midwest Conference on Wetlands
Values and Management. June 17-19. St. Paul, MN.
Willenbring, PR. 1985. Wetland treatment systems—why do
some work better than others? In Lake and Reservoir Man-
agement: Practical Applications. Proc. Int. Symp. Oct. 16-19,
1984. N. Am. Lake Manage. Soc., Washington, DC.
381
-------
QUENTIN D. SKINNER
JERROLD L. DODD
J. DANIEL RODGERS
MICHAEL A. SMITH
Range Management Department
University of Wyoming
Laramie, Wyoming
Overgrazing by domestic livestock and periodic flooding
are often cited as sources for increasing nonpoint source
pollution in streams within semi-arid rangelands. Riparian
zones along streams may help decrease nonpoint pollu-
tion if maintained in a healthy ecological condition. This
paper will address two research programs designed to
reverse desertification of streamside zones along cold
desert streams in Wyoming by: (1) manipulating livestock
grazing, (2) promoting regrowth of desirable vegetation,
(3) willow planting, (4) using instream flow structures to
store water in channel banks and trap sediment, and (5)
encouraging beaver damming. Research theory as well
as monitoring protocol will be discussed and related to
ease of use by management agencies and producer
groups affiliated with western rangelands.
Riparian zones are areas along streams supported by
high water tables because they are near surface or sub-
surface water. They have distinct soils and more highly
productive and diverse plant and animal communities
than adjacent xeric (dry) areas. Riparian zones normally
are ectones between xeric and aquatic ecosystems
(Brown et al. 1978).
Riparian zones have many users (Busby, 1978; Johnson,
1978; Tubbs, 1980; Kauffman and Krueger, 1984). Appar-
ently, multiple uses concentrate in riparian zones because
of the vegetation species diversity, productivity, and prox-
imity to open water. High plant species diversity in the
riparian zones is reported by Campbell and Green (1968),
Brown et al. (1978), Ewel (1978), and Kauffman et al.
(I983a). The vegetation stabilizes stream channels by cre-
ating a rough surface that reduces stream flow velocity;
roots hold bank material together (Li and Shen, 1973;
Heede, 1977; Platts, 1978; Andrews, 1982).
Because stream flow velocity is reduced and vegetation
traps sediment on banks, water quality improves
(Schumm, 1963; Andrews, 1982). Lowrance et al. (1985)
show how interflow between bank waters and streams in
riparian zones further improves water quality. Better water
quality promotes diverse aquatic habitat and thus im-
proves fisheries (Cummings, 1974; Duff, 1979; Platts,
1981). The value of riparian habitat to wildlife is also well
documented (Crothers et al. "1974; Johnson et al. 1977;
Thomas etal. 1979).
Increased edge effect for unit area occupied (Odum,
1978) and greater vegetation structural diversity are often
characteristic of riparian zones as compared to surround-
ing plant communities (Anderson et al. 1983). Both edge
effect and structural diversity are important for habitat to
maintain diverse wildlife species (Ohmart and Anderson,
1978). In addition, high vegetation production, free flowing
water, flat terrain, and shade are cited as reasons why
livestock use riparian habitat (Kauffman and Krueger,
1984).
Users of riparian zones may cause soil compaction,
slough off undercut stream banks, and remove vegetation
along channels. These actions can increase erosion, often
widening the stream channel, downcutting, or both (Peter-
son, 1950; Schmidly and Ditton, 1978; Meehan and Platts,
1978; Thomas et al. 1979). This erosive action may cause
loss of (1) floodplain water tables, (2) floodplain soil mois-
ture, (3) aquatic habitat quality, (4) fisheries, (5) plant vigor,
and (6) plant species diversity (Jahn, 1978; Campbell,
1970; McCall and Knox, 1978; Platts, 1981). Examples of
stream degradation and channelization with effects such
as described above are noted by Busby (1978), Meehan
and Platts (1978), Roath and Krueger (1982), and Kauf-
fman etal. (1983b).
Recovery of stream channels, aquatic habitat, fisheries,
and riparian vegetation after livestock have been removed
has been demonstrated by Keller et al. (1979), Duff (1979),
Bowers et al. (1979), Platts (1981), and Kauffman et al.
(1983a). These researchers have used exclosures to elimi-
nate grazing along stream reaches within different grazing
management strategies to try to document which strategy
best conserves riparian and aquatic habitat. Little re-
search has been done on reclamation of streams and ri-
parian zones to promote water storage and control non-
point source pollution. In contrast, studies have evaluated
removing riparian vegetation to increase waterflow down-
stream.
Water loss through evapotranspiration by streambank
vegetation cannot be denied. In areas like the southwest
United States, researchers have measured water yield fol-
lowing riparian vegetation removal from floodplains. While
large water savings were predicted by Gatewood et al.
(1950) and Robinson (1965), Culler (1970) found small
savings and Morton and Campbell (1974) found even less
when salt cedar was removed. Actual interest in removing
riparian plants to increase water yield has been minimal
during the last decade (Graf, 1980). This is perhaps be-
cause water yields decline again shortly as plants rein-
vade riparian zones (Horton and Campbell, 1974), cost to
benefit ratios are high (Graf, 1980), other user demands
for riparian zones exist (Campbell, 1970), and recovered
water may be lost to deep aquifers downstream (Daven-
port et al. 1982).
Certainly conflicts exist between users on how to best
manage riparian zones. Graf (1980) points out that saving
water by reducing transpiration is not currently as popular
382
-------
as habitat management for other uses. However, man-
agers must determine how to reduce floods when riparian
vegetation reduces stream flow to and through down-
stream areas. Conversely, if mature riparian zones cause
beneficial flooding, why not use this phenomenon to repair
degraded stream channels, store ground water, and con-
trol nonpoint source pollution?
HYPOTHESIZED MECHANISMS FOR
RECLAIMING STREAMS AND RIPARIAN
ZONES
Invading riparian plants stabilize stream bars, islands, and
floodplains. Often, bars become islands and channels
around islands close to form floodplains bordering one
channel. This occurs when flushing flows are not able to
remove established vegetation and when overbank flood-
ing deposits sediment. This normally occurs in low rather
than steep gradient stream reaches. When the flow re-
gime is in equilibrium with channel size and bank resistiv-
ity, mature riparian zones may indicate the geomorpholog-
ical character of the stream system (Leopold and
Langbein, 1966; Graf, 1978; Heede, 1981). The reclama-
tion process may begin in a wide, degraded stream chan-
nel when surface flow decreases. Low flow meanders
across a low gradient channel bottom, increasing stream
sinuosity and length.
Permanent aggradation occurs when sediment is de-
posited and vegetation stabilizes it. Andrews (1982) shows
aggradation occurring bank-first during overbank flooding
and through accumulation of bedload during lower flows.
Accumulated bedload may persist until the channel nar-
rows to meet the annual flow regime. Narrowing of the
channel increases flow velocity and accumulated bedload
is then transported downstream, deepening the channel.
Andrews (1982) also indicates that although a mature
stream maintains an average width and depth in equilib-
rium with the flow regime, it will move laterally from year to
year thus fitting Leopold and Langbein's (1966) descrip-
tion of meandering streams. Undercut banks along stable
streams are evidence of lateral movement of meanders
and do not necessarily show stream channel instability.
Reclaiming degraded streams to support mature ripar-
ian zones depends on sediment deposition and its stabili-
zation by vegetation. Water may be lost during a high flow
when water moves into downstream alluvium as shown by
Lane et al. (1970). Glymph and Holton (1969) show that
maximum transmission loss from any one runoff event in
semi-arid regions should occur near the mouth of a drain-
age basin or more extensively in large basins. Loss in flow
downstream should cause aggradation of sediment. Prac-
tices to reclaim degraded streams based on loss of flow
and aggradation of sediment should be placed at loca-
tions of maximum water travel time such as the mouth of
the drainage basin.
Instream structures, like check dams or trash collectors,
and biological damming by beaver or encroaching banks
and riparian zones cause (1) reduced flow velocity, (2)
stable bedload, and (3) storage of water in the stream-
banks near the dam. Heede (1978, 1982) discusses using
check dams to reclaim gullies by raising the local base
level in ephemeral stream reaches to decrease gradient
slope upstream. The lower upstream gradient reduces
sediment transport. Deposition occurs upstream in a
wedge shape.
According to Heede's 1978 and 1982 research, dams
should be placed downstream just above a tributary junc-
tion. To restore riparian habitat the dam also should be
located on a stream reach having a low gradient, where
meandering occurs and a stable floodplain exists. The
dam should then cause bank deposition and maximum
LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
filling in the upstream drainage network. Established ripar-
ian vegetation and narrowing of the channel may eventu-
ally cause water to spread over banks instead of eroding
them during floods (Graf, 1980).
STUDY APPROACH: RECLAMATION OF
COLD DESERT STEPPE STREAMS
Funding and Administration
Antidesertification of cold desert steppe streams through
reclaiming degraded channels and promoting riparian
habitat is a joint research effort of the Range Manage-
ment, Civil Engineering, and Zoology departments, and
the Wyoming Water Resources Research Center, Univer-
sity of Wyoming; the Bureau of Land Management; the
Department of Environmental Quality, State of Wyoming;
industry, and ranchers. Funding comes from the U.S. De-
partment of Agriculture (Science and Education Adminis-
tration) and the U.S. Environmental Protection Agency.
Funding has been requested from the U.S. Geological
Survey to document hydrologic response associated with
reclaiming degraded streams and associated riparian
zones. The Water Resources Research Center provides
the administrative umbrella for various research efforts
being conducted by the University. Individual departments
account for their own budgets, data sets, and report obli-
gations. Land and water management agency personnel,
industry, and ranchers work with individual University de-
partments when: (a) research is being planned, (b) re-
search sites are selected, (c) livestock are used in experi-
mental designs, (d) facilities are developed, (e) funds are
requested, and (f) data are gathered.
Integration of the multidisciplinary team just described
was critical in developing best management practices for
abatement of nonpoint source pollution to streams from
semi-arid western rangelands. Wyoming's research effort
did not start with research requests from the University to
State or Federal land and water management agencies.
On the contrary, field personnel of management agencies
and local ranchers urged the University to become in-
volved. Only after much deliberation with the involved in-
terest groups did University researchers make a commit-
ment. By following this philosophy, a strong bond has
developed between the multiple users and managers of
rangeland and water. Accountability is distributed to those
who need questions answered. The University is helping
to answer them. This process has resulted in State sup-
port for field level agency personnel and construction of
research facilities by the Bureau of Land Management,
the Water Resources Research Center, and ranchers.
Facilities
Because personnel, funds, and time are limited, research
facilities and experimental designs must apply to any
drainage basin or stream reach. Wyoming has selected
research facilities in a range of geographic areas and veg-
etation zonations. Examples of these facilities are: (1) The
Snowy Range Hydrologic Observatory, an instrumented
watershed including alpine and montane vegetation; (2) a
valley without a developed channel slated to divert water
for municipal supply; it will become a perennial stream
located above reservoir storage in montane to foothill veg-
etation zones; (3) an instrumented flood-irrigated meadow
complex with automated monitoring of return flow to tribu-
taries of the Green-Colorado River system in foothill
rangeland; (4) an ephemeral stream exclosure to docu-
ment using sediment, livestock grazing, and vegetation to
reclaim degraded streams and riparian zones located in a
cold desert steppe basin; (5) a perennial stream exclosure
383
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
and pasture system to document using sediment, live-
stock grazing, vegetation, instream flow structures, and
beaver to reclaim degraded streams and riparian zones in
a cold desert steppe basin; and (6) the wildland watershed
laboratory located on the University of Wyoming campus
to interface with field site data for modeling riparian eco-
systems.
Wyoming's research efforts are concentrated within se-
lect and well developed facilities, located in representative
. areas of vast Western rangelands. They are located on
streams characteristic of those flowing through semi-arid
region drainage basins and concentrated on streams
whose flow regime is typical of water conveyance from
mountain to basin.
Fifteen Mile Creek
The research facility for Fifteen Mile Creek is located
above the confluence of a downstream tributary. The max-
imum transmission loss of stream flow thus reduces veloc-
ity from any one runoff event. The stream reach is mean-
dering, has a low channel gradient, and loses water to
supporting alluvium. These characteristics, plus manipu-
lation of livestock grazing to leave vegetation on channel
bank slopes, minimize stream flow velocity. Reduced ve-
locity should cause bank-first deposition of sediment. Sta-
bilization of this sediment by vegetation and encroach-
ment of plants into the channel should further reduce
stream flow velocity and thus promote channel filling.
As the channel fills, peak flow events can be forced over
lioodplains left dry because of past downcutting. Flood-
plain vegetation again traps sediment and slows flow ve-
locity. Terrain depressions provide surface storage. De-
pression storage, after evapotranspiration, should
percolate into the alluvium, increasing soil and ground
water. Spreading of water over floodplains can be in-
creased by rounding of banks through livestock trampling.
Rounding promotes vegetation establishment on formerly
straight wall stream banks and decreases channelization
by floods. Hoof print depressions should cause surface
storage of water, increase infiltration, promote bank plant
production, lower bank profile, and help curb bank rill ero-
sion. As the process of channel filling progresses through
encroachment of stream banks, roots should cross the
narrow interim channel. If grasses and other vegetation
along the stream edges are rhizomatous, plants should
establish across channel. This vegetation should filter
sediment during flows and roots should stabilize bed load.
The channel bottom should rise, and as a result Fifteen
Mile Creek could have a natural dam 2 miles long. This
dam should cause channel filling upstream and the proc-
ess should repeat itself upstream.
Reclaiming this ephemeral stream channel depends on
management of vegetation along banks. Livestock and
wildlife are grazing this channel. The challenges for this
research effort are to determine how grazing affects
ephemeral stream channel vegetation and stream chan-
nel stability, and to develop ways to graze degraded and
reclaimed ephemeral streams while maintaining vegeta-
tion and channel stability and promote channel filling. An-
swers to these questions should help control nonpoint
source pollution from wildlands of semi-arid regions of the
world in a cost-effective manner over extended periods of
time. University Extension is a member of this research
team to insure that the techniques are transferred to land
and water management agencies and others.
Muddy Creek
Muddy Creek was selected as a study site because it is
located next to a highway with easy access and different
stream reaches within the 65 km study area depict differ-
ent degrees of channel stability and degradation. The
study area meets all geomorphological requirements used
for selecting the Fifteen Mile Creek site, although Muddy
Creek is perennial.
The Muddy Creek study area is divided into six hydrau-
lic response units, each with a different degree of channel
degradation. Instream flow structures (trash collectors),
willow and other vegetation planting, and beaver dams are
being used to create riparian zones.
The 3-mile downstream stream reach has been ponded
during low flow with 32 (45 cm high) trash collectors
placed on straight channel sections along the reach. The
trash collectors have withstood summer high flow events
and winter ice. Many are full of sediment and the channel
bottom between catchers is aggrading. Beaver are using
three of them as a base for constructing dams out of wil-
low and sagebrush. New sets of 32 (45 cm high) trash
collectors will be constructed each summer over filled col-
lectors until high flows flood adjoining areas left dry be-
cause of past channel downcutting. The resultant 5 km
constrictive dam will back sediment upstream, creating
riparian zones to be further enhanced with more trash
collectors. The 5 km reach immediately upstream is de-
graded because sediment is being filtered by a 16 km
reach of good riparian habitat just above it. Above this,
5 km of flood flats will be planted and grazed to cause
aggradation for an additional 10 km upstream.
RESEARCH QUESTIONS
Questions to be answered from this research effort are:
1. How does water storage differ between degraded,
natural, and improved riparian zones of high desert
steppe streams?
2. Do different stream reaches along improved cold
desert steppe streams have different water storage capa-
bilities?
3. Do improved riparian zones change flow regimes? If
so, is there a prolonged release of water for downstream
users during periods of shortage?
4. What are the hydrologic responses associated with
riparian zone improvement practices of cold desert steppe
streams such as damming by beaver and instream flow
structures, willow and grass establishment, brush control
(burning, spraying), and fertilization?
5. Can riparian zone improvement practices initiated
on cold desert steppe streams reduce nonpoint source
pollution downstream?
6. How does improving riparian zones on cold desert
steppe streams help control nonpoint source pollution?
7. What hydrologic responses are associated with graz-
ing of improved riparian zones of cold desert steppe
streams?
8. What are the economic costs and benefits of improv-
ing degraded riparian zones of cold desert steppe
streams?
Vegetation response, stream flow, soil moisture, ground
water recharge, stream channel morphology, root biomass
of stream banks, particle size distribution of channel
banks, animal grazing behavior, effect of season-of-use by
livestock, effect of stocking rate of livestock on riparian
vegetation, trash collector design, and techniques to rein-
force beaver dams are examples of data being collected
by the University of Wyoming. Change in stream channel
morphology is being monitored using cross-section tech-
niques. Permanent cross sections have been placed on
meander and straight stream reaches in all study units.
Vegetation response to change in stream channel mor-
phology is being determined by monitoring production,
species composition, and density at each cross section.
Encroachment of vegetation and banks across the interim
384
-------
LAND USE ISSUES: MANAGEMENT AND ASSESSMENT
channel is measured by decrease in width of the open
channel. Root biomass and particle size distribution of
banks are determined by coring techniques. Marked
plants along transects will help determine grazing prefer-
ences.
Soil moisture, water table change by season or flow
event, stream flow, and precipitation are being monitored
using neutron scattering techniques, well logging, gaug-
ing stations or peak flow techniques, and permanent
gauges, respectively. Soil moisture tubes and wells to
monitor changes in the water table are located at perma-
nent cross sections from the interim bank edge across
flood plains and into the uplands. Suspended sediment
will be collected at each stream gauging station. The Bu-
reau of Land Management and the Department of Envi-
ronmental Quality are collecting sediment load data in
stream flow.
Soon, the University of Wyoming will begin to evaluate
denitrification and sulfate reduction potential in riparian
zones and study planting of vegetation along degraded
channels.
REFERENCES
Anderson, B.W., R.D. Ohmart and J. Rice. 1983. Avian and veg-
etative community structure and their seasonal relationships
in the lower Colorado River Valley. Condor 85: 392-405.
Andrews, E.D. 1982. Bank stability and channel width adjust-
ments, East Fork River, Wyoming. Water Resour. Res. 18(4):
1184-92.
Bowers, W., B. Hasford, A. Oakley, and C. Bond. 1979. Wildlife
habitats in managed rangelands—the Great Basin of south-
eastern Oregon: native trout. Forest Serv. Gen. Tech Rep.
PNW-84. U.S. Dep. Agric., Washington, DC.
Brown, S., M.M. Brinson, and A.E. Lugo. 1978. Structure and
functions of riparian wetlands. Pages 17-31 in Proc. "Symp.
Strategies for Protection and Management of Floodplain Wet-
lands and Other Riparian Ecosystems. Forest Serv. GTR-WO-
12. U.S. Dep. Agric., Washington, DC.
Busby, F.E. 1978. Riparian and stream ecosystems, livestock
grazing, and multiple-use management. Pages 21-30 in Proc.
of Forum—Grazing and Riparian Streams Ecosystems. Den-
ver, CO.
Campbell, C.J., and W. Green. 1968. Perpetual succession of
stream-channel vegetation in semiarid region. J. Ariz. Acad.
Sci. 86-98.
Campbell, C.J. 1970. Ecological implications of riparian vegeta-
tion management. J. Soil Water Conserv. 25: 49-52.
Crothers, S.W., R.R. Johnson, and S.W. Aitchison. 1974. Popu-
lation structure and social organization of southwestern ripar-
ian birds. Am. Zool. 14: 97-100.
Culler, R.C. 1970. Water conservation by removal of phreatophy-
tes. Am. Geophys. Union Trans. 51: 684-9.
Cummings, K.W. 1974. Structure and function of stream ecosys-
tems. Bioscience 24: 631-41.
Davenport, D.C., P.E. Martin, and R.M. Hagen. 1982. Evapo-
transpiration from riparian vegetation: Water relations and ir-
recoverable losses for salt cedar. J. Soil Water Conserv. 37(4):
233-6.
Duff, D. 1978. Riparian habitat recovery on Big Creek, Rich
County, Utah—A summary of eight years of study. Pages 91-2
in Proc. Forum—Grazing and Riparian Stream Ecosystems.
Denver, CO.
Ewel, K.C. 1978. Riparian ecosystems: conservation of their
unique characteristics. Pages 56-62 in Proc. Symp.—Strate-
gies for Protection and Management of Floodplain Wetlands
and Other Riparian Ecosystems. GTR-WO-12. Forest Serv.,
U.S. Dep. Agric., Washington, DC.
Gatewood, J.S., et al. 1950. Use of water by bottom-land vegeta-
tion in Lower Safford Valley, Arizona. Water Supply Paper
1103, U.S. Geol. Survey, Washington, DC.
Glymph, L.M. and H.N. Holton. 1969. Land treatment in agricul-
ture watershed hydrology research. Pages 44-68 in: Effects of
watershed changes on streamflow, Water Resources Symp.
No. 2, Univ. Texas Press. Austin.
Graf, W.L. 1978. Fluvial adjustments to the spread of tamarisk in
the Colorado Plateau Region. Geol. Soc. of Amer. Bull 89:
1491-1501.
1980. Riparian management—a flood control per-
spective. J. Soil Water Conserv., 35:158-61.
Heede, B.H. 1977. Case study of a watershed rehabilitation proj-
ect: Alkali Creek, Colorado. Forest Serv. Res. Pap. RM-189.
U.S. Dep. of Agric.
. 1978. Designing gully control systems for eroding
watersheds. Environ. Manage. 2(6): 509-22.
_. 1981. Dynamics of selected mountain streams in
Western United States of America. Z. Geomorph. N.F. 25(1):
17-32.
1982. Gully control: Determining treatment priorities
for gullies in a network. Environ. Manage. 6(5): 441-51.
Morton, J.S. and C.J. Campbell. 1974. Management of
phreatophyte and riparian vegetation for maximum multiple
use values. U.S. Forest Serv. Res. Paper RM-117. U.S. Dep.
of Agric., Washington, DC.
Jahn, L. R. 1978. Values of riparian habitat to natural ecosys-
tems. Pages 157-60 in Proc. of Symp. Strategies for Protec-
tion and Management of Floodplain .Wetlands and Other Ri-
parian Ecosystems. GTR-WO-12. U.S. Dep. of Agric.,
Washington, DC.
Johnson, R.R., L.T. Haight, and.J.M. Simpson. 1977. Endan-
gered species versus endangered habitats: A concept, in Im-
portance, Preservation and Management of Riparian Habitat.
Forest Serv. Gen. Tech. Rep. RM-43: 68-79. U.S. Dep. of
Agric., Washington, DC.
Johnson, R.R. 1978. The lower Colorado River: A western sys-
tem. In Proc. of Symp. Strategies for Protection and Manage-
ment of Floodplain Wetlands and Other Riparian Ecosystems.
Forest Serv. GTR-WO-12: 41-55. U.S Dep. of Agric., Wash-
ington, DC.
Kauffman, J.B., W.C. Krueger and M. Vavra. 1983a. Effects of
cattle grazing on riparian plant communities. J. Range Man-
age. 36: 685-91.
1983b. Impacts of cattle grazing streambanks in
northeastern Oregon. J. Range Manage. 36: 683-5.
Kauffman, J.B., and W.C. Krueger. 1984. Livestock impacts on
riparian ecosystems and streamside management implica-
tions: A review. J. Range Manage. 37:430-8.
Keller, C., L. Anderson, and P. Toppel. 1979. Fish habitat
changes in Summit Creek, Idaho, after fencing. Pages 46-52
in Proc., Forum—Grazing and Riparian/Stream Ecosystems.
Trout Unlimited, Inc.
Lane, L.J., M.H. Diskin, and K.G. Renard. 1970. Input-output
relationships for an ephemeral stream channel system. J. Hy-
drology 13: 22-40.
Leopold, L.B. and W.B. Langbein. 1966. River meanders. The
Physics of Everyday Phenomena. Sci. Am. 79-9287: 28-38.
Li, Ryh-Ming, and H.W. Shen. 1973. Effect of tall vegetation and
flow sediment. J. Hydraulics D., ASCE, Vol. 9, No. HY5, Proc.
Paper 9748.
Lowrance, R., R. Leonard, and J. Sheridan. 1985. Managing
riparian ecosystems to control nonpoint pollution. J. Soil Wa-
ter Conserv. 40(1): 87-91.
McCall, J.D., and R.F. Knox. 1978. Riparian habitat in channeli-
zation projects. Pages 125-8 In Proc. of Symp. Strategies for
Protection and Management of Floodplain Wetlands and
Other Riparian Ecosystems. Forest Serv. GTR-WO-12. U.S
Dep. of Agric., Washington, DC.
Meehan, W.R., and W.S. Platts. 1978. Livestock grazing and the
aquatic environment. J. Soil Water Conserv. 33(5): 274-8.
Odum, E.R 1978. Ecological importance of the riparian zone In
Proc. of Symp. Strategies for Protection and Management of
Floodplain Wetlands and Other Riparian Ecosystems. Forest
Serv. GTR-WO-12: 2-4. U.S Dep. of Agric., Washington, DC.
Ohmart, R.D. and B.W. Anderson. 1978. Wildlife use values of
wetlands in the arid southwestern United States. Pages 278-
95 in Proc. National Symp. on Wetlands, Am. Water Resour.
Assn.
Peterson, H.V. 1950. The problem of gullying in western valleys.
Pages 407-34 in Applied Sedimentation. John Wiley & Sons,
Inc., New York.
Platts, W.S. 1978. Livestock grazing and riparian/stream ecosys-
tems in Proc., Forum—Grazing and Riparian/Streams Eco-
systems. Trout Unlimited, Inc. Dennis, CO.
385
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
1981. Effects of sheep grazing on a riparian-stream
environment. Forest Serv., Res Note Int. 0-307. U.S. Dep. of
Agric., Washington, DC.
Roath, L.R. and W.C. Krueger. 1982. Cattle grazing influence on
a mountain riparian zone. J. Range Manage. 35(1): 100-3.
Robinson, T.W. 1965. Introduction, spread and areal extent of
salt-cedar (Tamarix) in the western states. Prof. Paper 491-A.
U.S. Geol. Surv., Washington, DC.
Schmidly, D.J. and R.B. Ditton. 1978. Relating human activities
and biological resources in riparian habitats of western Texas.
Pages 107-16 In Proc. of Symp. Strategies for Protection and
Management of Floodplain Wetlands and Other Riparian Eco-
systems. Forest Serv. GTR-WO-12. U.S Dep. of Agric., Wash-
ington, DC.
Schumm, S.A. 1963. Sinuosity of alluvial rivers on the great
plains. Geological Soc. of Amer. Bull. 74:1089-1100.
Thomas, J.W., C. Maser, and J.E. Rodiek. 1979. Wildlife habitats
in managed rangelands—The Great Basin of southeastern
Oregon. Riparian Zones. Forest Serv. Gen Tech. Rep. PNW-
80. U.S. Dep. of Agric., Washington, DC.
Tubbs, A.A. 1980. Riparian bird communities in the Great Plains.
Pages 419-38 in Proc. Workshop—Management of Western
Forests and Grasslands for Non-game birds. Forest Serv.
Tech. Rep. lnt-86. U.S. Dep. of Agric., Washington, DC.
386
-------
Case Studies
HIGHWAY RUNOFF/DRAINAGE IMPACTS
BYRON N. LORD
Federal Highway Administration
McLean, Virginia
ABSTRACT
The Federal Highway Administration has undertaken a
four-phased research program on nonpoint source pollu-
tion from highway operations. The objectives of this pro-
gram are to characterize highway stormwater runoff;
identify sources, mechanisms of transport, and fate of
pollutants within the right-of-way; determine the magni-
tude and extent of impacts to receiving waters; and de-
velop guidelines to implement cost-effective measures to
protect water resources. A multistudy program has devel-
oped a significant national highway stormwater runoff
data base. The objective of these studies was to identify
and quantify pollutants found in runoff from operating
highways. Research is underway to develop a design pro1
cedure to predict pollutant loading from highways. This
study also identified factors that influence the transport
and fate within the right-of-way. Results of this research
were used to identify cost-effective measures to mini-
mize impacts. An extensive receiving water study has
investigated the effects of highway runoff on receiving
water environment. Guidelines were developed for as-
sessing water resources impacts from highways. An eval-
uation and synthesis of nonpoint source runoff mitigation
has identified cost-effective measures for highway
stormwater and guidelines to assist highway agencies in
identifying potential problems and implementing mitiga-
tion. An important product of this study was identification
of ineffective measures. Research is underway to de-
velop performance criteria and design specifications for
retention, detention, and overland flow systems. A high-
way runoff training program is being prepared to transfer
the technology developed from this program.
Interactions of the highway system with the Nation's water
resources are continuous and far reaching. Every mile of
highway daily affects adjacent watersheds. Each stage of
the highway process (planning, location, design, construc-
tion, operation, and maintenance) may have an impact on
water resources. Protecting the quality of water, wetlands,
and related aquatic resources is an important facet of
improving the total environment; thus, it is essential that
pollution from highway sources be identified and reduced
as much as possible. To accomplish this objective, the
environmental effects must be identified and measured.
Construction, operation, and maintenance of the high-
way system can contribute a wide variety of pollutants to
surrounding surface and subsurface waters through natu-
ral runoff. The sources of runoff pollutants and the immi-
gration pathways and transport mechanisms from the
roadway surface to the receiving water must be deter-
mined to enable efficient mitigation. This will minimize
unnecessary or ineffective treatment procedures. Where
necessary, methods of treating or minimizing the pollu-
tants in runoff must be devised.
A cooperative Federal and State research and develop-
ment program was begun to identify and quantify the ef-
fects of highway runoff and develop measures for protect-
ing the environment from any adverse effects. The FHWA,
charged with the responsibility for protecting the environ-
ment from pollution from highway sources, has ap-
proached the problem in a four-phase research program.
The objective of each phase is as follows:
1. Identify and quantify the constituents of highway run-
off.
2. Identify the sources of these pollutants and migra-
tion paths from the highway to the receiving water.
3. Analyze the effects of these pollutants in receiving
waters.
4. Develop the necessary abatement/treatment meth-
odology for objectionable constituents.
Phase 1 included not only identification and quantifica-
tion of highway runoff constituents, but also development
of a predictive procedure to be used analytically for prepa-
ration of Environmental Impact Statements (EIS). An ex-
tensive literature review was conducted at the beginning
of the Phase 1 study, and a current state-of-the-art report
387
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
was prepared at the and of the study in March 1978. A
Procedural Manual, written for State highway personnel,
details procedures for establishing and conducting a high-
way runoff monitoring program.
Results of the Phase 1 study have been documented in
a six-volume publication titled, Constituents of Highway
Runoff. The predictive procedure developed from this
study's data will provide highway designers and other in-
terested individuals/agencies with a simplified tool to pre-
dict the quantity and quality of rainfall-generated highway
runoff. The procedure is made up of four components
corresponding to the following functions:
1. Rainfall—runoff,
2. Pollutant buildup,
3. Pollutant washoff,
4. Constituent loadings.
Rainfall-Runoff! The predictive procedure calculates the
volume of runoff for a given rainfall volume using equa-
tions developed from monitoring data and regression
analyses. Since the rainfall-runoff relationship depends
on site characteristics, an equation was developed for
each of three basic site types defined from general site
characteristics of the six monitoring locations. The three
site types are:
Type 1: all pavement, bridges or overpasses
Type 2: partially paved with curbs and inlets along the
paved area (30 to 40 percent paved)
Type 3: rural sites with flush shoulders, grassy ditch
conveyance to inlets (20 to 30 percent paved).
The resultant equations relating runoff volume to rainfall
volume for the three site types are as follows:
•
Type I
Q = 0.969 R-0.019
Type II
Q = O.470 R1 •389 DO-0088
Type III
Q = 0.845 R1-892 DD-0654
where; Q = runoff volume (inches)
R = rainfall volume (inches)
: DD = dry days to last storm event
A linear equation produced the most significant correla-
tion when regression analysis techniques were applied to
monitoring data for the Type I site. For the Type II and Type
III sites, the most significant equations were obtained by
log-normalized values for total rain and dry days.
The average runoff rate in inches per hour has the high-
est correlation with the actual pounds of pollutant dis-
charged as determined through regression analyses. For
the predictive procedure to calculate duration runoff, rain-
fall duration is used in a series of equations to produce the
runoff duration for each event. Prestorm history and site
characteristics are again used to predict runoff duration.
Equations relating rainfall, runoff duration, dry days and
site characteristics were developed from the monitoring
program's extensive data base. The duration of runoff and
runoff volume are then used as average runoff intensity,
the mechanism for washoff of pollutants from the highway
area.
PoHutent Buildup and Washoffi. Buildup and washoff of
pollutants from the highway drainage area are predicted
using a carrier pollutant as the mechanism of each proc-
ess. The carrier pollutant is total solids since it had the
best correlation with 16 other commonly used water qual-
ity parameters. Buildup of total solids on the drainage area
is simulated in the model using a buildup rate K1( calcu-
lated in the following manner:
l, = 0.007 (APT08")
where; K, = total solids in Ib/mi/day
ADT = average daily traffic in vehicles per day
The K, factor is then used at the modeling site to accumu-
late solids on the drainage area surface during the pre-
storm dry period.
Removal of the carrier pollutant from the highway area
is predictively accomplished using the standard washoff
equation with the following format:
where; PD = pounds of total solids washed off
P0 = initial surface loading (pounds of total solids)
= K, x dry days x site length in miles
K2 = washoff coefficient
r = average runoff intensity in in/hr
K2 values are selected in the predictive procedure based
upon site characteristics and range from 5.0 to 12.0.
Constituent Loadings. The predictive procedure has, to
this point, determined the mass of total solids washed off
for each rainfall event. The transformation of total solids
into pounds of biochemical oxygen demand (BOD), heavy
metals, nutrients, or any other of 16 available parameters
is performed using individual equations for each site.
These equations have been developed from more than
1,000 individual chemical analyses from the monitoring
program for correlation of parameters. Total loading (Ib)
and concentration (mg/L) of each parameter is listed in the
predictive procedure output.
Predictive Procedure Results. The predictive procedure
was incorporated into a set of equations for input to a
computer. This format allows simulation of runoff quantity
and quality from continuous precipitation records covering
months or years of data. Output was compared to mea-
sured quality data from five of the study's monitoring sites.
Accuracy for predicting the total solids load for the entire
monitoring period at each site was 12 percent low at Mil-
waukee I-794, 15 percent high at Milwaukee Hwy. 45, 15
percent high at Nashville, 1 percent high at Harrisburg
and 37 percent high at Denver. The model output was
verified with independent data from another FHWA project
in Dallas, TX. For the Texas site, the model predicted total
solids loadings 34 percent higher than the measured load-
ings. Further research should refine the predictive proce-
dure as additional data become available.
Procedure Limitations. Because of complex
interactions of rainfall, runoff, and traffic on a highway,
following are some limitations of the predictive procedure
needing improvement during future research.
1 . The predictive procedure assumes the highway area
to be uniformly characterized by the three site types listed.
Significant variations in a site may have widely varying
results.
2. The predicted pounds of total solids washed off dur-
ing a rainfall event depend on model prediction of surface
load at the start of the storm. If surface load is underesti-
mated, the pounds discharged will be low.
3. The use of average runoff intensity to remove pollu-
tants is the quickest method and easiest to calculate.
Peak runoff intensities during the runoff hydrograph may
be more accurate, but are too involved for this procedure.
4. Long dry periods and overlapping storms present
predictive problems in determining the prestorm surface
load.
5. Construction activities are difficult to simulate unless
monitoring data are available to determine K, values.
To accomplish the objectives of Phase II, a literature
search and field monitoring program was conducted. The
field monitoring program was divided into two categories:
pollutant source investigations and migration/transport/
fate studies.
388
-------
CASE STUDIES
Sources of many highway pollutants were adequately
documented in the literature, while pathogenic indicator
bacteria, asbestos, and PCB's were further investigated.
As part of FHWA's Phase I study of the constituents of
highway runoff, significant data were collected with re-
spect to the presence and quantification of highway runoff
constituents; however, a gap remained in understanding
the origin and fate of these constituents within the high-
way environment.
Data were collected to evaluate the qualitative and
quantitative aspects of background pollutant loading to
the highway system, pollutants originating from the high-
way system, and the mechanism of pollutant dispersion
within the transfer out of the highway system. Variables
affecting pollutant deposition, accumulation, and removal
were also measured. These variables include traffic char-
acteristics, highway design, maintenance activities, sur-
rounding land use, and climatic conditions. Field studies
were conducted for a minimum 12-month period to evalu-
ate seasonal effects on these processes. Field monitoring
can be categorized as follows:
1. Atmospheric deposition
2. Total suspended particulates
3. Saltation
4. Highway surface loads monitored through sweep-
ing/flushing studies
5. Runoff quantity and quality
6. Groundwater percolation monitored by lysimeters
7. Soil and vegetation studies
8. Traffic characteristics
9. Highway maintenance data
10. Climatological data
Precipitation at Milwaukee, Wisconsin, I-94, the most
urbanized and industrialized of the sites monitored, had
the highest maximum and median value for most quality
parameters (mg/L). However, for many constituents Sac-
ramento, California, U.S-50, Efland, North Carolina, I-85,
and Harrisburg, Pennsylvania, 1-81 had larger loadings
per precipitation event (mg/m2) than Milwaukee, probably
because of the larger total volume per rainfall event ob-
served. Deposition of chlorides via precipitation was
higher during the winter than summer at Milwaukee and
Harrisburg, possibly attributable to chloride aerosols from
street and highway salting activities.
Although enteric bacteria (total coliform, fecal conform,
and fecal streptococci) were present in paved and un-
paved runoff at the Milwaukee site, they were not detect-
able in precipitation, dustfall or ambient air samples. Ap-
parently, the roadway surface is periodically seeded with
debris containing enteric bacteria. A possible source
could be trucks carrying livestock and stockyward waste;
the FC/FS ratios monitored in runoff indicate the enteric
bacteria present on roadway surfaces to be largely of ani-
mal origin. Bacteriological data also indicated that fecal
conforms remained viable within roadway dust and dirt for
relatively long periods of time (at least 7 weeks). Fecal
coliform and fecal streptococcus bacteria remain viable in
stagnant storm sewer water for at least 13 days.
Asbestiform material was not detected in precipitation,
runoff, dustfall, or air samples at the Milwaukee I-94 site.
These results, consistent with those of FHWA's Phase 1
study on runoff constituents, indicate that the quantity of
asbestiform material present in the highway systems is
either below detection limits or is difficult to detect.
Low levels of polychlorinated biphenyls were detected
in soil, vegetation, precipitation, highway surface dust,
and dirt and runoff samples. Runoff studies indicated that
PCB's in the highway environment are transported from
that environment via runoff during storm events. Also, in-
dicated sources of PCB in highway runoff include precipi-
tation, highway surface dust and dirt, and contaminated
soil eroded by the runoff from unpaved surfaces adjoining
the highway.
Field studies evaluated the quantitative and qualitative
aspects of background pollutant deposition to the highway
system (source studies), pollutant accumulation within the
highway system, and the mechanism of pollutant trans-
port within and out of the highway system.
Bulk precipitation data (wet and dry atmospheric depo-
sition) were collected at each site to establish the level of
pollutants migrating from the highway to the surrounding
environment through atmospheric processes. The area
adjacent to the highway receiving TPM (total particulate
matter) and associated metals (impacted area) was de-
fined using bulk precipitation and 1-cm soils data. One-cm
soils data were used because accumulation of highway-
related metals from atmospheric deposition should be re-
flected in the topsoil layer of areas adjacent to the high-
way. The impact area was defined to be approximately
35 m from the edge of pavement at Milwaukee, 35 m at
Sacramento, 15 m at Harrisburg, and 12 m at Efland. The
smaller impact areas at Harrisburg and Efland are proba-
bly a function of average daily traffic (27,800 and 25,500
vehicles per day at Harrisburg and Efland, respectively,
and 116,000 and 85,900 vehicles per day at Milwaukee
and Sacramento, respectively). Urban sites showed TPM
loadings four times higher than rural sites. Background
metals deposition was similarly higher.
Data indicated that bulk precipitation as monitored dur-
ing this study provided generally precise measurements
of TPM deposition, but that localized effects of vehicular
turbulence and severe meteorological conditions can de-
crease the accuracy.
Another mechanism for removing pollutants from the
highway through the atmosphere is saltation. The quantity
of saltating particles (sand sized particles injected into the
atmosphere by vehicular turbulence) and highway-gener-
ated TPM reaching areas adjacent to the highway ap-
pears to be related to:
1. Average daily traffic
2. Wind speed and direction
3. Available surface load (seasonal variation)
4. Highway drainage design
5. Proximity of travel lanes to right-of-way area
6. Landscape features near the highway affecting wind
patterns.
Monitoring of runoff from the paved and unpaved areas
was segregated to determine pollutant loadings leaving
the highway drainage system and to examine the hydrau-
lics of pollutant movement in the drainage scheme. At the
Milwaukee and Sacramento sites (curb and gutter drain-
age design), the unpaved area contributed negligible
amounts to the total constituent load, while at Harrisburg
(flush shoulder drainage design) the unpaved area con-
tributed approximately 17 percent of the total load for most
constituents.
Runoff data indicated that the highway system has a
large capacity to buffer the runoff of acid precipitation
before it reaches the surrounding environment. Ground
water percolation data also indicated that the soil system
adjacent to the highway sections monitored at Milwaukee
and Sacramento had considerable buffering capacity
against acid rain while the Efland and Harrisburg soil sys-
tems had limited buffering capacity. The prevalence of
acid rain in the United States and the apparent ability of
highway systems to buffer it may have important implica-
tions when considering pollutant migration from the high-
way. The solubility of metals is a function of pH (generally
higher solubilities occur at the extremes of the pH scale)
and the quantity of anionic complexing agents and or-
ganic matter present. Soluble metals would be easier to
remove from the highway surface, would tend to migrate
389
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
further, and would be readily assessed for bioaccumula-
tion.
Highway runoff at Milwaukee, the site with the highest
average daily traffic, had the highest solids loadings, and
generally the highest loadings for most parameters. Sites
where deicing agents were applied showed increases in
total solids, sodium, and chloride loadings during winter
periods. The deicing salt used at Milwaukee was analyzed
for contaminants. The salt analyzed contained lead, zinc,
chromium, copper, cadmium, nickel, and cyanide. The
loading of cyanide, an anticake compound used to keep
salt granular, was approximately 0.79 kg/km/yr. Based
upon loading values, rock salt was also an important
source of cadmium and nickel. At Efland, deicing agents
(rock salt and calcium chloride/sand mixture) were also
analyzed for contaminants. Lead, iron, chromium, copper,
and cyanide were present in the sample analyzed, but in
generally lower levels than at Milwaukee. The contami-
nants associated with deicing agents vary with the source
of the deicing agent and additives used.
Solids and associated pollutants tend to accumulate in
the distress and median lanes, while more soluble pollu-
tants tend to be uniformly distributed across the distress,
median and travel lanes. Apparently, vehicular turbulence
moves any solids from the travel to the outer lanes. Lateral
variation in surface load at a given time appears to be a
function of profile grade and other factors including: inlet
placement, seasonal characteristics, maintenance activi-
ties, and traffic patterns.
Commercial sweeper efficiency studies performed at
Milwaukee showed that commercial sweepers' efficiency
was generally highest for solids and their associated con-
stituents and lowest for the more soluble constituents.
Sweeper efficiency was also higher in summer than in
spring. The surface load, more compacted in the spring
than in summer, was presumably more difficult to remove
by surface sweeping.
Metals and sodium concentrations were generally
higher in the topsoil layers (major rooting zone usually 10-
cm deep) than substrate layers, and decreased with dis-
tance from the highway. Chlorides did not show this gradi-
ent. Lysimeter data indicated that chlorides are removed
from the topsoil layer shortly after spring thaw.
At the Milwaukee site litter was highest for the near
highway samples although biomass production was
slightly higher for samples obtained further from the high-
way. At the Sacramento site no vegetation grew near the
highway. This would be expected since the soils next to
the highway were sandy, low in organic matter, high in
soluble salts, low in nutrients, and high in lead and zinc.
Normal ecosystem processes may be affected in areas
immediately adjacent to the highway (1-2 m), especially
near highways with high average daily traffic (ADT greater
than 85,000 vehicles per day).
The third phase of FHWA's administrative contract re-
search program, Effects of Highways on Water Re-
sources, is underway. The objective of this research is to:
1. Determine the magnitude and extent of the impacts
of highway stormwater runoff on receiving water quality
and aquatic biota;
2. Formulate procedural guidelines for assessing im-
pacts of highway stormwater runoff;
3. Provide guidelines for conducting field studies to de-
termine effects of highway runoff on receiving waters.
An extensive field monitoring and laboratory analysis
program has been conducted at three sites: a stream site
in southeastern Wisconsin (Wisconsin Highway 15), a
stream site adjacent to I-85, west of Efland, North Caro-
lina, and a lake site north of Milwaukee, Wisconsin. In-
terim results indicate no significant impact to receiving
water ecology for highway facilities with low to medium
ADT (less than 30,000).
Acute toxicity bioassays of undiluted highway runoff
from Wisconsin Highway 15 (ADT 12,000) and I-94, Mil-
waukee (ADT 120,000) simulated worse case shock load-
ing on the receiving waters for several days. Some as-
sumptions are implied: (1) the quality of the receiving
water may be temporarily degraded by runoff from a storm
event, but will rapidly return to its previous state; (2) detri-
mental substances in the runoff water are flushed out of
the receiving waters and do not linger; and (3) detrimental
effects on the biota of the receiving water would be from
direct toxicity, not indirect or delayed effects of assaults on
other components of the system.
Neither of the undiluted highway runoff samples was
acutely toxic to the fathead minnow. The fish exposed to
the runoff water swam sluggishly in comparison with those
in the control water, thereby implying that the runoff water
imposed a sublethal stress on the fish. The stress was
removed when the fish were returned to the control water
at the end of the exposure period.
The isopod Asellus was insensitive to exposure to undi-
luted highway runoff water from either source for 4 days.
No delayed responses were observed. The size of the
organisms did not influence the results of the assay.
The amphipod Gammarus was sensitive to exposure to
undiluted water from Highway 15. Of the observed mortal-
ity, 40 percent was attributed to direct toxicity of the water.
Fewer deaths were observed in the assay of I-94 water
than in the control water within the same experiment, al-
though the difference was not statistically significant. No
additional mortality because of delayed effects was dem-
onstrated. This organism was the most difficult to maintain
in the laboratory, and mortality of the control group was
always greater than 20 percent after 4 days. Cannibalism
was observed in the test vessels, indicating possible nutri-
tional stress. Any toxic stress from the runoff water would
then be superimposed on the physiological stress. The
conditions under which the assays using Gammarus were
performed therefore represent a worst case. The size of
the animals did not influence the results of the assay.
Mayfly nymphs, genus Hexagenia, were slightly sensi-
tive to undiluted highway runoff water, but no more than
20-percent mortality was observed in any test vessel. No
delayed response to toxins in the water was demon-
strated, and mortality was not confined to a single size of
the nymphs. Exposure to I-94 water did not inhibit the
nymphs from hatching into adults.
The cladoceran, Daphnia, was not lethally sensitive to
Highway 15 runoff water in a 96-hr flowthrough assay, nor
to I-94 water in a 48-hr static assay. The animals were
under stress when exposed to the I-94 water, relative to
the control group, but they were not dead after 48 hours.
Algal assays, using Selenastrum, demonstrated ad-
verse effects of undiluted runoff water from both Highway
15 and I-94 on algal growth. The algal assay is not a
short-term, acute toxicity test, but rather a chronic toxicity
test. The time frame for chronic tests is not realistically
representative of actual conditions; however, as a bioas-
say tool it may indicate that a water source inhibits algal
growth, and may implicate the causative agent for the
inhibition.
Snowmelt runoff water from Highway 15 contained both
heavy metals and remarkably elevated concentrations of
salt ions from deicing chemicals application. The results of
the algal assay demonstrated a complex interaction be-
tween the metals, salts, a metal chelator, and a phos-
phorus nutrient. Metals inhibition may have been present
but was not clearly resulting from compounding effects of
salt stress. The results of the assays of (-94 water demon-
strated a probable heavy metal stress on the algae. When
metals were chelated, the algae were phosphorus limited.
390
-------
ROCK CREEK RURAL CLEAN WATER PROGRAM:
THE EXPERIMENT CONTINUES
MICHAEL J.NEUBEISER
Rural Clean Water Program
U.S. Soil Conservation Service
Twin Falls, Idaho
INTRODUCTION
Rock Creek in Twin Falls County, Idaho, has long been
recognized as one of the most severely degraded streams
in the State. Both point and nonpoint sources of pollution
have contributed to this problem. The 1972 Federal Water
Pollution Control Act (PL. 92-500) stimulated pollution
abatement efforts, and since then both State and Federal
programs have been directed toward pollution abatement
in Rock Creek. The National Pollution Discharge Elimina-
tion System (NPDES) program essentially has eliminated
point source discharges from food processing plants, fish
hatcheries, and the Twin Falls sewage treatment plant.
Removing these point sources reduced bacterial contami-
nation and nutrient loading, increased the dissolved oxy-
gen level, and improved aesthetics in Rock Creek (Martin,
1984).
Agricultural nonpoint sources, however, continue to
cause severe pollution problems within the Rock Creek
drainage. Irrigation return flows to the creek contain high
concentrations of suspended sediment and related agri-
cultural pollutants such as phosphorus, nitrogen, and fe-
cal coliform bacteria. During the irrigation season, the wa-
ter from Rock Creek can be traced as a brown muddy
streak at its confluence with the Snake River.
This paper will present and briefly discuss the history,
major activities, and progress in restoring the health of
Rock Creek through the Rural Clean Water Program.
PROGRAM EVOLUTION
The Idaho Department of Health and Welfare, Division of
Environment (IDHW) has conducted several water quality
studies over the years in Rock Creek. An initial survey in
1960 identified public health problems (Idaho Dep. Health
Welf. 1960). Water quality studies from 1972 to 1974
(Clark, 1975) have identified the impact of point source
discharges in the Twin Falls area on Rock Creek. Another
water quality survey in 1977 recorded the status of the
upper segment from the townsite of Rock Creek to the
Forest Service boundary (Schaefer and Bauer, 1979). In
addition, a water quality trend station has been sampled
monthly since 1969.
Because of the continuing water quality problems, the
Idaho Agricultural Pollution Abatement Plan (Idaho Soil
Conserv. Comm.) identified Rock Creek as a priority
stream segment in 1979. That year, the Snake River and
Twin Falls Soil Conservation Districts applied for and ob-
tained a planning grant, under Section 208 of the Federal
Water Pollution Control Act, to develop a detailed water
pollution abatement plan for Rock Creek.
In 1980 Rural Clean Water Program (RCWP) funds were
obligated to begin the Rock Creek project. The RCWP
came about as a result of a provision in the 1977 Clean
Water Act amendments to FWPCA. Twenty-one water-
sheds throughout the country were eventually selected for
planning, implementation, and monitoring under the
RCWP.
The Rock Creek RCWP was approved for funding with
the intention that specific agricultural pollutants be signifi-
cantly reduced This program's objectives are to reduce
sediment loading 70 percent, total phosphorus 60 per-
cent, total nitrogen 40 percent, fecal coliform bacteria 70
percent, and pesticides 65 percent through best manage-
ment practices (BMP's). The BMP's were to reduce the
amount of sediment and sediment-related pollutants en-
tering Rock Creek from agricultural lands and the amount
of animal wastes entering the creek by applying animal
waste management systems.
DESCRIPTION OF PROJECT AREA
AND IRRIGATION SYSTEM
Rock Creek is in the south central part of Idaho, in Cassia
and Twin Falls counties. From its headwaters in the Saw-
tooth National Forest in western Cassia County the creek
flows northwest, approximately 67 km (41.6 mi), to the
Snake River, north of the city of Twin Falls. The entire
watershed contains 80,292 ha (198,400 acres). The proj-
ect area includes 18,211 ha (45,000 acres). Elevation
within the entire watershed ranges from 2,432 m (7,977 ft)
at the headwaters to 912 m (2,991 ft) at the mouth of Rock
Creek. The stream gradient is fairly constant down to river
kilometer 27. At this point the gradient substantially in-
creases, being steepest the last 1.5 km (0.9 mi) before the
confluence with the Snake River.
Soils in the lower watershed from the mouth to river
kilometer 47 (mile 25) generally have thin, dark, silt loam
and loam surface layers and very strongly calcareous sub-
soils. These soils vary in depth and are underlain by frac-
tured basalt. The soils developed under arid conditions
and in their natural state are low in organic matter. They
are highly productive and highly erodible.
The Rock Creek project area watershed contains ap-
proximately 350 farms. The basic crops are dry beans,
small grains, sugar beets, corn, dry peas, and alfalfa. All
crops are irrigated because of low (8" to 10") annual pre-
cipitation. The majority of cropland is furrow irrigated.
Irrigation water is diverted from the Snake River and
delivered at a regulated and measured rate through a
network of canals owned and maintained by a private
company, the Twin Falls Canal Company, from about mid-
April through mid-October.
The developers of the irrigation tract used natural
streambeds and other drainages as much as possible to
deliver water to the fields and to carry runoff water away.
Water is diverted from the high line and low line canals to
the laterals. Laterals are the heads of the drainages that
deliver irrigation water to the farms and also carry runoff
return flows to Rock Creek. Water is diverted from a lateral
to the first field, and runoff from the first field is used on
the next field with whatever additional water is needed
from the lateral to provide adequate water supply for the
second field. The method continues through each farm
until the water reaches the next farm. The wastewater, or
391
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
runoff, may enter the next farm directly or some portion
may reenter the lateral (along with newly accumulated
sediment and nutrients) and be diverted onto another
farm further downstream. This system of water delivery
continues through the length of the lateral until it dis-
charges into Rock Creek.
Before the irrigation tract was developed, Rock Creek
was historically fed mainly by snowmelt. Low flows oc-
curred in summer and fall and high flows in late winter and
spring. Irrigated agriculture changed these flow trends,
greatly increasing the summer stream flow. Peak flow now
occurs in September (Martin, 1984).
LANDOWNER PARTICIPATION
IN THE RCWP
Without question, the critical key to success is landowner
acceptance of this program, or any similar voluntary par-
ticipation program, and the willingness to commit funds to
implement practices which may not provide an immediate
tangible benefit to the landowner's bank account or farm
operations.
When the Rock Creek RCWP began in 1980 the agricul-
tural market was healthy and initial interest and response
for planning assistance were overwhelming—so over-
whelming, in fact, that the local Soil Conservation Service
staff were playing catch-up from the beginning. With some
changes in staff along the way, it took almost 31/2 years to
service the backlog of requests for planning assistance.
About 1 year after the program began, the farm econ-
omy faltered and the number of requests steadily de-
clined. By spring of 1984 there were no requests for plan-
ning assistance on file.
Surprisingly, the economic condition has not seriously
crippled the project's goals, although a fourth straight year
of depressed market prices could adversely affect remain-
ing contracting and implementation goals. However, a
sound foundation has been established by the local soil
conservation districts and SCS staff. The project has been
successful to date because landowners recognize the
need for improvements. Most individuals knew that Rock
Creek was degraded before the project was ever con-
ceived. The farmers within the project are to be com-
mended for their participation despite the poor economy.
COMPREHENSIVE MONITORING
AND EVALUATION
The Comprehensive Monitoring and Evaluation phase of
the Rock Creek RCWP is undoubtedly more technical and
complex than the contracting and best management plan
implementation phase. Idaho's Division of Environment is
the principal agency responsible for monitoring stream
flow and the water quality parameters outlined in the plan
of work. The division monitors trends in water chemistry,
benthic macroinvertebrates and game fish populations.
Other aspects of the monitoring program are being car-
ried out by various State and Federal agencies. The U.S.
Department of Agriculture (USDA)—Economic Research
Service (USDA-ERS) and University of Idaho Agricultural
Economics Department are jointly evaluating the social
and economic impacts of installing best management
practices. The Idaho Energy Resources Research Insti-
tute and University of Idaho Agricultural and Civil Engi-
neering Departments, in cooperation with the USDA-Agri-
cultural Research Service and University of Idaho
Cooperative Extension Service, have attempted to de-
velop a sediment generation and routing model for irriga-
tion return flow. The Agricultural Research Service is also
evaluating the effectiveness of individual sediment reten-
tion BMP's at the Snake River Conservation Research
Center.
STRATEGY FOR MONITORING
IMPLEMENTATION EFFECTS
The RCWP project watershed is divided into 10 subwa-
tersheds, or subbasins, for individual analysis and com-
parison of data in relation to different levels of BMP imple-
mentation (Fig. 1). Monitoring stations were set up on the
priority subbasins (1, 2, 4, 5, and 7). Monitoring stations
were also established at key locations along Rock Creek
(Fig. 2). Idaho Division of Environment staff have been
collecting data at these locations since 1981.
The RCWP subbasins receive water from the High Line
and the Low Line Canals. Subbasins also receive water
from seeps occurring throughout the project. However, not
all of these seeps contribute irrigation water and as such
are not considered indicative of upstream water quality.
RCWP subbasins 4,5, and 7 are monitored where water is
diverted from the canals into the major laterals. Subbasins
1 and 2 receive a combination of seep and canal water
and are monitored as close to the source as possible. The
laterals are also monitored near their points of discharge
into Rock Creek. The amount of water delivered to the
subbasins is controlled and the way it is used and distrib-
uted throughout the subbasins is controlled. Therefore,
the hydrologic cycle that affects erosion and subsequent
irrigation runoff is almost totally man-induced during the 6-
month-long irrigation season.
The subbasins are monitored only during the irrigation
season, biweekly from mid-April to mid-May, weekly from
mid-May to early August, and biweekly again from early
August to mid-October. The heaviest pollutant load from
irrigation return flows occurs from mid-May to early Au-
gust. Irrigation continues until a week or so before har-
vest, and most crops are removed by mid-October. The
irrigation return flows move through drainageways that
have basalt bedrock at or very near the surface and slope
gradients that allow little deposition of sediment. There-
fore, most of the pollutant load reaches Rock Creek. The
resultant changes in water quality from the implementa-
tion of BMP's should be detectable at the downstream
stations for each lateral where they discharge into the
creek (Martin, 1984).
MONITORING RESULTS
AND CONCLUSIONS
By the end of the 1983 irrigation season, the percent of
land benefiting from BMP implementation ranged from
none in the nonpriority subbasins (9 and 10) to 84 percent
in subbasin 7. Annual levels of implementation will always
fluctuate because of factors such as crop rotations, eco-
nomics, and tillage practices. Along with changes in the
weather, evaporation rates, irrigation demands and incom-
ing (upstream) water quality, these factors will influence
outgoing (downstream) water quality
Sediment Loading
The most obvious effects of BMP's are seen in the drain-
ageways closest to the site of implementation. Significant
(p < 0.01) reductions in suspended sediment concentra-
tions were measured at five of the six subbasin stations
that discharge into Rock Creek. The suspended sediment
concentrations at those five sites in 1983 averaged 55
percent less than the concentrations measured in 1981.
Altogether, 45 percent of the acreage of those five sites
392
-------
CASE STUDIES
R.17E.|R.18E.
R.18E.|R.19E.
ROCK CREEK RCWP BASINS
TWIN FALLS COUNTY, IDAHO
Figure 1.—Rock Creek RCWP basins, TWin Falls County, Idaho.
benefitted from BMP implementation upstream. Three of
the Rock Creek monitoring sites had significant (p < 0.01)
reductions in suspended sediment concentrations, with
an average of 50 percent reduction from 1981 to 1983. It
appears that the project objective of 70 percent reduction
in the suspended sediment contributed to Rock Creek
from the subbasins may be attained when all of the con-
tracted BMP's are installed.
Phosphorus
The changes in concentrations of phosphorus (total and
dissolved orthophosphate) in the study area were not as
pronounced as suspended sediment. A significant (p <
0.05) reduction in total phosphorus concentration was
measured at only one Rock Creek site, S-1. This site was
the only one on Rock Creek to exhibit a significant (p <
0.01) decrease in dissolved orthophosphate from 1980 to
1983. Two subbasin sites, 2-2 and 4-3, had significant (p
< 0.01) reductions in total phosphorus concentrations, 48
percent and 53 percent, respectively. Significant (p <
0.01) changes in dissolved orthophosphate concentra-
tions were measured at subbasin sites 4-2 (35 percent
increase) and 4-3 (42 percent decrease). The reason for
the increase at site 4-2 is not yet understood. Both these
stations also exhibited significant (p < 0.01) reductions in
393
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
suspended sediment concentrations. The expected rela-
tionship of total phosphorus and suspended sediment has
been inconsistent in the analysis to date. The project ob-
jective of 60 percent reduction of total phosphorus, how-
ever, still appears to be realistic and should be retained
until further BMP implementation is attained and addi-
tional analyses are conducted.
Analysis of the nitrogen data is divided into two compo-
nents: Kjeldahl (organic) and inorganic. Significant (p <
0.01) reductions in Kjeldahl nitrogen concentrations were
measured at all Rock Creek monitoring sites from 1981 to
1983, with an average decrease of 58 percent. Similar
significant (p < 0.01) decreases in this parameter were
measured at three subbasin stations, 2-2, 4-3, and 7-4,
with an average of 55 percent reduction in Kjeldahl nitro-
gen concentration. The trends for inorganic nitrogen are
inconsistent with suspended sediment concentration
trends at both the Rock Creek and subbasin sites.
Rock Creek had two sites, S-3 and S-4, with significant
(p < 0.05) increases that averaged 211 percent of 1981
values. Martin (1983) and others have noted that BMP's
targeted at reducing suspended sediment appear to have
insignificant effect on reducing inorganic nitrogen concen-
trations, but do have a dramatic impact on reducing
Kjeldahl nitrogen concentrations. If the project objective of
reducing total nitrogen by 40 percent is to be achieved it
will apparently have to be met by the reduction of Kjeldahl
nitrogen alone. It may be time to reevaluate this objective.
Feeal ©olofoirmri) Bacteria
Bacterial contamination of Rock Creek from fecal con-
forms was significantly (p < 0.01) reduced at Rock Creek
monitoring sites S-3 and S-4. From 1981 to 1983 the con-
centration at these sites decreased 63 percent and 72
percent, respectively. From 1981 to 1983 only subbasin
site 4-3 had significant (p < 0.01) reductions in fecal coli-
form bacterial concentrations. This pollutant decreased 70
percent at subbasin site 4-3 after an RCWP participant
removed his cattle from the corrals through which this
lateral flowed. The reason for reduction at Rock Creek site
S-3 is perhaps the same. A large livestock operation was
eliminated during this period; although no specific data
were collected, this is very likely the cause for the mea-
sured difference at S-3. The probable cause for the reduc-
tion in fecal conform bacterial concentrations at S-4, and
in part the other noted locations, is a general decline in
numbers of cattle owing to poor market prices. Since a
comprehensive study of livestock numbers is not con-
ducted annually this is only an educated guess. The proj-
ect objective of reducing fecal coliform bacteria concen-
tration by 70 percent is attainable. However, this is an area
which will require intensified information and education to
the landowner.
Samples of game fish for pesticide analysis were collected
in March 1982 at Rock Creek sites S-1 and S-6 (Martin
and Bauer, 1982). Site S-6 is above the irrigation drainage-
ways. Rainbow and brown trout were analyzed for pesti-
cides to determine whether pesticide usage in the project
has an adverse effect on trout populations and to assess
whether they posed a possible hazard to anyone consum-
ing the fish.
Four pesticides, or their analogs, were detected in the
samples: DDT, PCB, toxaphene, and dieldrin. Minimum
detection limits for these toxins are 0.001 mg/kg, which is
equivalent to one part per million (ppm).
Residues in trout at both stations were low, and well
below Food and Drug Administration (FDA) standards for
human consumption. DDT averaged 0.106 ppm in rain-
bow trout near the mouth (S-1), and 0.012 ppm in rainbow
trout above the project area (S-6). Large suckers at the
mouth of Rock Creek contained the highest concentration
of pesticide residue, but these concentrations were still
well below the FDA standards.
Another sampling of game fish for pesticide analysis
was conducted in March 1985. The results of those tests
are pending.
It appears from this small sample that pesticide resi-
dues are not a problem in Rock Creek and will not inter-
fere with the RCWP efforts to enhance the fishery. How-
ever, if drastic changes in tillage operations occur (see
Program Changes, this paper) the need to continue pesti-
cide monitoring must be considered.
Figure 2.—Map of the Rock Creek Rural Clean Water Pro-
gram study area, Twin Falls County, Idaho.
Flow
Studies of sediment transport in Rock Creek showed that
no computer models of sediment routing could determine
the impact of changes in quality of irrigation return flow
from the Rock Creek watershed. Analysis of flow rate and
sediment data indicated that 85 percent of the sediment
transported in the stream is wash load (irrigation return
flow sediment) and could not be modeled (Sterling, 1983).
Analysis of sediment particle sizes, inflows, and deposi-
tion patterns in the lower reach of Rock Creek and appli-
cation of transport models showed that irrigation return
flow sediments are generally washed through the stream
on an annual basis.
A furrow irrigation sediment yield model based upon
stream power concepts was developed and tested (Brock-
way et al. 1985). The model, sensitive to furrow end slope
and to furrow roughness and using actual runoff hydro-
graphs, was able to predict annual sediment yields within
10 percent of measured yields. Driven by runoff hydro-
graphs, the sediment yield model was linked to a kine-
matic-wave furrow irrigation model. The combined models
394
-------
CASE STUDIES
can predict the annual sediment yield from a field in a
watershed when given information concerning number of
irrigations, set times, furrow stream sizes, slope, rough-
ness, length, and infiltration.
Sediment retention treatment models were also devel-
oped (Brockway et al. 1985) to define potential sediment
retention rates of the BMP's. According to these models,
the sediment retention practice construction designs used
by SCS have potential retention rates in excess of those
required by the project guidelines. The vegetative filter
strip model predicts a potential sediment removal effi-
ciency based upon ideal management. The management
of the filter strip will significantly influence the operational
sediment removal efficiency. Whereas the filter model pre-
dicts efficiencies of 80 to 90 percent measured on ideally
managed strips, normal farm management of filter strips
will produce operational efficiencies of 50 percent.
The watershed sediment yield model estimates the av-
erage sediment yield at the field edges. The model does
not account for sediment deposited in drains or on fields
due to reuse of irrigation return flows. However, the model
does adjust sediment yields to account for BMP's installed
at field edge by using their sediment removal efficiency.
BMP's the model takes into account include filter strips,
sediment basins, l-slots, T-slots, minibasins, and buried
pipe runoff control systems.
PROGRAM CHANGES
The sediment yield problem appears to be under control,
relative to water quality in Rock Creek, especially with
many more sediment retention BMP's planned for imple-
mentation. A logical question is: how long will the reduced
sediment yield and improved water quality in Rock Creek
last? Will landowners realize the usefulness of BMP's and
continue to install and maintain them after contract obliga-
tions have expired, and without cost-share funds?
A study of irrigation return flow was carried out from
1976 through 1980 by USDA-Agricultural Research Serv-
ice and the University of Idaho Agricultural Engineering
Department, in conjunction with the Snake River Soil Con-
servation District, IDHW, Division of Environment, and
U.S. EPA (Brockway et al. 1980). BMP's were developed
for reducing sediment and nutrient concentrations in irri-
gation return flows, and these practices were applied to a
small drain, the "LQ," which is adjacent to the Rock Creek
RCWP project and discharges directly into the Snake
River. Because of BMP implementation, the sediment dis-
charged to the Snake River was reduced by 80 percent,
from 10,000 tons per year in 1977 to 2,000 tons per year in
1979.
The 1983 post-project evaluation of the LQ Drain water-
shed indicated a significant improvement in water quality.
However, there had been a reduction in water quality
when compared to the end-of-project levels. The amount
of sediment leaving the watershed in 1983 (5,500 tons)
was 55 percent of 1977 (10,000 tons), but 275 percent of
the 1980 (2,000 tons) end-of-project level. During the 1981
through 1983 period, governmental assistance was very
low for BMP's, as were information and education efforts.
As a result of no government assistance and a depressed
farm economy, practice implementation was minimal and
water quality in the LQ Drain is again declining.
With the LQ Drain project as an example, it is clear that
steps are needed to ensure that Rock Creek will enjoy
long-term benefits from the RCWP.
The original philosophy of the Rock Creek RCWP con-
tended that improved water quality could be achieved by
implementing BMP's that improved the irrigation system,
for example concrete ditch and gated pipe, and BMP's
that trapped end of furrow sediment yield, such as vegeta-
tive filter strips, sediment basins, and buried pipe runoff
control systems. These BMP's would be supplemented by
a strong information and education effort in irrigation wa-
ter management, also a planned and contracted BMP.
Monitoring data indicate that the program is achieving the
main objective—sediment reduction in Rock Creek—and
that the creek is gradually improving. However, these ac-
complishments are shortsighted, and, as was learned in
the LQ Drain project, will probably be short-lived.
One of the problems with the original philosophy was
that it did not adequately address erosion-control prac-
tices and problems associated with the continued reduc-
tion of crop yields where topsoil depth is at a critical point.
Another problem that has come to light is that this philoso-
phy is too expensive.
Economic analysis of the BMP implementation phase of
the RCWP strongly suggests that emphasis on less costly
field management practices will be necessary to ensure
an economically justifiable project (Gum et al. 1984). The
initial emphasis on structural irrigation and sediment re-
tention practices interested farmers in participating, but
the current need, as dictated by cost-to-benefit ratio, is to
reemphasize irrigation water management and to initiate
field management practices such as reduced tillage, mini-
mum tillage, and crop residue. These types of practices
(1) reduce farmer costs; (2) reduce in-furrow erosion, re-
sulting in greater long-term benefits; (3) reduce end of
furrow sediment yield more cost effectively; (4) provide
increased protection of the resource base; and (5) have a
high probability of continued farmer use after the project is
completed and the contracts have expired.
At this stage we now recognize that improving irrigation
systems and irrigation water management techniques,
while important and effective in reducing erosion and sub-
sequent sediment yield, does not technically and econom-
ically address the water quality problem in the best way
possible. BMP's that will improve in-furrow erosion control
must be incorporated into the Rock Creek RCWP.
This recognition was included in the 1984 Rock Creek
RCWP Annual Report. A request to cost-share conserva-
tion tillage practices was made in the report. That request
received full concurrence by the Local Coordinating Com-
mittee and State Coordinating Committee. The National
Coordinating Committee has recently approved the con-
servation tillage BMP for cost-share under the Rock Creek
RCWP.
SUMMARY
The Rock Creek Rural Clean Water Program has signifi-
cantly reduced sediment loading into Rock Creek using a
combination of sediment retention and irrigation improve-
ment practices, including irrigation water management.
Participation has been higher than expected because of
a successful information and education effort, cost-shar-
ing, and the positive results of BMP's already used that
stimulated further landowner interest. The project has ap-
proximately 11,400 ha (28,159 ac) identified as critically
eroding or highly subject to erosion. The goal is to treat 75
percent, or 8,550 ha (21,119 ac), through a combination of
BMP's carried out under long-term contracts. As of Sep-
tember 1984, 146 contracts planned to treat 7,092 ha
(17,517 ac) were in effect, approximately 83 percent of the
contracting goal. One year of contract preparation re-
mains. Since the first contract was approved in October
1980, about 1,025 individual BMP's have been imple-
mented, among them 17 sediment basins exceeding 765
m3 (1,000 yd3) each. Practices have been applied mostly
on schedule, and even though some landowners are de-
laying implementation because of poor economic trends,
BMP implementation has progressed well overall. Few re-
395
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
quests for planning assistance are received. After only 4
years the Rock Creek RCWP has significantly helped
meet targeted chemical parameters. As BMP implementa-
tion expands to influence and benefit more acres in the
project area, the resultant reduced pollutant loads dis-
charged from the subbasins should become even more
pronounced. Rock Creek will be slow to establish stable
long-term changes in its chemical and biological composi-
tion, but when the remaining sources of pollutants in irri-
gation return flows are subjected to the proposed level of
BMP implementation, positive long-term improvement in
the aquatic ecosystem should come about.
A tremendous challenge comes with the decision to im-
plement conservation tillage on furrow irrigated cropland.
Practically no information exists about conservation tillage
practices under this type of irrigation, even though millions
of acres are irrigated in this fashion in the western United
States.
It is a challenge to the agricultural institutions to adapt a
technology long in use in some regions, but in its infancy
here. The service agencies—Soil Conservation Service,
Agricultural Research Service and Cooperative Extension
Service—face a challenge in communicating this new in-
formation to land users. But mostly the challenge is to the
farming community, a community now struggling to im-
prove its chances for survival that must now accept, de-
velop, and implement new ideas more quickly than it ever
has in the past. The RCWP, as an experimental program,
has an opportunity to promote conservation tillage as a
method of controlling in-furrow erosion. We hope to an-
swer some important questions, and we hope to be able to
apply the answers to more than just the Rock Creek
RCWP.
REFERENCES
Brockway, C.E., G.S. Johnson, C.W. Bobbins, and D.L. Carter.
1980. Irrigation Return Flow Study Final Report, 1976-1979.
Agric. Eng. SEA/ARS, Univ. Idaho, Kimberly.
Brockway, C.E., F.J. Watts, C.W. Robison, and R.R Sterling.
1985. Development of a sediment generation and routing
model for irrigation return flow. Idaho Water Energy Resour.
Inst. Dep. Agric. Eng. Civil Eng. Univ. Idaho. Unpubl. rep.
Clark, W.H. 1975. Water Quality Status Report, Rock Creek,
Twin Falls County, Idaho, 1972-74. Div. Environ. ID Dep.
Health Welf.
Gum, R., R. Magelby, and J. Kasal. 1984. Interim economic
evaluations of the Rock Creek, Idaho RCWP Project. Pages
18-19 in Rock Creek Rural Clean Water Program Annual Pro-
gress Report, Executive Summary. Soil Conserv. Serv. Twin
Falls, ID.
Idaho Department of Health and Welfare. 1960. Report on pollu-
tion problems in Rock Creek, Cassia and Twin Falls County,
Idaho. Div. Environ. Boise.
Idaho Soil Conservation Commission. 1979. Idaho agricultural
pollution abatement plan. Idaho Soil Conserv. Comm. Boise.
Martin, D., and S.B. Bauer. 1982. Rock Creek Rural Clean Water
Program, Water Quality Monitoring Assessment. First year ba-
seline rep. Div. Environ. ID Dep. Health Welf. Boise.
Martin, D. 1983. Rock Creek Rural Clean Water Program, Com-
prehensive Monitoring and Evaluation. Ann. rep. Div. Environ.
ID Dep. Health Welf. Boise.
1984. Rock Creek Rural Clean Water Program, Com-
prehensive Monitoring and Evaluation. Ann. rep. Div. Environ.
ID Dep. Health Welf. Boise.
Schaefer, A., and S.B. Bauer. 1979. Water quality status report,
1976-77, Upper Rock Creek. Water Qual. Serv. No. 38. Div.
Environ. ID Dep. Health Welf. Boise.
Sterling, R.P. 1983. Stream channel response to reduced irriga-
tion return flow sediment loads. M.S. thesis, Univ. Idaho, Mos-
cow.
396
-------
REGULATING NONPOINT SOURCES OF POLLUTION FROM TIMBER
HARVESTING—A CASE HISTORY OF THE CALIFORNIA
EXPERIENCE
CARLTON S. YEE
Member, California State Board of Forestry
Forest Engineering and Watershed Management
Humboldt State University
Arcata, California
ABSTRACT
California possesses over 16,800,000 ha of forest land,
making it one of America's top timber producing States.
To meet the mandate of the 1972 Federal Water Pollution
Control Act, section 208, the State of California has had
to initiate significant institutional and regulatory changes
to comply with this legislation. Regulation of forest prac-
tices in California on private timber lands involves a com-
plex process of rule-making, mandatory timber harvest-
ing permits prepared by State-licensed professional
foresters, interdisciplinary review and approval of timber
harvesting plans by the State's Department of Forestry,
and ongoing enforcement inspections during the opera-
tional life of the plan. In addition, logging operators in the
State must also be licensed, adding additional leverage
for rule compliance during logging. This paper summa-
rizes the history of the last 8 years involving the California
State Board of Forestry and the California State Water
Resources Control Board efforts to institute a silvicultural
nonpoint source pollution control program.
INTRODUCTION
California's forest resources are among the most valuable
in the Nation. For 20 years California has ranked as one of
the top three States in annual harvested volume of forest
products. Over 42 percent of the State, 16.8 million ha (42
million acres) is forest land, an area larger than 30 of the
50 States. The forest lands produce over 70 percent of the
annual water yield, on which California's agricultural and
urban economies depend. Eight million ha (20 million
acres) of this forest land is in private ownership and is
subject to regulation by State forest practice rules when-
ever harvested commercially.
The subject of timber harvesting regulation and its rela-
tionship to water quality and other environmental con-
cerns has provoked heated political controversy in Califor-
nia for almost two decades. More than the usual animosity
between the timber companies and environmentalists,
fishing interests, and local communities reflects a basic
disagreement abut how the State's forests are to be used.
Because California probably epitomizes the growing con-
flict of values at an urban/forest interface (Vaux, 1982),
California's handling of such political confrontations may
inform other States facing forest/urban conflicts in the fu-
ture.
The 1972 Federal Water Pollution Control Act (FWPCA)
placed responsibility for water pollution control in forest
management at the State level, with oversight and na-
tional administration by the U.S. Environmental Protection
Agency. The FWPCA did not give EPA direct authority over
control of pollution from nonpoint sources, but dictated
that nonpoint sources of pollution had to meet State-devel-
oped water quality standards that were approved by EPA.
These became known as Basin Plans (Gefrath, 1984).
Although the FWPCA was probably one of the most
complicated measures ever passed by Congress, the writ-
ers correctly assessed that nonpoint sources had to be
treated by modification or elimination of practices that
caused pollution. These best management practices
(BMP's) were to be favored over "end of pipe" remedies
so common to point source treatments. As a result, under
section 208, local governments were required to develop
Areawide Waste Treatment Management Plans for both
point and nonpoint sources of pollution. Also under this
section, the governor of each State could require that agri-
cultural, silvicultural, mining, and related nonpoint source
activities become part of a State's control program. If the
State assumed control of nonpoint source enforcement,
any plan for control must include: "... processes to (i)
identify, if appropriate, agriculturally and silviculturally re-
lated nonpoint sources of pollution, and (ii) set forth proce-
dures and methods (including land use requirements) to
control to the extent feasible such sources."
Since FWPCA was passed, California has had two
agencies with an interest in regulating silviculturally re-
lated nonpoint sources, the State Water Resources Con-
trol Board and the Board of Forestry. California's 1967
Porter-Cologne Water Quality Control Act designated the
Water Resources Control Board as the State's water pollu-
tion control agency for all purposes stated in the FWPCA.
The Board of Forestry, on the other hand, was the forest
practices rulemaking body, directing policy for the State's
Department of Forestry, the enforcement agency for forest
practice regulation. In 1976, sensing that a battle over
regulatory turf would be best avoided, Governor Edmund
Brown Jr. assigned the Federal-State coordination and
contractual duties to the Water Resources Control Board,
but allowed for that agency to subcontract to the Board of
Forestry for development of a set of BMP's for silvicultural
operations. The interagency agreement called for the
Board of Forestry to review and revise rules relating to
watercourses and erosion control, erosion hazard rating,
and silvicultural and cutting methods; to provide for in-
creased public notice of harvesting operations; and to for-
malize an interdisciplinary review team procedure for tim-
ber harvesting permits. It was the State's intent to
eventually certify the Board of Forestry as the agency
designated to carry out the mandates of section 208 on
private timber lands.
In early 1979, the Board of Forestry set up a committee
of its own and Department of Forestry personnel, industry
and public agency foresters, other land use specialists,
environmental organization representatives, and the gen-
eral public to study forest practices as they related to wa-
ter quality. They investigated whether the structure of
mandatory timber harvesting permits and licensing pro-
fessional foresters and loggers, along with an aggressive
enforcement program would provide an adequate frame-
work for a BMP program. They also examined the
397
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
changes required to correct potential deficiencies. This
report was submitted in its final form to the Water Re-
sources Control Board in October 1983. In March 1984,
after lengthy hearings and much heated testimony, the
Water Resources Control Board granted certification to
the Board of Forestry for a limited term, 4 years. Because
the Board of Forestry study presented new rules and pro-
cedures, the Water Resources Control Board made certifi-
cation contingent upon the development of a Monitoring
and Assessment Program to evaluate the Board of For-
estry program. At the end of the 4-year term, further hear-
ings will determine if an unlimited time certification will be
granted.
THE CALIFORNIA PROGRAM-
OVERVIEW
The California BMP program for forestry relies on more
than just rules. Even though the rules are acknowledged
by most authorities as the toughest forest practices laws in
the United States, California's program involves an inte-
grated process involving legislation, administrative regula-
tions, licensing of professional foresters and timber opera-
tors, and an active enforcement program. California chose
an integrated process for three reasons:
Timber Harvesting Is Very Site-Specific
Forested terrain in California varies greatly in physical fea-
tures, site quality, vegetative species, and climatic factors.
Ownership objectives and past logging history serve to
compound these variations, making it very difficult to set
narrowly prescribed standards for performance, as might
be common in regulating point sources.
Public concerns over timber harvesting are usually
greater in more heavily populated forest areas. Consider-
ations such as traffic, noise, and timing of logging become
important, in addition to questions of water quality.
In response to the adverse environmental effects of tim-
ber harvesting, many mitigation measures are possible.
However, practices that work in one place may not work
elsewhere, may not be needed, or may even cause dam-
age.
In view of the wide variety of conditions, the Board of
Forestry chose to adopt general, flexible rules and then
find a process to make them specific. This is an extremely
important point and will be elaborated upon later in this
paper.
A Mixture of State Laws Influence Forest
Practices
The Board of Forestry operates under the Forest Practices
Act, the Professional Foresters Law, the California Envi-
ronmental Quality Act (CEQA), the Coastal Act (a coastal
zoning law), the Wild and Scenic Rivers Act, and a variety
of provisions in the State's Fish and Game Code, the Wa-
ter Code and the Government Code.
The Professional Foresters Law and a strong new For-
est Practices Act went into effect in 1972-73. A court deci-
sion in 1975 held that provisions of CEQA applied to tim-
ber harvesting. Ultimately the concept of functional
equivalency under CEQA was developed. Board of For-
estry rules and procedures were changed and certified as
functionally equivalent to the environment impact review
procedures under CEQA for other types of projects. Multi-
disciplinary review of proposed harvesting operations
were included.
Until 1978, the director of the Department of Forestry
could require the necessary mitigation measures under
CEQA even if such measures were not spelled out in the
Forest Practice Rules. The Legislature in 1978 mandated
that the Board of Forestry include standards in its rules for
the director to exercise discretion in their application. This
mandate changed the types of general rules that could be
used by the Board of Forestry and led to more emphasis
on procedures and the timber harvest plan review process
to make the rules specific.
Maximum Flexibility in the Field Was Needed
California has the strongest law to license and discipline
foresters in the United States. Foresters working for both
the Department of Forestry and private timber companies
and landowners are usually licensed or in training for li-
censing. Timber operators must also be licensed.
To harvest timber commercially (for sale or exchange) in
California requires that the registered professional for-
ester prepare a timber harvest plan. This document goes
through an interdisciplinary review process and eventually
contains specific enforceable conditions to protect the en-
vironment. These conditions interpret and make the For-
est Practice Rules specific.
The Board of Forestry has adopted the philosophic
premise that rules will be written rather generally and be
made specific in the timber harvest plan. This is very simi-
lar to the practice of the U.S. Forest Service to adopt very
general statements in policy and to make them specific in
technical manuals and handbooks. Recent rulemaking by
the Board of Forestry has heavily emphasized registered
professional foresters. Less reliance is placed on prescrip-
tive standards in rules to control loggers and more reli-
ance is placed on the timber harvest plan.
THE CALIFORNIA PROGRAM IN DETAIL
The Forest Practice Act
This act requires the Board of Forestry to adopt rules to
assure the continuous growing and harvesting of commer-
cial tree species and to protect soil, fish, and wildlife re-
sources. Board of Forestry rules must deal with soil ero-
sion, water quality, watershed control, flood control, and
the control of timber operations that will adversely affect
the uses of waters in the State.
Before harvesting, timber owners must have obtained a
timber harvest permit prepared by a registered profes-
sional forester and approved by the director of the Depart-
ment of Forestry. It is a mandatory permit, analogous to a
building permit, and it is a criminal offense in California to
commercially harvest timber without one. Proposed per-
mits are noticed publicly and reviewed by an interdisi-
plinary review team that may recommend acceptance of
the plan with or without modification. It may also recom-
mend rejection if suggested additional mitigation mea-
sures meeting the standards set in the rules are not incor-
porated by the submitting registered professional forester.
The Timber Harvest Plan Process
Figure 1 illustrates the complete timber harvest permit
process in California. The process is broken up into three
periods: the filing period; the preharvest inspection per-
iod, which is not compulsory where no environmental
questions exist; and the review period, when the review
team determines if additional mitigation measures are
needed.
Note that the time periods listed are maximums. The
longest time it would take a timber harvest permit to pass
through this process is 35 days. A longer time may be
involved if the permit submitter and the Department of
Forestry agree to an extension, as in circumstances in-
volving lengthy negotiations on mitigation measures. In
practice, the average permit is processed in about 20
398
-------
CASE STUDIES
THP Received by CDF
(Beginning of Public Review Period)
Public and
Neighbors Notified
THP Reviewed for Accuracy,
Review Team inspects THP
for Preharvest Inspection Need
THP Returned As
Deficient
^ THP Accepted for Filing]
I
No Preharvest Inspection,
Review Team Makes
Recommendation on THP
Preharvest Inspection
Conducted**
Review Team Makes
Recommendation on THP **
CDF Director Makes
Determination on THP
&
i§
23
Q
§S
fc SI
[THP Approved f
Appeal to POP
|THP Abandoned
** Additional Measures May Be Incorporated into THP
Figure 1 .—California's timber harvesting plan review process.
days, and no less than 10 days, so that public comment
can be heard, even on a simple, very remote plan.
For the State as a whole, an average of over 1,400
permits per year has been processed in the recent past.
This is down from as high as 1,900 plans during the late
1970's. Review teams work at each of the five Department
of Forestry regional offices. However, a great majority of
the permits, both in numbers and volume harvested, oc-
cur in the northern half of the State.
If a permit is denied by the Department Director, the
submitter may appeal to the Board of Forestry. At present,
no other party may appeal the issuance or denial of a
timber harvest permit.
Enforcement
As has been mentioned previously, Board of Forestry rules
are enforced by the Department of Forestry. The Forest
Practice Enforcement Process is shown in Figure 2. The
Department has many legal enforcement tools as shown
in Table 1. Obviously, the Department inspectors attempt
to use persuasion and less severe tools to obtain compli-
399
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Harvesting Begins
Vith Approved THP
Inspection by CDF
No Violation Observed
Inspection Report
Sent to
Reponsible Person
Minor Correctable
or Non-correctable
Violation Observed
Serious Correctable or
Non-correctable
Violation Observed
Inspection Report
Sent to Reponsible Person.
Describes Violation &
Due Date
Inspection Report Sent
To Responsible Person;
Describes Violation.
Corrective Vork &
Due Date
I
RPF Disciplinary Action
if Serious Unprofessional
and/or
_L
Inspection
Shows Violation
Corrected
1
1
Inspection
Shows Violation
Not Corrected
I
1
Misdemeanor
Action
Inspection Report
Sent to
Reponsible Person
Clearing Violation
Administrative
Action
I
Violation
Corrected bv
Responsible Person
1
Violation
Not corrected by
Responsible Person
Civil
Action
Violation Corrected
by Responsible Party
or CDF
Figure 2.—California's forest practice enforcement process.
Table 1.—Forest Practice Enforcement Actions and
Possible Resolutions.
Actions
available
Possible
resolution
1. Warnings
2. Notice of intent to take
corrective action
3. Stop work orders
4. Injunctive process
5. Misdemeanor action
6. Timber operator license
action
7. RPF disciplinary actions
Via inspection reports, verbally;
and/or administrative letters
Violation corrected by
responsible person; violation
corrected by CDF
Violation corrected—operations
continue; agreement entered into
for continuance of operations;
injunctive process started
Temporary restraining orders;
preliminary injunctions;
permanent injunctions; court
ordered correction; stipulated
agreements
Fine; probation; incarceration;
dismissal
Warning; suspension;
revocation; dismissal
Warnings; letters of reprimand,
suspension; revocation
ance with the rules and conditions in the timber harvest
plan, such as warnings, before resorting to judicial and
misdemeanor remedies.
Continuous Rule Review
As section 208 anticipated, the Board of Forestry continu-
ously reviews its rules and institutional procedures. Con-
sultation with Federal and other State agencies is manda-
tory. This is important because no set of rules and
procedures is perfect. There must be ways to incorporate
improvements, experience, changing technology, and
evolving institutional needs. From the beginning of its in-
volvement with the FWPCA, the Board of Forestry has
believed that providing for such changes is implicit in the
term best management practices.
OUTLOOK FOR THE FUTURE
California is one of America's leading timber States. Yet it
was probably the last major timber State to develop a 208
plan for managing nonpoint sources of pollution from silvi-
cultural operations. It based its program around a previ-
ously developed regulation program rather than around
voluntary or educational programs as in other States. As a
result, the Board of Forestry estimates that its program
increased logging costs by up to $13 per thousand board
feet. When compared to stumpage prices ranging from
about $100 to $150 per thousand on a statewide average,
this is not an inconsequential cost.
The Board has established a system of general water
quality and environmental protection, leaving its field per-
sonnel considerable flexibility at specific sites. This flexi-
400
-------
bility and reliance on a process has made California water
quality personnel uneasy, especially since their predomi-
nant experience is in regulating point sources of pollution.
This difference in philosophies was largely responsible for
the delay in certifying California's silvicultural BMP pro-
gram. In fact, the story is not yet complete as the results of
the 4-year Monitoring and Assessment Program will not
be available until after 1989, at the earliest.
Appropriately this paper ends with the unfinished Cali-
fornia Experience. In the field of water quality regulation,
the final chapter may never be written. However, for stu-
dents of this field, California's experiences will undoubt-
edly give many examples of what to do and what not to do.
Many States are experiencing, or are about to experience,
CASE STUDIES
the urban/forest interface problem similar to California's.
Water quality-forest harvesting concerns will only be the
"point man" for many other heated concerns. Forestry
and water quality regulators will need to work together
very closely to avoid expensive and highly divisive tests of
political will.
REFERENCES
Grefrath, B.C. 1974. The Federal Water Pollution Control Act
and forestry. J. Forestry 72(12): 757-9.
Vaux, H.J. 1982. Forestry's hotseat. The urban/forest interface.
Am. Forests. 88(5): 37: 44-6.
401
-------
F. J. HUMENIK
University of North Carolina
Raleigh, North Carolina
Annual nitrogen, phosphorus and sediment loadings for 2
to 4 years of field sampling in three geographic regions of
North Carolina are presented. Predictive equations devel-
oped to expand the concentration data for use with the
more complete flow record to calculate annual loading
rates are described. Comparisons between the five
Coastal Plain, six Piedmont, and one mountain water-
shed, and with literature values are made. Producer sur-
veys were conducted 4 years for the Coastal Plain study
and 2 years for the Mountain study. Questions were de-
signed to evaluate farmer awareness of technical and
cost-sharing assistance available. Information on atti-
tudes concerning the need and adoption of BMP's to con-
trol nonpoint sources at the beginning and end of the
study period was also collected. Results show the need
for and challenges that exist for producer education and
motivation to implement and maintain appropriate recom-
mendations. Techniques are recommended for evaluat-
ing nonpoint source control program effectiveness or wa-
ter quality changes over time and space based on
experiences with judgmental and statistical grab and in-
strumental sampling, modeling techniques, and plot ver-
sus watershed studies.
yields a surprisingly similar value of about 1 .0 ttVsec/mi*.
Thus good confirmation exists between state-of-the-art
and measured average total flow for very different geocli-
matic regions ranging from the Costal Plains to the Blue
Ridge Mountains in North Carolina.
Studies have been conducted in North Carolina since
1975 to determine techniques for measuring areawide wa-
ter quality and to evaluate the effectiveness of agricultural
nonpoint source control programs. Resulting data add to
state-of-the-art values for flow and concentrations and
thus annual loading rates or yields. Such data sets also
provide a better technical base for estimating areawide
water quality and nonpoint source impacts at a fraction of
the time and cost of extensive field monitoring programs.
Results for measured average total flow and measured
average constituent concentrations are shown in Table 1;
total nitrogen (TN), total phosphorus (TP), and chemical
oxygen demand (COD) yields in Table 2. Data from the
project, Probability Sampling to Measure Pollution From
Rural Land Runoff (Humenik et al. 1980), are noted as
Chowan River I in these tables. Data listed for Chowan
River II, Coastal Plain, and Piedmont were derived from a
series of studies to evaluate strategies for reducing agri-
cultural nonpoint source impacts (Humenik et al. 1983;
Humenik and Foreman, 1984). Data for the Mountain area
were from similar watershed studies conducted by North
Carolina State University (NCSU) workers (Kilmer et al.
1974), a cooperative NCSU-Tennessee Valley Authority
(Tenn. Valley Auth. 1963), and the Dunn Creek Agricultural
Nonpoint Source Control Project (Humenik et al. 1985).
Flow
The literature state-of-the-art value of 0.8-1.2 ft3/sec/mi2
sets the basis for evaluating flows presented in Table 1.
The simple numerical average shown for the measured
average total flow in each region, as well as for all regions,
A general gradient is apparent in the average constituent
concentrations listed in Table 1 ,' with the highest ranges
being for Mountain, intermediate for Piedmont, and lowest
for the Coastal Plain watersheds, which include the Cho-
wan River. Similar ranges are noted for total Kjedahl nitro-
gen (TKN) and nitrite-nitrate (NO2.3). Thus the level of av-
erage TN, TR and COD concentrations seem to be related
to land type, stream gradient, and potential runoff trans-
port. The average nitrate values range from 2.09 for the
Piedmont to 0.74 for the Coastal Plain and 0.63 mg/L for
the Chowan River in spite of the higher infiltration capacity
and ground water inputs in the Coastal Plains. Denitrifica-
tion in the high organic and saturated Coastal Plain and
Chowan River watershed soils can contribute to the lower
nitrate levels recorded for this area.
These data can be compared with values from a na-
tional estimate of nonpoint source-related nitrogen and
phosphorus concentrations presented in the U.S. Environ-
mental Protection Agency report, Nonpoint Source-
Stream Nutrient Level Relationships: A Nationwide Study
(Omernik, 1977); these concentrations are also listed in
Table 1. While some of the nutrient concentrations in
streams listed by the nationwide survey are similar to the
arithmetic average of values listed in Table 1, such as
about 1 .0-1 .5 mg/L total nitrogen in the Coastal Plains
area, other survey versus measured values are much dif-
ferent, such as TN in the Piedmont area, 0.7-1.4 versus
4.02 mg/L; TP in the Coastal Plains area, 0.051-0.7 ver-
sus 0.18-0.21 mg/L; and TP for the Piedmont, 0.31-0.07
versus 0.59 mg/L.
Additionally, the gradient in nitrogen and phosphorus
concentrations measured for North Carolina watersheds
as listed in Table 1 is not as apparent in the concentrations
shown in the topographical type mapping for North Caro-
lina (Omernik, 1977). Nevertheless, such estimates of av-
erage constituent concentrations provide helpful informa-
tion that can be refined by local studies to provide
increasing reliable state-of-the-art values to evaluate am-
bient water quality and the effects of nonpoint sources on
receiving waters.
Annual yields or loadings in kg/ha/yr can be calculated
based upon flow and concentration data. It would be most
desirable to develop good relationships between flow and
concentration for a given sampling station to use with the
more easily obtained flow data from stage recorders or
U.S. Geological Survey type records to calculate constitu-
ent loadings. However, very poor relationships between
concentration and flow have been recorded at watersheds
studied in North Carolina and these are corroborated in
402
-------
Table 1.—Measured average nitrogen,
Watershed Acres
CASE STUDIES
phosphorus, chemical oxygen demand and flow for North Carolina watersheds.
Year TP TKN NO2-3 TN COD Flow
Western N.C.— 1
Western N.C.— 2
Dunn Creek
Parker Branch
Nationwide Study
(Omernik, 1977)
4.67
3.65
614
68-72
68-72
MOUNTAINS
0.08
0.1
0.85
0.03-
0.07
4.66
1.2
3.9
1.73
1.6
4.6
6.53
0.5-
1.4
204
0.57
0.75
1.0
0.70
PIEDMONT
Gourdvine Creek
Lanes Creek
Wicker Branch
Arithmetic Average
Nationwide Study
(Omernik, 1977)
5,621
3,727
3,087
79
80
81
82
79
80
81
82
79
80
81
82
0.72
0.45
0.77
1.6
0.55
0.32
0.71
0.37
0.4
0.27
0.58
0.37
0.59
0.31-
0.10
1.76
1.27
2.57
4.17
2.08
1.49
2.73
1.44
1.4
0.98
1.92
1.22
1.92
2.4
2.89
4.49
5.61
1.48
0.85
1.76
1.09
0.85
0.96
1.92
0.79
2.09
4.17
4.17
7.22
9.62
3.53
2.4
4.49
2.57
2.24
1.92
4.0
1.92
4.02
0.7-
1.4
163
45.7
132
202
57.7
38.6
64.8
41.4
74.4
52.4
112
49.9
132
1.5
0.85
0.29
1.3
1.42
0.99
0.31
1.18
1.3
1.16
0.33
1.1
0.98
COASTAL PLAIN
Beaverdam Creek
Daily's Creek
Bear Creek
Forested Piedmont
Agricultural Piedmont
Well-drained Coastal Plain
Poorly-drained Coastal Plain
Well-drained agricultural
Bells Branch
Cypress Creek
Panther Swamp
Poorly-drained agricultural
Cutawhiskie Lateral
Poorly-drained forested
Big Woods
Arithmetic average
Nationwide Study
(Omernik, 1977)
Arithmetic Average for all flow
data
(4 sites)
(3 sites)
(4 sites)
(4 sites)
498
4,590
6,953
752
799
79
80
81
82
79
80
79
80
81
82
75-76
75-76
75-76
75-76
79-80
81-82
79-80
80-81
81-82
79-80
81-82
81-82
0.03
0.04
0.10
0.19
0.27
0.32
0.24
0.16
0.26
0.16
CHOWAN
0.12
0.10
0.12
0.22
CHOWAN
0.86
0.08
0.06
0.12
0.08
0.16
0.55
0.06
0.2
0.051-
0.07
0.48
0.32
0.48
1.36
0.77
1.25
0.66
0.66
0.82
0.53
RIVER I
1.12
1.0
1.10
1.18
RIVER II
0.82
0.65
0.43
0.64
2.45
0.65
1.82
0.5
0.9
0.48
0.53
0.74
0.55
1.6
2.08
0.42
0.66
0.18
0.14
0.04
0.11
0.75
0.53
2.04
2.8
0.16
0.04
0.18
0.52
0.27
0.08
0.68
0.09-
1.1
0.96
0.85
1.28
1.92
2.4
3.37
1.07
1.31
0.99
0.67
1.16
1.11
1.85
1.71
3.1
3.5
0.61
0.69
0.64
1.19
2.1
0.61
1.5
42.8
34.8
73.9
151
30.6
32.2
44.6
26.6
23.5
17.8
26.1
26.6
1.36
1.14
0.79
1.02
1.15
1.19
1.53
0.94
0.55
1.17
0.93
1.14
1.57
0.31
0.98
0.70
1.02
0.46
0.75
1.76
1.83
0.67
1.04
0.99
403
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 2.—Nitrogen, phosphorus and COD yields for North Carolina watersheds.
Watershed
Western N.C.— 1
Western N.C.— 2
Dunn Creek
Gourdvine Creek
Lanes Creek
Wicker Branch
Beaverdam Creek
Daily's Creek
Bear Creek
Forested Piedmont (4 sites)
Agricultural Piedmont (3 sites)
Well-drained Coastal Plain (4 sites)
Poorly-drained Coastal Plain (4 sites)
Well-drained Coastal Plain
W-3
W-4
W-8
W-10
Poorly-drained Coastal Plain
P-8
P-10
P-11
P-13
Well-drained Agricultural
Bells Branch
Cypress Creek
Panther Swamp
Poorly-drained Agricultural
Cutawhiskie Lateral
Poorly-drained Forested
Big Woods
Acres Time
MOUNTAINS
4.67 1968-1972
3.65 1968-1972
614 5/81-6/83
PIEDMONT
5,621 1979
1980
1981
1982
3,727 1979
1980
1981
1982
3,087 1979
1980
1981
1982
COASTAL PLAIN
2,855 1979
1980
1981
2,568 1979
1980
2,986 1979
1980
1981
1982
CHOWAN RIVER I
6/75-11/76
6/75-11/76
6/75-11/76
6/75-11/76
4,003 6/75-11/76
124 6/75-11/76
2,125 6/75-11/76
4,077 6/75-11/76
2,891 6/75-11/76
2,397 6/75-11/76
3,138 6/75-11/76
24,339 6/75-11/76
CHOWAN RIVER II
498 7/79-6/80
10/81-9/82
4,590 7/79-6/80
10/80-9/81
10/81-9/82
6,953 7/79-6/80
752 10/81-9/82
799 10/81-9/82
N
kg/ha/yr
3.2
12.1
22.4
21.4
12.3
7.0
44.4
17.6
8.0
4.8
10.3
10
7.8
4.5
7.6
4.5
3.4
3.5
9.4
14.0
5.7
4.3
1.9
2.7
3.8
4.3
8.6
1.8
6.4
5.1
3.6
3.0
3.4
2.3
4.1
4.9
10.5
8.4
2.1
1.18
1.7
7.2
13.6
1.3
P
kg/ha/yr
0.15
0.27
6.4
3.7
1.3
0.75
7.5
2.7
1.1
0.74
1.5
1.8
1.1
0.65
1.4
0.12
0.16
0.25
1.1
1.3
1.2
0.50
0.50
0.67
0.39
0.38
0.56
0.22
0.71
0.17
0.34
0.25
0.26
0.31
0.22
0.38
2.9
0.20
0.21
0.19
0.20
0.95
3.57
0.13
COD
kg/ha/yr
703
844
133
130
926
282
131
68
168
334
209
127
189
200
137
202
165
152
164
104
85
108
76
69
122
27
404
-------
CASE STUDIES
the literature. Therefore, predictive equations were devel-
oped to relate concentration to flow so that concentrations
could be generated for use with the more complete flow
record to calculate instantaneous constituent flux and
then annual loading rates.
The methods for computing short- and long-term sedi-
ment yield given in the Field Manual for Research and
Agricultural Hydrology and the American Society of Civil
Engineers Sediment Engineering Manual were found to
provide suitable predictive tools for determining constitu-
ent yields. These procedures used to calculate constituent
flux on an instantaneous basis are based on the straight
line relationship resulting from a log-log plot of instantane-
ous flux versus flow.
The coefficient of determination (R2) can be determined
to indicate the closeness of fit or reliability of the straight
line equation on the log-log plot to predict constituent
concentration from flow data. The standard error of the
estimate for regression coefficients can also be deter-
mined, allowing designation of statistical significance.
Predictive regression equations can be developed for
total flow, stormflow, or baseflow. If the stormflow and
baseflow models are better (higher R2, better fit from ex-
amination of residuals), those equations should be used
for baseflow and stormflow conditions rather than using
an equation for total flow. Water quality samples can be
grouped by season for the total data set or by season for
each year in an effort to obtain equations with better fit or
a higher coefficient of determination (R2). As equations
are developed for more specific flow regimes and time
periods, the R2 factor should increase, thus justifying us-
ing more equations with a higher level of accuracy for
developing relationships between flow and concentration
and predicting instantaneous loads. Computer programs
or other modeling techniques can be used to calculate the
area under the curve formed by plotting instantaneous
loads in kg/hr versus time to determine loads or yields
over time in kg (Humenik et al. 1983,1985).
The TN, TP, and COD loads in kg/ha/yr calculated ac-
cording to this procedure are listed in Table 2. Such load-
ing values are used for water quality planning and evalua-
tion by State government in North Carolina with the
nitrogen loading rate being 6.3 kg/ha/yr for the Coastal
Plain and the. Piedmont, while the total phosphorus load-
ing is 1.1 kg/ha/yr for the Piedmont and 0.55 kg/ha/yr for
the Coastal Plain (N.C. Dep. Nat. Resour. Comm. De-
velop. 1982,1983). Such data as listed in Table 2 serve to
verify or update and refine such areawide loading guide-
lines for nonpoint source control water quality manage-
ment.
Producer Surveys
A producer and field practice survey was taken each win-
ter from 1979 to 1982 to cover the previous crop year, and
a project evaluation survey was conducted for the Cho-
wan River II study. A similar survey was conducted in
conjunction with the Dunn Creek nonpoint source techni-
cal assistance and evaluation watershed study. Selected
portions of these surveys are shown in Tables 3 and 4.
Results of the Coastal Plain watershed survey, which
covered 95 percent of the 12,793 acres in study water-
sheds (Humenik et al. 1983), revealed that the level of soil
testing remained fairly high, between 75 and 80 percent,
during the study period. However, an evaluation of fertil-
ization data shows that in 1981,14 tons of excess nitrogen
and 37 tons of excess phosphorus were applied to 4,600
acres of corn, peanuts, and soybeans. The nitrogen and
phosphorus value of this unnecessary fertilizer was about
$23,000. One suggested reason for this overapplication is
that chemical fertilizer dealers do not offer customized
blends for different fertilization needs.
The survey revealed a rapid growth in broilers, with over
100,000 in the study watersheds, and also that of the sev-
eral hundred swine, some 25 to 100 had stream access at
various periods throughout the year. None of the pro-
ducers surveyed indicated having animal waste tested for
nutrient content, although survey results showed fertilizer
rates were reduced to account for manure application. An
inventory of animal production and waste management in
the five counties bordering the Chowan River in North
Carolina indicated that the broiler litter produced in 1980
had a value of $2.7 million. Costs used were 25
-------
PERSPECTIVES ON NONPO1NT SOURCE POLLUTION
table 3.—Coastal plain watershed producer survey results.
QUESTIONS
1. Which of the following would have the most effect in getting you to use good conservation and management practices?
First number is importance rank: 1—most important; 2—2nd in importance, 3—3rd in importance; other number is
respondents. Not all producers assigned a 1, 2, 3 importance rank to questions resulting in differing number of respondents.
Technical Assistance
Cost-Sharing
Educational Activities:
(such as on-farm tours, economic data on
benefits, informational literature)
Advice of neighbor
Tax credit or incentive
Lease agreement
(for those renting or leasing)
1/21
1/37
1/2
1/0
1/6
1/1
2/13
2/23
2/3
2/1
2/25
2/4
3/14
3/2
3/13
3/4
3/9
3/20
2. Who should be responsible for controlling the agricultural pollution problem (soil erosion, fertilizer loss, animal waste runoff)?
No. of respondents %
Individual Landowner 36 56
Local Government 2 3
Local Soil & Water Conservation District 8 12
State Government 0 0
Federal Government 6 9
Don't know 14 20
3. Who should bear the costs of practices which reduce erosion, fertilizer loss and other forms of agricultural pollution?
No. of Combination
Respondents Responses
Individual Landowner
Local Government
State Government
Federal Government
Don't Know
6
2
2
15
3
Individual Landowner and
Federal Government
Individual Landowner and
State Government
Individual Landowner and
State and Federal Governments
State and Federal Governments
All 3 Governments
31
3
2
1
1
Table 4.— Dunn Creek Mountain Watershed Producer Survey results.
1981
1 . Are you aware of the technical assistance by the Soil and Water Conservation
District?
2. Do you have a farm conservation plan prepared by the District?
3. Would you like to have one developed?
4. Would you be willing to carry out conservation practices developed with the
assistance of the District?
5. Do you think you have conservation problems — erosion, stream siltation,
fertilizer loss, etc.?
6. Do you participate in an Integrated Pest Management Program?
7. Are you aware of the availability of no-till equipment?
8. Are you interested in trying no-till equipment?
9. Are you familiar with ASCS cost-sharing practices?
10. Have you ever participated in ASCS programs?
1 1 . Did you use any conservation practices this year?
Yes
8
0
1
3
0
3
4
2
5
4
No
3
11
10
8
11
8
7
9
6
7
1982
Yes
10
4
1
11
0
9
11
1
11
4
No
2
8
7
1
12
3
1
11
1
8
and pesticides into nutrient sensitive waters by increasing
production efficiency and on-farm management.
Monitoring Programs
The goal of many field monitoring programs is to deter-
mine specific relationships between BMP systems in-
stalled and resulting water qualtiy changes. However, the
work and resources required to accurately account for all
conservation practices on a watershed and procure a total
record of stream flow and associated constituent concen-
trations are generally grossly underestimated. Even if nor-
mal hydrologic conditions existed, most studies do not run
long enough to detect these changes. Even more impor-
tantly, background water quality generally overwhelms
changes that result from BMP implementation. The results
of these and similar studies (Atkins, 1984) emphasize that
BMP performance may be more cost-effectively deter-
mined by monitoring individual practices, fields, or farms,
and not large watersheds. Too many variables and uncon-
trolled factors make it very difficult and expensive, if not
impossible, to establish cause and effect relationships and
document statistically significant water quality changes on
a watershed basis.
Monitoring programs can provide good estimates of
concentrations and yields for a particular site during the
evaluation period. This data can be expanded over space
if a statistically-based sampling program is developed to
406
-------
expand data to the whole sampling universe. The'only
way to expand monitoring data over time is to employ a
conjunctive modeling effort. Actual field monitoring serves
as the best basis for obtaining rate coefficients and other
in-stream data required for modeling. Monitoring prob-
lems in obtaining a complete flow record and sufficient
constituent samplings can be alleviated by developing
predictive equations or models to relate concentration to
flow on an instantaneous basis and then to annual yields.
Coordinated monitoring and modeling programs ranging
from simplistic loading analyses to detailed evaluations
over time and space provide added tools for assessing
BMP effectiveness.
Increased emphasis has to be directed to watershed
characteristics in determining annual yields and evaluat-
ing reasons for high or low loading rates. The state-of-the-
art is developing to the point that general yield estimate
ranges can be made on the basis of watershed character-
istics with primary emphasis on agricultural activities, soil
types, watershed characteristics, and geoclimatic condi-
tions. The systems approach should be taken to evaluate
watershed loading rates and in-stream concentrations by
conjunctive use of watershed characterization data, moni-
toring programs, and modeling techniques, depending
upon the type of data required and the time and resources
available. A very important conjunctive judgment is
whether desired information is truly worth the cost in
terms of time and resources—and whether the required
level of precision can be obtained.
CASE STUDIES
REFERENCES
Atkins, J.B. 1984. Agricultural nonpoint source control case
studies in North Carolina. IV,Wake County demonstra. farm.
Biol. Agric. Eng. Dep., N.C. State Univ., Raleigh.
Humenik, F.J., and J.M. Foreman. 1984. Agricultural nonpoint
source control: case studies in North Carolina. II. Wayne-Le-
noir and Union Counties priority watersheds. Biol. Agric. Eng.
Dep., N.C. State Univ., Raleigh.
Humenik, F.J., B.A. Young, and F.A. Koehler. 1983. Investigation
of strategies for reducing agricultural nonpoint sources in the
Chowan River Basin. UNC-WRRI-83-211. Water Resour. Res.
Inst. Univ. N.C., Raleigh.
1985. Agricultural nonpoint source control project:
Dunn Creek Watershed, Henderson County, North Carolina.
Land-of-Sky Region. Counc. Proj. Biol. Agric. Eng: Dep., N.C.
State Univ., Raleigh.
Humenik, F.J., et al. 1980. Probability sampling to measure pol-
lution from rural land runoff. EPA-600/3-80-035. U.S. Environ.
Protect. Agency, Athens, Ga.
Kilmer, V.J., et al. 1974. Nutrient losses from fertilized grassed
watersheds in western North Carolina. J. Environ. Qual. 3(3):
July-Sept.
North Carolina Department of Natural Resources and Commu-
nity Development. 1982. Chowan River Water Quality Man-
agement Plan. Div. Environ. Manage., Raleigh.
1983. Nutrient management strategy for the Neuse
River Basin. Rep. No. 83-05. Div. Environ. Manage., Raleigh.
Omernik, J.M. 1977. Nonpoint source-stream nutrient level rela-
tionships: a nationwide study. EPA 600/3-77-105. U.S. Envi-
ron. Prot. Agency, Washington, D.C.
Tennessee Valley Authority. 1963. Parker Branch Watershed
Res. Proj. Final Rep. Knoxville, TN.
407
-------
JEFFREY B. MAHOOD
Soil Conservation Service
U.S. Department of Agriculture
Winooski, Vermont
In 1979, the LaPlatte River Watershed became the first
land treatment only PL. 83-566 project. The purpose of
the project was to reduce the runoff of agricultural non-
point source pollutants into Shelburne Bay of Lake
Champlain. Being the first project of its kind, there were
few established guidelines to follow for either planning or
implementation. As the program evolved, its strengths
and weaknesses have become more obvious. These les-
sons resulted in improvements for the ongoing project as
well as for new projects. The LaPlatte River Watershed
program is described. Both the watershed treatment proj-
ect and the water quality monitoring and analysis pro-
gram are discussed. Recommendations for the planning,
implementing, and monitoring of similar programs are
presented.
In the early 1970's, Shelburne Bay of Lake Champlain
began showing signs of culturally accelerated eutrophica-
tion and excessive sedimentation. The Vermont Agency of
Environmental Conservation determined that phosphorus
from three municipal wastewater treatment facilities, as
well as from nonpoint sources, was the primary nutrient
responsible for the eutrophication. The LaPlatte River con-
tributed as much as 63 percent of the nonpoint source
phosphorus and virtually all of the agricultural nonpoint
source pollutants to the bay. To improve the bay's water
quality, a reduction in pollutants from point sources would
have to be accompanied by a similar reduction in pollu-
tants from nonpoint sources (Vt. Agency Environ. Con-
serv. 1977).
Vermont made the Shelburne Bay drainage basin one
of its top priorities for agricultural nonpoint source pollu-
tion control (Vt. Agency Environ. Conserv. 1978). An ac-
celerated program of cropland and streambank erosion
control and proper animal waste management was pre-
scribed for the LaPlatte River Watershed. (Soil Conserv.
Serv. 1978; Vt. Agency Environ. Conserv. 1978).
Such a program was available through the Watershed
Protection and Flood Prevention Act, as amended (PL. 83-
566). Enacted into law in 1954 and administered by the
USDA Soil Conservation Service (SCS), PL. 83-566 tradi-
tionally has been used for flood prevention. Though flood
prevention projects often included watershed protection, a
project had never been authorized solely for watershed
protection. SCS began considering such projects in the
mid 1970's because of national water quality concerns
identified by the U.S. Environmental Protection Agency
(Gallo, 1985).
The LaPlatte River PL. 83-566 Watershed project was
approved for planning in 1978. During planning, the proj-
ect sponsors and SCS recognized the need to evaluate
the effectiveness of this project in improving water quality.
Therefore, they incorporated a Water Quality Monitoring
and Analysis plan. The watershed protection and monitor-
ing project, authorized for implementation in 1979, was
the first of its kind in the Nation.
DT SET™©
The LaPlatte River Watershed is located in Chittenden
County, Vermont, just south of the city of Burlington. This
13,815-ha (34,100-acre) watershed drains westward into
Lake Champlain's Shelburne Bay (Fig. 1).
The eastern headwaters region is in the Green Moun-
tain foothills. This area represents about 20 percent of the
watershed and is dominated by steep slopes, glacial till
soils, and forests. The remainder of the watershed is pri-
marily in the Champlain lowlands, where terrain ranges
from rolling to almost flat; soils are lacustrine sands, silts
and clays, and agriculture predominates (Cassell and
Meals, 1981).
Overall land use is about 50 percent agricultural, 40
percent woodland, and 8 percent residential. About 60
active farms are in the watershed. Dairy farming domi-
nates, with herds averaging 120 head. Silage corn and
hay are the principal crops. Residential land is expanding
in some areas as growth continues in the city of Burlington
and adjacent communities.
TOE
Studies by the Vermont Agency of Environmental Conser-
vation (1977) and the SCS (1978) provided much of the
information needed to develop the watershed protection
plan. These studies identified the water quality problems
of excessive sediment and phosphorus delivery into
Shelburne Bay. Sources of these pollutants were deter-
mined to be excessive cropland erosion and the resulting
sediment deposition, along with insufficient control of ani-
mal wastes.
Figure 1.—Project location map. LaPlatte River Watershed.
408
-------
CASE STUDIES
to improve water quality in the LaPlatte River and
Shelburne Bay, three goals were established:
1. Reduce the average annual rate of cropland erosion
from 26 metric tons/ha (11.6 tons/acre) to 11 metric tons/
ha (5 tons/acre).
2. Reduce annual sediment flow into Shelburne Bay
from 7,830 metric tons (8,630 tons) to 4,280 metric tons
(4,710 tons).
3. Reduce the amount of nutrients from manure reach-
ing the LaPlatte River by increasing winter storage of ma-
nure from 7,300 metric tons (8,000 tons) to 32,000 metric
tons (35,000 tons).
To meet these goals, SCS recommended developing
conservation contracts for 41 farms. These contracts
would call for conservation treatment of 1,070 ha (2,650
acres) of cropland, grassland, and forestland; installation
of 30 animal waste storage facilities; and protection for
760 m (2,500 ft) of critically eroding streambank.
The plan described specific best management practices
(BMP's) to be used, that is, those conservation practices
considered most effective and practical for the desired
pollution control. In addition, allowable rates of Federal
cost-sharing were established for each BMP. Farmer par-
ticipation in the program was strictly voluntary.
WATER QUALITY MONITORING
AND ANALYSIS PLAN
The Water Quality Monitoring and Analysis plan was de-
veloped by the Vermont Water Resources Research Cen-
ter, University of Vermont. This plan would evaluate the
overall effectiveness of BMP's in improving surface water
quality The objectives of this program were (Soil Conserv.
Serv. 1979):
1. To evaluate and document the impact of the water-
shed protection project on the export of sediment and
nutrients from the LaPlatte River to Shelburne Bay;
2. To evaluate and document the reduction in the runoff
of sediment, phosphorus, and animal wastes as a result of
implementing particular BMP's; and
3. To provide a detailed analysis of the project useful to
the long-term program needs of SCS and the Vermont
Agency of Environmental Conservation.
To accomplish these objectives, river flow and water
quality would be continuously monitored at five automated
stations over 11 years; a series of other studies were de-
signed to determine relationships between surface water
quality and specific BMP's (Soil Conserv. Serv. 1979). The
water quality monitoring plan is described in greater detail
in another paper presented at this conference (Meals,
1985).
The plan identified monitoring site locations, water qual-
ity parameters to be measured, a schedule of operations,
and responsibilities of cooperating agencies. This plan
also established a Project Advisory Council consisting of
representatives of all cooperating agencies. The Council
meets quarterly to discuss programs and problems.
IMPLEMENTATION PROGRESS
Development of conservation contract agreements and
BMP implementation began in 1980. As can be seen in
Figure 2, all of the contract agreements and most of the
BMP implementation occurred in the first 4 years. At
present, 27 farms have signed contracts covering over
2,850 ha (7,000 acres). These farms control an estimated
70 percent of the animal wastes produced in the water-
shed. Most of these farms have installed various compo-
nents of waste management systems and conservation
cropping systems. Streambanks were protected on some
farms. No new contracts are anticipated (Meals, 1984).
REPORTING YEfiR
Figure 2.—Workload distribution by year. LaPlatte River Wa-
tershed project.
Analysis is underway to estimate the expected changes in
average annual export of phosphorus and sediment.
Monitoring also began in 1980, although limited water
quality data were collected during 1979. Except for
amendments in study years 3 and 5, the monitoring and
analysis program is proceeding as planned. In year 3, land
use monitoring was intensified to produce the data
needed to relate water quality changes to land use and
land management changes (Meals and Cassell, 1982). In
year 5, a special study was added to obtain more detailed
information on nutrient and sediment export from barn-
yards, milkhouses, and manured haylands, with and with-
out conservation treatment (Meals, 1984). All data col-
lected since 1979 are being analyzed to detect water
quality trends and relationships.
DISCUSSION
Watershed Protection Program. Osteen et al. (1981) rec-
ommended water-based land management for water-
sheds having severe water quality problems or where criti-
cal resources must be protected. Water-based land
management often requires either monitoring or modeling
to identify water quality problems and the sources and
movement of the pollutants. The increased planning time
results in better allocation of resources to improve the
water quality.
The LaPlatte River Watershed plan was developed us-
ing water-based land management principles. The tools
used to relate water quality changes to land management
changes, however, were crude. The expected water qual-
ity changes were not well documented, especially for ani-
mal waste BMP's.
Experience, early monitoring results, and new research
have refined these tools. Subsequent projects are using
these new techniques to determine treatment alternatives
and to choose farms for treatment. These techniques are
described in another paper at this session (Keeler, 1985).
In the LaPlatte project, farms were earmarked for treat-
ment as applications were received. The selection meth-
ods were somewhat subjective, and timing was a problem.
Sometimes, because of a lag in application submittal,
some medium-priority farms were serviced before higher-
priority ones.
Pollutant control generally becomes more efficient and
cost effective if the most severe sources are treated first. A
thoroughly prepared water-based land management plan
allows land management units to be ranked according to
severity before project startup. Project funding can be
contingent upon receiving enough high- to medium-prior-
409
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 1.—Contract and Implementation summary, LaPlatte River Watershed project.
CONTRACTED IMPLEMENTED
Waste Mgmt. Systems
Totals to date
Projected
No. of
Farms
27
27
Hectares
2,851
2,851
Manure
Storage
26
26
Barn-
yards
9
14
Milk-
houses
17
19
Conservation
Cropping
(ha)
943
1,000
Streambank
Protection
(m)
350
715
ity applications to ensure the desired water quality effects.
Regardless of the planning approach, an accelerated
long-term program of this type will require more personnel
than annual programs. This fact was evident in the La-
Platte project as technical assistance needed from SCS
increased during project implementation (Fig. 3). Further-
more, contract preparation demands extra time. As legal
documents, contracts require attention to detail and multi-
ple meetings with the landowner.
Technical personnel in such a project often have a wide
range of knowledge and experience. Additional preproject
training on water quality, the estimated influence of BMP's
on pollutant transport, the preparation and legal aspects
of contracts, and the design and installation of complex
practices will result in a more efficient and effective pro-
gram.
During project implementation, unanticipated situations
and conditions are likely to arise. Continual assessment of
project progress and problems is essential to avoid the
•recurrence of similar problems and to redirect activities if
necessary (Natl. Water Qual. Eval. Proj. 1983). When the
LaPlatte project began, administrative procedures were
new to everyone. Problems with certain practice designs
developed. However, administrative procedures are being
continually streamlined. Designs have been improved. As
a result, this project, as well as new projects, is proceed-
ing with fewer problems.
Finally, it is very important to keep the public informed
of progress (Natl. Water Qual. Eval. Proj. 1983). A project
of this type arouses much public interest. Releasing pro-
gress reports to the public will help ensure public aware-
ness, increase program participation, and aid in the ad-
ministration of future projects.
Monitoring and Analysis Program. The Water Quality
Monitoring and Analysis Program was fully operational a
few months after the watershed protection program be-
gan. Some preproject water quality data were available.
from previous studies. These data, however, were col-
lected for only 1 year, and station locations were not com-
patible with the current monitoring locations. Preproject-
postprpject comparisons are not possible. Data analysis
and evaluation depend primarily on the observation of
year-to-year trends in water quality. Climatic variability
makes this a complex task, requiring long-term study. De-
Demand for Technical Assistance
60
40
1330
1381
198E
1983
1384
1985
1386
1330
REPORTING YEftR
Figure 3.—Percent Increase In demands for technical assistance resulting from the LaPlatte River Watershed project.
410
-------
CASE STUDIES
tailed preproject data would be desirable in future projects
to more readily determine water quality changes.
A factor that has made trend analysis difficult in the
LaPlatte project is the discharge from a municipal waste-
water treatment facility. This point source discharge has
complicated the interpretation of water quality trends at
the main monitoring station, obscuring the nonpoint
source water quality changes related to BMP implementa-
tion. Future projects designed to monitor these changes
would benefit if monitoring were located in areas unaf-
fected by point source discharges.
Another factor complicating the interpretation of water
quality trends was the lack of detailed land use informa-
tion. Though BMP implementation was carefully tracked,
details on the time, location, and magnitude of farm activi-
ties were needed as well. Farmers were asked to record
activities such as manure spreading, fertilizer application,
plowing, and planting on a field-by-field basis. Collected
twice each year, these data are being analyzed through a
geographic information system (GIS), a specialized com-
puter-based mapping overlay system for the analysis and
display of spatial data. The use of detailed land use data in
GIS is greatly improving the ability to correlate land use
activities and water quality.
The large volume of information generated by the moni-
toring and analysis effort should be developed and put in
practice in a manner consistent with the users' needs. It is
essential that each cooperating agency commit at least
one individual to the program. This person should be able
to devote the time necessary to maintain good communi-
cations and actively participate in Project Advisory Coun-
cil meetings. The Council has helped ensure that data
collection and evaluation are proceeding desirably by pro-
viding a mechanism for program review, revision, and if
necessary, redirection.
Finally, the monitoring program is intended to provide
useful information both locally and nationally. The SCS
national office has organized a task force to evaluate the
project and to use the results to improve other such proj-
ects nationwide. Therefore, the wide dissemination of
these results is assured.
SUMMARY AND CONCLUSIONS
The LaPlatte River Watershed was identified as the pri-
mary agricultural nonpoint source of pollution reaching
Shelburne Bay, Lake Champlain. In 1979, the LaPlatte
River Watershed project became the first land treatment
only P.L. 83-566 project in the Nation. Both the ongoing
program and future programs can benefit from this proj-
ect.
Watershed Protection Program:
1. Develop detailed water-based land management
plans using the best available data, models, and tech-
niques.
2. Prioritize farms before initiating project.
3. Provide enough experienced personnel to conduct
the program effectively.
4. Provide preproject training on water quality, pollutant
transport, contract administration, and design and instal-
lation of complex practices.
5. Continually assess project progress and problems,
and revise and redirect the program if necessary.
6. Keep the public informed.
Monitoring and Analysis Program:
1. Collect sample preproject data.
2. Consider locating new monitoring projects in areas
unaffected by point source pollution.
3. Collect detailed land use data from the beginning of
the project.
4. Organize a project advisory council and meet regu-
larly
5. Organize a task force to help disseminate monitoring
results to advance new projects.
The LaPlatte River Watershed project is only 5 years
old. Much has been learned, and surely much remains to
be learned. However, the success of implementing this
project and the opportunity to improve new projects is
apparent. Watershed projects (P.L. 83-566) that involve
only land treatment appear to be acceptable to decision-
makers and the public as an effective approach to imple-
menting agricultural nonpoint source pollution control
practices.
REFERENCES
Cassell, E.A., and D.W. Meals. 1981. LaPlatte River Watershed
Water Quality Monitoring and Analysis Program. Rep. No. 1. A
description of the Watershed and Water Quality Monitoring
and Analysis Program. VT Water Resour. Res. Center. Univ.
Vermont, Burlington.
Gallo, R.A. 1985. Personal commun. Soil Conserv. Serv. U.S.
Dep. Agric. NC State Off., Raleigh.
Keeler, P.M. 1985. Agricultural land treatment project planning
for off-site phosphorus reduction. Soil Conserv. Serv. U.S.
Dep. Agric. VT State Off., Winooski.
Meals, D.W. 1984. LaPlatte River Watershed Water Quality Mon-
itoring and Analysis Program. Rep. No. 6. Proj. Year 5. VT
Water Resour. Res. Center. Univ. Vermont, Burlington.
1985. Monitoring changes in agricultural runoff qual-
ity in the LaPlatte River Watershed, Vermont. VT Water Re-
sour. Res. Center. Univ. Vermont, Burlington.
Meals, D.W, and E.A. Cassell. 1982. LaPlatte River Watershed
Water Quality Monitoring and Analysis Program. Rep. No. 4.
Program Achievement Rep. Year 3. VT Water Resour. Res.
Center. Univ. Vermont, Burlington.
National Water Quality Evaluation Project. 1983. The Model Im-
plementation Program, lessons learned from agricultural wa-
ter quality projects. North Carolina State Univ. Raleigh.
Harbridge House, Inc., Washington, DC.
Osteen, C., W.D. Seitz, and J.B. Stall. 1981. Managing land to
meet water quality goals. J. Soil Water Conserv. May-June:
138-41.
Soil Conservation Service. 1978. Erosion and Sedimentation,
Nonpoint Pollution Sources and Controls, LaPlatte River Wa-
tershed. LCBS-19. U.S. Dep. Agric. Natl. Tech. Inform. Serv.
Springfield, VA.
1979. Watershed Plan for LaPlatte River Watershed,
Vermont, U.S. Dep. Agric. Burlington, VT.
Vermont Agency of Environmental Conservation. 1977. Nutrient
loading to Shelburne Bay and St. Albans Bay, Lake Cham-
plain, Vermont 1975-76. Montpelier.
1978. A State Water Quality Plan for Controlling Agri-
cultural Pollution. Montpelier.
411
-------
Making Decisions About
Nonpoint Source Pollution
POINT/NONPOINT SOURCE TRADING PROGRAM FOR DILLON
RESERVOIR AND PLANNED EXTENSIONS FOR OTHER AREAS
TOM ELMORE
Northwest Colorado Council of Governments
Frisco, Colorado
JOHN JAKSCH
U.S. Environmental Protection Agency
Washington, D.C.
DONNA DOWNING
The Skylonda Group, Inc.
Menlo Park, California
INTRODUCTION
The Clean Water Act was designed to protect the quality
of the Nation's waters. Much progress has been made in
water pollution control since its passage in 1972. Although
the population and economic activity have grown mark-
edly, water quality has improved throughout the United
States (U.S. Environ. Prot. Agency, 1984).
What approaches might be used for remaining prob-
lems? Sophisticated controls have already been placed on
most point sources. Further improving water quality will
require either additional, incrementally more expensive
point source controls, or an expansion of focus to include
nonpoint sources. In many cases, point sources may al-
ready be controlled to the degree that it is more cost effec-
tive to control pollutants from nonpoint sources.
Nonpoint sources play a major role in our remaining
water quality problems. Virtually all States have some wa-
ter quality problems because of nonpoint sources, and
half those States say nonpoint sources are the major or
significant cause of degraded water quality (Elmore et al.
1984). Generally speaking, nonpoint sources are the larg-
est remaining uncontrolled pollution problem. Controlling
nonpoint source pollution from forestry, agriculture, min-
ing, and urban runoff has been studied extensively. Now
we need to implement that knowledge.
The basic approach taken by the Clean Water Act for
managing point sources—application of uniform techno-
The views and opinions expressed in this paper are those of the authors and not to be
taken as the official policy of the U.S. Environmental Protection Agency or the North-
west Colorado Council of Governments.
logical control to classes of dischargers—is not appropri-
ate for managing nonpoint sources. In fact, the Act offers
little incentive to manage nonpoint source pollution. The
diversity of the nonpoint source problem makes national
guidelines difficult. As EPA's recent report to Congress on
nonpoint sources points out, flexible, site- and source-
specific decisionmaking is the key to effectively controlling
nonpoint sources (U.S. Environ. Prot. Agency, 1984b).
This poses a challenging opportunity as State and local
governments begin to address nonpoint source problems.
One innovative approach to encourage nonpoint source
control is to tie nonpoint sources into the existing National
Pollution Discharge Permit System (NPDES). This is a
feature of a new program in Colorado that will clean up
runoff and save local governments millions of dollars an-
nually. The Dillon Reservoir Trading Project in Summit
County grants wastewater treatment plants credit for
cleaning up runoff (nonpoint source) pollution. This pollu-
tion trading program is the first of its type in the Nation.
This paper will describe the Dillon Reservoir Trading
Program in detail, discuss ongoing analyses related to the
Dillon concept, and evaluate Dillon's implications for other
regions of the country
DILLON RESERVOIR'S POINT/NONPOINT
TRADING PROJECT
Problems of Nutrient Overload
Dillon is a 20-year old reservoir located in Summit County,
Colorado. Constructed as Denver's primary West Slope
413
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
water supply reservoir, Dillon quickly became a recreation
center for fishing, camping, and boating. One of the reser-
voir's main attractions is its reputation for clear, deep blue
water.
Summit County grew to be a popular, year-round recre-
ation-oriented community upon the completion of several
major ski areas. The area is one of the fastest growing
counties in the Nation, with a permanent population of
10,000 and peaks exceeding 60,000 during winter
(Northw. Colo. Counc. Gov. 1984a). This popularity may
be Dillon's downfall. Its deep blue color changes to green
as algae bloom in the summer, fed by phosphorus enter-
ing the lake from natural and manmade activities. If algae
growth continues the lake will become eutrophic. Dillon
would lack the oxygen necessary to support a high quality
water recreation experience. A clean Dillon Reservoir is
very important to Summit County's recreational economy
and key to Denver's water supply.
Approximately half the phosphorus entering Dillon is
from background runoff and direct precipitation to the
Reservoir. These natural sources exist in the absence of
any human population. Difficult or impossible to control
with existing technology, natural sources would in them-
selves not cause the lake to become eutrophic (West. En-
viron. Anal. 1983).
The other half of Dillon's phosphorus load is from hu-
man activities. Significant sources include: point source
discharges from municipal wastewater treatment facilities,
runoff from parking lots, erosion from construction sites,
seepage from septic systems, and other nonpoint sources
(West. Environ. Anal. 1983). This half of Dillon's phos-
phorus load is controllable.
The four municipal waste dischargers to the reservoir
have installed advanced treatment equipment to control
for phosphorus (less than 0.2 mg/L discharge). The Dillon
1983 Clean Lakes Study found, however, that even if
these dischargers were kept to zero by complex and ex-
pensive treatment methods, nonpoint sources of phos-
phorus from manmade activities would cause eutrophica-
tion (West. Environ. Anal. 1983). Control of manmade
nonpoint sources was necessary to inhibit Dillon's eu-
trophication. The Colorado Water Quality Control Com-
mission asked local agencies to develop plans for ad-
dressing the phosphorus problem in the Dillon Reservoir
basin (Northw. Colo. Counc. Gov. 1984a). A moratorium
on sewer taps—effectively a growth freeze—appeared im-
minent.
The Northwest Colorado Council of Governments is the
water quality planning agency for Summit County. The
Council had assembled a group representative of the pub-
lic and private sector to develop a phosphorus control
strategy. Calling itself the "Phosphorus Club," the commit-
tee consisted of officials from the County, the six incorpo-
rated municipalities, two unincorporated urban areas,
three ski areas, four municipal dischargers, the Denver
Water Department, U.S. EPA, the Colorado Department of
Health, and Amax, a major molybdenum mine (Northw.
Colo. Counc. Gov. 1984b).
The Phosphorus Club members were wary of "hidden
agendas" and of a strategy development process that
could decrease any source's discharge allowance. Lower
discharge allowances, for example, could limit a munici-
pality's growth potential. Initial competitiveness was less-
ened by the decision that all committee conclusions had to
be unanimous. While realizing that difficult compromises
would probably be necessary, Phosphorus Club members
also realized they had common goals: preservation of Dil-
lon Reservoir and avoiding a tap moratorium. Early ten-
sions dissipated as members saw they must cooperate to
protect Dillon (Northw. Colo. Coun. Gov. 1984a). The com-
mittee realized a successful strategy would have to en-
courage the control of nonpoint sources.
The committee met weekly for 6 months to discuss
sources of phosphorus in the basin, control options, ad-
ministration, monitoring, and costs of different strategies.
They soon focused on the point/nonpoint trading concept
(Northw. Colo. Coun. Gov. 1984a). Under the strategy,
point source dischargers receive credit for cleaning up
existing nonpoint sources of phosphorus. This credit can
be traded for increased phosphorus discharge from the
wastewater treatment plant. Local governments must re-
quire state-of-the-art controls on growth areas. This ac-
commodates growth with no overall increase in Dillon's
phosphorus loadings. Equally important, the strategy af-
fords a built in incentive to clean up the nonpoint sources
and maintain Dillon's current, high level of water quality.
The first issue raised in developing the trading strategy
concerned the target level of phosphorus loadings. The
group decided that phosphorus in Dillon should be main-
tained at 1982 levels. The goal was viewed both as realis-
tic and as maintaining the water quality level (Northw.
Colo. Coun. Gov. 1984a).
A second issue was how to allocate the phosphorus
load among municipal wastewater treatment plants. After
much discussion, each plant was given a share of the
available load based on its total flow for 1983. This pro-
vided an equivalent growth margin for each community
through 1990 (Northw. Colo. Coun. Gov. 1984c).
The Colorado Department of Health was concerned
about future nonpoint sources. If a Dillon nonpoint source
control strategy allowed continued population growth, that
growth in itself would create more nonpoint sources. The
Phosphorus Club agreed that a successful point/nonpoint
source trading program would require that old nonpoint
sources be cleaned up and new ones minimized. To ac-
count for uncertainties in the system, a 2:1 tradeoff ratio
was established. For each pound of credit assigned to a
point source, 2 pounds of phosphorus must be removed
from a nonpoint source that existed prior to 1984 (Colo.
Water Qual. Control Comm. 1984).
Another issue was the long-term management of a
point/nonpoint source control strategy. Local agencies
were reluctant to create a new agency that might be costly
to operate and possibly reduce local control of land use
decisions. The Colorado Department of Health wanted a
stable oversight agency for the trading program. The ap-
proach, accepted by the State, was a committee estab-
lished by intergovernmental agreement among local
agencies to manage the trading program on a day-to-day
basis (Colo. Water Qual. Control Comm. 1984). The State
would oversee activities and document all trades in a
NPDES permit.
The Phosphorus Club believed that a demonstration proj-
ect was necessary to test nonpoint source control aspects
of the proposed trading system. A demonstration project
provided data on both cost and effectiveness of technol-
ogy for nonpoint source control. Low technology methods
can remove phosphorus from runoff. Settling ponds can
remove roughly half the phosphorus in urban runoff, while
rapid sand filters can remove another quarter. Percolating
runoff through unsaturated soil can remove virtually all the
phosphorus. Settling ponds and percolating pits are much
less expensive to construct and operate than wastewater
treatment plants. They also demand much less energy
and generate little sludge, in contrast to advanced treat-
ment. The group believed that such controls might be a
414
-------
MAKING DECISIONS ABOUT NONPOINT SOURCE POLLUTION
viable, low-cost alternative to new high technology con-
trols on point sources. In the spring of 1982, the Council of
Governments applied for U.S. EPA funds for a pilot facility.
EPA became interested in Dillon's point/nonpoint trading
strategy, and agreed to provide partial analytical support
and funding for the demonstration project.
Low technology nonpoint source controls were tested at
Frisco, Colorado. Urban runoff from a 32-ha (81-acre) wa-
tershed was collected in a stilling basin that overflowed
into a settling pond (called a hole-in-ground treatment sys-
tem, or "HIG"). The sand and rock in the HIG were sepa-
rated by a geotextile erosion control fabric that lined the
device entirely. In eight runoff events, settling removed 45
percent of incoming phosphorus. Filtration raised the total
removal to 68 percent. The Council believes using a soil
treatment matrix would raise removal efficiency to virtually
100 percent (Northw. Colo. Coun. Gov. 1984a).
EPA and the Phosphorus Club were very interested in
the relative costs of point versus the HIG nonpoint source
control device. The pilot system cost $4,200 to build, with
iand and much of the labor donated by local groups. When
land and excavation costs are added, a more practical
cost estimate would be $50,000. This initial cost could be
lower if public land is used for the control device. The 2-
year pilot project indicated the pond will need to be
drained and cleaned at least every 5 years. The sand and
filter fabric will need to be replaced annually. The ex-
pected annual labor and materials cost is $980 (Northw.
Colo. Coun. Gov. 1984a).
EPA commissioned an economic study comparing the
cost of low technology nonpoint source controls with the
advanced treatment alternative (higher levels of control)
for municipal wastewater plants. The study found substan-
tial cost savings with HIGs, up to 88 percent cheaper than
upgrading existing point source controls (IEC, 1984).
The Dillon Management Plan
The Phosphorus Club now had (1) a cost-effective and
environmentally sound nonpoint source control technol-
ogy, and (2) the outline of a point/nonpoint trading system
to encourage its use and that of other nonpoint source
controls. Realizing that Dillon's water quality relies on the
long-term management of phosphorus, the group de-
signed a detailed control strategy. The key element in the
trading strategy is immediate control of future nonpoint
sources in growth areas. As future nonpoint source load-
ings are minimized, older nonpoint sources will be con-
trolled, allowing for point source growth in the future. In
this manner, nonpoint controls will be traded for point
sources growth (Table 2). As mentioned, these reductions
in nonpoint sources can be achieved relatively inexpen-
sively. The control strategy has six major elements:
1. 1982 levels of phosphorus in the reservoir are the
baseline for water quality.
2. Point sources will continue to receive advance
Table 1 .—Point and nonpoint controls for phosphorus
removal.
Annual
Cost Annual
Typical Phosphorus Cost
Plant Removal Per Pound
Efficiency Phosphorus
$ % $
wastewater treatment and will meet individual annual
phosphorus totals.
3. State-of-the-art nonpoint source controls will be re-
quired by local governments for all new developments. No
credits will be granted for these controls. New develop-
ments must also contribute to an NPS Control Facilities
Investment Fund. The fund will be used by the Summit
Water Quality Committee, established by the Phosphorus
Club, to construct controls for pre-1984 nonpoint sources
of phosphorus and oversee the entire Dillon Nonpoint
Management Plan.
4. Point source growth beyond 1990 will be accommo-
dated by reducing nonpoint sources that existed prior to
1984. The management plan encourages cleaning up old
sources and minimizing new sources.
5. The Summit Water Quality Committee will monitor
the success of the control program and manage phos-
phorus trading between point and nonpoint sources. The
trading ratio will be 2:1. This committee provides for long-
term management of the control program, while continu-
ing strong local input in decisionmaking.
6. The Water Quality Control Commission has autho-
rized effluent trading and will use their NPDES program
for enforcement where necessary. The Water Quality Con-
trol Division, the operating arm of the Commission, docu-
ments a trade in a revised (NPDES) permit after the non-
point source control device's effectiveness is determined
and the 2:1 ratio has been applied. A specific discharger is
given the phosphorus credit and responsibility for main-
taining the nonpoint source control device. Failure to oper-
ate and maintain these controls results in enforcement
action for an NPDES violation (Northw. Colo. Coun. Gov.
1984b).
The Dillon Water Quality Management Plan relies on a
Federal/State/local regulatory partnership. Local land use
authorities hold future runoff sources to an acceptable
level by requiring erosion and runoff controls and stream-
side setbacks. The State water quality agency uses the
NPDES to ensure that the lake receives long-term protec-
tion. The NPDES permit revisions assign phosphorus
"credit" and nonpoint operation and maintenance respon-
sibilities. The Summit Water Quality Committee, formed
by intergovernmental agreement, runs the trading pro-
gram on a daily basis.
Public hearings on Dillon's proposed trading plan were
held by the State of Colorado in May 1984. The State
formally approved the plan in June; EPA Region VIII ap-
proved it in July. With that approval, Dillon Reservoir be-
came the first approved point/nonpoint source trading sys-
tem in the United States (Colo. Water Qua). Control
Comm. 1984).
Table 2.—The Dillon management strategy.
Land treatment
Activated alumina
Reverse osmosis
Pilot project "HIG"
451,000
353,000
3,871,000
8,708
100
75
90
68
824
' 860
7,861
67
415
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
NATIONAL APPLICATIONS OF DILLON
Dillon Reservoir showed that point/nonpoint source trad-
ing can be a cost-effective means of pollution abatement
for lakes and impoundments. Trading provides an eco-
nomic incentive for local communities to control new and
existing sources of nonpoint pollution, where no controls
formally exist under the Clean Water Act. As in the case of
Dillon, trading can achieve water quality standards when
even zero point source discharge cannot.
Several circumstances at Dillon Reservoir helped make
the demonstration project successful.
1. All parties had common goals: preserving Dillon
Reservoir and avoiding a tap moratorium. If degradation
continued, economic growth would slow because of fewer
recreational opportunities. Drinking water quality would
drop. An unambiguous goal—1982 levels of phos-
phorus—was readily decided upon.
2. All point sources were already at advanced treat-
ment levels. No inexpensive conventional control ap-
proaches were available. Studies indicated that additional
equipment leading to zero point source discharge would
slow but not halt eutrophication. Nonpoint sources had to
be controlled to maintain water quality.
3. Stakeholders had continuing input into the design
and operation of the demonstration project. The State,
county, municipalities, and local industries were all repre-
sented in the Phosphorus Club.
4. Water quality data and projections were available to
highlight implications of various control strategies. Types
and magnitude of phosphorus sources were well under-
stood throughout the demonstration project's design and
implementation. The 1983 Clean Lakes Study found that
both point and nonpoint sources must be included in a
strategy to protect Dillon Reservoir.
5. The Phosphorus Club designed a trading system
which outlined clear liability for maintaining controls. Am-
biQuous responsibilities could have had a disastrous effect
on Dillon's water quality.
6. The system relies on good communication and coop-
eration between Federal, State and local agencies. Differ-
ent aspects of the trading system are overseen by the
agency with the most appropriate regulatory authority. Be-
cause each level of government was involved in the devel-
opment stage, no unexpected objections arose during im-
plementation and formal adoption.
Dillon has fostered thinking on the wide application of
trading, given the phasing out of the Construction Grants
Program and the increasing recognition of nonpoint
source pollution. Control of pollution discharges to ad-
vanced treatment levels will be required on many stream
segments to achieve water quality goals. However, ad-
vanced treatment requirements can be achieved in a num-
ber of different ways.
An enforceable trading arrangement could result in a
cheaper and more stringent control system than using
advanced treatment technology on a point source dis-
charger. First, through trading ratios, nonpoint source
BMP's can remove more pollution from the environment
than the additional amount a point source might dis-
charge. Second, the enforcement agency could require
that credit not be given for planned or existing BMP's—
resulting in more cost-effective control than by traditional
means. Third, the municipality would be required to lease
or buy the land on which the BMP is located, and to con-
struct, operate, and maintain that BMP at no cost to the
landowner.
Can this trading approach be applied to other locations,
or is Dillon unique? The quality of virtually all lakes is
controlled by a delicate balance of nutrients such as phos-
phorus. Many coastal rivers and bays are also affected by
phosphorus pollution. Trading may present a cost-effec-
tive pollution control option for all such waterbodies and is
currently being considered by several localities for lakes
and reservoirs with problems similar to Dillon's.
While trading shows tremendous promise, Dillon left
several questions unanswered. For example, EPA is ex-
ploring whether or not trading will work on other waterbo-
dies—free flowing streams or estuaries—or for other types
of nonpoint sources such as agriculture. Can the regula-
tory, enforcement, and institutional framework developed
at Dillon be adapted to other locations? Dillon is an epit-
ome of Federal/State/local cooperation. When goals are
less well defined, can local groups reach consensus on
desired means and ends? Will other types of nonpoint
source BMP's prove to be as cost effective as those at
Dillon?
The Office of Policy, Planning, and Evaluation (OPPE) at
EPA Headquarters, in cooperation with EPA's Office of
Water, Region III, the Chesapeake Bay Program, and the
States of Maryland, Virginia, and Pennsylvania, is examin-
ing the application of a "Dillon type" point/nonpoint ap-
proach to the Chesapeake Bay. The Bay has a major non-
point source problem resulting from agricultural activity in
its stream drainages. Analysis should be completed by
late summer 1985.
OPPE is also providing technical assistance to several
State/local governments interested in using trading to
solve current or future water quality problems caused by
both point and nonpoint sources. OPPE is exploring, par-
ticularly with respect to agricultural nonpoint sources, the
legal and institutional implications of a point source con-
structing, operating and maintaining BMP's at no cost to
the contributing source to take credit for pollutants con-
trolled in the NPDES permit. This approach would be use-
ful, for example, to a POTW with secondary treatment
facing an upgrade to higher levels of treatment where
nonpoint source is a major, controllable contributor to wa-
ter quality problems.
In conclusion, point/nonpoint source trading appears to
offer a cost-effective approach to controlling nonpoint
sources. Many questions remain after the successful im-
plementation of point/nonpoint trading at Dillon Reservoir.
However, trading and other innovative approaches are re-
ceiving increasing attention from States who view non-
point source as a significant yet uncontrolled problem.
REFERENCES
Colorado Water Quality Control Commission. 1984. Written
summation: In the manner of control regulations and numeric
standard for the Dillon Reservoir watershed.
Elmore, T. et al. 1984. Trading between point and nonpoint
sources: a cost-effective method for improving water quality—
the case of Dillon Reservoir. Presented at 57th Ann. Conf.
Water Pollut. Control Fed., New Orleans.
IEC, Inc. 1984. Case Studies on the Trading of Effluent Loads:
Dillon Reservoir. Prep. Reg. Reform Staff, U.S. Environ. Prot.
Agency. Washington, DC.
Northwest Colorado Council of Governments. 1984a. Point
Sources-Nonpoint Sources Trading in the Lake Dillon Water-
shed. Prep. Reg. Reform Staff, U.S. Environ. Prot. Agency
and Colo. Dep. Health. Frisco.
. 1984b. Summit County Phosphorus Control Commit-
tee Recommended Water Quality Management Plan for the
Colorado Water Quality Control Commission. Frisco.
U.S. Environmental Protection Agency. 1984a. Environmental
Progress and Challenges: An EPA Perspective. Washington,
DC.
. 1984b. Report to Congress: Nonpoint Source Pollu-
tion in the United States. Off. Water Progr, Water Plann. Div.,
Washington, DC.
Western Environmental Analysts, Inc. 1983. Dillon Clean Lakes
Study Final Report. Prep, for Northw. Colo. Counc. Gov.
Frisco.
416
-------
OPTIMIZING POINT/NONPOINT SOURCE TRADEOFF IN THE
HOLSTON RIVER NEAR KINGSPORT, TENNESSEE
MAHESH K. PODAR
JOHN A. JAKSCH
STUART L. SESSIONS
U.S. Environmental Protection Agency
Washington, D.C.
JOHN C. GROSSMAN
RICHARD J. RUANE
GARY HAUSER
Tennessee Valley Authority
Chattanooga, Tennessee
DAVID E. BURMASTER
Industrial Economics, Inc.
Cambridge, Massachusetts
INTRODUCTION
The Office of Policy, Planning, and Evaluation of the U.S.
Environmental Protection Agency is studying a variety of
innovative approaches for controlling water pollution from
point and nonpoint sources. Among them is the trading of
effluent loads of water pollutants. This approach differs
from current EPA water policy in that the State or regional
authority responsible for issuing discharge permits for pol-
lutants under the Clean Water Act may modify those limits
if two or more dischargers propose a reallocation, or a
trade. The reallocation allows dischargers with lower treat-
ment costs to control more pollution and those with higher
costs to control less. Thus, the dischargers comply with
the same total load limit and achieve in-stream water qual-
ity standards at a total lower cost.
Three types of trading are possible: within a plant with
multiple outfalls, between or among plants located on the
same stream, and between point and nonpoint sources.
This regulatory approach is similar to the "multifacility
bubble concept" adopted by EPA's air program. However,
it differs from that concept in that technology-based permit
limits, required under the Clean Water Act, continue to
apply to individual outfalls, rather than to the outfalls as a
group.
Since the passage of the Federal Water Pollution Con-
trol Act of 1972 (PL. 500), as amended), dischargers of
waste waters along the Holston River near Kingsport, Ten-
nessee, have invested heavily in facilities to treat their
wastes; but portions of the stream remain designated as
"water quality limited." This term generally indicates the
likelihood for increasingly stringent waste treatment con-
trols and restrictions on the construction of new and ex-
pansion of existing facilities.
For this reason, EPA's Office of Policy, Planning and
Evaluation and the Tennessee Valley Authority's Office of
Natural Resources and Economic Development initiated
this study of trading among point and nonpoint sources of
pollution along a 32-km (20-mile) reach of the Holston
River. Local dischargers and the Division of Water Man-
agement of the State of Tennessee cooperated. The major
The ideas In this paper do not reflect the official policies of either the U.S. Environ-
mental Protection Agency or the Tennessee Valley Authority.
dischargers in the study area are four point sources-
Tennessee Eastman Co., Mead Paper Co., Kingspbrt's
publicly-owned treatment works, and the Holston Army
Ammunition Plant—and one nonpoint source—Fort Pa-
trick Henry Dam.
We selected dissolved oxygen (DO) as the variable of
interest and examined several means to enhance DO in
the river: further restrictions of the discharges of oxygen-
demanding wastes (for example, five day Biochemical Ox-
ygen Demand (BOD)), varying flow regimes, injecting air
into the stream, and trading between point and nonpoint
sources. With these various treatments available, the
study evolved into a question of cost-effectiveness: For a
given stretch of waterway with specified upstream and
downstream boundaries, what mix of point and nonpoint
sources and sinks of oxygen will achieve desired DO con-
centrations at key times and places at the lowest total
annual (incremental) cost?
METHODOLOGY
The study's methodology can be divided into two major
parts: selecting the study area and simulating treatments
for increasing DO concentrations in the river.
Selecting the Study Area
In selecting the study area, we used the following criteria:
. • The site must have been designated as a water qual-
ity limited stream;
• All point source dischargers must be in compliance
with effluent limitations set forth in their permits;
• Baseline information must be available for both point
and nonpoint sources in the study area; and
• Future economic growth and development are or
have the potential of being adversely affected by poor
water quality.
The 32-km reach of the Holston River met these criteria.
The study area extended from South Fork Holston River,
RM + 8 (upstream) to Holston River, RM-12 (down-
stream). Industrial and municipal discharges enter the
Holston and South Fork Holston Rivers as shown in Figure
1. This stream reach is subject to flow regulation by Fort
Patrick Henry Dam near RM + 8. Downstream from three
417
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Figure 1.—Location map.
of the major point sources, the South Fork Holston River is
joined by the smaller, unregulated North Fork Holston
River (RMO). Previous investigations have shown that a
DO sag develops under certain late summer low-flow con-
ditions in the reach 4.8-11.2 km (3-7 miles) downstream
of the confluence of the two rivers. Definite DO recovery is
evident at the downstream end of the study reach (RM -
12).
Simulating Treatments
Setting Water Quality Goals
The Division of Water Management of the Tennessee De-
partment of Health and Environment (1982) sets forth the
following criteria for DO concentrations:
Dissolved Oxygen—The dissolved oxygen shall be a min-
imum of 5.0 mg/L except in limited sections of streams
where it can be clearly demonstrated that (i) the existing
quality of the water due to irretrievable man-induced con-
ditions cannot be restored to the desired minimum of 5.0
mg/L dissolved oxygen; (ii) the cost for application of ef-
fluent limitations more stringent than those defined
through Section 301 (b) of the Federal Water Pollution
Control Act (PL. 92-500) is economically prohibitive when
compared with the benefits to be obtained; or (iii) the
natural qualities of water are less than the desired mini-
mum of 5.0 mg/L. Such exceptions shall be determined
on an individual basis but in no instance shall the dis-
solved oxygen concentration be less than 3.0 mg/L. ...
The dissolved oxygen concentration of recognized trout
stream shall not be less than 6.0 mg/L....
The Division has generally interpreted the 5.0 mg/L
value as a monthly average, and does not offer guidance
on how to average the 3.0 mg/L value.
Faced with a range of DO values, we generalized the
problem by computing the most cost-effective methods (or
combination of methods) to meet DO values of 3,4,5, and
6 mg/L for various points in the river, to achieve:
1. The minimum 6-hr running average of DO concen-
tration in the sags downstream of RM-3;
2. The daily average of DO concentration in the sags
downstream of RM-3;
3. The minimum 6-hr running average of DO concen-
tration from RM + 8 to RM-12 (the whole 32-km reach);
and
4. The daily average of DO concentration from RM + 8
to RM -12 (the whole 32-km stretch).
Selecting Technologies
Through model simulation, the study considered a num-
ber of treatments:
• Seven traditional treatments that would further re-
strict the discharge of oxygen-demanding wastes, thereby
reducing sinks of DO;
• Eight innovative treatments that would add oxygen or
flow to the stream, thereby creating sources of DO; and
• Eight combinations of these traditional and innova-
tive treatments.
Establishing an Analytical Sequence
We identified and analyzed the cost-effective combina-
tions of traditional and innovative treatments using the
following sequence:
• We specified one scenario for computer simulation by
enumerating the various inputs.
• We used the water quality model to simulate (1) the
daily average DO concentration profile and (2) the mini-
mum of the 6-hr running average DO profile for the entire
32-km study area.
• We computed the total incremental annual cost for
the methods in place for this particular scenario.
• For each particular scenario with its given set of treat-
ments, we used the profiles of the DO concentration and
the cost calculation to develop a cost-effectiveness plot.
Each plotted point represents the results of one complete
simulation (72 hrs) and one cost calculation.
Developing the Modeling Approach
DO concentrations were predicted using a state-of-the-art
unsteady mathematical model calibrated with field data
and process rates from two earlier field surveys. The mod-
eling system consists of an unsteady flow model and a
mass transport water quality model (Hauser and Ruane,
1984). The flow model provided flows, velocities, and
depths at short time intervals for the water quality model.
The water quality model predicted temperature, carbona-
ceous and nitrogenous biochemical oxygen demands
(CBOD and NBOD, respectively). DO concentrations were
also simulated over time to denote diurnal variations.
Modeled sources and sinks of DO include upstream and
lateral inflow sources, natural reaeration, macrophyte
photosynthesis and respiration, CBOD, NBOD, and resid-
ual sediment oxygen demand (SOD).
Improvement strategies were explored by comparing
the predicted movement in DO regime to a base that cor-
responded to the release of current permitted waste loads
to the stream during critical low-flow conditions. Critical
low-flow (base case) on the South Fork Holston River is a
daily average of 750 cfs, provided by pulsing Fort Patrick
Henry Dam under contractual agreement with one of the
industries downstream.
Simulating DO Sinks. The relative influence of each
DO sink in the calibrated model is shown in Figure 2. In a
series of simulations, each DO sink was removed sequen-
tially from the base case (lower line in each plot) until DO
saturation levels were approached. The simulations were
of several days' duration and included diurnal variations in
DO.
In Figure 2, results of the base case simulation are
shown as the lower curve in each plot. Release DO from
the Dam (RM + 8) was assumed to be 3.0 mg/L. Moving
downstream, the predicted daily average DO in the base
case reaerated to around 4.5 mg/L at the diversion weir,
RM - 4.5. The small dip in the predicted DO at this loca-
tion was due to Tennessee Eastman Co.'s withdrawal of
cooling water from a diversion weir pool. Below the pool,
this withdrawal reduced the amount of water, and the flow
volume reached nearly zero when the Fort Patrick Henry
turbines were off, creating nearly stagnant conditions be-
tween the point of withdrawal and the point of return. The
model predicted a significant drop in saturation DO just
below the weir because the industrial cooling water was
returned at a temperature elevated approximately 10°C.
418
-------
Figure 2.—Relative influence of DO sinks.
Downstream, another pool beginning at RM -1.0 and
extending to a shoals section at RM-4.3 contributed
greatly to DO depletion because of the assumed high
SOD and longer residence time. At the shoals section at
RM - 4.3, a third important pool begins. The model pre-
dicted about a 1 mg/L recovery across the shoals due to
natural reaeration, a value insensitive to the exact form of
the reaeration equation assumed. In the base case, a pre-
MAKING DECISIONS ABOUT NONPOINT SOURCE POLLUTION
dieted DO minimum of 2.5 mg/L (daily mean) occurred at
RM - 4.3, and a predicted DO minimum (minimum of 6-hr
running average) of 1 mg/L occurred at RM - 7.0.
DO sinks were removed in the following order: waste
load, SOD, photosynthesis and respiration of macro-
phytes, and background CBOD and NBOD. A deficit re-
mained because cumulative reaeration was insufficient at
this location to bring the large upstream deficit to satura-
tion. Under critical base case conditions, approximately
30 percent of the deficit in daily mean DO at the predicted
sag was from SOD, and about 25 percent from back-
ground CBOD and NBOD and the residual deficit. Aquatic
weeds played a greater role in the 6-hr minimum DO be-
cause each night they use oxygen with no compensating
oxygen-producing photosynthesis, thereby creating an
early morning minimum. For base case conditions, ap-
proximately 30 percent of the predicted deficit in the 6-hr
minimum DO was from the weeds, 25 percent from SOD,
25 percent from waste loads, and 20 percent from back-
ground CBOD and NBOD and the residual deficit.
Simulating Waste Loads. All municipal and industrial
effluent loadings are within current permit limitations.
These permits allow a monthly average load and a maxi-
mum daily load. Although the maximum daily permit level
represents the heaviest loading, it was considered ex-
tremely improbable that all dischargers would be at maxi-
mum day permit levels simultaneously. We decided it was
more plausible to simulate one of the larger dischargers at
maximum day permit load and all others at monthly aver-
age permit loads. Wasteload permit and average waste-
loads are shown in Table 1. Although Tennessee Eastman
Co. currently holds the highest maximum day permit load,
its actual discharges seldom reach the monthly average
permit load. Mead holds the second largest permit and
has an effluent that more frequently approaches its maxi-
mum day permit load. The loading scenario used for the
base case, therefore, was Mead at maximum day permit
load and all other dischargers at their monthly average
permit load.
NBOD loads were assumed to be 598 kg/day (1,328 Ib/
day), 269 kg/day (598 Ib/day), and 195 kg/day (433 Ib/day)
from Tennessee Eastman, Mead, and the Holston Army
Ammunition Plant, respectively. NBOD from the treatment
works was not included because of the lack of data, and
Table 1.—Industrial and municipal BOD discharges in the study area.
(Ibs/day)
SFHRM
HRM
Permitted (current)
Maximum day (May-Sept.)
Monthly average (May-Sept.)
Maximum day (Oct.-April)
Monthly average (Oct.-April)
Actual (current)
Monthly average (May-Sept.)
SFHRM
HRM
Actual (past)
July 1969 survey average
July 1977 survey average
TEC
3.5
8,500
4,000'
13,000
6,000
1,540
4.5
69,300
2,330
MEAD
2.35
6.0001
3,500
7,200
4,800
2,900
2.35
1 1 ,700
3,920
KPOTW
2.3
4,670
2,335'
4,670
2,335
600
2.3
500
1,160
HAAP
141.6
1;620
8101
2,430
1,215
220
4.0
23,480
270
TOTAL
20,790
10,645
27,300
14,350
139.1
32,220 137,200
3,530 11,210
'Base case loadings (total loading = 13,145 Ib/day).
SFHRM = South Fork Holston River mile
HRM = Holston River mile
TEC = Tennesse Eastman Co.
MEAD = Mead Paper Co.
KPOTW = City of Kingsport Publicly-Owned Treatment Works
HAAP = Holston Army Ammunition Plant
419
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
we thought its effect on DO was insignificant. Critical DO
conditions would be expected to occur during late sum-
mer when DO at the Dam was 3.0 mg/L, flow was 800 cfs,
and water temperature was 30.5°C downstream of TEC
discharge. Seasonal permit loads for May-September
were therefore selected for the base case.
Calculating Costs
To calculate the incremental annual costs, we worked with
each participant to identify possible methods by perform-
ance, technology, and economic life (in years) (Indus.
Econ., 1984). With the exception of the city of Kingsport,
for which we made a separate calculation, each of the
participants submitted the incremental costs of each op-
tion—in 1983 dollars—disaggregated into three cost com-
ponents (as applicable):
• Capital costs: one-time (before tax) cost to design,
purchase, and install the equipment;
• Operation and maintenance costs: the recurring (an-
nual) costs for routine maintenance;
• Power costs resulting from changes in release pat-
tern: the recurring annual difference in costs between cur-
rent operation and possible future operation at the dam,
using on peak and off peak (replacement) prices for a
kilowatt hour of electricity.
SIMULATION RESULTS
From the individual simulation results, we were able to
identify opportunities for cost-effective treatments.
Individual Simulations
We conducted the analysis in four steps. First, we devel-
oped a base case that depicted 1983 conditions in the 32-
km reach. Second, we assessed the opportunity for cost-
effective interplant trading among the four industrial and
municipal dischargers. Third, we assessed the cost-effec-
tiveness of three sets of innovative treatments that added
oxygen or flow: aeration at the dam, changes in the sched-
ule of releases from the dam, and in-stream aeration near
the confluence of the South Fork and North Fork Holston
River. Lastly, we assessed the cost effectiveness of the
different combinations of the traditional and innovative
methods.
Base Case
To assess changes in DO concentrations, we simulated
the DO profile for five scenarios: maximum permit day—
when all four plants discharged their permitted values for
maximum day BOD simultaneously; typical permit day—
when all four plants discharged the amounts of BOD as
listed on their permits for average day; average permit
Table 2.—Traditional treatment options.
Discharger
Tennessee Eastman1
Mead Paper2
Holston Army Ammunition Plant3
City of Kingsport Sewage Treatment Plant4
Economic
Life
(yrs)
20
25
(25)
20
Capital
Costs
30,000
12,000
0
1,662
O&M
Costs
24,000
2,500
0
263
Annual
Costs
™/
27,203
3,702
0
440
Hypothetical
Permit Value
for BOD
(Ibs/day)
1,500
3,600
0
600
'Treatment technology is the company's confidential information and hence not listed here.
2New activated sludge treatment plant and sludge handling system.
'Total recycle at no extra cost.
'Mix-media filter.
Table 3.—Scenarios with innovative methods.
Methods
Economic
Life
(yrs)
Capital
costs
O&M
costs
Replacement
power costs
Annual
costs
(thousands of S)
Aeration at Fort Patrick Henry Dam'
1. Low option
2. Medium option
3. High option
Increased flow at Fort Patrick Henry Dam2
4. Low option
5. Medium option
6. High option
In-stream aerators3
7. Single
8. Double
25
25
25
25
25
25
25
25
187
382
662
0
0
0
350
700
18
73
199
0
0
0
59
142
0
0
0
49
129
233
0
0
37
111
265
49
129
233
94
212
'Install pumps and diffusers to release pure oxygen in the reservoir just above the
turbine intake to achieve DO in the tail waters of 4.0 mg/L at low option, 5.0 mg/L at
medium option, and 6.0 mg/L at high option.
'Modify the operation of the Dam to release extra pulses so that more than 750 cfs of
water flows on critical days: low option-875 cfs flow: medium option—1,000 cfs flow;
high option—1,125 cfs flow.
Install one or two aerators, each capable of delivering. 16,000 Ibs/day of DO, via
supersaturating side-stream diffusers: single-operate 15 days/yr; double-operate
128days/yr.
day—when three plants discharged the amounts of BOD
as listed on their permits for average day and one plant
discharged the amount equal to its maximum day; aver-
age actual day—when all four plants discharged the
amounts of BOD that are the actual average of their long-
term discharge; and zero discharge—when all four plants
totally eliminated their discharge of BOD.
As the base case, the typical permit day scenario best
represents a typical day of BOD loads. Figure 3 shows the
calculated profiles for daily average DO and the minimum
420
-------
MAKING DECISIONS ABOUT NONPOINT SOURCE POLLUTION
It
•
•
7'
3'
r •
§ 4
t
2
I
KASE CASE
MAX DAY PERMIT LOADS
MONTHLY AVB PERMIT LOADS
ACTUAL AVC LOADS
M LOADS. CURRENT (00
MO LOADS. •ACKMOUND SOD'
DAILY MAN 00
-//.> -8.0 -v.t MILE
Si1!-^
-n.it -f
i»
• OASE CASE
_ i MAX DAY PERMIT LOADS
9 ' MONTHLY AYE PERNIT LOADS
e
7 -
e.o //.a
5'
r«
S <
»
2
14-
ACTUAL AVC LOADS
NO LOADS. CURRENT SOD
MO LOADS. lACKCROUHD 1^
changes above
^-W
-» -1 -1
-/'.a .8.0 -y.s NILE
Figure 3.—Calculations for the base case.
of the 6-hr running average DO. For the base case, the
lowest daily average value for the whole 32-km reach is
2.6 mg/L, and the lowest 6-hr minimum is 1 mg/L.
Traditional Treatments
We estimated the annual cost for incremental treatment
based on the information that each discharger provided
and assumed hypothetical permit values for BOD for each
discharger (Table 2). From the base case scenario, we
used these suggested treatments to compose seven sce-
narios. Of these new scenarios, the first four correspond
to the base case loadings of BOD except that the sug-
gested treatment is turned ON one at a time for each of
the four dischargers. The fifth scenario, called ALL ON,
corresponds to having all treatments turned ON simulta-
neously
The last two scenarios, called Trading No. 1 and Trading
No. 2, have almost identical BOD but very different annual
costs. In Trading No. 1, Tennessee Eastman is OFF, and
the other three are ON. This scenario has a total BOD load
of 3,690 kg/day (8,200 Ibs/day) and an annual cost of
approximately $4.1 million/yr. In Trading No. 2, Mead is
OFF, and the other three are ON. This scenario has a total
BOD load of 3,685 kg/day (8,100 Ib/day) and an annual
cost of approximately $27.6 million/yr.
For each of these seven scenarios, the predicted DO
profiles are slightly better than the base case profiles be-
low the confluence but not changed above RM + 3. The
results summarized in Table 5 show that further restric-
tions on the four dischargers create modest water quality
improvements at high costs because the dischargers have
already installed equipment that removes over 90 percent
of BOD in the raw wastewater. These treatments affect the
— (USE CASE
— 00-«
... OO'S
00-«
. . DO-I
DAILY MEAN DO
-9 -7 -S
-1 1
HILE
10
9
•
7
5§
7 t
§ «
3
3
i
— OASE CASE
— DO-4
... DO*S
00>*
- - 00-0
=11 -9 -7 -1 -J
lile
-i—T-
Figure 4.—Aerating Fort Patrick Henry releases.
DO profile below the outfalls; they produce no change
aboveRM + 3.
Innovative Methods
Based on the information that TVA provided, we estimated
the annual cost and performance of three innovative
methods to improve DO in the river: aeration at the dam,
increased flows from the dam by the release of extra
pulses of water, and in-stream aeration near the conflu-
ence. All of these innovative methods cost less annually
than the traditional treatments.
We used these suggested innovations to compose the
eight scenarios shown in Table 3. Quite simply, each new
scenario consists of the base case conditions, with each
innovation turned ON one at a time to simulate the re-
sponse in water quality. These results are summarized in
Table 5. The model shows that aeration at the Dam im-
proves water quality near the Dam but has little lasting
effect below the Dam (Fig. 4). Increasing flow at the Dam
creates little DO improvement immediately below the Dam
but substantial improvement below the confluence (Fig.
5). We found that in-stream aeration can simply increase
the DO concentrations, but that these concentrations at-
tenuate rapidly downstream (Fig. 6).
We considered several options for augmenting flow, in-
cluding increasing pulsing frequency to once every 3rd
hour, adding flow to the current 4th-hour pulses, and add-
ing pulses between the usual 4th-hour pulses. The model
predicted that adding the first 125 cfs resulted in approxi-
mately 1 mg/L DO improvement, with diminishing im-
provement for the additional flow increments. Subsequent
to this modeling effort, a field study verified the suspected
DO improvement from flow augmentation. Because the
421
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
"Able S.—Summary statistics
Below the
confluence
Treatment options
Base case
Traditional treatments
A. Tennessee Eastman
B. Mead Paper
C. Holston Army
D. Kingsport
E. All
F. Trading No. 1
G. Trading No. 2
Innovative methods
H. Dam — low aeration
I. Dam — medium aeration
J. Dam — higt) aeration
K. Dam — low augmentation
L. Dam— medium augmentation
M. Dam— high augmentation
N. In-stream — single
O. In-stream — double
Combinations
P. No. 1
Q. No. 2
R. No. 3
S. No. 4
T. No. 5
U. No. 6
V. No. 7
W. No. 8
Minimum
6-hr
average
DO
mg/L
1.0
1.3
1.3
1:2
1.3
1.9
1.6
1.7
1.2
1.3
1.4
2.0
2.8
3.3
1.9
2.0
2.3
3.0
2.4
3.3
3.3
3.2
3.6
4.0
Minimum
dally
average
DO
mg/L
2.6
2.8
2.8
2.6
2.7
3.3
3.0
3.1
2.7
2.8
2.9
3.6
4.3
4.6
3.6
4.6
3.9
4.6
4.3
4.9
4.4
4.7
5.1
5.4
Whole 20-mile
reach
Minimum
6-hr
average
DO
mg/L
1.0
1.3
1.3
1.2
1.3
1.9
1.6
1.7
1.2
1.3
1.4
2.0
2.8
3.0
1.9
2.0
2.3
3.0
2.4
3.3
3.0
3.2
3.6
4.0
Minimum
daily
average
DO
mg/L
2.6
2.8
2.8
2.6
2.7
3.0
3.0
3.0
2.7
2.8
2.9
3.0
3.0
3.0
3.0
3.0
3.9
4.6
4.3
4.9
3.0
4.7
5.1
5.4
Annual
cost
($ thousands)
—
27,203
3,702
0
440
31,345
4,142
27,643
37
11
265
49
129
223
94
212
98
178
133
213
148
178
282
268
field results did not support the initial model results, we
consider conclusions on flow augmentation as tentative
pending further investigations and modeling.
We simulated the effect of adding oxygen at a rate of
7,200 kg/day (16,000 Ib/day) for a 12-hour period (not
3,600 kg/day (8,000 Ib/day)) at locations upstream of the
predicted sag. The model predicted that the DO improve-
ment from instream aeration rapidly diminished down-
stream from the aeration source.
Combination of Methods
Each of the innovative methods has either (or both) rela-
tively lower cost or higher performance than the traditional
methods; but because no single innovative method per-
forms well at both the upstream and downstream ends of
the 32-km segment, we explored various combinations of
both the traditional and innovative methods. We com-
posed eight scenarios based on low cost and methods
that complement each other. While we make no claim that
these scenarios will be ultimately optimal, they show
promise in and of themselves; and combinations of inno-
vative methods are much more attractive than traditional
methods.
Table 4 shows cost components and estimated annual
cost for each combination. Combinations No. 1 through
No. 4 improve DO both above and below the confluence
because the methods complement each other: aeration
improves DO above the confluence and flow augmenta-
tion improves DO below the confluence. Combination No.
5 leaves DO concentration essentially unchanged from
the base case above the confluence but substantially im-
proves DO concentration below the confluence. The
promise of the complementary innovative methods, Com-
binations No. 6, No. 7, and No. 8, is shown in Table 5. For
example, Combination No. 8 achieves a daily average DO
concentration of more than 5 mg/L everywhere along the
32-km reach. We generally conclude that complementary
combinations of innovative methods can achieve desir-
able DO concentrations in the stream not reached by tra-
ditional methods or by single innovative methods.
Water temperature plays a significant role in modeling
stream water quality for two reasons. First, the driving
force for aeration is the DO deficit below saturation, and
saturation DO decreases with an increase in temperature.
Second, the rates of important DO sinks and sources,
such as DO demands from stabilization of organic wastes
or weed respiration and natural reaeration, increase with
an increase in temperature. Because removal of heat load
implies the use of cooling towers and in TVA's experience,
cooling towers are not a cost-effective alternative, we have
not simulated heat removal in this study.
Opportunities for Cost-Effective Innovations
In Table 5 we summarize the results of the analyses con-
ducted for the base case, for the seven traditional sce-
narios labeled A through G, the eight single innovations
labeled H through O, and the eight combination scenarios
labeled P through W. Given annual costs and a measure
for water quality—the minimum 6-hr average and the daily
average DO concentrations both below the confluence
and for the entire 32-km stretch, we can find the most cost
effective way to reach the chosen concentration of DO.
For example, the cheapest option for achieving minimum
daily average DO concentration of 3 mg/L for the whole
32 km is innovative method K. Similarly to achieve a daily
average DO concentration of 5 mg/L or better for the 32
422
-------
MAKING DECISIONS ABOUT NONPOINT SOURCE POLLUTION
•AM CAM <0>7M> ADO PULSE 7,-i, 10*11*1)
AOO PULSE •»-§» IO.»«7*> —-ADD PULSE •»-•« _
•All CASE
AC* «»-*. • « -1.12
AIR ••-•• • < -1.12 •"< ».«•.
• tASE CAft (Q*7M> ADO PULSE ?•-•• (QMI*»
•ADD PULSE ••-•• <0>S7S> AOO PULSE •»-•• IOM12S>_
•ASE CASE
AC« §.-«. .1 -1.12
AEH •«-*« .1 -(.12 «»< •.«
Figure 5.—Flow augmentation with additional evening Figure 6.—In-stream aeration.
pulses.
km, we find that Combination No. 8 costs the least. No
other methods that we evaluated could achieve this DO
concentration.
To examine more closely various combinations of inno-
vative methods, we have plotted points P through W: con-
centrations of DO on the X-axis and annual costs on the Y-
axis. For daily average of DO concentration throughout
the 32-km reach, Figure 7 shows that the estimated costs
rise steeply as an increasing function of DO concentra-
tions. A point lying wholly to the right of and below a
second point is preferred in terms of cost effectiveness.
That is, the first point has both higher performance and
lower costs than the second. We have drawn the most
cost-effective frontier possible, that is, the straight-line
segments connecting the points lowest and farthest to the
right. Points lying on the cost effectiveness frontier are
more economically efficient than points to the left of or
above the curve. The cost-effective frontier in this chart
consists of Base, P, R, U, S, and W. A few other observa-
tions are:
• Point W is always preferred to point V on purely eco-
nomic grounds because the single in-stream aerator more
than compensates for the lower augmentation, and
• Point U is always preferred to point Q because the
zero-net-cost option at Holston Army Ammunition Plant
does improve water quality.
OVERALL OBSERVATIONS
This exploratory study has yielded us two major conclu-
sions that apply regardless of the water quality measure:
• For the same results, innovative techniques generally
have annual costs at least an order of magnitude lower
than the traditional methods. For a given DO concentra-
tion, the estimated cost savings can range as high as
several millions of dollars per year.
• Complementary combinations of certain innovative
methods can achieve DO concentrations that traditional
methods cannot. For example, even if the four industrial
and municipal dischargers stopped discharging BOD, the
DO concentrations in some segments of the river would
not reach the concentrations achievable with combina-
tions of innovative methods, such as in-stream aeration
and increased flow.
Taken together, all the scenarios and simulations in this
explanatory study form a "menu for opportunities." With
care, one may find the lowest cost set of treatments that
meets the desired ambient water quality standards. Two
examples will illustrate the method and show the magni-
tude of the potential cost savings:
• If the State set the ambient Do standard at a daily
average of 3 mg/L for the entire 32-km reach, the most
cost-effective combination of traditional treatments has an
annual cost of over $4 million/year, while the most cost
effective innovative treatment has an annual cost of under
$50,000.
• If the State set the ambient DO standard at a daily
average of 5 mg/L for the entire 32 km reach, no combina-
tion of traditional treatments could achieve this goal, but
the most cost-effective combination of innovative treat-
ments has an annual cost of under $275,000.
Next Steps
In the Clean Water Act, the Congress did not anticipate
such opportunities as this study has explored. Federal
and State laws and regulations focus almost exclusively
423
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 4.—Combination scenarios.
Label
P
Q
R
S
T
U
V.
w
Method
Combination No.
Combination No.
Combination No.
Combination No.
Combination No.
Combination No.
Combination No.
Combination No.
1
2
3
4
5
6
7
8
Economic
Life
(yrs)
25
25
25
25
25
25
25
25
Capital O & M
costs costs
(thousands of $)
248
248
400
400
350
248
248
598
73
153
93
173
64
153
257
208
Replacement
power costs
0
0
0
0
49
0
0
0
Annual
costs
98
178
133
213
148
178
282
268
NOTE: For a combination, the capital costs and the O & M may be less than the sum
of the corresponding costs for the component methods due to savings from
parttime operation. These combinations are:
Combination No. 1: Low augmentation (875 cfs) and high aeration (6 mg/L)
at the Dam.
Combination No. 2: Medium augmentation (1,000 cfs) and high aeration (6
mg/L) at the Dam.
Combination No. 3: Low augmentation (875 cfs) and superhigh aeration (8
mg/L) at the Dam.
Combination No. 4: Medium augmentation (1,000 cfs) and superhigh aera-
tion (8 mg/L) at the Dam.
Combination No. 5: Medium augmentation (1,000 cfs) at the Dam and in-
stream single aeration at RM - 3.12 for 16 hrs/day.
Combination No. 6: Zero discharge at HAAP, and medium augmentation
(1,000 cfs) and high aeration (6 mg/L) at the Dam.
Combination No. 7: Zero discharge at HAAP, and high augmentation (1,125
cfs) and high aeration (6 mg/L) at the Dam.
Combination No. 8: Zero discharge at HAAP, medium augmentation (1,000
cfs) and high aeration (6 mg/L) at the Dam, and in-
stream single aeration at RM-3.12 for 16 hrs/day
GE^E
0123456
COHCEHTKATIOR OF DISSOLVED OXTGEN IN MG/L
Figure 7.—Cost-effectiveness of combinations of Innovative
methods (P-W) using the lowest value of the dally average
of dissolved oxygen concentration throughout the 20-mile
reach.
on reducing discharges from industrial and municipal
plants (basing permit limits on either categorical stand-
ards or ambient water quality standards), and various Fed-
eral court decisions have exempted dams from discharge
requirements. More narrowly, Federal and State laws and
regulations discourage flow augmentation as a way to re-
duce pollution, and they say little, if anything, on aeration.
The realization of these opportunities will require negoti-
ation among the participants, including the regulatory au-
thorities. Their general acceptability may be based on
such factors as efficiency-least cost, equity-fairness, en-
forceability, and ease of administration. The participants
will then need approval through regulatory, administrative,
or legal channels to implement any proposed agreements.
The final plan may well allow for cash payments among
the participants or even the creation of a new nonprofit
corporation to own and operate in-stream or at-the-dam
aerators. Realizing these complementary combinations of
innovative methods will require bold thinking about the
institutions and their interdependencies.
REFERENCES
Mauser, G.E., and R.J. Ruane. 1984. Model exploration of Hol-
ston River water quality improvement strategies. Draft rep.
Off. Nat. Resour. Econ. Develop., Div. Air Water Resour.,
Tenn. Valley Author.
Industrial Economics, Inc. 1984. Exploratory Study of Improving
Dissolved Oxygen Concentrations in the Holston River near
Kingsport, Tennessee. Draft rep. Prepared for joint proj.: Off.
Policy Plann. Eval., U.S. Environ. Prot. Agency; Off. Nat. Re-
sour. Econ. Develop., Tenn. Valley Author.
Tennessee Department of Health and Environment. 1982. Crite-
ria for fish and aquatic life. In General Water Quality Criteria.
Div. Water Manage. Nashville.
424
-------
PROTECTING TILLAMOOK BAY SHELLFISH WITH POINT/NONPOINT
SOURCE CONTROLS
JOHN E. JACKSON
Oregon Department of Environmental Quality
Portland, Oregon
Tillamook Bay, located on the North Oregon Coast 96 km
west of Portland, is Oregon's second largest estuary but
produces the State's largest amount of commercially
grown oysters. Because of its close proximity to the metro-
politan area of Portland, it is also a popular recreational
area for clam diggers, fishermen, swimmers, and sight-
seers.
In 1979, the bay waters and many of the streams drain-
ing into the bay were found to be contaminated by fecal
material from, at that time, sources unknown. The water
quality conditions threatened closure of the Bay to the
shellfish harvesting that supported a portion of the local
economy of approximately 13,000 people.
Under U.S. Environmental Protection Agency Section
208 funds the Oregon Department of Environmental Qual-
ity conducted a project from July 1979 to June 1981. The
goal was to establish a comprehensive Tillamook Bay Fe-
cal Waste Management Plan for protecting the beneficial
use of the water, that is, Tillamook Bay's shellfish re-
source. The objectives of the project were to: (1) analyze
existing and new data to quantify the problem, (2) identify
the fecal bacteria sources, and (3) develop a plan to pro-
tect the shellfish resource by establishing necessary best
management practices (BMPs), rules, and standards to
minimize fecal waste discharges to the surface waters of
the basin.
The intent was to preserve and protect the shellfish, a
natural resource, as a beneficial use and, at the same
time, to allow activities identified as sources of bacterial
pollution to continue in a sensible, sanitary manner. The
management plan would not achieve zero bacteria dis-
charge from identified sources.
During the investigation, six major fecal sources were
examined: sewage treatment plants (five located in the
bay watersheds), recreation, forestry activities, industries,
agricultural operations (120 dairies; 19,100 cows; 256,360
metric tonnes (282,000 tons) of manure annually), and on-
site subsurface sewage disposal systems (serving approx-
imately 40 percent of the population).
The project identified malfunctioning sewage treatment
plants, some malfunctioning or inadequate on-site subsur-
face sewage disposal systems, and some agricultural op-
erations discharging fecal material to the streams and bay
that created a health hazard for consumption of bay shell-
fish and endangered swimmers in the tributaries.
Once the fecal source types were identified, corrective
actions had to be determined. Existing control programs
and new actions were investigated to determine the best
suited corrective method for each fecal source type of
water pollution problems. Tradeoffs of control came into
play.
Alternatives were considered that required tradeoffs in
timing implementation (do everything now or sequence
the cleanup over a number of months or years), tradeoffs
in identifying controls for the sake of human health or
stream health, tradeoffs in what was to be corrected (the
point sources or the nonpoint sources) and finally trade-
offs in strategy of controls (control the land activity and
sources of the problem or control the water activity by
closing the bay).
The word "tradeoffs" used in the context of this panel
discussion suggests that point or nonpoint controls of wa-
ter pollution can be traded back and forth to fit the situa-
tion, that such controls depend on a person's likes and
dislikes. This might be appropriate up to a point. When
choosing effective controls, the decisionmaker(s) must
have a clear picture of the problem, its specific occur-
rences and the ultimate correction goal.
In the Tillamook Bay situation a number of factors dic-
tated or limited the tradeoff choices. A compendium of
controls resulted and have proved to be very effective in
improving the water quality of Tillamook Bay and its rivers.
Current health risk was the key element to the Tillamook
control strategy. This effectively eliminated the option of
maintaining a status quo, in other words, doing nothing.
The luxury of months and years to correct the problem
was unavailable; however, correcting the fecal contamina-
tion problem from identified sources would take time. A
tradeoff was identified. Instead of immediately closing the
bay to further shellfishing until corrective actions could be
completed at the pollution sources, shellfish harvesting
was allowed when the known major fecal discharges were
not contaminating the bay and was prohibited when they
did discharge.
To institute the cleanup of the pollution problems, the
who, when, and where factors of the dischargers had to be
known. Placement of the cleanup emphasis became an-
other tradeoff decision with the cleanup goal firmly in
mind. Oregon, as is the case in most places, strives to get
the biggest cleanup for the least dollars. This becomes an
easy task if the interaction of pollution sources is known.
In the Tillamook situation, the primary problem of storm
runoff from dairy barnyards occurred during every storm
no matter how saturated the ground. Because of the per-
vasiveness of the problem, inadequate on-site subsurface
sewage disposal systems became a secondary problem.
The third was that occasional sewage treatment plant
breakdowns caused raw sewage to enter the bays.
Hence, cleanup emphasis was placed on the dairy waste
management. This did not preclude action on the serious
raw sewage bypass problem if and when it occurred.
The water quality is improving basinwide from cleanup
activities dealing with dairy wastes and on-site subsurface
sewage problems in localized areas. The infrequent sew-
age treatment plant malfunctions have been monitored
when they occur.
What does this all mean? An accurate assessment of
the problems, use of that information, and subsequent
tradeoff decisions enable the State to make the biggest
improvement towards alleviating the health risk.
No one has had to close a business. Shellfishing oc-
curs, but with the knowledge of the bay water quality con-
ditions. The local area has gained a cleaner bay and riv-
ers.
425
-------
BRUCE BAKER
STEVEN SKAVRONECK
Bureau of Water Resources Management
Wisconsin Department of Natural Resources
Madison, Wisconsin
Recognition that water quality problems can result from
combined point and nonpoint source pollution impacts is
important in making cost-effective management deci-
sions. Information gathered to assess whether combined
impacts are important can be useful both in planning gen-
eral water quality program strategies and evaluating site-
specific pollution control options. Program areas where
such information would be useful include: standards and
effluent limits development, facilities planning, point and
nonpoint source pollution abatement grants, and enforce-
ment. The focus of the Wisconsin program is to determine
what controls are needed for both point and nonpoint
sources in achieving our water quality objectives.
Two types of point/nonpoint source issues have become
apparent in Wisconsin. The first type occurs when waste-
water treatment plants are upgraded to maintain water
quality standards in the receiving stream, yet when the
new plant goes on line the standards and beneficial uses
are not achieved because of nonpoint sources. The typical
situation here involves a relatively small (less than 5 mgd)
treatment plant providing advanced treatment and a small
receiving stream (less than 5 cfs) impacted by agricultural
runoff or other nonpoint sources.
An example of this issue is the south fork of the Lemon-
weir River at Tomah. The Toman wastewater treatment
plant discharges to the stream 1.4 miles below the outlet
from Lake Tomah, a highly eutrophic impoundment of the
Lemonweir River. Lake Tomah is shallow and algae
choked and receives the runoff from an agricultural water-
shed. The wastewater treatment plant was upgraded in
1981 to provide advanced treatment for biological oxygen
demand, suspended solids, and ammonia. However, dis-
solved oxygen levels remain severely depressed both up-
stream and downstream of the effluent discharge because
of the dead algae in the outflow from Lake Tomah.
The second type of issue involves the achievement of
nonpoint source control objectives in the presence of a
point source discharge, typically a large wastewater treat-
ment plant. The key question is whether water quality im-
provements from a priority watershed project to control
nonpoint sources would be negated by water quality deg-
radation by point sources. This has become an important
issue in Wisconsin's Nonpoint Source Grant Program.
An example of this second type of issue occurs on Turtle
Creek in southeastern Wisconsin. A project was devel-
oped for the Turtle Creek watershed under the Wisconsin
Fund Nonpoint Source Grant Program following signifi-
cant demonstrated local support. It was chosen as a Prior-
ity Watershed. The main water quality objective for this
watershed project was phosphorus reduction through ag-
ricultural best management practices (BMP's). However,
the Walworth County Metropolitan (Walco Met) wastewa-
ter treatment plant discharges to Turtle Creek. This plant
was built in 1981 to divert effluent from Lake Delavan and
discharge if downstream to Turtle Creek. The lake previ-
ously acted as a sink for the phosphorus in the wastewater
effluent. Some of the questions that arise from this situa-
tion are:
° Can the water quality objectives of the priority water-
shed project be achieved given the presence of the Walco
Met discharge?
° Should phosphorus control be required at the Walco
Met plant?
° Can we remove enough phosphorus from Turtle
Creek through voluntary BMP's to not worry about the
phosphorus load from Walco Met, that is, are tradeoffs
involved?
Two types of interfaces can occur. The difference is sig-
nificant. In one instance, both sources have similar effects
on the stream, that is, the impacts are additive. A dis-
solved oxygen sag caused by both sources is an example.
In the other instance, the point and nonpoint sources af-
fect the stream differently, that is, the impacts are indepen-
dent. Physical habitat degradation, by sedimentation due
to nonpoint sources, coexisting with chemical degradation
from point source discharges (biochemical oxygen de-
mand or nutrients) is an example. Additive impacts may
involve tradeoffs between the two pollutant sources since
both use the same stream assimilative capacity. Indepen-
dent impacts do not allow for the same tradeoffs; pollution
control in both sectors is necessary to achieve the
stream's beneficial uses.
Wisconsin Department of Natural Resources (DNR) de-
cided to study point/nonpoint source interface issues in
more detail and produced a report, The Role of Nonpoint
Source Information and Control Programs in Achieving
Water Quality Improvements at Point Source Discharge
Sites. Prepared by Steven Skavroneck and John Render,
this report represents a joint effort of the Water Resources
Planning and Policy Section and the DNR Southern Dis-
trict Office.
The overall goal of the project was to integrate informa-
tion concerning point and nonpoint source pollution im-
pacts, controllability of these pollution sources, and the
ability to attain water quality standards under different
control options into pollution control strategies. Major ob-
jectives of the study included the following:
1. development of a site assessment procedure to as-
sess relative impacts in a stream from point and nonpoint
related problem sources,
2. development of a method to establish municipal en-
forcement priorities based on potential water quality im-
provement, taking into account the effects of nonpoint
sources of pollution,
3. development of a method for determining back-
ground water quality in wastewater treatment plant (WTP)
impact zones coimpacted by nonpoint sources,
4. development of a framework for determining target
water quality criteria and the significance of water quality
improvements resulting from different point source treat-
ment levels, and
5. development of a method for evaluating the control-
lability of nonpoint sources.
A site assessment procedure was developed to deter-
mine which stream reaches are actually or potentially im-
pacted by both point and nonpoint sources of pollution.
The procedure should be viewed as a way for water re-
sources staff to "order their thinking" about a stream
426
-------
. MAKING DECISIONS ABOUT NONPOINT SOURCE POLLUTION
reach, using all available information. This procedure aids
in the identification of water quality problems, determina-
tion of the relative roles of point sources and nonpoint
sources in causing the water quality problems, and provid-
ing a general indication of whether or not desired water
quality improvements can be attained through point
source controls alone. This site assessment procedure
has been incorporated into the water quality management
plan update process.
A three-step process for evaluating the water quality of
selected stream reaches is envisioned. The first step is to
determine which stream reaches to assess. The second
step is to apply the site assessment procedure to those
stream reaches. The third step is to perform a detailed
analysis of attainable goals for those stream reaches iden-
tified as priorities based on the site assessments.
The point/nonpoint source issues discussed in this pa-
per are being addressed through several of Wisconsin's
Water Resource Management programs including Water
Quality Management Plans, Effluent Limit Setting, and the
Nonpoint Source Grant Program through Priority Water-
shed Plans. Regardless of the program, the overall frame-
work for addressing point/nonpoint source issues should
be:
1. problem identification,
2. analysis of the relative contribution of point and non-
point pollution sources to the water quality problem,
3. definition of water quality improvement objectives
based on the identified problems,
4. assessment of the ability to meet water quality im-
provement objectives with various combinations of point
and nonpoint source controls, and
5. development of a management plan recommending
appropriate point and nonpoint source controls.
If necessary, modifications might be made to the previ-
ously established water quality improvement objectives to
reflect their attainability.
427
-------
Data Availability
and Needs
A DATA MANAGEMENT SYSTEM TO EVALUATE WATER QUALITY
IMPACTS OF NONPOINT SOURCE POLLUTION CONTROL
M. D. SMOLEN
S. A. DRESSING
R. P. MAAS
J. SPOONER
C. A. JAMESON
A. D. NEWELL
F. J. HUMENIK
North Carolina State University
Raleigh, North Carolina
ABSTRACT
The data base and data management system of the Na-
tional Water Quality Evaluation Project (North Carolina
State University) were developed specifically to analyze
the effects of various approaches to land treatment in the
context of diverse hydrologic, geologic, meteorologic,
and socioeconomic factors. The data base consists of two
parts: (1) an annotated bibliography and (2) a water qual-
ity project file, containing summaries of agricultural water
quality projects. The project file contains subfiles for gen-
eral project data, detailed descriptions of projects' water
quality problems and land treatment programs, and, sum-
maries of the projects' water quality results. Land treat-
ment data are indexed by project code, date, and a Land
Area Resource Code (LARC), and water resource and
water quality data are indexed by project code, date, and
a Water Resource Code (WARC). Water quality results
are then related to land treatment by associating LARCs
with WARCs. The data base is used to evaluate water
quality projects, but future developmental effort will be
directed toward a BMP-decision matrix to access all avail-
able information for making and planning nonpoint
source pollution projects.
The National Water Quality Evaluation Project (NWQEP)
is charged with examining agricultural nonpoint source
control efforts to evaluate the efficacy of best manage-
ment practices (BMPs) and implementation strategies in
terms of water quality. The factors affecting water quality
and the NPS control programs vary widely from one loca-
tion to another. The primary problem in conducting the
national evaluation is, therefore, to assemble highly di-
verse types of information into a common data base to
compare results and methods and develop recommenda-
tions.
To address this problem we have developed a two-part
data base consisting of: (1) a water quality project file to
store primary information, and (2) an annotated bibliogra-
phy of nonpoint source-related publications to access the
relevant scientific literature. The project file is restricted to
those agricultural water quality projects that include both a
land treatment component and a water quality monitoring
component. This restriction reduces the problem of man-
aging a great deal of unrelated information and eliminates
many demonstration projects that have implementation
without water quality monitoring.
In developing a data management system, the designer
needs to consider what types of data will reside in the data
base, who will be the users of the data base, and what
analyses will be requested. The analyses considered for
the NWQEP data management system (NWQEP-DMS)
included testing hypotheses concerning: project manage-
ment and BMP implementation, cost of BMP implementa-
tion and cost efficiency, immediate effects on quality of
runoff or ground water, and the ultimate impact on quality
of an impaired water resource. For this reason, the data
base includes parameters that describe each project's
429
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
goals and objectives, implementation costs, cost sharing,
and the value of the impaired water resource, as well as
the physical setting and water quality impact.
STRUCTURE OF THE PROJECT FILE
IN THE NWQEP-DMS
The project file stores NPS project information and param-
eters that identify other data bases for supporting informa-
tion, such as individual water quality observations. The
project file includes detailed information on climate, geog-
raphy, land use distribution, BMP implementation goals
and accomplishments, a description of the water re-
sources of the project area, and summaries of water qual-
ity monitoring results.
The structure of the NWQEP-DMS project summary file
is depicted in Figure 1. It consists of a land information file
and a water information file. The land information file has
three data sets: a General Descriptive Data (GDD) set
containing information pertinent to the whole project area,
a Land Treatment Data (LTD) set containing land use infor-
mation, and a BMP data set containing information on
BMP implementation goals, accomplishments, and costs.
The water information file consists of two data sets: a
Static Water Quality (WQS) file and a Dynamic Water
Quality (WQD) file. The static file describes the receiving
water resources of the NPS project, and the dynamic file
contains water quality parameters that change with time,
such as pollutant concentration and loading values.
The GDD data set is indexed by Project Code (for exam-
ple, R21 for RCWP 21, the Rural Clean Water Program
project in Virginia). It contains such nonvarying informa-
tion as the project's location (latitude and longitude), the
average monthly rainfall, snowfall, and temperature, and
lists land areas that contribute to each water resource
within the project. Each land area is identified by a Land
Area Resource Code (LARC), and each water resource is
identified by a Water Resource Code (WARC). Every proj-
ect in the NWQEP-DMS has one GDD file. Figure 1
shows the relationship of LARCs and WARCs in an actual
nonpoint source project.
Every NPS project has at least one land area as a
unique LARC. The boundaries of the LARC correspond
with a natural watershed divide so that all of its surface
runoff contributes to a single water resource. A single wa-
ter resource (WARC) may have several LARCs, but a
LARC can contribute to only one WARC. For analysis of
cause-effect relationships in a project, the data from sev-
eral LARCs may, therefore, be pooled if water quality data
are available at few monitoring stations.
If land use and land treatment information are available
for each water resource within a nonpoint source project
area, the project area may be divided into subareas with
different LARCs. Because the workload to provide land
use and BMP implementation data on a subwatershed
basis increases rapidly with the number of LARCs, most
projects in the NWQEP-DMS at this time have fewer than
four LARCs. LARCs range in size from about 400 ha
(1,000 acres) to as much as 12,000 ha (30,000 acres).
Land use information for a LARC is placed in the LTD
file, which is indexed by project code, LARC, and date.
The file contains a description of the land use distribution
within the specified project subarea. Each land use type is
VCIIK
BASIN
WARC LARC
NANSEMOND HIVER
CHUCKATUCK CREEK
UNAN NANS
WCllK CHKT
LAND USE DATA
[LTD io
ILTD io/83
LTD 10/84
Projcod = R21
LARC =NANS
Land Use
^82
[LTD 10
ILTD io/83
LTD 10/84
Projcod = R21
LARC = CHKT
Land Use
IS2
IBMP 10/82
IBMP 10/83
BMP 10/84
Projcod = R21
LARC = NANS
BMP Goals
BMPs Installed
|BMP 10/82
(BMP io/83
BMP 10/84
Projcod = R21
LARC. = CHKT
BMP Goals
BMPs Installed
WATER QUALITY DATA
1/83
IWQD 10/82
a 11/82
2/82
IWQD i/83
IWQD 10/82
D 11/82
2/82
Figure 1.—Structure of the project summary file in the NWQEP-DMS.
430
-------
DATA AVAILABILITY AND NEEDS
identified by family, genus, and species. The family level
refers to such classifications as agricultural, urban, or for-
est; the genus level to such classifications as animal oper-
ation, residential, or harvested forest; and the species
level refers to such classifications as specific crop, con-
struction, or unconfined livestock area. The area of the
LARC, the percent area for each specific land use type,
crop yield information, estimates of fertilizer usage by
crops, and estimates of animal waste production are in-
cluded. A quality parameter rates each estimate on a
scale of 0 to 3 (unsubstantiated to precise and accurate).
Information on BMP implementation is contained in a
BMP file, indexed by project code, LARC, and date. The
file contains parameters to specify BMP reporting units,
implementation goals, and implementation accomplish-
ments. The BMP file also contains the amount of project
money spent on implementation and estimates nonproject
money spent on BMP implementation. New observations
are added to both the LTD and the BMP files annually.
Of the two water-related information files, the WQS (Wa-
ter Quality Static) file contains the descriptions of water
resources such as lakes, impoundments, rivers, and aqui-
fers, indexed by project code and WARC. The description
of a water resource includes its physical and hydrological
characteristics, relevant water quality standards, and pa-
rameters identifying the intended uses, the use impair-
ments, and the type and strength of documentation availa-
ble. The types of documentation include social
perceptions and economic data as well as chemical, phys-
ical, and biological data. This file also contains param-
eters to retrieve water quality recores from the EPA STOr-
age and RETrieval system (STORET).
The WQS data set also includes a list of pollutants and
the designated pollutant reduction goals specified by the
nonpoint source project. These data may be retrieved by
project code and WARC. Annual entries to this file are not
anticipated.
The WQD file contains pollutant concentrations and
loads indexed by project code, WARC, and date. In addi-
tion to pollutant names, such as nitrate-N, the STORET
analysis code is included. Observations in the data set are
entered as monthly means unless specified otherwise. Pa-
rameters are also reserved for standard deviations, medi-
ans, and number of samples. A procedure will be devel-
oped to extract monthly summary values from primary
data in STORET or other sources of water quality observa-
tions to move these values into the NWQEP-DMS.
DATA ANALYSIS AND DATA SOURCES
Procedures written in the language of the Statistical Anal-
ysis System (SAS, 1982) perform data management and
analysis. This system allows a high degree of program-
ming flexibility and direct use of the data in statistical anal-
yses.
Most of the data in the NWQEP-DMS project file were
extracted from the reports of agricultural nonpoint source
projects sponsored by State and Federal agency pro-
grams. These programs include: RCWP, the Model Imple-
mentation Program, Agricultural Conservation Program-
SOURCE
ROW CROPS
DISTRIBUTION
OF VALUES
Figure 2.—Related data planes in the BMP decision matrix.
431
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Special Water Quality Projects, the Great Lakes 108
Program, statewide 208 Programs, PL. 566 Programs, the
Chesapeake Bay Program, and others. These data will be
supplemented by information from the literature, from US-
DA's Conservation Reporting and Evaluation System
(CRES), and from any other sources we are able to iden-
tify
DEVELOPMENT OF A BMP DECISION
MATRIX
The NWQEP-DMS has been developed primarily for proj-
ect evaluation. Our future goal, however, is to use the data
to address basic questions about BMP's and BMP imple-
mentation strategies. Further efforts in this respect will be
focused on development of a BMP decision matrix (BMP-
DM).
The BMP-DM is illustrated in Figure 2. Conceptually, it
will be a multidimensional structure that may be thought of
as a series of related data planes. Each cell in a data
plane will contain a summary of water quality responses
for the combination of indexing variables that locates the
cell in the data matrix. The individual cells in the data
matrix will be filled by observations from the NWQEP-
DMS project file and information from the project library
(accessed through the annotated bibliograpy). If one is
interested in nitrogen reduction by a particular BMP on a
specific crop in a specific region, the distribution of values
will be retrievable from the BMP-DM as shown in Figure
2. Recommendations on application of BMP's to specific
problems will be obtainable by combining the probable
reduction of the pollutant by implementation with the prob-
able implementation density. Related information on costs
and expected lifetime could also be examined to identify
the optimal choice of BMP's for the specific situation.
CONCLUSION
Achieving the full potential of the NWQEP-DMS and the
BMP-DM will require considerably more data and a con-
certed programming effort. Most of the data presently in
the project file are from RCWP project reports. The anno-
tated bibliography contains about 1,800 references, re-
trievable by author or topic. Nevertheless, a data system
like this is essential to transfer the knowledge obtained
from the diverse nonpoint source projects conducted
throughout the country to future nonpoint source control
efforts. The efficiency of this information transfer may, to a
large extent, determine how successful we are in control-
ling agricultural nonpoint source pollution.
REFERENCES
SAS Institute. 1982. SAS User's Guide: Basics. SAS Inst. Inc.,
Gary, NC.
432
-------
DEVELOPMENT OF A NONPOINT SOURCE DATA CENTER
CLAIRE M. GESALMAN
The Synectics Group, Inc.
Washington, D.C.
ABSTRACT
One challenge faced by managers of nonpoint source
control programs is tracking progress made in solving
problems. Program managers and staff often do not have
ready access to the latest information on, for example,
best management practices, water quality effects and
benefits, and other State programs. Such information is
needed for problem assessment, designing solutions,
and evaluating progress. This paper presents the design
of a prototype nonpoint source data center. The center is
being established based on needs identified through in-
terviews and questionnaires and is based on data from
Chesapeake Bay States. The center's information will be
available to all States and local governments and to the
public, and the database will be gradually expanded to
cover all States. Existing data and information centers
such as the Conservation Tillage Information Center
were evaluated during the study to allow the new non-
point source data center to build upon their experiences.
INTRODUCTION
There can be little argument that, in many areas, control-
ling nonpoint sources is the key to further water quality
improvements. However, we need to be able to build on
the experience and results of existing programs. Where
efforts have been made to control nonpoint source pollu-
tion, it has become clear that progress is difficult to docu-
ment. Furthermore, few mechanisms exist for sharing in-
formation and results among the many workers in this
field.
The emphasis on cleanup of nonpoint source pollution
in the Chesapeake Bay dramatizes the need for effectively
documenting best management practices (BMP's) and
other activities. Although an effort to coordinate the wide
variety of information about the Bay is underway, the effort
does not extend to the systematic tracking of BMP's, and
it is only concerned with the Bay. A central point for coordi-
nating information about all types of nonpoint source prob-
lems (not just agriculture) nationwide would serve not only
the Chesapeake Bay effort, but the other nonpoint source
programs underway in most areas of the country.
We need solid data sources upon which to base non-
point source assessments. Our focus is on how decision-
makers can use the data in establishing programs and
assessing progress. The experiences of the Lake Erie
Wastewater Management Study, for example, can be used
to guide similar studies elsewhere (U.S. Army Corps Eng.
1982).
This paper presents the design of a prototype nonpoint
source data center and some of the information on which
this design is based. In particular, the paper:
• Addresses the types of information program man-
agers say they need,
• Reviews the extent to which some existing informa-
tion systems and centers may meet needs for nonpoint
source data, and
• Outlines a proposal for a new nonpoint source data
and information center to provide tracking and information
services.
The proposal suggests possible funding sources and
staffing needs, but these are secondary to the main is-
sues: who is to be served by the center, and how their
needs are to be met.
METHODS
The National Association of Conservation Districts
(NACD) studied the needs for a nonpoint source data cen-
ter with the assistance of an advisory committee drawn
from several Federal agencies, States, and groups work-
ing in nonpoint source control. These advisors included
individuals with expertise in various types of nonpoint pol-
lution, since the goal of such a center, if proposed, would
be to include all nonpoint sources, not just agriculture.
Following an initial meeting to discuss study goals and
methods, we began investigating data sources, conduct-
ing interviews, and considering possible ways of structur-
ing a data center.
The study team interviewed State water and soil conser-
vation program managers to gather information on pro-
gram needs and to test questions to be asked of key State
conservation district officials and others through a written
questionnaire.
Another area of inquiry was the availability and useful-
ness of some of the existing and planned data and infor-
mation centers. The study team visited the Conservation
Tillage Information Center and the U.S. Environmental
Protection Agency's Chesapeake Bay Program. We tele-
phoned others, such as the Susquehanna Basin Commis-
sion, to ask questions about the audiences reached and
availability of data bases and to request copies of annual
reports and lists of publications. We are currently investi-
gating other information sources.
RESULTS: DATA NEEDS AND USERS
Do Existing Systems Meet Needs?
Participants in the interviews identified several needs for
tracking capability and information handling that could be
effectively served by a new data center. We also found
aspects of State programs in the Chesapeake Bay area
that could be adapted and used by other States, such as
Virginia's system of tracking and calculating reductions in
sediment and nutrients reaching water courses because
of BMP installation. However, when needs are considered
for the whole country, existing and planned information
and data centers leave some gaps in tracking progress
and coordination of information sources related to non-
point source control.
What Needs Might Be Met by a Data Center?
Respondents described many ways a data center could
assist their programs. The needs they identified spanned
a wide range of topics. Some of them are discussed in the
following paragraphs.
Tracking BMP Implementation and Maintenance.
State program managers want to track BMP implementa-
tion and maintenance, be they publicly or privately
433
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
funded. Such a system could include information on soil
characteristics and biomes or biological communities that
would aid in evaluating BMP effectiveness. Data must be
keyed to hydrologic units as well as to specific project
locations. Also, information must be coordinated and re-
ported among States (for example, a common system for
all the States in the Chesapeake Bay Basin), and data
must satisfy requests from legislators and regulators.
Linking Soil Conservation and Water Quality Im-
provements. This item is closely related to BMP tracking.
Virginia soil and water conservation officials used a model
that can predict reduction of phosphorus transported to
receiving waters, given sufficient knowledge of soil type
and BMP effects. The Idaho batholith BMP effectiveness
study by the U.S. Forest Service also demonstrates this
type of linkage. Ability to link soil conservation and water
quality improvements will help fill another need: that of
demonstrating progress to farmers and legislators. A com-
mon database will make available progress reports in any
area of interest, no matter how large or small.
Monitoring Atmospheric Deposition of Pollutants.
Although it has not received much public recognition, at-
mospheric deposition is a major cause of water pollution
and acidity in many areas. Sources include airborne soil,
drift of aerially applied pesticides, and washout or fallout
of pollutants from industrial emissions. Calculated loads of
phosphorus from the atmosphere in the Great Lakes
range from 500 to 1,600 metric tons/yr, depending on the
size of the lake (U.S. Environ. Prot. Agency, 1983).
Accessing Various State Guidelines. Some States
have developed guidelines for making judgments about
BMP's and water quality effects of nonpoint sources.
These guidelines can be adapted by other States to serve
local needs. Information on measures of water quality may
also be shared, for example, surrogate indicators that may
be less expensive than some standard tests. Common
definitions of BMP's and watershed size could be main-
tained by a data center to provide a uniform basis for
communication. Data on the cost effectiveness of various
BMP's also fall under this heading.
Sharing Research Reports. All respondents sug-
gested that an important service of a nonpoint source data
center would be to collect information on current and past
nonpoint source research. Many institutions are conduct-
ing research projects, but until the results are published
many others are not even aware that such projects are
underway, and duplicative research might result. Early
results and a central listing of projects and researchers
could be extremely useful to State and local agencies.
Communicating State and Local Program Informa-
tion. Many State nonpoint source programs are evolving
to allow for new cost-sharing. State staff want to review
the experiences of other States and choose the best for
their own programs. Items needed include laws and regu-
lations for various watershed, land use, and management
situations (both actual and model), BMP success and fail-
ure stories, and evaluation of various types of incentives.
Providing Technical Assistance. Technical assistance
is always identified as an important need by State and
local program managers. For example, States and conser-
vation districts want access to experts in various fields,
which could be provided through an information center.
Who Will Use Nonpoint Source Data Center?
The primary users of a nonpoint source data center are
expected to be State soil and water agency and local con-
servation district staff and managers. They have the most
pressing needs for nonpoint source data and information,
especially to prepare program reports for EPA and State
legislatures. Conservation districts in particular serve as
local nonpoint source control and management agencies
with cooperation and support from groups such as the Soil
Conservation Service and Extension Service. Depending
on the ultimate functions supported, a coordinated data
center might provide far-reaching reports and augment
State staff in collecting and analyzing data. If a data center
collected nonpoint source information from a wide area, it
would aid each State in evaluating its program and as-
sessing progress in nonpoint source control.
Federal agency officials would be able to use the center
in developing national reports of progress in improving
water quality through nonpoint source control and for de-
veloping and evaluating policies and national programs.
The U.S. Geological Survey, EPA, and the U.S. Depart-
ment of Agriculture are among potential users. The avail-
ability of data collected according to standard reporting
items and subjected to quality assurance procedures will
improve the confidence of program managers—both Fed-
eral and State—in their ability to assess progress.
The public, including farmers, homeowners, and oth-
ers, is the third group that will certainly use a nonpoint
source data center, both directly and indirectly. The center
can answer inquiries directly, but it will also respond to
State and local agencies that receive information requests
from their constituencies. An important function of any
nonpoint source center will be to serve as a network focus
for these users and to develop effective mechanisms to
inform potential users of the services and information
available from the center.
Do Existing Systems Meet Some Needs?
Some existing systems do meet many needs for nonpoint
source information. Four of them are described in the sec-
tions that follow. In general, though, existing systems treat
nonpoint source data incompletely or lack the focus
needed by decisionmakers. Many of them limit them-
selves to a restricted geographic area, and most are not
specific in their analysis of nonpoint source issues. In
most cases, scientific research related to nonpoint
sources is more heavily weighted than practical field solu-
tions; yet the latter are what most correspondents are af-
ter.
USDA Tracks Its Cost-Sharing. The Conservation Re-
porting and Evaluation System (CRES) is used by the Soil
Conservation Service and the Agricultural Stabilization
and Conservation Service. It includes information from:
• 335 counties statistically selected to report CRES
data;
• PL. 566 Watershed Protection and Flood Prevention
projects;
• Resource conservation and development projects
(RC&D) providing financial assistance for land treatment
cost-sharing; and
• Rural Clean Water Program (RCWP) projects carry-
ing out comprehensive monitoring and evaluation activi-
ties.
CRES is designed to produce a statistically reliable na-
tional progress report and will provide summary reports of
individual PL. 566 and RC&D projects. However, because
of the statistical nature of data, analysis of smaller areas
such as specific farms or counties is not valid. Also, data
are reported by county, not by hydrologic unit; thus land
area affected by BMP's cannot be readily tied to effects on
water quality. Current reporting in CRES does not account
for BMP's installed without Federal cost-sharing, although
it does track technical assistance by the Soil Conservation
Service.
The Chesapeake Bay Program Handles Many Inquir-
ies. EPA's Chesapeake Bay Program and related organi-
zations such as the Citizen's Program for the Chesapeake
434
-------
DATA AVAILABILITY AND NEEDS
Bay; Inc., provide central points of contact for information
specifically related to the Bay. Activities of the Citizen's
Program include publication of a newsletter. The EPA Bay
Program maintains a database of the Chesapeake Bay
research data, accessible through an information system
called CHESSEE. This database is not designed to pro-
vide analysis of small detailed areas, such as effects of
BMP implementation on a particular watershed.
EPA's Bay Program staff are developing a Bay-related
network to improve coordination of information among all
groups working on problems in the Bay drainage area,
with the goal of making better use of existing information
rather than generating new information. However, this pro-
gram does not track BMP implementation and other tech-
nical data. It is also not set up to provide program progress
and evaluation functions.
Susquehanna Basin Commission Serves Pennsylva-
nia and New York. The concern of the Susquehanna Ba-
sin Commission is reflected in its name. The Commission,
recognizing the contribution of the Susquehanna River
and its tributaries to pollution in the Chesapeake Bay, has
conducted monitoring programs to determine sources and
magnitude of nonpoint source problems. However, non-
point source pollution is only one of several mandates of
this commission; and, because it is concerned with a par-
ticular geographic location, its utility is limited for program
managers who consider the "big picture."
A National Water Resources Research Center and
Information Clearinghouse Has Been Proposed. A
study for the Council on Environmental Quality by the
Chesapeake Research Consortium reported several op-
tions for research and information functions related to wa-
ter resources. The information-related options include a
referral center, an information clearinghouse that would
obtain material for requestors, and a national coordinating
center for regional or State water information clearing-
houses (Chesapeake Res. Consor. 1984). Serving a very
broad audience, the center would include information on
all water topics, not just nonpoint source pollution. How-
ever, the information-related proposal describes a "pas-
sive" center, which would use various existing sources
and not collect its own data; one research center proposal
included data collection, but no plans for funding have
been made by Congress yet. A nonpoint source data cen-
ter would be a source on which a water resources informa-
tion clearinghouse would be likely to draw for some of its
information needs.
PROPOSAL: SPECIFICATIONS FOR
NONPOINT SOURCE DATA CENTER
What Structure and Functions Should the
New Center Have?
Location of a Center
A nonpoint source data center could be established in
several ways, each with its own advantages. The center
could be:
• Sponsored by a private organization,
• Established in conjunction with a State water re-
source center, or
• Located at a Federal or State agency.
A nonpoint source data center established as a compo-
nent of a private organization would have the advantage of
equal access to all. It would be independent and would be
supported by a board of directors representing a cross-
section of interests concerned with nonpoint source
abatement. Even if supported financially by some of its
constituent groups, having many sources of support
would tend to dilute the possible influence of any one
group on policy or activities.
A center established at a State water resource center
could easily tap existing data systems. The process of
beginning operation might be eased, presuming appropri-
ate space is available at the State center. However, State-
related information and problems might predominate, par-
ticularly if some staff are shared between the two centers.
Located at a Federal or State agency, a nonpoint source
data center would have access to some necessary exper-
tise. However, by establishing close ties to that agency it
might lose objectivity in data collection and analysis.
Funding and Staffing
A nonpoint source data center could be supported in a
variety of ways and probably should rely on a combination
of sources. The potential availability of these sources has
not been evaluated, so they are merely listed here: gov-
ernment grants, corporate contributions, and membership
fees.
The staff should include individuals with experience in
nonpoint source control, information management, and
computer data analysis; they could be hired full- or part-
time, or as consultants. Staff could be assigned from Fed-
eral agencies, as has been done with the Extension Serv-
ice and the Conservation Tillage Information Center.
Center Functions
The study has identified several key needs that a new
nonpoint source data center could fill. Some of the func-
tions are currently handled in States and other centers,
but existing mechanisms often serve only one State or
limited geographic area. If a State or other group has an
effective method of data collection or analysis, the non-
point source center would try to use or adapt it. Although
we cannot go into all aspects of the institutional structure
of a nonpoint source data center here, the results of this
study show that such a center should include at least the
following functions to be responsive to existing needs:
• Actively seek information on current research proj-
ects and reports of completed work, and maintain biblio-
graphic access systems (including information on atmo-
spheric deposition);
• Gather information on State nonpoint source pro-
grams and develop fact sheets and analyses to facilitate
access to the information;
• Compile State and local laws and regulations and
case studies of BMP successes and failures;
• Provide technical assistance to users through litera-
ture searches, advice on BMP's, assistance with develop-
ment of pollution potential maps, and other projects;
• Develop the capability to analyze BMP information
and provide reports on water quality and other effects to
State agencies, Federal agencies, individuals, and inter-
ested groups;
« Work with EPA to create or adapt a computer-based
system to track BMP implementation in selected areas;
and
• Perhaps most important, develop effective mecha-
nisms to inform potential users of the information and
services available from the center.
CONCLUSIONS
There is a need for a nonpoint source information center,
probably including a data component, although the latter
may be difficult and expensive.
A new nonpoint source data center should probably be
separate from existing and proposed water resource or
435
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
water quality centers. Some economies are associated
with integrating the new functions into an existing institu-
tional setting; however, such economies are outweighed
by the need for increased visibility of and emphasis on
nonpoint source pollution. Historically, nonpoint source
concerns have received lower priority than other water
quality programs. Establishment of a separate data center
devoted to nonpoint source concerns would help raise the
level of awareness about this issue.
A nonpoint source center must focus on information
and data for program management and decisionmaking,
as opposed to the strict research orientation of most exist-
ing and proposed data resources. In addition, it should
eventually collect and analyze data and produce reports
to meet the needs of its clients.
As outlined above, its functions may be an ambitious
undertaking, but they need not be accomplished simulta-
neously At first the center might concentrate on serving a
limited area and on only one source (e.g., agriculture or
siliviculture). In developing a final proposal for a center,
the need for various activities will be assessed, and the
top one or two will be initiated, with others to follow as
resources allow. Thus, a center can be developed and
grow in a planned, logical way.
ACKNOWLEDGEMENTS: The author thanks Bob Williams and
Mary Garner of NACD for their assistance during the project and
their comments on this paper. This project is supported by a
grant from the U.S. Environmental Protection Agency. The ideas
presented here are those of the author and various interviewees
and do not necessarily represent the views of the author.
REFERENCES
Chesapeake Research Consortium. 1984. Alternatives for a Na-
tional Water Resources Center and Information Clearing-
house. Pub. No. 120 CRC, Inc., Gloucester Point, VA.
U.S. Army Corps of Engineers. 1982. Lake Erie Wastewater
Management Study- Final Rep.
U.S. Environmental Protection Agency. 1983. Region 5 Environ-
mental Management Report, Parts 1 and 2.
436
-------
WATER QUALITY DATA AND URBAN NONPOINT SOURCE
POLLUTION: THE NATIONWIDE URBAN RUNOFF PROGRAM
DIANE NIEDZIALKOWSKI
DENNIS ATHAYDE
U.S. Environmental Protection Agency
Washington, D.C.
ABSTRACT
Prior to 1960, concerns about urban stormwater were
related primarily to flooding and drainage problems. More
recent studies have focused on characterizing and quan-
tifying pollutants in stormwater or developing methodolo-
gies for reducing loads. While such research contributes
to understanding urban nonpoint source runoff, it does
little to illuminate the cause/effect relationship between
stormwater and associated water quality of the receiving
water. Water quality planners need information to guide
their stormwater management decisions to help them de-
termine: (1) if there is a problem; (2) the significance of
the problem; and (3) whether they need to do anything
about it. Lack of good data and appropriate methodolo-
gies have often been cited as the reason it is difficult to
relate stormwater runoff to the water quality associated
receiving waters. This paper describes the Nationwide
Urban Runoff Program (NURP) sponsored by EPA which
was undertaken to respond to these needs by providing
support, data, and methodologies for urban nonpoint
source problem assessment and water quality planning.
INTRODUCTION
The reduction of pollution from industrial and municipal
point sources has been the main regulatory focus of water
pollution control efforts since the passage of the Clean
Water Act in 1972. Less emphasis has been placed on
nonpoint source pollution, those pollutants mobilized by
storm events and transported by runoff across the land
surface. Point source pollution controls alone are insuffi-
cient to meet the objectives of the Clean Water Act. Re-
search by Gianessi and Peskin (1981) shows that for some
waterbodies, nearly all of the phosphorus and nitrogen,
more than half of the biochemical oxygen demand (BOD),
and many toxic substances are contributed by rural and
urban nonpoint sources. In their most recent biannual re-
ports to the U.S. Environmental Protection Agency on pro-
gress toward achieving water quality objectives and desig-
nated uses, the majority of the States have identified
nonpoint sources as the principal cause of remaining wa-
ter quality problems (U.S. Environ. Prot. Agency, 1985).
The focus of this paper is on the Nationwide Urban
Runoff Program (NURP), a data collection effort spon-
sored by the Environmental Protection Agency from 1978
to 1984 to improve the information available for assessing
urban nonpoint source pollution.
NONPOINT SOURCE POLLUTION
General Overview
The nature of nonpoint source pollution makes it more
difficult to characterize, quantify, and control than pollution
from point sources. Even so, the following characteristics
The opinions expressed in this paper are those of the authors and do not necessarily
reflect any policies or decisions of the U.S. Environmental Protection Agency.
are associated with both rural and urban nonpoint sources
of pollution:
1. Pollution is generated by a wide variety, and large
number, of activities rather than discrete, identifiable
sources.
2. Pollution is conveyed to surface waters by runoff,
stormwater culverts, or groundwater percolation.
3. Pollution is intermittent because of its relationship to
storm events and the hydrologic cycle.
4. Pollution is difficult to detect because of the low fre-
quency and short duration of storm events.
Urban nonpoint source pollution is a widely occurring
problem, estimated to affect 20 percent of the river miles
located in more than half of the watersheds in the United
States (U.S. Environ. Prot. Agency, 1984).
The predictability and uniformity of effluents from point
sources (particularly discharges from sewage treatment
plants) allow end-of-pipe technologies to be used for meet-
ing effluent discharge limits. However, because urban run-
off is associated with storm events and the hydrologic
.cycle, it is generated in pulses. Therefore, no "technologi-
cal fix" can be uniformly applied to reduce nonpoint
source inputs to receiving water, even though nonpoint
source runoff may include many of the same pollutants
found in point source effluents.
Urban Runoff: Quality and Quantity
Urban development generally results in changes in land
use that reduce the land surface area, allowing water to
infiltrate while increasing impervious surface areas such
as roof tops, streets, and sidewalks. For a given storm
event, an urban area will contribute a larger volume of
runoff more quickly than an undeveloped area. Such in-
creases in runoff rate and total volume often considerably
effect erosion, flooding, and the quality of the receiving
water. Thus, urban runoff can be viewed as a two dimen-
sional problem resulting from the quantity and quality of
runoff produced by storms.
Effects on Water Quality
Many of the pollutants commonly found in stormwater run-
off, including sediment, nutrients, metals, toxics, and bac-
teria can affect the quality of urban streams and lakes.
Sediment and silt are carried in runoff from streets, con-
struction sites, and eroding land. The pollutants associ-
ated with the sediments from these sources and urban
activities are generally not as benign as the natural min-
eral sediments that result from soil erosion. Sediment is
often the major pollutant in urban runoff, and is associated
with the following problems: (1) decreased carrying capac-
ity in storm water sewer systems, resulting in greater flood
potential; (2) increased dredging costs for maintaining
navigation channels; (3) higher pretreatment costs for mu-
nicipal and industrial users depending on sources af-
fected by runoff; (4) aesthetic degradation; and (5) trans-
port of phosphorus, pesticides, and toxics.
The effects of surface particulates (from tire wear, auto
exhaust, and road deterioration) on receiving waters tend
437
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
to be more chemical than physical. Nutrients, particularly
phosphorus and nitrogen, are contributed to runoff by fer-
tilizers applied to parks and lawns. Additional sources of
nutrients (and bacteria) are pet wastes and malfunctioning
septic systems. In some areas, atmospheric deposition is
a significant source of phosphorus. Heavy metals, such as
lead, copper, and zinc are not uncommon in urban runoff.
Although heavy metals and organic chemicals do not usu-
ally produce an acute and immediately observable impact,
such as a fishkill, these pollutants may accumulate in liv-
ing tissue or sediment and may have long term detrimen-
tal effects on individual organisms and ecological com-
munities.
Effects From Quantity of Runoff
The potential impacts urban runoff volumes on receiving
water, particularly on streams, equal the water-quality ef-
fects in importance. Increased flow and velocity in
streams from stormwater inputs can erode streambanks
and resuspend bottom sediments, increasing turbidity in
the receiving water. Disturbed sediments may also in-
crease the availability of sediment-bound phosphorus, en-
couraging algal growth and accelerating eutrophication.
Toxic substances, which may be relatively innocuous
when bound to undisturbed sediments, can become avail-
able when resuspended.
The volume of runoff and corresponding pollutant load
is related to a number of factors, including intensity and
duration of the storm, length of time since the last storm,
and land use. Urban activities that disturb land cover and
alter natural drainage patterns also affect urban runoff.
Figure 1 from the Results of the Nationwide Urban Runoff
Program; Vol. 1—Final report (U.S. Environ. Prot. Agency,
1983) illustrates the relationship between paved land area
and runoff. As the percentage of paved surfaces in a given
area increases, the rate of runoff and corresponding pollu-
tant load also increases.
40%
EVAPO
TRANSPIRATION
NATURAL
GROUND
COVER
10% RUNOFF
25%
SHALLOW
INFILTRATION
THE NATIONWIDE URBAN RUNOFF
PROGRAM (NURP)
Early Urban Runoff Concerns
Drainage, be it nuisance flooding of basements or cata-
strophic floods resulting in loss of life and property, has
historically been the primary concern in urban runoff. Re-
cently, concern has expanded to include the potentially
deleterious effects on water quality. Many of the early 208
Areawide Water Quality Management Plans from the
1970's indicated that urban runoff contributed to water
quality degradation. However, because data were not of-
ten collected during the development of 208 plans, the
relationship between urban runoff and water quality was
often difficult to assess. Where data were available con-
founding, physical and chemical reactions in the receiving
water caused additional complications. In those communi-
ties where urban nonpoint problems had been clearly
identified, reluctance to commit money for control devices
of questionable effectiveness understandably existed.
The NURP Program
In 1978 EPA responded to the need for consistent and
verifiable data on urban runoff by initiating a multiyear
study called the Nationwide Urban Runoff Program
(NURP). NURP had three major objectives: (1) develop
consistent and verifiable data on the quality of urban non-
point source runoff and the effects on the receiving water;
(2) develop practical data on the relative costs and effec-
tiveness of control measures; (3) respond to Congression-
al concern over whether urban runoff was a large enough
problem to mandate a control approach similar to that in
the Construction Grants program for treating stormwater.
Data were collected at 28 separate project sites, chosen
from among the 1976 Areawide Plans that had identified
38%
EVAPO
TRANSPIRATION
JO-20%
PAVED
SURFACES
20% RUNOFF
OEP
INFILTRATION
21%
SHALLOW
INFILTRATION
DEEP
INFILTRATION
25%
21%
35%
EVAPO
TRANSPIRATION
30% RUNOFF
35-50%
PAVED
SURFACES
20%
SHALLOW
INFILTRATION
DEEP
INFILTRATION
15%
Figure 1.—Typical changes in runoff flows resulting from
paved surfaces.
30%
EVAPO
TRANSPIRATION
75-100%
PAVED
SURFACES
10%
SHALLOW
INFILTRATION
5%
DEEP
INFILTRATION
Source: J.T Tourbier and R. Westmacott, Water Resources Protection Technology: A
Handbook ol Measures to Protect Water Resources in Land Development,
p.3.
438
-------
DATA AVAILABILITY AND NEEDS
urban runoff as a significant problem (Fig. 2). The projects
were geographically dispersed throughout the United
States in an attempt to collect representative data on cli-
mate, soils, geology, and other factors which interactively
affect the quality and the quantity of runoff from the Na-
tion's diverse urban areas. Each NURP project was man-
aged independently at the local level with oversight and
technical assistance in developing standard data collec-
tion methodologies from EPA Headquarters. Consistency
in data collection ensured that differences in data among
the projects were related to the urban area, not methodol-
ogy
Five categories of standard pollutants were monitored
in stormwater runoff and ambient receiving water at each
of the 28 NURP sites (Table 1). At 20 of these locations,
samples were also collected for priority pollutants, includ-
ing pesticides, metals, PCB's, and organics. Nine of the
projects collected engineering and economic data on best
management practices (BMP's), and BMP systems for
controlling runoff. Data were collected on wet and dry
detention basins, street sweeping, and two "living filter"
approaches—grassy swales and wetlands.
RESULTS
Objective I: Data Base Development
The first objective of NURP was met by developing a data
base for characterizing urban runoff and evaluating the
impacts of runoff on receiving waters. Data were collected
from 2,300 separate storm events, at 22 NURP sites, in 81
drainage basins within the NURP sites. Pollutant concen-
trations were represented as averages using a summary
statistic called an "event mean concentration" (EMC),
chosen to .meet NURP's objective of characterizing runoff
by a pollutant's average value for a given storm and site,
not its fluctuations within that storm event.
Urban runoff is stochastic and is generated in pulses in
association with unpredictable storm events. The data
were analyzed using an innovative statistical approach
that explicitly considered the inherent variability of the run-
off data. This methodology is similar to that used in pre-
dicting the frequency of occurrence of floods. The NURP
data base was used to develop a screening model for
EPA
Region
1
II
III
IV
NURP
Code
MAI
MA2
NH1
NY1
NY2
NY3
DC1
MD1
FL1
NCI
SC1
TN1
Project Name/Location
Lake Quinsigamond
(Boston Area)
Upper Mystic (Boston
Area)
Durham, New
Hampshire
Long Island (Nassau
and Suffolk Counties
Lake George
Irondequott Bay
(Rochester Area)
WASHCOG
(Washington, D.C.
Metropolitan Area)
Baltimore, Maryland
Tampa, Florida
Winston-Satem, North
Carolina
Myrtle Beach, South
Carolina
Knoxville, Tennessee
EPA
Region
V
VI
VII
VIM
IX
X
NURP
Code
IL1
IL2
Mil
MI2
MI3
WI1
AR1
TX1
KS1
CO1
S01
UT1
CA1
CA2
OR1
WA1
Pro|ect Name/Location
Champalgn-Urbana,
Illinois
Lake Ellyn (Chicago
Area)
Lansing. Michigan
SEMCOG (Detroit Area)
Ann Arbor, Michigan
Milwaukee, Wisconsin
Little Rock, Arkansas
Austin, Texas
Kansas City
Denver, Colorado
Rapid City South
Dakota
Salt Lake City, Utah
Coyote Creek (San
Francisco Area)
Fresno, California
Springfield-Eugene,
Oregon
Bellevue (Seattle Area)
Figure 2.—Locations of the 28 NURP projects.
439
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 1 .—Standard pollutants and bacteria adopted by the
NURP study to characterize the pollutants in urban runoff.
1. TSS—Total Suspended Solids
2. BOD—Biochemical Oxygen Demand
3. COD—Chemical Oxygen Demand
4. TP—Total Phosphorus (as P)
5. SP—Soluble Phosphorus (as P)
6. TKN—Total Kjeldahl Nitrogen (as N)
7. N—NO2 + NO3 (Nitrite + Nitrate) (as N)
8. Cu—Total Copper
9. Pb—Total Lead
10. Zn—Total Zinc
11. Total Coliform
12. Fecal Coliform
predicting instream concentrations of pollutants contrib-
uted by urban runoff. Data from 30 to 40 storm events, 50
years of streamflow, and 50 years of rainfall are needed to
use the model as a screening tool. The reliability of the
model was tested at the NURP site in Rapid City, South
Dakota. The test results found the model very accurate in
predicting in-stream pollutant concentrations that approxi-
mated the concentrations actually found in the monitored
samples.
The data showed that geographic location, land use
category, and other factors such as slope or population
density cannot be used to predict pollutant concentrations
in runoff. However, land use category was useful for pre-
dicting the volume of runoff generated since it was shown
to be a function of the percentage of paved or imperme-
able surfaces in a given land area.
Oxygen-demanding substances and total suspended
solids were usually present in runoff, sometimes in con-
centrations comparable to effluents from secondary sew-
age treatment plants. Nutrients were generally present,
and with few exceptions, the concentrations were found to
be insignificant compared to other pollutants.
Those sites monitoring runoff for priority pollutants
found that heavy metals, particularly lead, copper, and
zinc, were by far the most frequently detected priority pol-
lutants. Acute water quality criteria were exceeded for
copper in 47 percent of the samples, and for lead in 23
percent. Exceedences of chronic water quality standards
for priority pollutant metals were detected in 94 percent of
the samples for lead, 82 percent for copper, and 77 per-
cent for zinc. These data represent runoff characteristics,
and do not necessarily imply that an actual violation of
ambient water quality standards occurred.
Objective II: BMP Costs and Effectiveness
The BMP's that were evaluated in the NURP program can
be grouped in four categories: (1) detention devices; (2)
recharge devices; (3) street sweeping; and (4) other, which
included the "living filter" approach—grassy swales and
wetlands. Local participants choose which BMP control
options to study. Their choices reflect local perceptions of
what may be feasible and practical to implement.
Six NURP projects evaluated detention devices. Ade-
quately sized wet detention basins demonstrated removal
efficiencies in excess of 90 percent for particulates, total
suspended solids, and lead. Biological processes pro-
duced significant reductions (50 percent or more) in solu-
ble nitrates and phosphorus. Dry basins, designed to at-
tenuate peak runoff, and therefore, only briefly detain
stormwater, are ineffective in reducing pollutant loads. Ap-
proximate costs for wet ponds sited within urban areas
with relatively smalr populations ranged from $500 to
$1500 for each acre of urban area served. Offsite ponds
serving larger urban areas were slightly less expensive,
ranging from $100 to $250 for each acre of urban area
served. The differences in costs relate to the economies of
scale for larger ponds.
The possible contamination of ground water by control
measures that enhance runoff infiltration has caused
some concern. Ground water was monitored from aqui-
fers underlying urban runoff detention basins on Long Is-
land, New York, and in Fresno, California. Monitoring data
showed that heavy metals, organic priority pollutants, pes-
ticides, and coliform bacteria were attenuated in the soil
matrix and prevented from reaching the ground water.
Street sweeping was evaluated at 5 of the 28 NURP
projects to determine the effectiveness in reducing the
accumulation of contaminants on streets. A statistical
analysis of the data showed no significant reduction in the
event mean concentration for lead, total suspended
solids, or chemical oxygen-demanding substances be-
tween swept and unswept streets for four of the sites. At
one site with pronounced wet and dry seasons, sweeping
just prior to the rainy season possibly reduced accumu-
lated pollutants and lessened the pollutant load washed
off streets in runoff. The unit cost of operating a sweeper
varied from $16.80 to $45.45 per hour of operation; and
from $5.95 and $23.36 per curb mile swept.
The two living filter approaches to urban runoff manage-
ment studied were grassy swales and wetlands. Grassy
swales were studied in two NURP projects. At one project,
monitoring data from three swales showed no effect in
attenuating pollutants. At the second site, a carefully de-
signed artificial swale reduced heavy metal pollutants by
approximately 50 percent. Chemical oxygen-demanding
pollutants, nitrates, and ammonia were reduced about 25
percent. These NURP results suggest that grassy swales
are a practical and potentially effective technique for man-
aging urban runoff if they are designed carefully. Wetlands
are considered by many to be a promising technique for
control of urban runoff water quality. One project moni-
tored a natural wetland, but the investigation was not ade-
quate to identify wetland characteristics or artificial wet-
land design specifications with performance capability.
Objective III: Need for Another Construction
Grants Program
The third objective of the NURP study was to respond to a
Congressional concern of whether urban runoff was sig-
nificant enough to mandate a stormwater treatment pro-
gram on the magnitude of the Construction Grants pro-
gram.
An analysis of rainfall records for a wide distribution of
locations across the United States showed that storm
events that produce urban runoff only occur about 10 per-
cent of the time, even for the wetter parts of the country.
Pollution from runoff is only a problem for a few hours per
month, at intervals of every several days or longer, de-
pending on the season and geographic location. The
NURP study concluded that urban runoff was not a high
priority problem needing a construction grants approach
because of the infrequency of storm events, the short
duration of the individual storm events, and the relative
harmlessness of runoff compared to other sources of wa-
ter pollution.
CONCLUSIONS
Since nonpoint source pollution is site-specific, a case-by-
case evaluation is needed to pinpoint problem areas and
identify solutions. The screening model developed during
the NURP study provides a tool for local planning agen-
cies to use with their own monitoring data to identify the
most significant runoff problems. The monitoring data can
be used with the model to simulate the effects of alterna-
tive strategies on the pollutant concentrations in the re-
ceiving waters. Local governments can use this informa-
440
-------
tion to develop urban nonpoint management strategies
that are the most environmentally sound and cost-effec-
tive solutions to the site-specific problems.
The NURP project has precipitated a number of issues
and questions that warrant further investigation. Uncer-
tainties remain about the relationship between sources of
pollutants, their fates, and downstream impacts. The po-
tential long-term cumulative effects of nutrients and toxic
pollutants in the sediments or urban lakes and streams
also warrants further investigation.
The NURP database, which is being put into a more
useable format contains a wealth of information that could
provide insights to these uncertainties. Specific areas of
future investigation might include:
• Is there a relationship between the pH of rainfall, the
pH of urban runoff, and the concentrations of certain pol-
lutants? What does this relationship tell us about acid pre-
cipitation?
• What are the historical trends in pollutant concentra-
DATA AVAILABILITY AND NEEDS
tions between basins at a given site, and among NURP
sites nationwide?
• Is there a relationship between lead levels in runoff
and consumption of lead-free gasoline?
• What are the long-term effects of priority pollutants
that exceed acute and chronic freshwater criteria?
REFERENCES
Gianessi, L.P., and H.M. Peskin. 1981. Analysis of National Wa-
ter Pollution Control Policies. Vol. 2. Agricultural sediment
control. Water Resour. Res. 17(24): 9-27.
U.S. Environmental Protection Agency. 1983. Results of the Na-
tionwide Urban Runoff Program; Vol. 1—Final Report. Wash-
ington, D.C.
1984. Report to Congress: Nonpoint Source Pollu-
tion in the U.S. Washington, O.C.
_. 1985. National Water Quality Inventory: 1984 Report
to Congress. Washington, D.C.
441
-------
THE RFF NATIONAL DATA BASE FOR NONPOINT SOURCE POLICY
ASSESSMENTS
LEONARD R GIANESSI
Resources for the Future
Washington, D.C.
ABSTRACT
This paper discusses a newly developed data base for
national quantitative assessments of the nonpoint source
water pollution problem. A variety of methods have been
used to integrate often disparate sources of information
into a unified inventory of nonpoint sources. The type of
quantitative information needed to locate and quantify
nonpoint source levels includes a wide variety of demo-
graphic, hydrologic, meteorologic, engineering, and field
measurement data. A variety of benchmark years has
been employed in the various sources of data. Lack of
data for certain parameters requires the assumption of
average values for very large regions. While it is desir-
able to integrate point and nonpoint sources of water pol-
lution into the same inventory, often this means conver-
sion of nonpoint source data using extremely simplified
procedures. The type of information required and actually
used for validation of the estimates contained in the RFF
data base is discussed. Finally, suggestions are made for
priorities in the future collection of data and information to
make it possible to improve the accuracy of the esti-
mates.
INTRODUCTION
The Resources for the Future Environmental Discharge
Inventory (REDI) describes the discharge of 17 pollutants
and nonpoint sources throughout the United States: 5-day
biochemical oxygen demand, total suspended solids, total
phosphorus, total Kjeldahl nitrogen, chemical oxygen de-
mand, lead, cadmium, chromium, copper, mercury, iron,
arsenic, zinc, oil, PCB's, coliforms, and other chlorinated
hydrocarbons.
The data reflect average daily and average annual dis-
charge levels for a recent year, circa 1977-81. All records
in the inventory are identified by county location, U.S.
Geological Survey hydrologic unit location, and Standard
Industrial Classification. Summaries are easily prepared
at the national, State, Aggregated Subarea (ASA), USGS
subbasin, and Standard Industrial Classification (SIC)
level.
For point sources, these estimates are made for approx-
imately 40,000 individual plants—municipal sewage
plants and industrial facilities. In addition, urban runoff
records are included for all cities in the United State with
more than 10,000 inhabitants.
The data sources and methods used to make the point
source and urban runoff discharge estimates have been
fully described in a recent report to the USGS (Gianessi
and Peskin, 1984).
This paper focuses on the methods and data used to
assemble the rural nonpoint source inventory developed
as part of an ongoing project sponsored by the Water
Resources Division of USGS, the Soil Conservation Serv-
ice of the U.S. Department of Agriculture, and the U.S.
Environmental Protection Agency. The data and results
support several activities included in USDA's 1985 Re-
soure Conservation Act Report to Congress.
SEDIMENT FROM RURAL LANDS
The major part of our rural nonpoint source discharge
inventory consists of records that estimate agricultural
sediment and pollutants associated with sediment that
reaches waterways. The basic data source is the 1982
National Resource Inventory conducted by USDA (Soil
Conservation Serv. 1984). The Inventory provides esti-
mates of gross soil erosion (in tons of soil lost per acre) at
approximately 799,000 sample points throughout the
United States by using the Universal Soil Loss Equation.
The sample data were statistically extrapolated to cover all
nonfederal land in each U.S. county, yielding a national
estimate of gross soil erosion for nonfederal cropland
acreage of 1.8 billion tons per year. We used the Inventory
to compute county level estimates of gross soil erosion,
according to rural land use (cropland, pasture, range,
woodland) and according to soil texture (sand, loam, silt,
clay).
Our next step was to estimate the fraction of gross soil
erosion that becomes sediment lost to waterways. For this
purpose, USDA provided us sediment delivery curves that
estimate the fraction of gross soil erosion that becomes
sediment as a function of river basin drainage density and
soil texture. The estimated sediment delivery ratios are
higher for areas with more stream miles per unit of area
and with finer soil particles such as clay. A comprehensive
set of State drainage density estimates was provided by
USDA, as derived from the Phase II, 1977 National Re-
source Inventory (which was not repeated as part of the
1982 Inventory). We assumed that all counties in a State
have the same average drainage density as the State
value. By using county estimates of gross soil erosion
classified by soil texture, the county average drainage
density values, and the USDA curves, we estimated sedi-
ment discharge to waterways by county and by rural agri-
cultural land use.
For the Nation as a whole, we estimate that approxi-
mately 40 percent of the gross soil erosion on agricultural
land becomes sediment lost to waterways. For nonfederal
cropland, the national sediment loss estimate is 1.8 trillion
pounds per year. Although these estimates may appear to
be high, our concept of sediment delivery ratio differs from
the definition widely used in other studies. Those studies
define the ratio as the amount of sediment exiting from a
basin as a fraction of the gross soil erosion occurring
within the basin. Thus, deposition of soil in reservoirs and
flood plains within the basin is not included in the ratio.
Since our concept of the sediment delivery ratio includes
all soil lost to waterways within a basin, our estimates will
be higher.
Our next problem was to estimate the amount of pollu-
tants associated with the sediment that moves into the
waterways with the sediment. These pollutants include
nutrients (nitrogen and phosphorus), organics (which ac-
count for 5-day biochemical oxygen demand) and individ-
ual heavy metals (such as lead, copper, chromium, and
442
-------
zinc). For this task we used two sources of data that char-
acterize the average surface layer content of soils for Ma-
jor Land Resource Areas (MLRA's) throughout the United
States. The USGS has sampled surface soils throughout
the country, in a systematic program to determine the nor-
mal elemental composition of surface soil (Shacklette et
al. 1971). We overlaid maps of MLRA's onto the USGS
survey maps and assigned each USGS sampling point to
an MLRA for chromium, lead, copper, zinc, and phos-
phorus. Once all the sampling points were assigned to
MLRA's, the average for each element in the MLRA was
computed.
The second source of data was the USDA series of
individual State reports of Soil Survey Laboratory Data.
From this source we extracted the fraction of organic car-
bon and organic nitrogen at each sample point along with
the identification of the MLRA where the soil survey was
made. An average carbon and nitrogen value was then
computed for each MLRA and for each county within the
MLRA.
Several adjustments were made to the data to estimate
ratios for pollutants attached to the sediment. We as-
signed an enrichment factor of 2 for phosphorus, organic
nitrogen, and organic carbon, to estimate the higher con-
centrations expected in eroded soil (Tubbs and Haith,
1980). We also estimated 5-day biochemical oxygen de-
mand as 0.1 times the soil organic matter content (Midw.
Res. Inst. 1977).
The estimates of sediment-related pollutant levels for
the Nation are very large. For nonfederal cropland, we
estimate 8.2 billion pounds a year of BOD5 and 6.2 billion
pounds a year of nitrogen are lost to waterways. In con-
trast, for all points sources in the United States, we esti-
mate 6.1 billion pounds a year of BOD5 and 2.5 billion
pounds a year of nitrogen discharged to waterways.
To complete the rural land sediment discharge inven-
tory, we made two other sets of estimates. First, we
needed to make estimates for Federal lands since the
1982 Inventory is limited to nonfederal lands only. This
limitation could be significant, since some states (particu-
larly in the West) have sizable Federal land acreage. For
example, two-thirds of Utah is federally owned.
From the General Services Administration Inventory
Rle of Property Owned by the Federal Government (Gen.
Serv. Admin. 1983), we estimated Federal rural land use
acreage by county. We then assumed that the county per-
acre gross soil erosion and soil texture distribution esti-
mates derived from the 1982 Inventory for nonfederal
lands also applied to Federal lands in the county. Once we
estimated the gross soil erosion for Federal lands by
county, we followed the same steps to estimate sediment
loss and estimates of sediment-attached pollutants.
We also estimated countywide sediment loss and pollu-
tant estimates for other sources of erosion. These include
streambanks, roadsides, gullies, and construction sites.
Data on these sources were included as part of Phase II of
the 1977 Inventory and are readily available at the State
level. Again, the estimates are quite large: 1.1 billion tons
a year in gross soil erosion nationally. We prorated the
State estimates back to counties in proportion to area. We
also extrapolated the estimates to include Federal lands in
proportion to acreage. We arbitrarily assume that 67 per-
cent of the gross soil erosion from these sources become
sediment and that pollutant concentration in the eroded
soil is equal to 10 percent of the level for surface soil. After
all these steps, the resulting nonpoint source inventory
includes estimates of sediment and associated pollutant
loss from cropland, rangeland, pastureland, woodland,
streambanks, gullies, roads and construction sites from
both Federal and nonfederal lands for all 3,150 U.S. coun-
ties.
DATA AVAILABILITY AND NEEDS
LIVESTOCK RUNOFF
We have also estimated the quantities of pollutants dis-
charged in livestock runoff for all U.S. counties. To do this
we used the 1982 Census of Agriculture (U.S. Dep.
Comm. 1984) to estimate the number of animals by spe-
cies on farms by county and multiplied the totals by USDA
estimates of the quantity of waste voided by livestock and
poultry species in terms of pounds per animal per year
(Van Dyne and Gilbertson, 1978). USDA has also esti-
mated for each State livestock waste loss rates resulting
from volatilization, runoff, and seepage (Van Dyne and
Gilbertson, 1978). By arbitrarily assuming that one-third of
the loss rate is attributable to runoff to surface waters, we
estimate for each State (and each county in a State) the
amount of livestock waste lost to waterways. For most
States the estimated livestock runoff rate is as low as 0 to
6 percent.
We use pollutant coefficients that describe the average
characteristics of livestock runoff (Morton and Loehr,
1980) to estimate the amount of associated pollutants
(such as organics and nutrients) carried into waterways.
Although the aggregate quantities of pollutants in live-
stock runoff are not as large as the quantities associated
with sediment, livestock runoff accounts for almost 100
percent of the fecal coliforms discharged to waterways in
rural areas.
NUTRIENT RUNOFF
We used the Cornell Nutrient Simulation Model (Tubbs,
1980) to estimate annual loading of dissolved nitrogen and
dissolved phosphorus to surface waterways from crop-
land for the entire United States for nonsediment sources.
This model estimates direct runoff, percolation, and
evapotranspiration losses. It also estimates a monthly soil
nutrient budget for N and P, as a function of fertilizer in-
puts, crop nutrient uptakes, and losses of dissolved N and
P to runoff and percolation. In addition, transformation of
plant nutrients from fixed to plant-available forms are esti-
mated according to soil and meteorologic conditions.
Necessary input parameters for the Cornell model are
separated into three categories: meteorologic inputs, soils
data, and crop practice data. Meteorologic inputs are
needed to operate the stochastic meteorologic model, and
include average monthly precipitation, average number of
days with precipitation by month, and average summer
and annual air temperatures. These data were collected in
central weather stations for each of 42 meteorologic re-
gions.
The Cornell model was applied for a 10-year simulation
of unit loadings of runoff (cm/yr) and dissolved nitrogen
and phosphorus losses (kg/ha/yr) for each modeled crop
in each MLRA. The results for all MLRA's were averaged
into Land Resource Regions (LRR's).
Total harvested cropland acreage estimates were drawn
from the 1982 Census of Agriculture for each modeled
crop for all counties. Based on a counties MLRA assign-
ment, each county was assigned to a Land Resource Re-
gion. By multiplying a county's crop acreage estimates by
the appropriate LRR loading coefficients, annual runoff
and loads of phosphorus and nitrogen runoff were esti-
mated by crop. By summing across crops, estimates are
made of countywide dissolved nutrient runoff losses from
cropland.
The national totals for nitrogen and phosphorus dis-
solved in overland runoff are not large compared to the
amounts of nitrogen and phosphorus attached to sedi-
ment reaching waterways. However, it is often suggested
that the dissolved nutrients are almost entirely available
443
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
for plant growth while the sediment-associated nutrients
are not so readily available.
PESTICIDE RUNOFF
We have assembled a county-based file of annual pesti-
cide usage estimates by crop for 184 individual widely
used substances. It was necessary to develop the usage
data to estimate the quantities of pesticides lost to water-
ways. All chemicals on the EPA list for the National
Ground Water Survey and the California Priority Pollutant
List are included in the inventory.
The county pesticide usage file was derived from a
number of State and national usage files. The Economic
Research Service of USDA (ERS) surveyed the annual
amounts of individual substances used in 1982 on 13 ma-
jor crops in 33 States. These crops include corn, cotton,
rice, and wheat (Duffy, 1983). ERS has also compiled na-
tional estimates of individual substances used in growing
potatoes, vegetables, fruits and citrus products (Fergu-
son, in prep.). Since the State of California was excluded
from the ERS surveys, we obtained a separate study from
California in which the annual quantities of pesticides ap-
plied by crop are estimated (State Calif. 1981).
Our computerized inventory covers all of the ERS and
California usage estimates with the exception of minor
substances, banned substances, and solvents. The esti-
mates are by crop and by substance, either at the national
or individual State level.
We used the 1982 Census of Agriculture to estimate
cropland acreage at the State or national level to compute
pesticide application rate coefficients in terms of pounds
per acre per year. For States not included in the ERS
survey, we assigned pesticide usage coefficients from
neighboring States. For crops not covered in the ERS
surveys, we assigned coefficients to States from the coef-
ficients derived for the State of California. In this manner,
we derived a comprehensive set of crop application rate
coefficients for all States (or at the national level) for all
crops. Using the 1982 Census of Agriculture data on
county cropland acreage, we then estimated annual pesti-
cide application amounts by crop and by county for the
184 substances in the inventory for all crops.
In addition to the cropland pesticide application esti-
mates, we have also estimated the amount of pesticides
applied to urban lawns and nurseries by county The start-
ing point for these sets of estimates was the national sur-
vey prepared for EPA's Office of Pesticide Programs, in
which annual pesticide application amounts were esti-
mated at the national level for urban lawns and nurseries
(Res. Triangle Inst. 1984a, 1984b). By dividing these na-
tional usage estimates by national estimates of the num-
ber of single unit housing structures and by the number of
nurseries, we computed per unit pesticide application
rates for urban lawns and nurseries, respectively. We ob-
tained county by county estimates of the number of single
unit housing structures and the number of nurseries from
the Census of Housing, (U.S. Dep. Comm. 1977) and the
Census of Agriculture (U.S. Dep. Comm. 1982), respec-
tively. These data allowed us to estimate pesticide applica-
tion amounts by county for nurseries and housing units for
individual pesticides. As a result, the pesticide inventory
includes, by county, annual application estimates for 184
substances applied to individual crops, housing units, and
nurseries.
The purpose in developing the pesticide application es-
timates was to estimate the amount of pesticides that
reaches waterways. Don Wauchope of Agricultural Re-
search Service has classified each of the 184 substances
according to a runoff potential classification which esti-
mates annual losses as a percent of amount applied. The
loss rate varies from .5 percent to 3 percent (Wauchope,
1978). By multiplying the annual application amount esti-
mate by the annual loss percent estimates, we estimate
the amount of pesticides that reach waterways.
SUMMARY
The RFF Environmental Data Inventory is unique as an
information source for nonpoint source policy assess-
ments. It is the only national data base in which discharge
estimates are made for both point and nonpoint sources
for all U.S. counties. The same suite of 17 pollutants is
accounted for comprehensively for both point and non-
point sources. The data base allows comparisons to be
made between States, water resource regions, counties,
and by discharge category. EPA in a recent report to Con-
gress used an earlier version of the data base to show
States with a preponderant share of pollutant discharge
from nonpoint sources (U.S. Environ. Prot. Agency, 1984).
The uniqueness of the inventory poses some con-
straints, however. Only small parts of the inventory can be
compared to other regional data sets for verification pur-
poses. Often such comparisons can be made for particu-
lar regions, industries, or pollutants.
Clearly many of the simplifying assumptions and meth-
ods we have used are much less sophisticated than those
found in the many micro models being applied in small
watersheds. Therefore, we do not advocate using our
methods in doing small regional studies. On the other
hand, we have found very few of the micro models that
can be applied for very large regions, even for a single
county. Large regional studies (such as those performed
for 208 agencies) often employ methods similar to ours to
make discharge estimates.
We intend to maintain and improve the inventory. One
reason we try to be explicit in stating our simplifying as-
sumptions is to elicit help, and perhaps data, from other
researchers. We welcome any improvements to our esti-
mates.
REFERENCES
Duffy, M. 1983. Pesticide Use and Practices 1982. Agric. Infor.
Bull. 462. U.S. Dep. Agric., Washington, DC.
Ferguson, W.L. In prep. National Pesticide Use on Selected
Crops, Aggregated Survey Data, 1977-80. Econ. Res. Serv.
Staff Rep., U.S. Dep. Agric., Washington, DC.
General Services Administration. 1983. Summary Report of
Real Property Owned by the United States Throughout the
World as of September 30,1982. Off. of Admin. Washington,
DC.
Gianessi, L.P., and H.M. Peskin. 1984. An overview of the RFF
Environmental Data Inventory: methods, sources and prelimi-
nary results. Rep. Water Resour. Div. U.S. Geolog. Surv.,
Washington, DC.
Midwest Research Institute. 1977. Estimation of pollutant loads
from nonpoint sources using the nonpoint calculator. Draft
interim rep.
Morton, J.H., and R.C. Loehr. 1980. Nonpoint source water qual-
ity problems related to animal agriculture. Prelim, rep. to U.S.
Environ. Prot. Agency.
Research Triangle Institute. 1984a. National Urban Pesticide Ap-
plication Survey: Final Report, Overview and Results. RTI/
2764/08-01F.
1984b. National Nursery Pesticide Usage Survey,
RTI/2766/06-01F.
Shacklette, H.T., J.C. Hamilton, J.G. Boerngen, and J.M. Bo-
wles. 1971. Elemental composition of surficial material in the
coterminous United States. U.S. Geolog. Surv. Prof. Pap.
574-D. Washington, DC.
Soil Conservation Service. 1984. National Resources Inventory:
A Guide for Users of 1982 NRI Data Files. U.S. Dep. Agric.
Statis. Lab., Iowa State Univ.
444
-------
State of California. 1981. Pesticide Use Report, Annual. Dep.
Food and Agric. Sacramento.
Tubbs, L.J. 1980. Estimation of nonpoint dissolved nitrogen and
phosphorus losses from cropland in the coterminous United
States. Final Res. Rep. to Resour. Future. Washington, DC.
Tubbs, L.J., and D.A. Haith. 1980. Estimating nonpoint source
nutrient loads from croplands. Rep. Dep. Agric. Eng. Cornell
Univ., Ithaca, NY.
U.S. Department of Commerce. 1984. Census of Agriculture,
1982, Final County File, Technical Documentation. Bur. of
Census. Suitland, MD.
1982.1978 Census of Agriculture, Vol. 5, special rep.
DATA AVAILABILITY AND NEEDS
part 7. 1979 Census of Horticultural Specialties. AC78-SR-7.
Bur. Census. Suitland, MD.
_. 1977. County and City Data Book. Washington, DC.
U.S. Environmental Protection Agency. 1984. Nonpoint Sources
of Pollution in the U.S. Rep. to Congress. Washington, DC.
Van Dyne, D.L., and C.B. Gilbertson. 1978. Estimating US Live-
stock and Poultry Manure and Nutrient Production. ESCS-12.
Econ. Res. Serv., U.S. Dep. Agric., Washington, DC.
Wauchope, R.D. 1978. The pesticide content of surface water
draining from agricultural fields, a review. J. Environ. Qua).
7(4): 459-72.
445
-------
Water Quality Criteria
and Standards
BACTERIAL WATER QUALITY AND SHELLFISH HARVESTING
ELAINE A. GLENDENING
Oregon Department of Environmental Quality
Portland, Oregon
This paper describes how the effort to enforce water qual-
ity standards for the commercial shellfishing industry in
Oregon's Tillamook Bay led first to a federally sponsored
208 water quality study and then to the development of
plans to reduce nonpoint source pollution inputs and to
establish criteria under which the bay would be closed to
harvesting.
First, the characteristics of the Tillamook drainage basin
will be briefly described and the problems and programs
that preceded the 208 efforts summarized. Next, the
results of the 208 research will be presented, focusing on
how the lack of results showing straightforward correla-
tions between shellfish contaminations and bay pollution
levels necessitated further analyses. Finally, the paper
presents the process of using the scientific conclusions
drawn from all these considerations in formulating pollu-
tion management plans for the bay.
BACKGROUND
The Tillamook Bay Drainage Basin is located on the north-
ern Oregon coast, 48 miles south of the Columbia River
and 60 miles west of Portland. Five major rivers drain the
basin's 363,520 acres and discharge to Tillamook Bay.
Ninety percent of the basin is mountainous, forested, and
sparsely populated. Eight percent of the basin is alluvial,
relatively flat, and devoted to agriculture and population
centers. Dairy farming is the primary agricultural activity.
There are about 120 of these dairy farms, whose nearly
19,000 animals generate over 280,000 tons of manure a
year. The basin's 13,000 people live in three small cities
and rural hinterlands. About 6,300 are served by sewers,
with the other 6,700 using on-site sewage systems. The
remaining 2.5 percent (9,150 acres) of the basin is occu-
pied by the bay itself.
The Tillamook area is characterized by a strong marine
influence. Seventy percent of the rainfall occurs during the
months of November through March. Winter storms com-
ing off the Pacific can bring intense periods of precipita-
tion, resulting in sudden rises in river flows and occasion-
ally flooding the alluvial plain low in the basin. The
average rainfall is 229 centimeters (90 inches) along the
coast and 381 cm (150 inches) inland in the mountains
(Natl. Oceanic Atmos. Admin., 1973). The mean annual
water yield for the five rivers in the basin is about 2.6
million acre-feet.
As the lowest point in the drainage, Tillamook Bay re-
ceives these seasonally-generated large inflows of fresh-
water. These high flows can result in cumulatively increas-
ing bacterial densities. As shown in Figure 2, the upper
bay is generally an area of lower salinity since four of the
five rivers enter the bay in that area. During the winter and
spring, salinities in the upper bay will approach zero due
to high river inflows. During the summer and fall, salinities
in the upper bay reach 15 parts per thousand because of a
low inflow of freshwater. The water temperatures in the
bay also vary seasonally, ranging from 12° to 18° C in the
summer to 7° to 9°C in the winter.
Although this paper focuses on the needs of the com-
mercial shellfishing industry, Tillamook Bay serves many
different needs. The bay is the receiving water for the
effluent discharged by five sewage treatment plants. The
bay also supports commercial and recreational fishing
and shellfishing and recreational boating. Shellfishing in-
cludes recreational and commercial clamming, and com-
mercial oyster cultivation and harvesting. The Pacific oys-
ter (Crassostrea gigas) is seeded on and harvested from
the shallow bay floor and sold without rely or depuration.
The Pacific oyster has been grown commercially in the
bay since the 1930's. Of the 2,084 acres available for
leasing from the State of Oregon for oyster cultivation,
about 950 acres are leased annually by three growers (see
Fig. 1) (Osis and Demowy, 1976). In 1975,142,144 pounds
of oysters were harvested for a value of $280,180 (Fors-
bergetal. 1975).
447
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
HISTORY OF WATER QUALITY
STANDARDS AND MONITORING
EFFORTS
Estuarine waters that support commercial oyster produc-
tion are subject to water quality standards developed by
the Food and Drug Administration (FDA) as part of its
National Shellfish Sanitation Program. (The total coliform
standard is the bacterial standard that applies in Tillamook
Bay, see Table 1.) The State of Oregon through its State
Health Division has conducted a Shellfish Sanitation Pro-
gram that follows the framework established by the na-
tional program. The State's program has been in effect
since the late 1940's. Prior to 1969, the Health Division
had full responsibility for monitoring the quality of the
growing waters, inspecting the sanitary conditions of proc-
essing facilities, and coordinating the State program with
the National Shellfish Sanitation Program. In 1969, how-
ever, the State legislature created the Department of Envi-
ronmental Quality and charged it with monitoring estua-
rine waters (growing waters) and operating sewage
treatment plants.
In 1972, the Health Division developed a bay closure
plan that would restrict harvesting of oysters whenever it
rained 2 inches or more in 24 hours, or whenever a sew-
age treatment plant upset or bypass occurred. Water sam-
pling done by the Department of Environmental Quality
and the FDA in the 1970's established that these stand-
ards were justifiable, as the bay did at times exceed the
total coliform standard, especially during periods of heavy
rainfall in the winter (Ore. Dep. Environ. Qual., 1981 a).
The FDA, however, expressed concern that the State
OREGON DIVISION OF STATE LANDS (1973)
Figure 1 .—Location of oyster leases and clam beds in Tilla-
mook Bay with DEQ sampling sites noted.
Health Division was not being adequately informed about
the operations of the five sewage treatment plants that
discharged into the bay, and could therefore not know
when to close the bay following a bypass or upset. The
FDA also expressed concern that the rainfall criteria were
not being enforced.
In December 1977, the FDA conducted a special bacte-
rial survey in Tillamook Bay during a major storm event.
Results snowed bacteria densities in the bay far exceeded
the National Shellfish Sanitation Program standard for
safe oyster harvesting (U.S. Health Edu. Welf., 1978). The
FDA, acting upon the results of the survey, strongly rec-
ommended closing the bay to oyster harvesting and de-
veloping appropriate control measures. The FDA threat-
ened to withdraw their endorsement of the Oregon
Shellfish Sanitation Program if appropriate actions were
not taken.
Recognizing the gravity of this situation, the Health Divi-
sion formed a task force of the State Health Division, the
Department of Environmental Quality, Oregon State Uni-
versity, and the shellfish industry to deal with the problem.
Their recommendations included: (1) hire a full-time sani-
tarian whose primary duty would be shellfish sanitation;
(2) assess bay water quality using the fecal coliform stand-
ard, concurrently sampling shellfish meat for fecal coli-
form and Salmonella organisms; (3) develop criteria for
closing and reopening shellfish growing waters based on
those analyses of shellfish meats; (4) intensify bay and
shellfish monitoring, especially in growing waters; and (5)
develop programs to reduce nonpoint source pollutants.
DESIGN OF THE 208 STUDY
Following the recommendations of the task force for more
water quality data on the bay, and for the development of a
program to reduce nonpoint source pollution, the Depart-
ment of Environmental Quality and the Tillamook Soil and
Water Conservation District secured U.S. Environmental
Protection Agency funding under section 208 of the Clean
Water Act of 1972, to conduct a study. The purpose of the
study was to identify fecal waste sources in the basin,
develop a plan to reduce their input to the bay, and de-
velop monitoring criteria that would allow shellfish har-
vesting under safe conditions.
Integral to the conduct of this study was the formation of
two committees: a Technical Advisory Committee and a
Citizens Advisory Committee. These committees provided
local input on the field surveys, reviewed the findings, and
participated in developing plans to reduce fecal wastes
entering the bay and to develop criteria under which the
bay would temporarily be closed to shellfish harvesting.
To identify fecal waste sources, field surveys were
scheduled flexibly by the Department of Environmental
Quality to coincide with conditions of ground saturation
and rainfall events. Water samples from the five river sys-
tems (71 sites) were collected at 8-hour intervals around
the clock for 2 to 5 days. This type of field survey was new
to the Department and was a learning experience for the
staff involved. Concurrent with the water sampling, river
flow measurements were taken, rainfall amounts re-
corded, and the five sewer treatment plants that discharge
to the bay were sampled. Bay water samples (14 sites, see
Fig. 1) and oyster samples (two sites) were collected on
high and low tides during daylight hours. A review of past
Department ambient bay monitoring showed that most
sampling had occurred during or around high tide, be-
cause Tillamook Bay is very shallow in the oyster growing
area, and it is easy to go aground while attempting to
sample at low tide. However, more data at or near low tide
was believed necessary to better understand the impact of
freshwater inflows on these areas. Field measurements
448
-------
for temperature and salinity were also made on the bay
and many of the river sampling points that were tidally
influenced.
The water samples collected from the rivers and the five
sewage treatment plants were analyzed by the Depart-
ment's laboratory using the membrane filtration method
for total and fecal coliform according to procedures in the
14th edition of Standard Methods (1976). Both the water
samples and the oyster meats collected from the bay were
analyzed by the Health Division's laboratory using the 5-
tube MPN (most probable number) method for total and
fecal coliform, according to Standard Methods (1976).
Both the total coliform and fecal coliform standards are
applicable in the Tillamook Bay Drainage Basin. The
Health Division uses a total coliform standard for shellfish
growing waters, as does the FDA (see Table 1). In 1980
the Department changed from the total coliform standard
to the fecal coliform standard for shellfish-growing waters
(see Table 1). The change was made to better measure
the effects of fecal sources.
SUMMARY OF 208 SURVEY SAMPLING
CHARACTERISTICS
Field surveys were designed to coincide with four particu-
lar weather and soil conditions:
1. Heavy rainfall when soils were saturated. This oc-
curred December 1979.
2. Rainfall after a period of dry weather when soils were
unsaturated. This occurred March 1980.
3. No rainfall during summer low river flows and unsat-
urated soils. This occurred July 1980.
4. First "freshet" storm after October 1, when soils be-
come saturated and overland runoff increases. This oc-
curred October 1980.
Each of the four field surveys is summarized in Table 2.
Each survey consisted of several individual sampling
runs. The table shows daily rainfall and mean daily Wilson
River flow for each individual run, while the bay bacterial
water quality data are summarized for each survey as a
whole. Ranges of median values for fecal coliform are
included for each survey to show the extent of bacterial
input associated with each climatic condition. Coliform
ranges for followup sampling in March 1980 reflect actual
values, as do salinity and temperature data. Salinity is
displayed as a range for each end of the oyster lease. The
north end data represent stations, 7, 13, and 11. The
WATER QUALITY CRITERIA AND STANDARDS
south end represents stations 6, 14, and 2 (see Fig. 1).
Ranges of salinity and temperature are included to show
how specific climatic and runoff conditions can influence
these physical parameters which in turn influence oyster
pumping. Most importantly, the number of oysters col-
lected and the number that exceeded the meat standard
are included. An evaluation of sewage treatment plant
operation is also included. Not included for lack of space
is a generalized statement of water quality in each of the
rivers for each survey.
In brief, the water quality in the lower part of the basin
generally exceeded the bacterial water standard when-
ever a rainfall event took place, because of the ready ac-
cess that animal wastes had to the water courses (Ore.
Dep. Environ. Qua!., 1981b). The next section covers the
conclusions that were drawn from these field surveys.
CONCLUSIONS DRAWN FROM 208 FIELD
SURVEYS
The field surveys provided further evidence to confirm two
basic hypotheses, held by most observers who had previ-
ously studied the bay, about what was occurring in the
Tillamook basin. First, the total and fecal coliform stand-
ards for the shellfish-growing waters were exceeded peri-
odically or even constantly during the wet weather
months, October through April. Second, the data showed
that the major source of this fecal contamination was input
from improper waste practices at dairy operations. This
source represented approximately 70 percent of all fecal
coliform input to the rivers and thus into the bay (Ore. Dep.
Environ. Qual., 1981b).
The extensiveness of the data collected, however, al-
lowed many more detailed inferences than these to be
made. In addition to the estimate that 70 percent of all
fecal coliform input to the rivers was from dairy animal
wastes, the data permitted estimating that improperly
functioning on-site sewage systems were responsible for
15 to 20 percent of the fecal coliform input. The remaining
5 to 10 percent was attributable to natural background
levels from numerous sources such as the wild animal
population (Ore. Dep. Environ. Qual., 1981b). Although
the sewage treatment plants operated within their permit
levels during all the surveys, the data gained from the
surveys increased the ability to assess the immediate and
disastrous effects malfunctions can have on water quality.
The data also helped provide a more detailed under-
Table 1.—Federal Food and Drug Administration and State of Oregon shellfish growing water and market oyster meat
standards applicable to estuarine and fresh waters in the Tillamook Bay drainage basin.
Agency
Marketed oyster meats
Estuarine shellfish
growing waters
Freshwater and
non shellfish
growing estuarine
waters
Food and Drug Administration
(FDA)
For 100 grams oyster meat:
total coliform 60,000
fecal coliform 230
standard plate
count 500,000
For 100 milliliters of sample:
median of 70 total coliform;
10% of samples not greater
than 230 per 100 milliliters
No Standard
Oregon State Health Division
(DSHD)
Same as FDA
Same as FDA
No Standard
Department of Environmental
Quality (DEQ)
No Standard
For 100 milliliters of sample:
median of 14 fecal coliform;
10% of samples not greater
than 43 per 100 milliliters
For 100 milliliters of sample:
log mean of 200 fecal
coliform for 5 samples in 30
days; 10% of samples not
greater than 400 for period
Oregon Dep. Environ. Qua). 1981.
449
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
standing of the potential disease transmission pathways
by which a pathogen might be transported to the shellfish-
growing waters. The data showed, however, that a con-
stant straight line correlation between the amount of pollu-
tion in the bay and the level of contamination in the
shellfish meats did not necessarily exist. While this could
be expected given the complexity of the physical and bio-
logical systems being studied, it complicated the develop-
ment of a management plan for the water quality in the
bay. The next sections describe how this plan was devel-
oped.
DEVELOPMENT OF THE FECAL WASTES
MANAGEMENT PLAN
With the completion of the field surveys and the identifica-
tion of the sources of the bay's fecal contaminants, two
tasks lay ahead. One was to develop a plan to reduce
BOTTOM AND FORSBERG (1978)
WINTER
Dec. - Feb
SPRING
Mar. - May
SUMMER
June - Aug.
FALL
Sept. - Nov.
Figure 2.—Average seasonal salinities (parts per thousand) in Tillamook Bay from samples taken near the bottom at high
tide.
450
-------
WATER QUALITY CRITERIA AND STANDARDS
fecal inputs at their source; the other to create a set of
criteria to decide when the bay should or should not be
open to shellfish harvesting. As this paper focuses on the
latter task, only a brief synopsis of the waste plan will be
included here. (For a more complete account, see Ore.
Dep. Environ. Qua!., 1981c.)
The Tillamook Bay Drainage Basin Fecal Wastes Man-
agement Plan was developed by the Department of Envi-
ronmental Quality and the Tillamook Advisory Commit-
tees. The plan addresses the identified fecal sources with
a course of action to correct their fecal input to the bay.
Major components of this plan include:
1. A notification procedure was developed for all sew-
age treatment plant malfunctions. Warning equipment
was also installed.
2. Best management practices (BMP's) were devel-
oped by the Tillamook Soil and Water Conservation Dis-
trict. These BMP's addressed improper or nonexisting ani-
mal waste management practices.
3. On-site sewage problem areas were identified and
those areas requiring further investigation and correction
were prioritized.
4. Commitment was obtained from all involved parties
to execute their identified plans for fecal input correction.
5. The need for annual recertification and reevaluation
of the plan was agreed upon.
All parties recognized that the fruits of the Fecal Waste
Management Plan, in terms of improved water quality for
shellfish growing, were 5 to 10 years away. Since com-
plete closure of Tillamook Bay was not an attractive option
Table 2.—Summary of 208 water quality data for Tillamook Bay collected during selected rainfall and soil saturation conditions.
Date of survey
Daily rainfall
for survey (in)
December 1979
79/12/02
79/12/03
79/12/04
79/12/05
79/12/06
79/12/07
March 1980
80/03/10
80/03/11
80/03/12
80/03/13
80/03/14
80/03/15
80/03/16
80/03/17
Followup
sampling*
Flow and rain
data only*
80/03/18*
80/03/19"
80/03/20*
80/03/21'
80/03/22*
80/03/23*
80/03/24*
80/03/25*
80/03/26*
80/03/27*
July 1980
80/07/28
80/02/29
80/07/30
October 1980
80/10/25
80/10/26
80/10/27
80/10/28
80/10/29
1.62
0.15
2.54
0.05
0.18
0.05
0.36
0.78
0.94
0.81
0.61
1.34
0.03
0.99
0.49
0.43
0.00
0.01
0.90
0.18
0.03
0.00
0.45
0.11
0.00
0.00
0.00
0.14
0.89
0.00
0.00
0.00
Daily mean
Wilson R. flow
(CFS)
4560
2560
5830
3440
2200
1630
729
845
1010
1540
1570
1460
1320
1510
1970
1860
2050
2020
1760
1610
1410
1250
1220
1190
93
90
88
Fecal coliform Range of
WQ stnd for salinities over
shellfish waters, beds north/
7 sites (range) south (ppth)
exceeded
for
entire
survey
((30 to 2400)
exceeded
for
entire
survey
(4 to 1100)
5 to 27/
3 to 20
Range of
temperature
over beds
(centigrade)
8 to 10
7 to 23/
Oto22
7 to 12
exceeded
(9 to 240)
exceeded
(1510150)
20/4.5
17/6.5
18/9
meet
(3 to 23)
4 of 7 exceeded 9/5
(4 to 23)
meet
for
survey
(3 to 43)
26 to 32/
27 to 30
95
250
200
169
120
CFS = cubic feet per second
4 of 7 exceeded 20 to 271
(3 to 240) 14 to 27
8 to 10
8 to 10
11 to 12
10 to 11
11 to 15
11 to 12
Oyster meat Sewage
quality, 2 sites treatment
no. of samples/ plants
no. exceeded
Soil saturation
1/0, several
meat
samples
violated
hold times
were
discarded.
7/2
2/0
2/0
2/0
2/1
6/0
6/2,1 at limit
meet saturated
permit ponding
overland
flow
meet unsaturated
permit becoming
saturated
not saturated
sampled
not saturated
sampled
not saturated
sampled
not
sampled saturated
meet unsaturated
permit
meet
permit
unsaturated
becoming
saturated
ppth = parts per thousand
Glendening, 1985
451
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
while corrective actions were taken, the final task of the
study was to develop the criteria for temporary bay clo-
sures.
RATIONALE FOR THE BAY CLOSURE
PLAN
Development of the bay closure plan involved two courses
of action. First, the results of the 208 studies together with
other research data were examined to determine when
oysters were most likely to be contaminated. These analy-
ses formed the basis for developing scenarios projecting
the probability of oyster contamination, given different
seasonal conditions. Second, these scenarios were dis-
cussed with the shellfish committees consisting of oyster-
men, university experts, and the State Health Division.
Proposed Criteria for the Temporary Closure of Tillamook
Bay to Shellfish Harvest (Ore. Dep. Environ. Qual., 1981d)
fully discusses the plan that resulted from these consulta-
tions.
ANALYSIS OF FACTORS CONTRIBUTING
TO OYSTER CONTAMINATION
Preliminary analysis
Initial analysis focused on determining if a correlation ex-
isted between physical parameters and bacterial densi-
ties. If a strong correlation could be established, it would
serve as an immediate gauge of when to close the bay to
shellfish harvesting. Correlations were sought for fecal
coliform versus salinity and versus river flow (the Wilson
River and for bay stations 12, 6, and 14, using data from
1970 to 1979). No clear correlation was apparent during
Table 3.—Risk assessment of oyster meat contamination for various physical, biological and climatic conditions.
October
November
December
January
February
March
Average
rainfall
18.6cm
7.3 in
35.1 cm
13.8 in
42.4 cm
16.7 in
40.1 cm
15.8 in
23.3 cm
9.2 in
26.7 cm
10.5 in
Average
river flow
low
to
moderate
moderate
to
high
high
to
very high
high
to
very high
high
high
to
moderate
Soil
saturation
unsaturated
(filling)
unsaturated
to
saturated
saturated
saturated
saturated
saturated
Bay water
quality
meets stnd.
except for
freshet
rainfalls
exceeds
stnds.
greatly
exceeds
stnds.
greatly
exceeds
stnds.
exceeds
stnds.
exceeds
stnds.
Salinity over
shellfish
beds
(high tide)
27 to
31 ppth
27 to
31 ppth
10 to
23 ppth
10 to
23 ppth
10 to
23 ppth
6 to
26 ppth
Temp, over
shellfish
beds
11 to
13°C
11 to
13°C
8 to
9°C
8 to
9°C
8 to
9°C
9.5 to
10.5°C
Oyster
pumping
activity
active
reduced
to
very
limited
very
limited
very
limited
very
limited
reduced
to
limited
Oyster
harvest
yes
peak
harvest
peak
harvest
peak
harvest
yes
yes
Risk of
contaminated
oysters
low to
high
(rainfall
events)
high
to
moderate
moderate
to
low
moderate
to
low
low
low
April
May
June
July
August
16.6cm
6.5 in
9.4cm
3.7 in
7.7cm
3.0 in
3.5cm
1.4 in
4.6cm
1.8 in
September 9.9 cm
3.9 in
moderate
moderate
to
low
low
low
low
low
saturated
saturated
to
unsaturated
(draining)
unsaturated
unsaturated
unsaturated
exceeds to 6 to
meets stnds. 26 ppth
exceeds to 6 to
meets stnds. 26 ppth
unsaturated meets
(draining) stnds.
meets
stnds.
meets
stnds.
meets
stnds.
22 to
31 ppth
22 to
31 ppth
22 to
31 ppth
27 to
31 ppth
9.5 to
10.5°C
9.5 to
10.5°C
12.510
16.5°C
12.5 to
16.5°C
12.5 to
16.5°C
12.5to
16.5°C
reduced
active
active
(potential
spawn)
very
active
very
active
very
active
yes
yes
reduced
yes
yes
no
low
ppth = parts per thousand C = centigrade
Qtendening, 1985
low to
moderate
(rainfall
event)
low to
moderate
(rainfall
event)
low
low
low to
high
(rainfall
event)
452
-------
WATER QUALITY CRITERIA AND STANDARDS
the wet weather period for these physical parameters and
bacteria densities. The same analysis was tried using data
gathered during the 208 study, and the results were the
same.
These analyses suggested that any effort to explain the
coliform densities in the bay must take into account multi-
ple factors including: (1) amounts of freshwater input and
its bacterial loading reflecting recent rainfall intensities
and soil saturations; and (2) the varying tidal stages and
movements of salinity gradients up and down the bay,
which seasonally change the dilution rate of freshwater.
The complex interactic.n of these and possibly other varia-
bles made modelling these interactions extremely difficult.
The ability to obtain the measurements necessary to use
such a model also was beyond reach. The focus of the
analysis thus shifted to evaluating those situations that
might produce the highest risk of contaminated oysters.
Analysis of oyster behavior
Review of the bacterial data analyses on the oyster meats
had shown that the meats did not always exceed the bac-
terial standard even when the ambient water quality ex-
ceeded the standard for shellfish-growing waters (see Ta-
ble 2). Factors other than ambient water quality affect
oyster meat bacterial quality. Therefore, to better under-
stand oyster meat quality, the conditions that affect oyster
feeding/pumping activity were investigated.
Research has shown that three conditions influence
oyster pumping activity: temperature, salinity, and turbidity
(U.S. Health Educ. Welf., 1966). Any one of these condi-
tions can cause oysters to substantially decrease or cease
pumping. Oysters will decrease or cease pumping when:
(1) water temperature is 10° C or less; (2) salinities- are 10
parts per thousand or less; or (3) turbidity is more tnan 20
Jackson Turbidity Units (JTU). All three of these conditions
occur during periods of high freshwater inflow during the
winter months. During this time oysters pump only small
volumes of water just to keep themselves oxygenated.
Oysters will most actively pump when: (1) salinities are
25 parts per thousand or greater; (2) temperatures are 15°
C or greater; and (3) turbidity is below 20 JTU. These
conditions occur from April through October, during low
freshwater inflow when salinities range from 20 to 30 parts
per thousand, and temperatures from 12° to 17° C. Water
quality in the bay is generally good during this period be-
cause of low stream inflows, which implies low runoff and
thus low bacteria densities.
Bay salinity and temperature
Since the review of oyster behavior had shown that these
were important factors, bay salinities, temperature data,
and freshwater inflow were reviewed for each of the four
field surveys and analyzed against oyster meat bacteria
data and bay bacteria data. The aim was to determine
how important these conditions were in accounting for the
level of bacteria in the meat. The same analysis was also
performed on three other studies (U.S. Health Edu. Weif.,
1974,1976,1978) that had collected oyster meat samples.
The conclusion was reached that these factors were in-
deed essential to determining the probable degree of oys-
ter contamination.
The two studies conducted by the FDA in 1977 and
1974 illustrate very well the issues of concern here (U.S.
Health Edu. Welf., 1974, 1978). The 1977 study shows
how factors combine to make a period of presumably high
contamination less so. The 1974 study illustrates the op-
posite—how high contamination risks can exist even
when ambient bacterial readings are low.
In December 1977, the FDA conducted a survey on
Tillamook Bay during a major winter storm. Bacterial wa-
ter quality in the shellfish-growing water ranged from a
fecal coliform median of 4,600 to 9,300 per 100 ml. Salini-
ties ranged from a median of 10 to 16 parts per thousand.
Temperatures ranged from 9° to 10° C. Forty-four oyster
samples were collected from the bay. Of these, 17 ex-
ceeded the bacterial meat standard and 11 were at the
limit of the standard, 230 fecal coliform MPN/100 grams.
Of the 17 that exceeded the standard, nine had values in
the 300's, four had values in the 400's, and four had a
value of 500 or greater.
Even though the oyster meat samples did exceed the
market standard, they did not reflect the extent of bacterial
densities that occurred over the oyster beds. In its report
the FDA stated, "The environment was adverse for the
oyster because of the low salinity values, the high turbidity
of the water, and the presence of the sediment. Their in-
ability to pump, with the near absence of concentration of
pollutants and resulting lower than expected fecal coliform
counts, made the oysters an unreliable indicator of pollu-
tion." (U.S. Health Edu. Welf., 1978)
In November 1974 the FDA conducted a survey on Tilla-
mook Bay shortly after the first "freshet" rainfall event of
the water year and continued sampling into and through
the second "freshet." (The Department of Environmental
Quality 208 survey conducted in October 1980 had similar •
climatic conditions.) While during the survey the shellfish-
growing waters did meet standards, it was conjectured
that the bay had received a large bacterial load as a result
of this first "freshet." Salinities over the oyster growing
area were in the range for pumping activity of 19 to 33
parts per thousand. Temperatures were 9° to 10° C. Of
the 12 oyster samples collected, 10 exceeded the meat
standard, and 9 of the 10 had fecal coliform values that
ranged from 4,700 to 92,000 MPN. The risk to public
health from consuming raw contaminated oysters in this
situation was high, and the data showed that oysters have
a high probability of becoming contaminated when a bac-
terial load is introduced to the bay, and salinities are within
the optimum range for active pumping (U.S. Health Edu.
Welfare, 1974). (The sampling done in October 1980,
when conditions for active oyster pumping also existed in
the bay, further supports this hypothesis. Three of the six
meat samples were either at the limit of or exceeded the
meat standard.)
REVISED METHOD OF ASSESSING THE
RISKS OF OYSTER CONTAMINATION
By reviewing all the relevant data, conclusions were made
about the risks associated with the harvesting of shellfish
under various seasonal conditions. These principal con-
clusions are as follows:
1. Rainfall conditions exist October through April and
cause large inflows of freshwater, which carry with them
large bacterial loadings to the bay. Shellfish-growing wa-
ters in Tillamook Bay exceed the bacterial standard under
these conditions.
2. During periods of high inflow, salinities and tempera-
tures are below the optimum for active shellfish pumping.
3. Shellfish meats do not reflect bacteria concentra-
tions in proportion to the water quality because they are
not actively pumping.
4. Tillamook Bay meets the shellfish-growing water
standard under conditions of very little or no rainfall. This
occurs May through October.
5. During periods of low inflows and optimum salinities
and temperatures, oysters will actively pump.
6. Shellfish meats can exceed the meat standard if sa-
linities and temperatures are optimum for pumping, and a
large bacterial load is introduced to the bay as a result of a
rainfall event. This can occur even though the bay waters
453
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
meet the shellfish-growing water standard.
These conditions were used to project the likely risk of
oyster contaminations at different times of the year under
various conditions. For example, by using these premises
and varying the expected rainfall amounts (1.0,1.5, or 2.0
inches) together with the corresponding expected river
flows (based on historical data), it was possible to project
the risk in harvesting. Table 3 was created for this paper to
portray, in summary form, the types of criteria used to
assess risks for a hypothetical water year.
TILLAMOOK BAY SHELLFISH
MANAGEMENT PLAN
The conclusions and assessments of risk that were drawn
from the water quality studies were discussed with the
Shellfish Committees for use as criteria in constructing a
bay closure plan. A consensus was reached that any plan
would have to insure that the risk of contaminated oysters
was low. Concern was also expressed about what each
proposed criteria would mean in terms of the number and
length of possible bay closures. Important input was also
made at this time about how the impact of different fecal
sources was to be considered in devising the plan. The
committees recommended that the plan address itself to
specific events such as major rainfalls or sewage treat-
ment plant malfunctions. During these discussions a con-
sensus was also agreed upon that, according to the best
available evidence and expert opinion, a 5-day purging
period for the oysters after a period of probable contami-
nation would give the best margin of safety With the com-
mittees' assistance, a management plan was developed.
The plan was submitted to the State Health Division and,
with minor modifications, adopted. The plan as executed
by the Oregon State Health Division is as follows:
1. Close the bay when a sewage treatment plant by-
pass or malfunction occurs. Closure would be variable
depending on the magnitude of the problem.
2. Close the bay for 5 days when the Wilson River's
flow reaches 8,500 cubic feet per second(cfs) for 12 hours
or more. This situation represents the river's flow at flood
stage and will cause flooding of fecal sources low in the
basin.
3. Close the bay for 5 days when the Wilson River's
flow doubles between April 1 and August 31.
4. Close the bay for 5 days when the Wilson River's
flow reaches 500 cfs or greater from Sept. 1 to Dec. 1 for
the first "freshet" of the water year.
5. Close the bay for 5 days for the second rainfall event
between Sept. 1 and Dec. 1 that increases the Wilson
River's flow to 1,500 cfs or greater.
MANAGEMENT PLAN EXECUTION
AND ASSESSMENT
The Tillamook Bay Management Plan has been in effect
since October 1981. In 1981, it resulted in three bay clo-
sures, two in the fall and one during the winter because of
high river flows. In 1982, a higher than normal rainfall year,
it resulted in nine closures, two in the fall, four in the winter
resulting from high river flows, and three caused by sew-
age treatment plant malfunctions. In 1983, it resulted in
three closures, one in the fall and two in the winter be-
cause of high river flows. In 1984, it resulted in two clo-
sures in the fall only. And in 1985, no closures to date have
occurred because of a drier than normal winter.
In discussion with Health Division staff the management
plan has been viewed as a good starting point for control-
ling the harvest and sale of fecally-contaminated shellfish.
However, review of the oyster meat data collected in con-
junction with bay closures indicates that 5-day closure af-
ter "freshet" rainfall events is not a sufficient length of
time to allow oysters to purge themselves (Chaceran,
1985).
A followup nonpoint source study of the Tillamook Bay
Drainage Basin has been scheduled by the Department
for 1985 and 1986. The planned surveys will be smaller in
scope but are intended to gather data for use in assessing
the effects of animal waste BMP's that were put in place
using Rural Clean Water Act funding. Each survey will
coincide with specific rainfall events, and river, bay and
oyster meat bacteriological samples will be taken. To bet-
ter assess oyster purging abilities, bay and oyster meat
sampling will be continued for 7 to 10 days beyond each
rainfall event. This information will allow the Tillamook Bay
Shellfish Management Plan to be fine-tuned.
CONCLUSIONS
Total and fecal coliform water quality standards can indi-
cate sources of fecal pollution in fresh and estuarine wa-
ters. In the absence of better data, they are often used as
indicators of shellfish pollution. However, a complex rela-
tionship exists between densities of coliforms in shellfish-
growing waters and the densities of coliforms in shellfish
meats. Shellfish management plans should investigate
these complex relationships and use them to develop clo-
sure criteria for shellfish harvesting based on the probabil-
ity of high coliform densities in shellfish meats. Further
research is needed to establish an indicator directly re-
lated to the incidence of disease. To this end, recently
proposed EPA bacterial water quality criteria for fresh and
marine water contact recreational waters attempt to estab-
lish a more direct relationship to the incidence of disease.
It is hoped that the FDA will investigate and establish bac-
terial criteria that are more directly related to the incidence
of disease for use in shellfish-growing waters.
REFERENCES
Chaceran, G. 1985. Personal Commun. Ore. State Health Div.
Forsberg, B.D., J.A. Johnson, and S.M. Klug. 1975. Identifica-
tion, distribution, and notes on food habits of fish and shellfish
in Tillamook Bay, Ore. Ore. Fish Comm. Portland.
National Oceanic and Atmospheric Administration. 1973.
Monthly normal of temperature, precipitation and heating and
cooling degree days, 1941-1970—Oregon. In Climatography
of the United States. No. 81. U.S. Dept. Commerce. Natl.
Climatic Center. Asheville, NC.
Oregon Department of Environmental Quality. 1981 a. 208 Tilla-
mook Bay bacteria study—background data review report.
State of Ore. Portland.
: 1981b. 208 Tillamook Bay bacteria study—source
summary report. State of Ore. Portland.
_. 1981C. 208 Tillamook Bay bacteria study—Tillamook
Bay drainage basin fecal waste management plan. State of
Ore. Portland.
_. 1981d. 208 Tillamook Bay bacteria study—proposed
criteria for the temporary closure of Tillamook Bay to shellfish
harvest. State of Ore. Portland.
Osis, L. and D. Demony. 1976. Classification and utilization of
oyster lands in Oregon. Ore. Fish Wild). Dep. Portland.
Standard Methods for the Examination of Water and Wastewa-
ter. 1975. 14th ed. Am. Pub. Health Ass., Washington, DC.
U.S. Department of Health, Education and Welfare. 1966. Depu-
ration Plant Design. Pub. Health Serv., Washington, D.C.
1974. Tillamook Bay, Oregon—comprehensive sani-
tary survey, November 1974. NE Tech. Serv. Unit, Food Drug
Admin. Davisville, Rl.
_. 1976. Tillamook Bay, Oregon—pollution source eval-
uation with classification and management considerations,
May, 1976. NE Tech. Serv. Unit, Food Drug Admin. Davisville,
Rl.
.. 1978. Sanitary survey of shellfish water Tillamook
Bay, Oregon, Nov.-Dec. 1977. NETech. Serv. Unit, Food Drug
Admin. Davisville, Rl.
454
-------
EVALUATION OF NONPOINT SOURCE IMPACTS ON WATER QUALITY
FROM FOREST PRACTICES IN IDAHO: RELATION TO WATER
QUALITY STANDARDS
STEPHEN B. BAUER
Idaho Department of Health and Welfare
Division of Environment
Boise, Idaho
ABSTRACT
An interdisciplinary task force was appointed by the
Board of Health and Welfare to determine the impacts of
forest operations on protected uses and make recom-
mendations on water quality standards. Twenty five forest
operations were inspected by the Task Force in 1984 for
compliance with the Idaho Forest Practices Act (FPA) and
their potential for impacting salmonid fish habitat. Seven
of the 25 operations were considered a major impact or
hazard to salmonid habitat due to direct delivery of sedi-
ment associated with roads or skid trails. At the remaining
sites impacts on protected uses were prevented either by
site conditions—low geologic hazard, streams with no
protected uses—or by good practices. U.S. Forest Serv-
ice timber sales met or exceeded the Forest Practices
Act. Noncompliance on State and private lands was asso-
ciated primarily with reuse of existing roads near stream
channels, failure to identify and use appropriate logging
systems in hazardous geologic conditions, and lack of
timely installation of erosion control measures.
INTRODUCTION
In 1980, the Idaho Water Quality Standards (Idaho Dep.
Health Welfare, 1980) were revised to include specific lan-
guage for control of nonpoint source pollution, including
silviculture. The Forest Practices Act Rules and Regula-
tions (Idaho Dep. Lands, 1979) administered by the Idaho
Department of Lands were identified as best management
practices (BMP's) for silviculture. The Aritidegradation
Policy was deleted during this revision, but language that
requires protection of designated uses was retained.
In 1982 the Idaho Division of Environment commented
unfavorably on a timber sale environmental assessment
report prepared by the U.S. Forest Service. The Division
of Environment held that the potential reduction in fish-
eries, as estimated in the environmental assessment,
would violate the standard protecting beneficial uses. The
predicted impact was based on the cumulative effects of
sediment on fisheries habitat. The Forest Service replied
that strict interpretation of this standard would set a prece-
dent that could severely curtail timber harvest opportuni-
ties in the national forests—with consequential impacts on
the State economy.
The Forest Service petitioned the Board of Health and
Welfare to change the standards relating to injury of pro-
tected uses. Additional conflicting petitions were submit-
ted by environmental and industry groups, and public
hearings on these petitions were held. As a result of the
hearings, the Board of Health and Welfare adopted a com-
promise position in revising the standards. At the same
time, the Board directed the Division of Environment to
establish an interdisciplinary task force to study the prob-
lems of nonpoint source pollution from forest practices.
The Board established the task force to provide a tech-
nically sound answer to questions that arose during public
debate regarding the water quality standards:
1. Do BMP's provide adequate water quality protection
for protected uses defined in the Water Quality Stand-
ards?
2. Are current forest practices affecting water quality
(protected uses) and to what extent?
3. Are the existing regulatory controls for silvicultural
operations adequate to prevent water quality impacts?
METHODS
Eight task force members were selected to represent the
major agencies and interest groups involved in the issue
of nonpoint source pollution on forested lands and to pro-
vide technical expertise in the following fields: silviculture,
hydrology, geology/soil science, forest road construction,
fishery biology, and water quality. Agencies and interest
groups represented included: Idaho Department of Health
and Welfare-Division of Environment, Idaho Department
of Lands, Idaho Fish and Game Department, Idaho Con-
servation League, American Fisheries Society, Idaho For-
est Industry Council, and both Forest Service regions in
Idaho.
The task force made onsite evaluations of 25 silvicul-
tural operations in 1984. Sampling design incorporated
consideration of geographic location, geologic land type,
logging methods, proximity to streams, and the need to
examine forest operations after the first runoff season.
Site selection was stratified based on land ownership
categories. Forty-five percent of the timber volume in
Idaho is harvested from 10 national forests, 45 percent
from private industrial and nonindustrial forests, and 10
percent from State school endowment lands. The 25 eval-
uations were divided approximately by timber volume to
include 10 Forest Service timber sales, 10 private opera-
tions, and 5 State timber sales. Sites were selected ran-
domly from a list of candidate operations. Although 25
sites do not comprise a statistically valid sample of forest
operations in Idaho, observed trends of compliance with
practices, of impacts on streams, and of administrative
procedures used by land management agencies are con-
sidered to be representative.
Site evaluation was based on compliance with proposed
revisions of the Idaho Forest Practices Act Rules and Reg-
ulations (FPA). These revisions resulted from a section
208 project (Braun, 1979), and include 19 individual rules
for timber harvesting and 30 rules for road design, con-
struction, and maintenance. The proposed rules clarify
vague wording in the current rules, but do not differ sub-
stantially in intent. Therefore, ratings of compliance with
the proposed rules also apply to the current rules.
A task force consensus rated each applicable rule sub-
jectively for compliance with the rule for water quality im-
pact using a rating system from 1 to 5, with 1 being a low
rating and 5 a superior rating.
Analysis of water quality impacts was based primarily
on the effects of sedimentation on fisheries habitat. A site
was rated by observation of direct sediment delivery to
streams and the potential for continuing impacts from the
455
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
site. Observation of cobble embeddedness estimated the
existing status of sediment impacts in the drainage.
RESULTS
Compliance with the proposed rules varied by land owner-
ship category (Fig. 1). Forest Service administered lands
had a high compliance rate. Only 5 percent of the individ-
ual ratings (n = 371) were judged as a minor departure
from the intent of the rule. Noncompliance ratings were
higher on State and private lands. On State lands, 21
percent of the individual ratings were considered a minor
departure, and 12 percent a major departure. On private
lands 10 percent were judged a minor departure, and 8
percent a major departure.
State Lands
Administrative procedures and management practices
used by the Idaho Department of Land did not provide an
adequate level of water quality protection. Three of the
five inspected timber sales resulted in major impacts or
potential hazards to fisheries habitat (Table 1). Recurring
practices that caused water quality impacts or potential
hazards were associated with roads and skid trails. Reuse
of existing roads located too close to stream channels,
poor road construction and maintenance practices, and
incomplete stabilization of cut and fill slopes before the
runoff season were among the hazardous practices.
Ground skidding occurred during wet weather and on
steep erosive slopes. Skid trails on sorm sites paralleled
tributary channels so that erosion control was ineffective
in preventing sediment delivery.
United States Forest Service
Seven of the 10 sites inspected met or exceeded the FPA
rules. Minor departures from a limited number of rules
were noted in three sales. A minor water quality impact
occurred at only one site when poor road drainage prac-
tices at a stream crossing and a culvert installation cre-
ated a fish passage barrier.
Overall administration of forest practices by the Forest
Service helps prevent water quality impacts. Roads are
planned, constructed, and maintained to appropriate
standards. Erosion control practices are extensive and up
to date. Logging systems that minimize soil and stream
disturbance are applied in sensitive land types. Extensive
planning and consideration of environmental effects are
major positive factors in achieving water quality protection
* CIMn INNCt
I NiNMMPUNNCt
5
II
LEGEND
EXCEEDS S COMPLIANCE
RULE
MINOR • MAJOR
DEPARTURE
Figure 1 .—Comparison of compliance with proposed Forest
Practices Act Rules at 10 U.S. Forest Service, 10 private,
and 5 State silviculture! operations. The percentage is
based on ratings of individual rules at a site, then summed
for the sites within the land management category.
not evident in other land ownerships. The costs associ-
ated with Forest Service administration of timber sales is
much higher than under State or private ownerships.
Private Operations
Protection of water quality values was considered ade-
quate at 6 of the 10 sites (Table 1). This was due, in part, to
the low hazard land types and minimal stream values in
some of these operations, as well as to the forest practices
conducted. At each of the remaining four sites, a major
departure from the FPA rules resulted in a major impact
(one site) or potential major hazard (three sites) to fish-
eries habitti.
Recurring reasons for violation of the FPA rules in-
cluded inadequate planning in location and design of
roads, reuse of existing roads and skid trails located close
to stream channels, inadequate road drainage and stream
crossing structures, and erosion control practices not
completed before the runoff season.
Existing Stream Conditions
Cobble embeddedness was used as an indicator of the
existing substrate condition with respect to cumulative ef-
fects of watershed activities. Of the 25 sites inspected, 14
were near a Class I stream, that is, a stream that could be
used by resident or adfluvial trout. Of these 14 streams,
obvious cobble embeddedness was observed in 9 (Table
2). At these nine sites (60 percent of Class I streams),
sediment delivery from past or ongoing activities may
have already caused sustained damage to the fishery
habitat.
DISCUSSION
The current controversy in Idaho is over the impact of
logging on a watershed basis, that is, over the additive
effects of nonpoint source sediment produced on a sensi-
tive protected use. Specifically, the issue has focused on
plans for building roads in unroaded areas (usually consid-
ered de facto wilderness by conservation groups) in the
Idaho batholith. These watersheds are generally high haz-
ard lands that historically supported runs of salmon and
steelhead trout. Restoring these anadromous salmonid
runs is considered a high priority by the public in Idaho.
The risk that a site presents to continuing impacts de-
pends on a number of site-specific factors together with
the way in which the operation is administered. Key site-
specific considerations are the geologic erosion hazard,
the stream's capability to support protected uses, and the
stream energy in regard to transporting sediment out of a
critical habitat.
Cumulative impacts change habitat conditions at a criti-
cal stream reach by the addition of individual impacts over
space and time; recovery do?s not occur before the next
individual practice. The issue of cumulative impacts of
forest practices has received a great deal of attention. The
Washington Forest Practices Board has recently com-
pleted a summary of the literature (Geppart et al. 1984).
Cumulative effects can only be quantified by costly and
time-consuming monitoring after the fact, or by predicting
the impact through modeling.
The task force speculated on the cumulative impact po-
tential of the inspected operations independently from the
existing watershed conditions, because we were inter-
ested in the implication for recommendations on future
management. Sixteen sites were considered low risks, six
moderate risks, and three high risks for contributing to
cumulative watershed impacts (Table 2). If proposed
BMP's were fully complied with, the task force believed
these risks would be substantially lowered. Four sites
would still pose moderate risks for cumulative effects de-
456
-------
WATER QUALITY CRITERIA AND STANDARDS
Table 1.—Summary of compliance with proposed Forest Practice Act Rules and water quality Impact rating at 25
silviculture! operations in Idaho, 1984.
Stream
Class1
Site
Compliance with
proposed rules
Water quality
impact rating
Idaho Department of Lands
1
1
1
II
II
Lightning Point
Trapper Creek
Willow Creek
Crazy Creek
Killarney Lake
Major departure
Major departure
Major departure
Minor departure
Minor departure
Severe hazard
Major hazard
Major hazard
Adequate protection
Adequate protection
United States Forest Service
1
1
1
II
II
II
II
II
II
Bryan Creek
Camp Eleven
Cedar Creek
Decorah
M.F. Weiser River
Bonaparte
Deer Creek
Olson Tunnel
Spring/Done Creek
Tollgate
Exceeds requirements
Exceeds requirements
In compliance
In compliance
Minor departure
Minor departure
Exceeds requirements
In compliance
Exceeds requirements
Minor departure
Adequate protection
Adequate protection
Adequate protection
Adequate protection
Minor hazard
Adequate protection
Adequate protection
Adequate protection
Adequate protection
Adequate protection
Private Operations
II
II
II
II
Bellgrove Creek
French Creek
Gold Fork Creek
N.F. Grouse Creek
Mica Creek
Thomas Creek
Laffinwell
Little Meadow Creek
Little Mud Creek
Little Salmon Creek
Minor departure
Major departure
Major departure
In compliance
In compliance
Gross neglect
In compliance
Minor departure
In compliance
In compliance
Major hazard
Major hazard
Major hazard
Adequate protection
Adequate protection
Major hazard
Adequate protection
Adequate protection
Adequate protection
Adequate protection
'Stream Class: I—Important for spawning, rearing, or migration of fish.
II—Not used by fish; principal value is downstream influence on class I streams.
Table 2.—Geology, land type hazard, and sediment impacts at 25 silviculture! operations in Idaho, 1984.
Land Pre-Existing Cum.
type sediment Project impact
Site Geology hazard condition1 sediment3 potential3
Idaho Department of Lands
Lightning Point
Trapper Creek
Willow Creek
Crazy Creek
Killarney Lake
United States Forest Service
Bryan Creek
Camp Eleven
Cedar Creek
Decorah
M.F. Weiser River
Bonaparte
Deer Creek
Olson Tunnel
Spring/Done Creek
Tollgate
Private operations
Bellgrove Creek
French Creek
Gold Fork Creek
N.F. Grouse Creek
Mica Creek
Thomas Creek
Laffinwell
Little Meadow Creek
Little Mud Creek
Little Salmon Creek
Mica schist
Glacial outwash
Batholith
Glacial Till
Hard metamorphics
Batholith
Altered granitics
Hard metamorphics
Hard metamorphics
Basalt
Batholith
Batholith
Altered granitics
Glacial till
Batholith
Soft metamorphics
Batholith
Alluvium
Glacial till
Basalt
Basalt
Basalt
Alluvium
Basalt
Basalt
High
High
High
Low
Low
High
Mod.
Low
Low
Low
High
High
Mod.
Low
High
Mod.
High
Low
Low
Low
Low
Low
. Mod.
Low
Low
Yes
Yes
Yes
N.A.
N.A.
Yes
Yes
No
No
Yes
Yes
Yes
No
No
Yes
Yes
No
N.A.
Yes
Yes
No
N.I.
N.I.
N.A.
No
Yes
. No
Yes
N.A.
N.A.
No
No
No
No
Yes
No
No
No
No
No
Yes
Yes
Yes
No
No
?
No
Yes
Yes
No
High
Mod. ••
High
Low
Low
Low
Low
Low
Low
Low
Mod
Low
Low
Low
Low
High
Mod.
Mod.
Low
Mod.
Mod.
Low
Low
Low
Low
'Pre-existing sediment condition: stream has been severely impacted by watershed activities as shown by observed cobble embeddedness.
2Project sediment: observed sediment delivery of damaging magnitude from the current forest operation.
3Potential for sustained damage to fishery habitat based on contribution to cumulative impacts.
457
-------
PERSPECTIVES ON NONPO1NT SOURCE POLLUTION
spite these practices because of hazards associated with
the land type.
The difference between administration of forest prac-
tices by the Forest Service and administration on State
and private lands is a major consideration regarding the
potential risk for cumulative impacts. Management prac-
tices based on watershed objectives under Forest Service
administration are the key to this process. The ability to
schedule forest practices in a watershed over space and
time is critical to prevent cumulative impacts. This process
reasonably assures that sustained damage to a protected
use will not occur in low and moderate hazard land types
on Forest Service lands. The potential for cumulative im-
pacts is considered moderate on high hazard land types,
however, because of the unknown risk associated with
mass failure from roads in these areas.
On State and private lands, implementation of proposed
FPA rules is expected to eliminate most problems identi-
fied during this study. However, the potential for cumula-
tive impacts is much higher than under Forest Service
administration because no mechanism addresses water-
shed planning and the ability to schedule forest practices
in a watershed over time.
Relation to Water Quality Standards
The Federal Clean Water Act (PL. 92-500) provides the
framework and goals for nonpoint source pollution control.
Water quality standards are the statutory and regulatory
basis for achieving the goal of the Clean Water Act (sec-
tion 101 (a) and 303(c)). Section 208 set up a means to
develop water pollution abatement plans for nonpoint
sources, and a method to identify a reasonable set of
BMP's for meeting water quality standards.
EPA has promulgated the Antidegradation Policy (U.S.
Environ. Prot. Agency, 1983) based on the Clean Water
Act. States are expected to include this policy in their
water quality standards. Interpretation of the Antidegrada-
tion Policy in relation to nonpoint sources is unclear.
Strictly interpreted, the Antidegradation Policy could be
construed to prohibit any ground disturbing activity from
which sediment would be delivered to a stream—regard-
less of the magnitude of resulting impact on protected
uses.
In Idaho, the Health and Welfare Board has substituted
more workable language for the Antidegradation Policy.
The bottom line in these standards is protection of benefi-
cial uses. The Forest Service has developed and con-
tinues to refine the technology by which cumulative effects
of timber harvest on salmonid habitat can be predicted
and monitored (Platts et al. 1983; Cline et al. 1981). The
Division of Environment is working with the Forest Service
to establish guidelines for the protection of beneficial
uses. These guidelines will establish specific watershed
objectives based on the sensitivity and importance of the
fishery. This process is a practical attempt to apply the
Clean Water Act to State standards and nonpoint sources
of pollution.
ACKNOWLEDGEMENTS: The author is grateful for the individ-
• ual efforts of the task force members and the contributions of the
participating agencies: Dewey Almas and Don Jones, Idaho De-
partment of Lands; Virgil Moore, Idaho Fish and Game Depart-
ment; Doli Obee, Idaho Conservation League; Jack Griffith,
American Fisheries Society; Dale McGreer, Idaho Forest Indus-
tries Council; Michael Cook, Nez Perce National Forest; and
Philip Jahn, Payette National Forest.
REFERENCES
Braun, R.L. 1979. State of Idaho forest practices water quality
management plan. Idaho Dep. Health Welfare. Div. Environ.
Boise.
Cline, R., G. Cole, W. Megahan, R. Patten, and J. Potyondy.
1981. Guide for predicting sediment yields from forested wa-
tersheds. Intermountain Forest Range Exp. Sta. U.S. Forest
Serv, Boise.
Geppart, R.R., C.W. Lorenz, and A.G. Larson. 1984. Cumulative
effects of forest practices on the environment, a state of the
knowledge. Ecosystems, Inc. for Dep. Nat. Resour. Olympia,
WA.
Idaho Department of Health and Welfare. 1980. Idaho water
quality standards and wastewater treatment requirements.
Div. Environ. Boise.
Idaho Department of Lands. 1979. Forest Practices Act and
rules and regulations pertaining to forest practices of the
State and Idaho. Coeur d'Alene.
Platts, W.S. et al. 1983 Guide for predicting salmonid response
to sediment yields in Idaho batholith watersheds. Intermoun-
tain Forest Range Exp. Sta. U.S. Forest Serv. Boise.
U.S. Environmental Protection Agency. 1983. Water quality
standards regulations. 40 CFR parts 35, 120, 131. Fed. Reg.
48 (217): 51400-13.
458
-------
ILLINOIS AGRICULTURAL SOIL EROSION CONTROL STANDARDS:
A USEFUL TOOL FOR NONPOINT SOURCE POLLUTION CONTROL
HARRY HENDRICKSON
GEORGE DEVERMAN
Association of Illinois Soil and Water
Conservation Districts
Springfield, Illinois
JIM PENDOWSKI
Water Pollution Control Planning Section
Illinois Environmental Protection Agency
Chicago, Illinois
ABSTRACT
Illinois' primary nonpoint source pollutant is sediment
from agricultural land. The Illinois Water Quality Manage-
ment plan estimated costs of sediment pollution. The leg-
islature and subsequently all 98 soil and water conserva-
tion districts adopted a soil erosion and sediment control
program with the following provisions: (1) progressive
standards aimed at reducing agricultural erosion to "tol-
erable" soil loss levels by the year 2000; (2) a complaint
program tied to the standards; (3) a cost share program
tied to complaints. State priority areas, especially lake
watersheds, are addressed through a cooperative selec-
tion process and watershed work plans. Water quality
goals are incorporated in watershed work plans. This ap-
proach has provided a balance between both State man-
dated soil conservation and water quality conservation
goals. Our presentation will present examples of this ap-
proach.
BACKGROUND
To fully appreciate the usefulness of Illinois' Agricultural
Soil Conservation Standards, Illinois' soil and water re-
source base and setting needs to be understood.
Illinois is the Prairie State. It is a rich agricultural empire
blessed with rich topsoil and a very favorable climate for
crop production. Only about 170,000 acres (or Vz of 1
percent of cropland) is irrigated. With about a $3.3 billion
share of exports, Illinois ranks as the number one agricul-
tural export state. Of 35 million total acres, about 28 mil-
lion is farmland. Almost 25 million acres is cropland with a
corn and soybean rotation being predominant. About 10.2
million acres of this cropland exceeds the tolerable soil
loss or "T" discussed in this paper. About 5.5 million of
these acres can be brought to "T" through conservation
tillage and other practices to manage concentrated wa-
terflows. Only .5 million acres can be treated solely with
conservation tillage. On another 3.7 million acres where
soil loss exceeds 2 "T", structural practices such as ter-
races are necessary. Finally, about 1 million acres may
need conversion to less intensive land uses such as pas-
ture or forest crops to reduce soil loss to tolerable levels.
Two thirds of Illinois' approximately 11 million citizens
live in the Chicago metropolitan area. 10 million citizens
depend on surface water for their drinking water supply.
Few of Illinois' 2,900 lakes are natural; these are primarily
in Northern Illinois or along river channels. Many im-
poundments have been built for drinking water as well as
other benefits such as recreation, flood control, and cool-
ing water.
Illinois' primary nonpoint source pollutant is sediment
from agricultural land. The Illinois Environmental Protec-
tion Agency estimated in 1977 that 7 million tons of sedi-
ment are deposited each year in Illinois lakes and that it
would cost about $18 million/year to dredge out that depo-
sition.
Another important facet of the Illinois economy is water
based transportation. The Mississippi, Ohio and Illinois
Rivers carry a great deal of barge traffic. Grain from five
states moves through the Illinois River waterway. The Illi-
nois River from Hennepin to its confluence with the Missis-
sippi is of special concern since it is like a 200-mile long
lake. Its average velocity is only 0.6 mile per hour and it
drops only 21/2 inches per mile. This resulted from the last
glacier, which relocated the Mississippi channel from the
present Illinois River to its present course. About 1.2
inches of sediment deposits in this section each year—
about 14 million of 25 million tons entering the Illinois
River annually. This process has accelerated environmen-
tal changes. The Illinois River was once a sportsman's
paradise and the second largest fresh water fishery after
the Great Lakes. It is now sediment clogged and its once
deep backwater lakes are incapable of supporting wildlife
as in previous times.
Lakes and waterways are integral parts of the Illinois
economy and their maintenance is vital. The potentially
impaired uses are of great economic significance. The
sediment itself may be a nuisance to normal lake or water-
way maintenance but the nutrients, oxygen demanding
organic material, and agrichemicals carried with the soil
particles are probably of more concern in water pollution.
Illinois has about 100,000 farm operating units. Owners
farm approximately 40 percent of farmland, but much is
professionally managed for trusts and investors.
CONSERVATION DISTRICTS
Illinois has 98 Soil and Water Conservation Districts
(SWCD) generally set up on county boundary lines. The
State Water Quality Management Plan assigned districts
the primary responsibility for reducing nonpoint source
sediment pollution. Districts develop rapport and working
partnerships with land operators through (1) resource in-
formation and education; (2) technical assistance, (3) in-
centives, (4) recognition, and (5) evaluation based on mu-
tually developed goals.
Districts work cooperatively with many other conserva-
tion agencies to achieve their natural resource manage-
ment objectives, but primary support comes from the Illi-
nois Department of Agriculture and the U.S. Department
of Agriculture (Fig. 1). The Soil Conservation Service pro-
459
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
ORGANIZATIONAL CHART
Drainage
Districts
IL. i
Wi
Sui
Dep
Enei
Nat.
i
1 — ~\
7Regl
Offlc
State
ter ||
•»y En
t. of Prota
,y ft *«•
Res.
i
L
Municipal
Governments C
i
_ 1 _.
Foresters
oral Fish and
»s Wildlife
Mgrs.
IL.
rir. Dept. of
ncy tlon
t i
County
iOVCI HflMlllS
1
98 SWCDS
COOPERATORS
5 SWCD DIR/
SWCD
SWCD Staff
S Regional
Represtntslhres
Dlv. of Nat.
Resources
IL. Dept of
Agriculture
SWCD Adv.
Board
Direct administrative relationship
Limited relationship for specific pu
poses
18 Land Use
Councils
AISWCD Directors
AISWCD
10 Officers
18 Directors
7 Departments
8 Committees
8 Staff
North Central
Regional Council
8 States
Natl. Assoc. of
Conservation
Districts
SCS ASCS CES
District
Conservationist
(Tech AsslsL)
i
County Exec.
Director
(Cost Sharing)
Co. Committee
7 Area
Conservationist!
l_
9 District
Directors
Slate
Conservationist
Technical «
Admin Staff
Champaign
State ASCS
Exec. Dlr.
State ASC
Committee
Springfield
i
Soil
Conservation
Service
SERVICE
LEVEL
Extension
Advisor
(Education)
Ext. Council
i
10 Regional
Directors
1
Director
IL. Coop. Ext.
Service
Research a
Educ. Services
Urbana
i
Ag. Stablllz.
and Conserv.
Service
Coop.
Extension
Service
U.S. DEPARTMENT OF AGRICULTURE
COUNTY
AREA
STATE
REGIONAL
NATIONAL
Figure 1.—Soil and Water Conservation District (SWCD) and Association of Illinois Soil and Water Conservation District
(AISWCD) Organizational Chart.
ILLINOIS WATERSHED SELECTION AND IMPLEMENTATION PROCESS
LANDOWNERS :•
ASSOCIATIONS •
mpnriULITUI
WCI
1 .,, , I««0«UATIO»
' 1 AI1WCD/1WCO L«U
1 WMwitad Inncilen
1 application 1
16 LAND USi COUNCILS
REGIONAL PRIORITY
SETTING
STATE WATERSHED
PRIORITY COMMITTEE
AISWCO ASCS SCS. IDOT
rSWS. IOOA. IEPA. IDOC
USEPAWRC
I
^^^S" *£m£oa!t*
SCS ASCS
As" Looe Migrant Sim Law
Spring Lake
Racoon Laka
Ljfct Spruigflifei
PROGRAM NAME
ADMINISTERING
AGENCIES
EXAMPLES Of
LAKE WATERSHEDS
TREATED
JU LAKES
WATERSHED
TREATMENT
USEPA • IEPA - SWCD
Ljke Le Aqua Na
WM Lau< A mini
1
ACP SPECIAL
ASCS
LJke U Aqua Nl
Sam Date L«K*
LaUPmMll
wawrty LJM
Krtukd Lau
LakvManoon
ParuLaka
Sicpfwn Forbes Lakt
NairmDe Rraancw
Watfiingtoo County Ljto
CanunLaka
inr-ji luTMU^n BUILD ILLINOIS
LOCAL WATERSHED WATERSHED
HornarMMia^f AaaociaMn IOOA - SWCDa
Muric«alM> - SWCOa ^^ •->--
*HZ™ <*££%£*»
H"*-S£"° krmMMajk
•JJJJ^JJjJJ^ FV nSI-IWlMHil
1 AISWCO
1 ***
Figure 2.—Illinois watershed selection and Implementation process.
460
-------
WATER QUALITY CRITERIA AND STANDARDS
vides technical assistance. The Agricultural Stabilization
and Conservation Service (ASCS) provides farmer incen-
tives, and the Cooperative Extension Service develops
educational programs. District employees provide techni-
cal assistance, information programs, and demonstra-
tions. By next year we hope to have about 200 SWCD
employees: at least one resource conservationist and a
secretary in each SWCD office.
District directors are direct links to the most important
part of the system—land operators. Illinois district direc-
tors contribute over a million dollars in time each year.
Since districts do not have taxing authority, directors are
frequently searching out funds. As in other States, dis-
tricts have organized themselves into councils and a State
association, which helps develop a consensus on regional
and State conservation issues. The association empha-
sizes an action program of assisting SWCD's to develop
local watershed projects. Over 30 local watershed projects
with a variety of funding mechanisms are now operating.
Many projects have local funds. A State watershed selec-
tion committee reviews projects for possible applications
of State or Federal funds (Fig. 2). All projects are geared to
meet the SWCD soil erosion standards in specific water-
sheds above affected waterbodies.
The districts' arrangements with numerous agencies
and organizations are complex, but in Illinois the soil ero-
sion control standards help to unite the conservation famiy
and give it direction and a measurable goal. SWCD's have
significantly improved water quality and reduced sediment
through targeting lake watersheds.
ILLINOIS SOIL AND WATER
CONSERVATION DISTRICTS SOIL
EROSION AND SEDIMENT CONTROL
STANDARDS
Goals were adopted and endorsed on the State level
through the water quality management planning process
19(3
SWCO'i ofepMdprognm m*rtosdonfa to M opprond hf
DBnots Deportment of Ayiitwtw
Pragma BFKTIW (b iuni|jB»i(«)
nO tat a or Mow 4 "T"* (4-20 ton/oc.)
STATE EROSION t SEDIMENT CONTtOl GUIDELINES ADOPTED
April IS, 1912
January 1,
January 1,1988
January 1,1994 '^^ i.s"T»orbn
January 1,2000 '^•^<
Und 5% dop» or bn - "T" or leu (1-5 ton/it.)
OKnr M • 2 "T" or Ira (2-10 ton/oc.)
•I ton Mr ocrt of sot bn.
Figure 3.—Schedule for Illinois Soil and Water Conservation
District's Soli Erosion and Sedimentation Control Program
and Standards.
(Fig. 3). The State goal is to reduce the average annual
erosion rate on agricultural land to or below the "tolera-
ble" soil loss or "T" for that soil type by the year 2000. "T"
is based on maintaining long term agricultural productivity.
Normally 3 to 5 tons of soil/acre/year, on certain fragile
soils it is only 1 ton/acre/year. Average soil loss is mea-
sured by the Universal Soil Loss Equation (USLE). The
standards are progressively more stringent. Till 1988 soil
loss should be under 4 "T." To comply after 1988, the goal
will be 2 "T" on slopes greater than 5 percent and "T" on
less than 5 percent slopes. By 1994 soil loss should be
under 1.5 "T." Farm conservation planning is aimed at the
year 2000 goal of being at or under "T."
As mandated, all 98 SWCD's adopted the state guide-
lines as their standards by the 1983 deadline. Sixteen
districts adopted more stringent standards. Wisconsin,
Minnesota, Ohio, and Indiana are considering similar
standards or already have them.
A complaint program was adopted with the district
standards for offsite problems caused by agricultural soil
erosion (Fig. 4). Anyone can file a complaint. The SWCD
notifies the land operator, investigates to determine com-
pliance with the standard and assists the land operator to
Erosion
problem -
sighted
P Complaint filed Investioation undertaken
m S 4 WCD Ofi.ce ' by qualified technician 1
1
S 4 WCD Board
determines
'compliance or non-compliance
1
EROSION A SEDIMENTATION standards
CONTROL COMPLAINT ™t-no
PROCESS ,- action necessa-y
1
Violation
exists
Violations affecting Violator notified
water quality can be Reasons for (landowners 4 operator)
enforced by actions noncompiiance -technical assistance offered
before the Illinois resolved • ' -cost sf
Polution Control Board f -schedi
1 | Conservation __ Violator
plan implemented ' cooperates
Reasons for
noncompiiance «— •
resolved
Formal hearing
held m S 4 WCD Violator does
" -findings published * not cooperate *~~^
Form.! hearing «"d made pubnc
—held by IDOA «_
-findings published
and made public
anng sought
jle lor compliance agreed upon
Figure 4.—Complaint process.
461
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
develop a schedule of compliance and implement it. If a
violation goes unattended for a year, the district and the
Illinois Department of Agriculture can conduct hearings. If
there is still no action the case may be referred to the
Illinois Pollution Control Board. The combined forces of
reason, public opinion, and cost sharing have resulted in
no outstanding violations. Of 73 complaints filed to date,
47 were found out of compliance with the district stand-
ards. In all cases the landowners entered into a compli-
ance schedule.
A State cost share program for addressing complaints
started with a modest $50,000 in FY 1984 and remained
the same in FY 1985. Some friendly complaints were filed
to help secure cost sharing. Road commissioners also
filed a number of complaints. The complaint fund will
probably be doubled in FY 1986 and a $4 million "Build
Illinois" cost share program for structural conservation
practices will probably be enacted. This program was
based partly on the success of and interest generated by
the complaint program, but primarily on the districts' docu-
mentation of their needs to meet their standards.
USEFULNESS OF STANDARDS
To understand the usefulness of the soil erosion control
standards, three levels must be considered: the land-
owner or local level, the county office, and the State of-
fices.
First, if the land operators did not judge the standards
useful they would likely fail. Farmers must make a profit. A
soil erosion standard based on productivity encourages
the land operator to maintain his land's productivity and
profitability. Saving soil is in the land operator's best inter-
est.
The land operator also knows that a conservationist was
present and actually measured slope and calculated the
USLE, which gives these standards more credibility. Using
an Illinois Extension publication, #1220, the land operator
can measure and verify his own soil loss. The standards
and compliance schedules also provide a reasonable per-
iod for land users to develop management skills in new
soil saving methods such as conservation tillage.
For those landowners affected by sedimentation, the
standards and complaint program is a reasonable process
for solving a problem beyond their control. Many com-
plaints have been filed by neighboring farmers. The com-
plaints and their documentation emphasize the offsite
benefits and the public role in finding and funding solu-
tions. An example of a cooperative program for reducing
soil loss to tolerable levels above a lake watershed is
found in Jo Daviess County. The Apple Canyon Lake
Homeowners Association worked cooperatively with land-
owners, the Jo Daviess District, and USDA agencies and
provided $25,000 to help install conservation practices
above the lake.
At the county office level, soil erosion standards are
extremely useful. District evaluations, staffing, work plans,
and priority area designations are based on the districts'
standards. The documentation of their needs to meet "T"
by 2000 was crucial in the development of the Build Illinois
program.
The soil erosion standards provide a defensible goal to
county level soil conservationists when assisting land op-
erators with conservation planning. Conservationists also
document their work based on meeting their district's
standards. With a productivity based standard they also
have a more marketable program and a higher potential
for land user's acceptance.
County and State level information programs have suc-
cessfully built an understanding and popularity of "T." In-
formation programs virtually always mention the county
standards. Agribusiness frequently mention soil conserva-
tion standards in promoting their products and services.
"T farming is for our kids" is a popular bumper sticker.
Many other "T by 2000" promotional items have been
produced by districts.
County and State educational programs emphasize the
Universal Soil Loss Equation, conservation tillage, and
maintaining production while practicing conservation
farming. Numerous slide, newsletter, and news column
presentations reference the standards.
Lake watershed projects' land treatment goals are typi-
cally based on reaching "T" standards. To reduce sedi-
mentation to desired levels in some lakes, standards less
than "T" sediment retention structures, or shoreline ero-
sion practices may be needed. At the State level, the
standards are integrated with the reporting and allocation
systems. The Illinois Department of Agriculture allocates
State funds for conservation districts based on staff pro-
ductivity in meeting "T" goals, land not meeting "T", and
other factors. State training programs for SWCD directors
and annual conferences aim at meeting "T" goals. State
cost share funds allocations are based in part on land with
soil losses greater than "T." All State agencies were or-
dered by Governor Thompson to bring soil loss to these
standards on land they control.
For the Illinois Environmental Protection Agency, the
agricultural soil erosion control standards foster coopera-
tion with the agricultural community. The standards have
helped put a greater emphasis on critical area treatment
and have helped direct Federal funds to water quality
problem areas. The standards have helped various State
groups to come together through participation in each oth-
er's decisionmaking processes.
Federal agencies, too, utilize the "T" goal. All SCS con-
servation planning is based on meeting "T" goals. ASCS
utilizes "T" goal data in allocating funds and setting priori-
ties for special projects.
"T" standards and specifically the "T by 2000" goals
have permeated the entire soil and water conservation
and nonpoint source water quality programs in Illinois.
Everyone benefits from clean water, navigable waterways,
and the maintenance of productive farmland. Agencies
have designed their programs to help SWCD's meet their
standards. Excellent cooperation and hard work by the
conservation family have helped districts increase the
probability of meeting their soil erosion standards.
ON THE HORIZON
Anticipating problems and adjusting programs to meet
them are basic tenets of management. The easiest prob-
lems, however, have already been solved. The wide-
spread adoption of conservation tillage has helped a great
deal. The bulk of the additional reductions in soil conser-
vation will require construction practices and difficult land
use conversions. Districts and the Illinois Department of
Agriculture have a plan but completing it will require about
$1 billion for enduring conservation practices, 200 addi-
tional technical staff, and 70 additional educational staff.
That plan's achievement will depend on citizens recogni-
tion of benefits.
Another key concern is the documentation of improved
water quality. Since the "T by 2000" agricultural standard
is now well established, the Illinois Environmental Protec-
tion Agency will be reemphasizing water quality rather
than soil loss.
As farmers bring their farms closer to "T," it will be
increasingly more difficult to sell farmers on conservation
practices. A 5 ton/acre soil loss amounts to only about
0.03 inch of topsoil evenly spread over the field. This is
462
-------
WATER QUALITY CRITERIA AND STANDARDS
almost invisible even to the farmer. Farmers can see the Finally, flexibility to changing market conditions and
gullies and rills associated with 4 "T" fields, but considers- technologies is required. Research efforts must be aimed
bly more education and promotion will be needed to get to at improving water quality and maintaining soil productiv-
the much less visible "T" level. ity while improving farm profitability.
463
-------
GROUND WATER QUALITY STANDARDS
U. GALE MUTTON
Nebraska Department of Environmental Control
Lincoln, Nebraska
ABSTRACT
Contaminated ground water is difficult, if not impossible
to restore. Water quality standards do not protect this
resource, but they should serve as th< basis for a pro-
gram preventing contamination. In I dbraska, where
ground water use as a drinking water supply is even
higher than the national average, standards and regula-
tions are being revised to support each other as part of a
more comprehensive water protection program. New
standards are being written, where needed, under a
Standards Implementation Strategy that identifies exist-
ing gaps in data needed to draw up better standards.
In discussing the approach we use in Nebraska to refine
ground water quality standards, I will not examine criteria
for contaminants or the rationales behind them. Instead I
will focus on institutional arrangements and policies we
are using to protect ground water.
Obviously ground water is an extremely important re-
source throughout the United States. It is reported that
ground water constitutes more than 96 percent of all fresh
water in the United States. It supplies 50 percent of the
U.S. population with drinking water. It is used by 95 per-
cent of the rural population as a potable supply. Principal
uses of ground water are irrigation, public drinking water,
and industry.
Statistics for Nebraska show an even greater reliance
on ground water. Only two communities within the State
rely solely on surface water instead of ground water as a
drinking water source. Nebraska ranks third nationally in
total ground water use, behind California and Texas, and
trails only California in ground water used for agricultural
purposes. On a per capita basis, this ranking is even
higher.
We have relied upon ground water for generations with
little thought to the possibility of exhausting the supply or
contaminating this resource. Many people have believed
that ground water is a nearly pristine resource insulated
from contamination.
We are becoming more aware, almost on a daily basis,
of ground water contamination and resulting use impair-
ments. For instance, on the national level, reportedly
8,000 private, public, and industrial wells have been
closed or in some way affected by contamination.
In Nebraska, we are aware of 136 ground water contam-
ination sites, most of which have been identified by our
Department of Health in its monitoring of municipal wells.
Synthetic organic compounds (SOC's) have been de-
tected in 10 public drinking water supplies. Nitrate-nitro-
gen levels have been identified to excess of maximum
contaminant levels (10 ppm) in 86 public water supplies
and numerous private wells. Thirty-seven communities
within Nebraska have taken corrective action to alleviate
nitrate problems, involving in most cases relocating the
water supply at considerable expense.
Sources of contamination in Nebraska are similar to
those identified nationally. Among the most likely sources
in Nebraska are agricultural chemicals (pesticides and fer-
tilizers), waste treatment sites and waste disposal sites,
chemical and fuel storage facilities, improperly con-
structed or abandoned water wells and test holes, indus-
trial facilities, and accidents along transportation corri-
dors.
Most of the sources I have just listed would by definition
fall into the point source category. However, the most
widespread contaminant in the State is nitrate-nitrogen.
Nitiates can originate from both point and nonpoint
sources. However, in several large areas throughout Ne-
braska nitrates appear to come primarily from nonpoint
sources (agriculture-related activities).
Although we in Nebraska certainly do not claim to have
all of the answers, through the past 4 years the Depart-
ment of Environmental Control, other agencies, and con-
sultants have devoted a great deal of thought and effort to
the issue of ground water quality protection and the devel-
opment of our strategy.
The goal of ground water quality standards, as we per-
ceive it, is to protect ground water quality for actual and
attainable uses. This protection should be prevention-ori-
ented. Contaminated ground water is difficult, in some
cases impossible, to restore to original baseline condi-
tions. Add to this the economic constraints to restoration,
and it becomes apparent that prevention is the best option
if existing quality is to be maintained for future genera-
tions.
Standards, in and of themselves, do not provide any
protection to the resource; however, they should serve as
the foundation on which preventive programming is
based. For this reason we believe strong, well-defined,
and comprehensive ground water standards are the hub
for developing effective pollution prevention programs.
Nebraska has had standards for ground water quality in
effect since 1978. They have remained unaltered since
then and, quite frankly, have not been used to the extent
necessary to protect aquifers. The lack of use does not
reflect lack of initiative by our Department, but rather the
very general and nondescriptive nanner in which the
standards are written. The existing standards were a good
first step and have allowed for some protection and en-
forcement in cases where gaps existed among program
authorities.
Basically, our existing standards consist of an antidegra-
dation policy. Uses for aquifers have not been identified,
and the specific criteria stated within the standards are
based upon human health. These standards contain both
general criteria and numeric criteria along with a reporting
requirement. The list of constituents for which specific cri-
teria have been established is not particularly comprehen-
sive.
We are presently involved in reviewing the standards
and intend to complete these revisions by the end of this
calendar year. We now believe that this will involve a major
renovation of existing standards.
Three primary objactives that have been developed to
guide our efforts reflect our philosophy toward the use of
standards in protecting ground water quality. The objec-
tives determine standards as:
1. A guide for program development (that is, reflect
sensitive areas and serve as a basis for program regula-
tions and permit limits);
2. A mechanism for identifying problems, triggering en-
forcement action, and prioritizing planning activities; and
464
-------
3. The yardstick for measuring cleanup needs and res-
toration levels.
Our philosophy in developing new standards to meet
these objectives focuses on two premises. First, if stand-
ards are to guide program development, the criteria estab-
lished within them should apply to the programs and must
be easily translated into program specifications (for exam-
ple, permit limits, monitoring requirements, and so on).
Secondly, standards should be enforceable in and of
themselves, although the most effective enforcement will
result when standards are used to trigger regulatory activi-
ties in programs involving fuel and chemical storage, un-
derground injection, chemigation, pesticide management,
municipal and industrial lagoons, septic tanks, solid waste
disposal sites, and agricultural management practices, to
name a few.
Many of the program areas already have regulations in
place. However, within the past 4 years we have identified
numerous program deficiencies within and between exist-
ing State and Federal programs. We are proposing modifi-
cation in several areas of deficiency and new program
development in others. We believe it is paramount, as a
first step, to establish well-defined standards to guide
these program modifications and developments.
Some specific areas of the standards that will be devel-
oped on a statewide basis include:
1. Ground water area delineations based upon existing
and potential uses.
2. Use designation for each ground water, including
classes within some use designations (for example, drink-
ing water).
WATER QUALITY CRITERIA AND STANDARDS
3. Maximum contamination levels for probable contam-
inants.
4. Triggering or trend criteria on the most probable con-
taminants: planning, monitoring and compliance activities
will be initiated.
5. Reporting and liability clause.
6. Strict antidegradation clause.
7. Implementation mechanism for the antidegradation
clause (Continuing Planning Process (CPP) document).
8. Narrative criteria for less probable or new contami-
nants.
9. Implementation mechanism for transfer from narra-
tive criteria to numeric criteria (CPP document).
10. Restoration policy and a protocol for cleanup and
restoration decisions (CPP document).
The fact that we do not have ground water quality data
from many areas of Nebraska and also lack data on envi-
ronmental fates and chronic health effects limits a com-
prehensive and well-defined coverage of many compo-
nents of the standards. With this understanding, we will
therefore be identifying data gaps and developing a stand-
ards implementation strategy concurrent with our revision
activities. This strategy will plot a course for establishing
the data necessary to further refine standards. We antici-
pate this will be a multiyear implementation schedule.
Obviously, protecting our ground water is a major con-
cern in Nebraska. Our approach to standards is to make
them a more useful tool in preventing ground water con-
tamination.
465
-------
Sources and Fates of Material
Influencing Water Quality
in the Agricultural Midwest
MANAGEMENT PRACTICES TO REDUCE FARM CHEMICAL LOSSES
WITH AGRICULTURAL DRAINAGE
JAMES L. BAKER
Department of Agricultural Engineering
Iowa State University
Ames, Iowa
ABSTRACT
Field losses of nutrients and pesticides with surface run-
off represent concerns for the quality of receiving water
with respect to aquatic life and human consumption.
Movement of nitrate-nitrogen and possibly pesticides
through the soil profile with excess precipitation can af-
fect the quality of ground water through deep percolation
or surface water resources through shallow subsurface
drainage. The fate of a soil-applied chemical relative to
these modes of loss is highly dependent on the chemi-
cal's persistence and interaction or absorption to the soil.
In this paper chemicals will be classified as to major
mode of loss (sediment, surface runoff water, or subsur-
face drainage) and reasonable management practices
will be considered to control losses for each class. In-
cluded in the discussion will be conservation tillage; tim-
ing and methods of chemical application; chemical, for-
mulation, and additive choices; and monitoring to reduce
chemical application.
INTRODUCTION
Efficient application of nutrients and pesticides to the soil
in row-crop agriculture is important not only in economic
and energy conservation terms, but also from the environ-
mental viewpoint. Losses to surface and ground water
resources represent a double concern. An understanding
of the mechanism and magnitude of chemical losses is
needed before methods of reducing losses are consid-
ered. The three most important factors affecting the fate of
a soil-applied agricultural chemical are its persistence, its
soil adsorption, and the hydrology of the soil to which it is
applied.
Relative to persistence, transformations in the soil and
plant uptake probably account for the major part of a pesti-
cide or nutrient's disappearance or removal. For nutrients,
it is estimated that each bushel of corn grain removes 0.8,
0.13, and 0.20 kg/ha of nitrogen, phosphorus, and potas-
sium, respectively; corresponding numbers for a bushel of
soybeans are 3.3, 0.34, and 1.0 kg/ha (Fertilizer Inst.
1976). A significant amount of inorganic nitrogen and
phosphorus taken up by plants is also converted to or-
ganic forms left in the field as crop residues. Additional
inorganic nitrogen and phosphorus can be tied up or im-
mobilized by microorganisms that decompose crop resi-
due. Of course, organic nitrogen and phosphorus previ-
ously formed is continuously being mineralized or
converted back to inorganic forms, with the overall net
effect depending on reaction rates in the soil.
For nitrogen, losses to the atmosphere can result from
the transformation of nitrate-nitrogen to nitrogen gas or
nitrogen oxides, usually in wet soils under anaerobic con-
ditions. The Fertilizer Institute (1976) estimates 10 to 40
percent of the nitrogen applied may be immobilized and 5
to 35 percent lost by denitrification. For phosphorus they
estimate immobilization to be 50 to 90 percent, but phos-
phorus immobilization includes formation of nearly insolu-
ble compounds as well as tie-up by microorganisms.
Keeney (1983) and Nelson and Logan (1983) have written
detailed reviews on the fates, respectively, of the nitrogen
and phosphorus in the soil.
For pesticides, the major pathway of dissipation is deg-
radation in the soil, where the original compound is con-
verted to other compounds through chemical or biological
reactions. With only a few exceptions (for example, atra-
zine), currently used pesticides dissipate from the soil to
the point that no obvious pesticidal properties remain
within a year. Pesticides are often assumed to degrade at
a rate proportional to the amount present (first-order reac-
tion), and the term half-life is used to express the time it
would take for half of the pesticide to degrade. For exam-
ple, alachlor with a half-life of 2 weeks would be 75 per-
cent dissipated in 4 weeks, 87.5 percent in 6 weeks, and
467
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
so on. Whether a first-order reaction represents pesticide
degradation or not, very low levels (ppb) of pesticides
could remain in the soil for some time. If so, little is known
of the fate or environmental impact of these small
amounts.
The degradation products of a pesticide are generally
considered to be less toxic than the parent pesticide with
some notable exceptions, such as aldrin (which is con-
verted in the environment to dieldrin; both of these chlorin-
ated insecticides are now banned). However, considering
the large number of degradation products possible for
each different pesticide, not enough is known about the
fate and toxicity of all these possible compounds. In the
past, emphasis has been placed in studying the parent
pesticide. Plant uptake of pesticides is usually small com-
pared to other dissipation processes. In a study on the fate
of soil-applied dieldrin, Caro and Taylor (1971) estimated
that uptake by corn plants was about 5 percent of that
applied.
When a chemical is applied to the soil, which is a mix-
ture of organic and inorganic materials along with water
and air, the chemical attempts to distribute itself among
these compartments. Depending somewhat on soil prop-
erties, but to a much greater degree on its own properties,
a chemical will interact with or be adsorbed by the soil.
One measure of the degree of adsorption of a chemical is
the ratio of its concentration in soil to that in water when
the chemical, soil, and water are mixed together and are
at equilibrium.
For the neutral organic pesticides adsorbed by soil or-
ganic matter, this ratio, often called K, usually is in the
range of 1 to 1000. For example, for a 2-percent organic
. matter soil, atrazine would have a value for K of about 10,
which would result in runoff water from surface soil with 3
ppm, initially having about 0.3 mg/L of dissolved atrazine
(or, K = 10 - sjnj-gs). Of course, concentrations in the sur-
surface soil and runoff water would decrease as atrazine
was removed from the surface soil by both runoff water
and water infiltrating into the soil.
Because of the fixed, negatively charged sites on clay
surfaces associated with the cation exchange capacity of
soils, cations (positive ions) such as ammonium are also
adsorbed by soil (the organic portion may also have ad-
sorption sites). For ammonium, K has been measured in
the vicinity of 50 (Baker and Laflen, 1 983a). For this high a
value of K and for a soil without surface nitrogen fertiliza-
tion, ammonium can actually be extracted from rainwater
by the soil. For several storms, Johnson and Baker (1982,
1984) measured about 1 mg/L ammonium-nitrogen in
rainwater, with concentrations in runoff from cropland be-
ing reduced to 0.1 to 0.2 mg/L.
To illustrate the importance of adsorption as opposed to
solubility in determining runoff losses of pesticides, para-
quat, a highly soluble but strongly adsorbed herbicide be-
cause of its cationic nature, was never observed in runoff
water although it existed in the soil and sediment at ppm
levels (Johnson and Baker, 1982).
For anions (negative ions) few fixed, positively charge
sites exist, compared to negatively charged sites, and
thus the soil capacity for adsorption is rather limited. In
addition, the soil exhibits an order of preference for an-
ions, with nitrate among the least preferred; therefore a K
value of zero can usually be assumed for nitrate. The
phosphate also forms slighly soluble compounds with alu-
minum and iron, and therefore, K values for phosphate
(relative to available phosphorus in the soil) can exceed
1000.
Relative to hydrology, one of the key factors determining
how much chemical is lost in surface runoff, particularly
for the less strongly absorbed chemicals, is the amount of
infiltration that occurs at the beginning of a storm before
runoff begins. It is generally believed that only a thin layer
of surface soil, perhaps 1 cm, interacts with and releases
chemicals to rainfall and runoff. If a large amount of infil-
tration occurs, therefore, before surface runoff begins, not
only is the volume of runoff reduced, but the concentra-
tions as well—if infiltration moves some of the chemical
below the mixing zone. The overall hydrology of the soil
system will then determine how much excess water there
is and the route, surface or subsurface, it will take to a
water resource. Hydrology will also influence the potential
for erosion and sediment loss. Table 1 illustrates the im-
pact that initial infiltration, in conjunction with adsorption,
can have on chemical losses in surface runoff.
Depending on the volumes of sediment, runoff water,
and subsurface drainage, and on K or the degree of soil
adsorption, therefore, chemicals can be classified into
three groups: those lost mainly with sediment (strongly
adsorbed), those lost mainly with surface runoff water
(moderately adsorbed), and those lost mainly with leach-
ing or subsurface drainage water (weakly or nonad-
sorbed). Because chemical loss is a product of the volume
of carrier (sediment or water) and the chemical concentra-
tion in that carrier, a reduction in concentration or carrier
will reduce loss. Furthermore, knowledge of which group
a chemical belongs to is necessary if loss reduction is to
be achieved by reduction in the volume of carrier, be it
sediment, surface runoff, or subsurface drainage.
MANAGEMENT PlfMCTICES
Because of present economic restraints, the farmer has a
rather limited number of management alternatives he can
pick from for reducing sediment and chemical losses. This
is particularly true if he is locked into row-crop agriculture
and cannot use or make a profit with less chemical-inten-
sive crops such as forages and possibly small grains. Con-
sidering only row-crops, conservation tillage, soil incorpo-
ration of chemicals, improved timing and methods of
chemical application, use of special formulations or addi-
tives, and reduction in application rates seem to be high-
est on the list of possibilities.
Conservation tillage, which leaves some or all of the
prevous year's crop residues on the soil surface, has been
very effective in controlling erosion and sediment-trans-
ported chemical losses. The degree of erosion control is
highly correlated with the percentage of the soil surface
covered with residue. For example, in one study (Barisas
et al. 1978), a no-till system with 58 percent residue cover-
age reduced erosion by 98 percent relative to a mold-
board-plowed area with only 2 percent of the surface resi-
due covered.
Total nitrogen losses associated mainly with sediment
were reduced by 81 percent. The nitrogen loss reduction
was less than the soil loss reduction because the finer
sediments lost with no-till were enriched with nitrogen;
and solution nitrogen losses with no-till, although only a
small part of the total, were up because of decreased
incorporation of surface-applied fertilizer and possible
leaching from residue. This is one example of the possibly
counterproductive effects of conservation tillage.
Another concern is that the use of herbicides might
increase as tillage weed control is reduced. Also, leaching
might increase if infiltration is increased with conservation
tillage, although water quality benefits would result from
the reduced runoff. However in at least one study (Kanwar
et al. 1985), the amount of nitrate-nitrogen leaching in a
no-till plot was less than that in a moldboard-plowed plot.
The fate of herbicides broadcast-sprayed on crop residue
and susceptible to washoff and volatilization is another
concern (Baker and Laflen, 1983b).
For the chemicals lost mainly with surface runoff water,
468
-------
EFFECTS ON WATER QUALITY IN THE AGRICULTURAL MIDWEST
soil incorporation to decrease the concentration in the
rainfall-runoff mixing zone would be an effective practice.
However, tillage used for incorporation is a counterproduc-
tive practice if conservation tillage is needed, because
tillage destroys soil-protecting surface residue. It has been
shown that disk-incorporation of herbicides reduces sur-
face runoff losses by about a factor of three compared to
surface application without incorporation (Baker and
Laflen, 1979) and that plowing down fertilizer results in
runoff losses no greater than if no fertilizer were applied
(Timmons et al. 1973).
For chemicals lost primarily with subsurface drainage,
particularly nitrate, better timing of applications to match
crop needs should reduce the potential for leaching. In an
irrigation experiment, Timmons and Dylla (1981) found
that four nitrogen applications during the corn growing
season reduced nitrate leaching losses about 20 percent
compared to a single application of the same total nitro-
gen amount. Using mathematical modeling, Kanwar et al.
(1984) also predicted reduced nitrate leaching with de-
layed nitrogen applications. A reduced application rate
should also reduce leaching losses as a Minnesota study
showed (Gast et al. 1978) where nitrate-nitrogen leaching
losses increased nearly in proportion to applied nitrogen.
Others (Baker and Johnson, 1981; Zwerman et al. 1972)
have found similar results.
Formulations or additives used can potentially affect
losses with agricultural drainage, mainly as persistence
and adsorption are affected. For example, application of
nitrogen as ammonia or ammonium reduces the potential
for nitrogen leaching because of soil adsorption, hence
additives such as nitrapyrin are promoted because they
decrease nitrification or the conversion of ammonium ions
to nitrate ions. Formulations affecting adsorption also af-
fect the potential for surface runoff losses. In one study
(Barnett et al. 1967), 2,4-D applied as an amine salt had a
total runoff loss of about one-fifth that for a more strongly
adsorbed ester form. However, it should be again empha-
sized that the effect of adsorption on runoff losses is tied
to the amount of initial infiltration that occurs before runoff
begins. Had no infiltration occurred, runoff losses for the
less strongly adsorbed form would in all likelihood have
been higher than for the more strongly adsorbed form
(Table 1).
Reduction of chemical inputs to cropland through edu-
cation and monitoring has potential to reduce chemical
losses. Unnecessary application of insecticides to corn in
corn-soybean rotations is decreasing and use of monitor-
ing and prediction procedures for continuous corn should
result in further reductions of insecticide applications. Ac-
curate soil testing procedures also may reduce inefficient
fertilizer inputs.
RESEARCH NEEDS
To improve efficiency of uptake and to reduce nitrate
leaching losses, a reliable soil test for nitrogen would be
very valuable, particularly for humid regions. Such a test,
not now available, would have to take into account not
only the inorganic nitrogen present, but that which would
result from mineralization during the growing season. In
addition, some accounting for transformation, uptake, and
movement as affected by the variables of weather would
have to be part of the overall scheme. This would probably
require the use of mathematical models.
Logic suggests placing chemicals out of zones of higher
water movement to reduce chemical movement. For ex-
ample, there may be an advantage (currently unquanti-
fied) to placement of nitrogen in the ridge of a ridge-tillage
system where a larger portion of the water would infiltrate
in the valleys between the ridges. Herbicide banding just
over the row would not only reduce the amount of chemi-
cal applied, but would avoid application to the traffic-com-
pacted interrow areas where surface runoff losses are
known to be higher because more runoff occurs sooner.
On the other hand, chemical placement beneath a com-
pacted zone may reduce the leaching potential because of
less infiltration.
Baker et al. (1983) are working on a pointer-injector
fertilizer applicator that can incorporate liquid fertilizer
without incorporating crop residue. It also requires little
power to pull and has potential for multiple nitrogen appli-
cations to improve nitrogen-use efficiency. Others are
working on use of high pressure equipment to "blast"
liquid fertilizer into the soil. Work on equipment for appli-
cation of herbicides beneath crop residue in conservation
tillage systems is also ongoing. This would be particularly
useful for some of the herbicides on the market that are
strongly adsorbed (hence not leached and with runoff
losses controlled by erosion control) but require incorpora-
tion because of volatilization or photodegradation prob-
lems.
Finally, work on new equipment or concepts to control
runoff and erosion is still needed. For example, a new
tillage tool called the paraplow has some potential to in-
crease infiltration without increasing erosion by fracturing
the soil without inverting it and covering residue; and the
concept of slot-mulch tillage, used in the west to reduce
runoff and erosion from frozen soils, may have application
under other conditions.
REFERENCES
Baker, J. L., and H. P. Johnson. 1981. Nitrate-nitrogen in tile
drainage as affected by fertilization. J. Environ. Qual. 10:
519-22.
Baker, J. L., and J. M. Laflen. 1983a. Runoff losses of nutrients
and soil from gound fall-fertilized after soybean harvest.
Trans. Am. Soc. Ag. Eng. 26: 1122-7.
. 1983b. Water quality consequences of conservation
tillage. J. Soil Water Conserv. 38:186-93.
_. 1979. Runoff losses of surface-applied herbicides as
affected by wheel tracks and incorporation. J. Environ. Qual.
8:602-7.
Baker, J. L., T. S. Colvin, S. J. Marley, and M. Dawelbeit. 1983.
Use of a point-injector fertilizer applicator for better fertilizer
Table 1.—Percent chemical losses In runoff water as affected by depth of mixing zone, adsorption, and initial infiltration.1
Initial
infiltration2
cm
0.0
0.63
1.27
1.90
Depth of mixing zone = 0.63 cm
K = 0 K = 2 K = 10 K = 50
Depth of mixing zone = 2.54 cm
K = 0 K = 2 K = 10 K = 50
50.0
6.8
0.9
0.1
45.4
29.0
17.9
10.5
21.7
17.2
13.1
9.3
5.6
4.6
3.7
2.7
47.5
. 27.8
15.9
8.7
22.5
17.8
13.5
9.6
6.7
5.5
4.3
3.2
1.5
1.2
1.0
0.7
'From mixing zone assuming a 1-h rain at 3.8 cm/h and using a simple mixing model.
Infiltration that occurs before runoff begins; attar the initial infiltration period (0, 10, 20, or 30 min), it was assumed runoff was half of rainfall.
469
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
management with conservation tillage. Pap. MCR-83-114.
Am. Soc. Ag. Eng. St. Joseph, Ml.
Barisas, S. G., J. L. Baker, H. P. Johnson, and J. M. Laflen.
1978. Effect of tillage systems on runoff losses of nutrients, a
rainfall simulation study. Trans. Am. Soc. Ag. Eng. 21: 893-7.
Barnett, A. P., E. W. Mauser, A. W. White, and J. H. Holladay.
1967. Loss of 2,4-D in washoff from cultivated fallow land.
Weeds 15: 133-7.
Caro, J. H., and A. W. Taylor. 1971. Pathways of loss of dieldrin
from soils under field conditions. J. Agric. Food Chem. 19:
379-84.
Fertilizer Institute. 1976. The Fertilizer Handbook. Washington,
DC.
Gast, R. G., W. W. Nelson, and G. W. Randall. 1978. Nitrate
accumulation in soils and loss in tile drainage following nitro-
gen application to continuous corn. J. Environ. Qual. 7: 258-
62.
Johnson, H. P., and J. L. Baker. 1982. Field-to-stream transport
of agricultural chemicals and sediment in an Iowa watershed:
Part I. Data base for model testing (1976-1978). NTIS Rep.
No. PB 82-254 046. Natl. Tech. Inf. Serv., Springfield, VA.
1984. Field-to-stream transport of agricultural chemi-
cals and sediment in an Iowa watershed: Part II. Data base for
model testing (1979-1980). NTIS Rep. No. PB 84-177 419.
Natl. Tech. Inf. Serv., Springfield, VA.
Kanwar, R. S., J. L. Baker, and J. M. Laflen. 1985. Effect of
tillage systems and methods of fertilizer application on nitrate
movement in the soil profile. Pap. 85-2031. Am. Soc. Ag. Eng.
St. Joseph, Ml.
Kanwar, R. S., J. L. Baker, and H. P. Johnson. 1984. Simulated
effects of fertilizer management on nitrate loss with tile drain-
age water for continuous corn. Trans. Am. Soc. Ag. Eng. 27:
1396-9.
Keeney, D.R. 1983. Transformations and transport of nitrogen.
In. F. W. Schaller and G. W. Bailey, eds. Agricultural Manage-
ment and Water Quality. Iowa State Univ. Press, Ames.
Nelson, D. W., and T. J. Logan. 1983. Chemical processes and
transport of phosphorus. In: F. W. Schaller and G. W. Bailey,
eds. Agricultural Management and Water Quality. Iowa State
Univ. Press, Ames.
Timmons, D. R., R. E. Burwell, and R. F. Holt. 1973. Nitrogen
and phosphorus losses in surface runoff from agricultural land
as influenced by placement of broadcast fertilizer. Water Re-
sour. Res. 9: 658-67.
Timmons, D. R., and A. S. Dylla. 1981. Nitrogen leaching as
influenced by nitrogen management and supplemental irriga-
tion level. J. Environ. Qual. 10: 421-6.
Zwerman, P. J., T. Greweling, S. D. Klausner, and D. J. Lath well.
1972. Nitrogen and phosphorus content of water from the tile
drains at two levels of management and fertilization. Soil Sci.
Soc. Am. Proc. 36: 134-7.
470
-------
THE FATE OF MATERIALS EXPORTED BY THE BIG BLUE AND THE
BLACK VERMILLION RIVERS INTO TUTTLE CREEK
RESERVOIR, KANSAS
J. R. SHUMAN
G. R. MARZOLF
Division of Biology
Kansas State University
Manhattan, Kansas
J. A. ARRUDA
Division of Environment
Kansas Department of Health and Environment
Topeka, Kansas
ABSTRACT
Suspended sediments and dissolved nutrients in mid-
western reservoirs originate in agricultural watersheds.
Rivers carry the sediments and nutrients to reservoirs
formed by impounded rivers. Highest loadings occur dur-
ing storm flows, thus the hydrologic regime represents a
significant control variable. Most of the precipitation in the
Great Plains occurs in late spring and early summer, usu-
ally in the form of thunderstorms, but little similarity exists
from year to year. Once in the reservoir, wind generated
currents keep materials in suspension, thus establishing
turbidity gradients along the long axis of the reservoir.
These gradients of suspended materials control many bi-
ological processes, e.g., photosynthesis, decomposition,
particulate feeding by zooplankton and predation by zoo-
planktivorous fishes. These biological processes influ-
ence water quality; thus, while the original quality of the
water entering the reservoir is directly related to water-
shed conditions, watershed properties more subtly and
indirectly influence water quality.
INTRODUCTION
The number of reservoirs in North America has increased
most in the last three decades. In regions where no natu-
ral lakes existed a short time ago, there are now signifi-
cant numbers of large bodies of standing water. Reser-
voirs and lakes differ significantly in their morphometric
and hydrologic characteristics (Neel, 1963; Baxter, 1977;
Thornton, 1984; Marzolf, 1984), so that biological proc-
esses and their controlling variables also differ, being
more dependent upon the river and watershed events.
Since larger reservoirs are formed by impounding major
rivers, a reservoir is, in some ways, a lacustrine feature of
its river; therefore, a river flowing into a reservoir has an
important influence on biological processes in the reser-
voir. Thornton (1981) and Marzolf (1984) established the
significance of river inflow on water quality parameters
and reservoir productivity.
Silts and clays often dominate the particulate materials
in the inflow to reservoirs in agricultural drainages. This
report relates a study of the effects of a pulsed inflow
resulting from storms in the summer of 1984 to the topic of
nonpoint source pollution. The report compares the storm
flow effects to water quality characteristics and biological
processes during stable inflows.
COMPARISON BETWEEN RESERVOIRS
AND LAKES
Lakes historically were the main focus of limnological in-
vestigations. Our conceptions of reservoir characteristics
have therefore been biased towards those of lakes. The
following comparison of lakes and reservoirs addresses
that misconception.
Watershed Area. Lakes and reservoirs differ in water-
shed area. Lakes rarely have a ratio of watershed area to
water surface area exceeding 10; the ratio for reservoirs in
the Kansas River drainage basin averages 507 (Marzolf,
1984). These large ratios reflect the potential for greater
nutrient and sediment loading to reservoirs. The water-
shed is the geomorphic unit drained by the river; the sizes
of rivers are proportional to the sizes of watersheds. Con-
sequently, the large ratio for reservoirs results from dam-
ming large rivers.
Tributaries. Reservoirs receive water primarily from the
rivers impounded to form them. Lakes typically receive
water from several small inflowing streams. Thus, a single
dominant inflow point provides the materials in suspen-
sion and solution for reservoirs.
Hydraulics. The turnover time, or hydraulic retention
time, is greater for lakes than reservoirs. Marzolf (1984)
found average retention times for lakes to be 1,622 and for
reservoirs, 427 days. Materials entering a reservoir, in
most cases, are transported more rapidly through the res-
ervoir than in lakes (Groeger and Kimmel, 1984). Marzolf
(1984) referred to reservoirs as "continuous flow proces-
sors" and to lakes as "batch processors."
Depth. Reservoirs are shallower than lakes, and are
shallower in relation to water surface area exposed to the
wind. Many reservoirs in the Great Plains do not stratify
because of wind mixing; the shallow depth and long fetch
allow complete circulation. Particulate materials entering
the reservoir from the impounded river remain sus-
pended.
Shape. Elongate reservoir morphology (maximum
length several orders of magnitude greater than width),
large watersheds, and river influences act together to cre-
ate "lentic rivers" in that the materials in suspension and
solution at any point in space are dependent upon those
materials entering from points upstream.
TUTTLE CREEK RESERVOIR:
A CASE STUDY
Tuttle Creek Reservoir, 8 km north of Manhattan, Kansas,
is convenient to the Kansas State University campus, and
it exemplifies the reservoir characteristics just described
(Fig. 1). The reservoir was formed by damming the Big
Blue River. The Black Vermillion River joins the Big Blue
471
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
TUTTLE CREEK RESERVOIR
Surface Area (SA)
Drainage Area (DA)
DA/SA
Retention Tiae
Mean Depth
Maxiaw Depth
Volu
63.9 tox ,
24.900 tai
390
62 dayi
8.2 m
16.8 m
42,000 ha-m
DAM
Figure 1 .—Map of Tuttle Creek Reservoir showing the sam-
pling sites and associated reservoir morphometric param-
eters.
near the upstream end of the reservoir. The watersheds of
both rivers are used primarily for row crop agriculture,
corn and milo being important crops. Clay and silt, eroded
from these watersheds, enter the two rivers and, eventu-
ally, Tuttle Creek Reservoir. These inflows establish the
characteristically high inorganic turbidity in the reservoir.
The reservoir is longer than it is wide and is oriented in a
northwest-southeast direction (Fig. 1). Winds in this re-
gion generally follow the longitudinal axis of the reservoir,
thus wind mixing prevents stratification.
Less primary productivity occurs in reservoirs with high
inorganic turbidity than in reservoirs and lakes with clear
water and comparable nutrient concentrations (Osborne,
1972). Suspended clays and silts limit light penetration.
Although the primary production is diminished, secondary
productivity is as high in these reservoirs as in others with
more normal levels of primary productivity. The nutritional
base for the secondary producers is thus open to ques-
tion.
Patterns of inflow suggest two broad types: stable, or
base, inflows and pulsed storm inflows. Pulsed inflows
generally last a short time and follow precipitation events
in the watershed. Stable inflows occur during periods of
little or no precipitation, and are of low magnitude. Obvi-
ously, greater concentrations of materials enter the reser-
voir during storm flows, thus hydrologic inflow patterns
are important control variables. Thornton (1984) and Mar-
zolf (1984) address reservoir response to stable inflows
from a modeling standpoint, but storm flow responses
have not been investigated.
TUTTLE CREEK RESERVOIR INFLOW
PATTERNS IN 1984
A large inflow occurred in June 1984. Inflows began to
increase on June 7 (Fig. 2), peaked on June 16 at 161.8
million m3/day, and returned to normal or base flow condi-
tions (10.6 to 1.4 million m3/day) on June 27. From this
f. < ',„„'-pr^T- < -f-1 .BOH" -]
Figure 2.—Reservoir volume, inflow, outflow, and retention
time during the pulse period and subsequent stable period
in the summer of 1984. Arrows indicate sampling dates.
latter date to the end of the sampling season (November),
inflows remained near base flow levels.
Inflow and outflow rates influence reservoir volume. The
Army Corps of Engineers determines outflow rates based
upon conditions in higher order rivers in the Kansas River
and Missouri River drainages. The Corps of Engineers
held water in Tuttle Creek Reservoir and reduced outflow
during this storm to baseline levels (2.4 million m3/day,
Fig. 2). Consequently, reservoir volume increased immedi-
ately after inflow increased (June 8), peaked at 149,672
ha-m on June 26 (1 day before inflow returned to baseline
conditions), and gradually decreased to conservation pool
levels (42,000 ha-m) by July 23. Outflow was increased
when reservoir volume peaked, and returned to baseline
flows on July 28. Reservoir volume was above normal for
a period of 6 weeks, having increased 365 percent at peak
volume. Retention times, by definition, follow volume and
outflow patterns (Fig. 2).
Water samples were collected at eight stations along
the longitudinal axis of the reservoir (Fig. 2) and analyzed
for parameters believed to influence zooplankton nutrition
(the investigation underway at the time). Standard sample
collection and analytical methods for these parameters
were employed; Shuman (in prep.) discusses the details.
The first sampling date, June 8, was the day after the
initial increase in inflow (Fig. 2). Two additional sampling
dates (June 13 and 22) fell during the period of high in-
flows and high reservoir volume (outflow was still at base-
line levels). The next dates (July 5, July 13, and July 18)
fell during high outflows, baseline inflows, and high vol-
ume. The final two sampling dates included in this analy-
sis (July 27 and August 2) were during stable reservoir
conditions (baseline inflows, outflows, and volume). Of the
eight sampling dates, the first six occurred during various
stages of the pulsed inflow and high reservoir volume (the
pulse period), and the final two dates occurred during sta-
ble hydrologic conditions (the stable period).
RESERVOIR RESPONSES TO THE
STORM INFLOW AND SUBSEQUENT
BASE INFLOWS
Suspended Paniculate Matter
Concentrations were extremely high at the river input end
of the reservoir (Fig. 3) on the first and second sampling
dates. Concentrations of suspended solids exceeded
1,300 mg/L. These high loads moved down the reservoir
as input continued to be high, decreasing upstream and
increasing downstream concentrations. From July 5 to
July 18, when outflow was high, the pulsed load continued
to move down the reservoir as the retention time averaged
about 19 days at peak outflow.
During the stable period, July 27 and August 2, gradi-
ents in suspended paniculate matter were stabilized.
These patterns typify data sets previously collected from
Tuttle Creek Reservoir during stable reservoir periods
(Dufford, 1970; Taylor, 1971; Osborne, 1972).
472
-------
JUNE 8
01
E
3
VJ
Q
z
1600
1200
800
400
0
1600
1200
800
400
0
400
300
200
100
0
•\ ^ :
' a*' \
JUNE 13
- -\ o a ;
- * A^i__ .
JUNE 22
: :
400
300
200
100
0
400
300
200
100
400
300
100
n
JULY 5
400
300
200
100
0
1
•
- ^ , ..---•.. '•
•
tlVER MID DAM
NPUT RESERVOIR END
END
400
300
200
100
0
EFFECTS ON WATER QUALITY IN THE AGRICULTURAL MIDWEST
JULY 13
200 h 1 400
100
75
50
25
0
80
60
40
20
0
-a—
JULY 18
JULY 27
300
200
100
0
400
300
200
100
0
400
300
200
100
0
O
en
o«
O
r-»
<
m
O
c/»
o
r™
O
3
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
JUNE 8
JULY 13
I
o*
u
0
£
Z
<
w
<
240
180 >
120
60 •
o. -o
cr''
JUNE 13
240 '
60
.0 a a--"'0 ° "
•tr" > i '
' "^^
JUNE 22
240
180
120
60
a o, o •» "a-.
JULYS
240
180
120
60
'•a
7.6
7.3
7.0
6.7
6.4
7.6
7.3
7.0
6.7
6.4
7.6
7.3
7.0
6.7
6.4
7.6
7.3
7.0
6.7
A A
240
180
120
60
a.
' * * —-—
JULY 18
240
180
120
60
''a*''
JULY 27
240
180
120
60
AUG 2
240 •
180
120 •
60 •
k*>^-— -—««;-,, .:
:
RIVER MID DAM RIVER Ml° °AM
INPUT RESERVOIR END '^1|T RESERVOIR END
END EMD
Figure 4. — Alkalinity and pH measured along the long axis of the reservoir during the pulse and stable period.
JUNE 8 JULY 13
ISSOIVED AMMONIA, „ g / 1 * *
SSOLVED PHOSPHATE, Mg / L D D
0 Q
200
150
100
50
200
ISO
100
50
0
200
100
50
0
200
150
100
50
0
',_ ' L^— -«^ ' • •
^"*Ss~ ..a a.. ..-a
^^ , " ° .-
JUNE 13
. a a.
JUNE 22
a... ^ .
JULYS
•n ' «- 1 ' -**' *
• • •^'••a a a a a- a '
RIVER MID DAM
INPUT RESERVOIR END
END
3600
2700
1800
900
0
3600
2700
1800
900
0
3600
2700
1800
900
0
2400
1 800
1200
600
0
200
150
100
50
0
200
150
100
50
0
200
150
100
50
0
200
150
100
50
0
a''
JULY 18
V /*-*• ^
. a-... ..a... ...a...
"•a- o " '-a-
JULY 27
*r ' • '
AUG 2
•S^cE:^
RIVER MID DAM
INPUT RESERVOIR ENO
END
7.6
7.3
7.0
6.7
6.4
7.6
7.3
7.0
6.7
6.4
7.6
7.3
7.0
6.7
6.4
7.6
7.3
7.0
6.7
6.4
2400
800
1200
600
0
2400
1 800
1200
600
2400
1 800
1200
600
2400
1 800
1200
600
0
1
Z
g
d
o
O
r-
<
m
O
Z
-H
m
Figure 5.—Dissolved ammonia, dissolved phosphate (SRP), and dissolved nitrate concentrations along the long axis of the
reservoir during the pulse and stable period.
474
-------
EFFECTS ON WATER QUALITY IN THE AGRICULTURAL MIDWEST
No obvious patterns in dissolved nitrate concentrations
(Fig. 5) occurred during pulse or stable inflow periods.
Concentrations were highest during the storm inflow, but
since nitrate concentration was uniformly high (2-3 mg/L),
changes through time represented only small percent-
ages of the total. Thus, changes, while they may be large
in an absolute sense, are not always detectable.
Phosphorus. During pulsed inflows, dissolved phos-
phate (Fig. 5) was highest at the river input end of the
reservoir (mean concentration of 181.8 /xg/L). Concentra-
tions were higher during the pulse period than during the
stable period. Concentrations during the stable period
(July 27 to August 2) were lower at the river input end;
measurements during August to November were similarly
low. Mean dissolved phosphate concentrations on these
two dates were 70.
Dissolved and Particulate Organic Carbon
Dissolved organic carbon (DOC) concentrations (Fig. 6) at
the river input end of the reservoir during storm inflows
were similar to other sites in the reservoir (mean concen-
tration of 6.8 mg/L). Similar results were found for the
storm inflows in 1 983. Adsorption of the organics to the
more concentrated clay particles during floods possibly
decreases the DOC concentrations in the incoming water.
During stable hydrologic periods, with lower concentra-
tions of inorganic particulates, higher DOC concentrations
were observed at the river input end of the reservoir. This
pattern persisted through November. The mean DOC con-
centration (averaged over all sites) on these two dates was
6.8 mg/L.
Particulate organic carbon (POC) was measured on five
of the eight sampling dates (Fig. 6). Levels of POC during
the periods of high inflow (June 8 and 22) were notably
higher than levels during base inflow (July 13-August 2).
The peak POC concentration detected during storm in-
flows was 42 mg/L at the river input end of the reservoir at
the onset of the pulse. During the stable period, a gradient
in POC is established, with highest concentrations at the
river input end, suggesting adsorption of dissolved organ-
ics unto clay and silt particles entering at the river input
end of the reservoir during pulse and stable periods.
Although the potential for high levels of DOC entering
with pulsed inflows seems likely, adsorption unto the abun-
dant particulates would remove these from solution and
increase the levels of POC. Further, the high suspended
sedimentation loads at this time are dominated by silts
and clays, normally inorganic particles. Without adsorp-
tion of organics unto these particles, POC concentrations
should be quite low at this time. Since the reverse is true,
and the ratio of DOC to POC is less than 1, adsorption
seems a significant process during inflows.
Chlorophyll
Chlorophyll levels (an indication of algal abundance) in
Turtle Creek Reservoir (Fig. 6) are normally low (< 10 ^g/
L). Occasionally, concentrations (such as at the river input
end on August 2) compare to reservoirs and lakes with low
abiotic turbidity. However, concentrations below 2 /*g/L are
common, with the highest concentrations occurring at the
river input end. Since light is most limiting there, the algae
present are probably allochthonous in origin (Taylor, 1976;
Swanson and Bachmann, 1976). Osborne (1972) showed
primary productivity was lowest at the river input end, al-
though chlorophyll was highest there. Since chlorophyll
levels are low throughout the reservoir (often below detec-
tion), algae are unlikely to be a significant nutritive source
for secondary producers in Tuttle Creek Reservoir.
JUNE 8
JULY 13
D
b
o>
E
Z
O
o
at
O
o
to
X
a.
O
oc
O
20
15
10
5
0
JUNE 13
JUNE 22
JULYS
20
15
10
5
n
•
^*--^ ^--•* •
•
' 0" o o o ° ° o o '
20
15
10
5
-
.A
." *•,
...rf' ''to ° "
. a o ° o-"
8
6
4
2
RIVER MID DAM
INPUT RESERVOIR END
END
20
15
10
20
15
10
5
20
15
10
5
n
-
-
«... o_*^£^ '
«- °' ° o 0..--0 v .
8
6
4
2
0
RIVER MID DAM
INPUT RESERVOIR END
END
O
X
O
>
z
n
O
Z
3
ID
I
Figure 6.—Particulate and dissolved organic carbon concentrations along the long axis of the reservoir during the pulse and
stable period.
475
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
JUNE 8
JULY 13
D
b
O
o
25
20
15
10
5
0
JUNE 13
JUNE 22
JULYS
50
40
30
20
10
0
25
20
15
10
5
0
p -q
•
•
u- - *~ b o
50
40
30
20
10
n
25
20
15
10
5
0
, . , .
•
•
•
"~*^.-n— -t^n:-*..-^- ,
50
40
30
20
10
0
25
20
15
10
5
•
•
-
' • n— - -*• o .
RIVER MID
INPUT RESERVOIR
END
.
-
-
-
50
40
30
20
10
DAM
END
25
20
15
10
5
0
25
20
15
10.
5
0
100
75
50
25
0
b
JULY 18
JULY
AUG
o--7
50
40
30
20
10
0
50
40
30
20
10
0
RIVER
INPUT
END
MID
RESERVOIR
200
150
100
50
0
DAM
END
n
O
o
o
o
Figure 7.—Zooplankton (cladoceran and copepod) abundance along the long axis of the reservoir during the pulse
stable period.
and
Zooplankton
Striking differences in the abundance of zooplankton oc-
curred between the pulse and stable periods (Fig. 7). Both
taxa were far more abundant during the stable period.
Highest concentrations were found at the river input end
of the reservoir, with a decreasing gradient moving to-
wards the dam. Suspended paniculate matter concentra-
tions were highest at the river input end, as well as POC
concentrations. Since both taxa use suspended particles
as food, bacteria and clays with adsorbed organic matter
probably function as significant sources of nutrition.
During the period of high reservoir volume, zooplankton
abundance was much lower. The extremely low abun-
dances throughout the reservoir on June 22 and July 5
suggest that dilution due to increasing reservoir volume,
reduced birth rates, or increased death rates are responsi-
ble for the reduction in abundance.
RESERVOIRS AND AGRICULTURAL
WATERSHEDS: IMPLICATIONS FOR
MANAGEMENT
Impoundment of rivers creates bodies of standing water in
areas with few or no lakes. These reservoirs are different
in morphometry and hydrologic patterns from lakes and
from the river that was impounded. Silts and clays domi-
nate the input of materials from rivers draining agricultural
watersheds. Concentrations are highest during storm
flows. Wind-generated currents maintain these materials
in suspension, creating gradients along the long axis of
the reservoir.
Since the input of these materials to reservoirs is a
single point in space, agricultural runoff to reservoirs is
actually a point source, that is, the impounded river. Run-
off is only a nonpoint source in terms of the river itself. The
magnitude of silt and clay loading depends on watershed
size, land use practices, and hydrologic patterns.
Since reservoir water quality and biological productivity
are strongly influenced by the inflow of these materials
and establishment of gradients, the reservoir is function-
ally linked to its watershed. For reservoir management
objectives, this implies that hydrologic patterns and land
use in watersheds must be understood to adequately pre-
dict water quality and biological productivity. Developers of
reservoir management strategies should recognize these
linkages.
The categorization of agricultural runoff as a nonpoint
source of pollution to reservoirs is ambiguous. Further,
classifying runoff as a reservoir pollutant connotes detri-
mental effects. Whether biological and water quality re-
sponses to these inflows are negative depends on their
importance in management strategies and goals.
REFERENCES
Baxter, R.M. 1977. Environmental effects of dams and impound-
ments. Ann. Rev. Ecol. Syst. 8: 255-83.
Dufford, D.W. 1970. Analysis of the causative agents of turbidity
in a Great Plains reservoir. Thesis. Kansas State Univ., Man-
hattan.
Groeger, A.W., and B.L. Kimmel. 1984. Organic matter supply
and processing in lakes and reservoirs. Pages 277-81 in Lake
and Reservoir Management. Proc. 3rd Ann. Conf. N.A. Lake
Manage. Soc. EPA 440/5/84-001. U.S. Environ. Prot. Agency,
Washington, D.C.
Marzolf, G.R. 1984. Reservoirs in the Great Plains of North
America. Pages 291-302 in F.B. Taub, ed. Lakes and Reser-
voirs. Elsevier Sci. Publ. B.V., Amsterdam.
476
-------
Neel, J.K. 1963. Impact of reservoirs. Pages 575-93. in D.G.
Frey, ed. Limnology of North America. Univ. Wisconsin Press,
Madison.
Osborne, J.A. 1972. The application of a photosynthetic model
for turbid reservoirs—field investigation. Dissertation. Kansas
State Univ., Manhattan.
Shuman, J.R. In prep. Gradient analysis of dissolved and sus-
pended materials and its importance to zooplankton commu-
nity structure in a turbid reservoir. Dissertation. Kansas State
Univ., Manhattan.
Swanson, C.D., and R.W. Bachmann. 1976. A model of algal
exports in some Iowa streams. Ecology 57:1076-80.
Taylor, M.K. 1976. Total, planktonic and benthic photosynthetic
EFFECTS ON WATER QUALITY IN THE AGRICULTURAL MIDWEST
production in the Kansas River. Thesis. Kansas State Univ.,
Manhattan.
Taylor, M.W. 1971. Zooplankton ecology of a Great Plains reser-
voir. Thesis. Kansas State Univ., Manhattan.
Thornton, K.W. 1981. Reservoir sedimentation and water qual-
ity—an heuristic model. Pages 654-61 in H.G. Stafan, ed.
Proc. Symp. Surface Water Impoundments. Am. Soc. Civil
Eng., New York.
. 1984. Regional comparisons of lakes and reservoirs:
geology, climatology, and morphology. Pages 261-5 in Lake
and Reservoir Management. Proc. 3rd Ann. Conf. N.A. Lake
Manage. Soc. EPA 440/5/84-001. U.S. Environ. Prot. Agency,
Washington, DC.
477
-------
THE INTERACTION OF BIOLOGICAL AND HYDROLOGICAL PHENOMENA
THAT MEDIATE THE QUALITIES OF WATER DRAINING NATIVE
TALLGRASS PRAIRIE ON THE KONZA PRAIRIE RESEARCH
NATURAL AREA
J. VAUN McARTHUR
Savannah River Ecology Laboratory
Aiken, South Carolina
MARTIN E. GURTZ
CATHY M. TATE
FRANK S. GILLIAM
Division of Biology
Kansas State University
Manhattan, Kansas
ABSTRACT
The quality of water from native landscapes is the base-
line against which the impact of pollutants on surface
water resources must be evaluated. The King's Creek
watershed has been a U.S. Geological Survey bench-
mark watershed since 1979. It represents the landscape
type that gave rise to much of the agricultural midwestern
United States. Hydrologic, chemical, and biological mea-
surements have been made by ecologists performing
long-term ecological research (LTER) in tallgrass prairie.
Streamwater chemistry varies seasonally with the
amount and movement of ground water and soil water,
with changes in prairie vegetation, and in response to
changes in surface water discharge. Concentrations of
organic carbon, organic and inorganic nitrogen, and
phosphorus increase during storm flows. The paniculate
fractions in transport and in storage in the stream bed
and on the flood plain vary seasonally with recent hydro-
logic history and changes in vegetation.
INTRODUCTION
The term wafer quality implies some standard against
which waters can be compared. Since a continuum of
physico-chemical conditions exists in nature, so does a
similar range of standards. Water quality is a function of
the natural watershed, and the condition of the watershed
is a function of the geologic history of the basin. There-
fore, each ecological region must have a characteristic set
of standards. To establish these standards, measure-
ments must be made on pristine systems within an eco-
logical region. Baseline measurements taken over long
time frames would provide an estimate of the true range of
chemical and physical parameters for these ecological re-
gions.
Recent research on stream ecosystem properties has
shown the linkage between terrestrial and aquatic ecosys-
tems (Likens and Bormann, 1974; Hynes 1975). In any
lotic system there exists a continuum (Vannote et al. 1980)
of physical, biological, and chemical conditions. These
conditions represent the most probable states of stream
reaches dependent on hydrologic and geologic con-
straints.
The import, transport and storage of organic and inor-
ganic material in a given stream reach depends on both
biotic and abiotic factors. Included in the biotic component
is the nature of the microbial assemblages. The microbi-
ota form the tightest link between terrestrial and aquatic
ecosystems (McArthur et al. 1985). Organisms that feed
directly on microorganisms or upon organic material im-
ported into the stream further alter the quality of the water
through feeding (Wallace et al. 1982). Seasonal changes
in the terrestrial ecosystem would further alter the
amounts and qualities of material imported into a stream
reach. Primary production both in the stream and within
the watershed would alter the nutrient loading to the sys-
tem.
While biological processes play an important role in the
chemical quality of the water, physical (e.g., hydrologic)
events control directly the import, transport and amount of
storage. Extreme hydrologic events may act as reset
mechanisms in flowing waters (Gurtz et al. 1982).
Tallgrass prairies once covered greater than 6 percent
of the coterminous United States. This ecological system
was exceeded in total area only by eastern deciduous
forests. The landscape once represented by tallgrass prai-
rie has been converted into much of the agricultural land-
scape of the midwestern United States. The Konza Prairie
Research Natural Area (KPRNA) is the largest representa-
tive tract of tallgrass prairie. Less than 2 percent of the
Konza Prairie has ever been plowed, and the KPRNA is
managed to provide a range of conditions encompassing
those of tallgrass prairie prior to settlement.
The 1,060 ha Kings Creek watershed is located entirely
within the Konza boundaries. This watershed has been a
U.S. Geological Survey benchmark watershed since
1979.
We present here preliminary data resulting from ecolog-
ical research in tallgrass prairie. Streamwater chemistry
has been monitored in the Kings Creek watershed for
NO3-N, dissolved and particulate organic matter, specific
conductivity and phosphorus. These parameters have
been measured over seasons and during storm events.
Our results will provide useful values to help determine a
meaningful measure of water quality in the midwestern
United States.
METHODS
Baseflow water samples were collected weekly from the
Kings Creek drainage system in 500 ml bottles. Storm
events were sampled using automatic water samplers
(ISCO Model 2100). The samplers collected every 30 min-
utes after being actuated by increased stage height. Wa-
ter samples were analyzed for nitrate-N and soluble reac-
478
-------
tive phosphorus on a Technicon AutoAnalyzer II and
specific conductivity was measured. Total persulfate nitro-
gen (TPN) was determined by the nitrate method after an
alkaline persulfate digestion using a modified method of
D'Elia et al. (1977). Dissolved organic carbon analyses
were done on a Beckman 915-B Total Carbon Analyzer
after the samples had been filtered (Gelman Type A/E),
acidified and sparged to remove inorganic carbon. Panic-
ulate organic carbon was separated into size fractions
gravimetrically and combusted at 450°C for 4 hours. The
results are expressed as ash free dry mass (AFDM). Dis-
charge measurements were made from data collected by
the U.S. Geological Survey and see Gurtz et al. (1982).
Diurnal samples were collected by grab samples on
hourly intervals in June 1983. The samples were analyzed
as described above. Water collected during storm events
was used as inoculum (10 /J) and spread on a minimal
salts media with glucose added as the sole carbon source.
RESULTS AND DISCUSSION
Hydrological measurements of water years 1980-84 on
Kings Creek are shown in Figure 1. The annual hydro-
graph varies considerably from year to year in terms of
duration of the flow period and the number, magnitude,
and timing of major storm flows. This temporally variable
hydrological regime imposes limits on biotic processes
while influencing physical channel processes that further
mediate biological phenomena. Life cycles, food require-
ments, and behavior of aquatic species reflect adaptations
to the relatively unpredictable harsh extremes of drought
and flood in prairie streams. Organic matter storage pro-
vides both structure and a potential food resource to
aquatic biota and is strongly influenced by recent hydro-
logic history. Floods can cause high rates of export as well
as movement of large organic matter from the channel to
the adjacent banks of the floodplain (Gurtz et al. 1982).
The import, transport, deposition and entrainment of
nutrients in stream ecosystems is strongly influenced by
ONDJFMAMJ J AS
1200-
1000-
800
600
400
200
1981
1
u^
« 1200
1000-
800-
600'
400
200
1982
61 19
D J F
M
M
=r"
EFFECTS ON WATER QUALITY IN THE AGRICULTURAL MIDWEST
the hydrology. The frequency and/or intensity of hydro-
logic events are important in determining whether mate-
rial is retained or transported through the system. Dy-
namics of dissolved organic carbon concentration differed
during two storm events in 1981 (Fig. 2).
The early storm (Fig. 2a) shows dissolved organic car-
bon levels tracking changes in discharge. This storm (May
1981) occurred early in the water year. The pattern of
transport changed after repeated storms (July 1981; Fig.
2b). Dissolved organic carbon concentrations peaked
prior to the peak in discharge and subsequently de-
creased. This may be due to at least one of two factors: (1)
dilution of the dissolved organic carbon by increased rain
or (2) an actual decrease in available soluble carbon as a
result of previous leaching.
Bacterial population densities in the water column fol-
low patterns similar to the dissolved organic carbon (Fig.
2). McArthur (1984) has shown experimentally that bacte-
ria respond immediately to inputs of dissolved organic car-
bon but these studies were not done during storm events.
The increases in DOC and bacteria during storms may be
from the same sources (floodplain). Dissolved organic car-
bon leached from the floodplain and bacteria associated
with forest floor litter may be washed into the stream dur-
ing storm events.
Particulate organic carbon enters the stream from direct
litter inputs from riparian vegetation as well as lateral
movement of matter at the soil surface. (Gurtz et al. 1982).
Smith (1982) showed that most of the decomposition of
coarse particulate organic matter (CPOM) was due to mi-
crobial and physical processing. Macroinvertebrate shred-
ders (Merritt and Cummins, 1978), organisms that con-
sume CPOM, were an insignificant component in Kings
Creek. In some streams shredding organisms can contrib-
ute significant amounts of fine particulate organic matter
(FPOM) through their feeding activity and fecal material.
Most of the particulate matter in transport in Kings Creek
is smaller than 53 /xm; since invertebrate activity is mini-
A S
Figure 1.—Mean daily discharge (Us) from Kings Creek, KPRNA, at the station of the U.S. Geological Survey. Period of
record began April 1979. Numbers in parentheses are instantaneous peak discharges for the major storm flows (after Gurtz
etal. 1982).
479
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
mal and input is predominantly CPOM in the lower
reaches we hypothesize that this FROM is originating in
the prairie or from bank erosion.
The senescent growth of prairie grasses remains as
standing dead material which is physically broken up by
wind and precipitation and subsequently transported into
the stream channel. Furthermore, when water is present
in the grassland reaches of the stream and nutrient avail-
ability is high, extensive algal mats are formed. The
sloughing of senescent algae further contributes to or-
ganic matter transport.
Nitrogen and phosphorus concentrations in surface wa-
ters draining the Midwestern States are among the high-
est in the country ranging from 0.3 to 5 mg N/L for nitro-
gen and 0.05 to 0.20 mg P/L for phosphorus and are
related to the amount of watershed used for agriculture
(Omernik, 1977). Despite this, nitrogen and phosphorus
concentrations draining Kings Creek, a tallgrass prairie
stream, are low enough to limit algal growth during the
summer months (Tate, 1985).
Phosphorus concentrations in stream water are gener-
ally less than 0.01 mg P/L in Kings Creek (Tate, 1985)
during base flow. Phosphate concentrations increase up
to approximately 200 to 300 mg P/L during storm events
(Fig. 3). The major export of phosphorus occurs during
storm flows.
Inorganic nitrogen (NO3-N) concentrations in Kings
Creek fluctuate over seasons (Tate, 1985) and diurnal and
30O
400
IU
£ 30O
a!
CO
K
§ aoo
8
Z 100
8OO-
7OO
6OO
-500
o
~ 4OO
52
§ 300
o
° 200
IOO
Q
(4.38)
(2£6)r4l
0500
/
'
f
\
X
X
, V
T1
\( 18.98) A
\
\
X
\
%
\
X
X
X
X
X
fl
V
(IS.2O)
rh(3.i2)
\
1
^
0730 I03O I6OO
TIME (hours)
t
/
t
ri-
,-
(7.86) „-'
^^ ^^*
^^-^**
B
•h
(4.13)
Q(2.72l"
IOOO
800 z
m
z
0
6OO w
o
x
o
m
400 „
i-
g
o
2OO
900
800
Z
7OO g
z
600 o
o>
500 g
3)
4OO m
300 »
Of
ZOO _"
IOO
f\
06OO 07CO 06OO 0900 IOOO
TIME (hours)
Figure 2.—Changes In discharge (Us), mean bacterial
counts (Colonies/10 ML), arid dissolved organic carbon (mg/
L) during two storm events in Kings Creek. (A = May 1981,
B = July 1981) Dash line = discharge. Histogram bars =
plate counts. Numbers in parentheses are dissolved organic
carbon concentrations.
2 4 6 8 10 12 14 16 18 20 22 24 26 28
TIME (hours)
Figure 3.—Changes in nitrate-N concentration (jxg/L) and
specific conductivity (^mhos/cm) during one storm event in
watershed 1D on KPRNA (May 1983).
storm events. Figure 4 depicts changes in nitrate concen-
trations and specific conductivity over time during one
storm event (May 1983). Specific conductivity decreases
in response to increased discharge and then increases as
the stream returns to baseflow conditions. Nitrate patterns
are the reciprocal of the specific conductivity measure-
ments: there is an initial increase and then subsequent
decrease. The major export of nitrogen occurs during
storm events.
Less nitrogen is exported during the summer when ter-
restrial plants are actively growing. Significant amounts of
nutrients are available for export in the fall after the grow-
ing season or in the spring prior to the growing season.
The primary mechanism to maintain these nutrients within
a stream reach is uptake by instream producers. Tate
(1985) has shown lower nitrate concentrations in late
spring and summer, a period corresponding to the grow-
ing season of terrestrial vegetation, and higher nitrate con-
centrations in fall, winter and early spring corresponding
to the dormant season of terrestrial vegetation.
Diurnal variation in nitrate concentrations depends on
location within the drainage. Figure 5 shows nitrate con-
centrations at two sites over a 24-hour sampling period.
Lowest nitrate concentrations at a site 2 km from channel
origin occurred during the daylight hours, corresponding
to the period when photosynthesis occurs. Night-time lev-
els of nitrate at this site were more than double the day-
light values. Farther downstream (4 km from channel ori-
gin), in the gallery forest, the diurnal pattern in nitrate was
much different. The stream channel at this lower location
had a dense canopy inhibiting instream primary produc-
tion and subsequently altering the transport of nitrogen.
Total persulfate nitrogen is composed of nitrate, nitrite,
ammonia, and organic nitrogen. Tate (1985) reported ni-
trite and ammonia to be a negligible component of total
nitrogen in the Kings Creek system. The TPN values in
Table 1, therefore, represent the higher concentrations of
organic nitrogen found in Kings Creek, a tallgrass prairie
stream. Total persulfate nitrogen concentrations are an
Table 1.—Baseflow mean concentrations (ng/L) of nitrate-N
(n = 17) and TPN in four watersheds on the Konza Prairie.
Mg NOj-N/L M9 TPN/L
Watershed X SE X SE
1D
2D
N1
N4
1.6
3.2
24.2
3.5
1.58
4.05
13.14
2.87
281
237
188
185
37
32
45
38 .
480
-------
EFFECTS ON WATER QUALITY IN THE AGRICULTURAL MIDWEST
O)
3
45
40
35
30
25
20
15
10
5
Nitrate
Conductivity
I I I I I I I I I I I I I I I I if T Tl I I
450
400
350
300
250
200
150
100
50
O
CO
»>»
O
10 15
TIME (HOURS)
20
25
Figure 4.—Diurnal changes In nitrate-N C*g/L) in two reaches of Kings Creek, KPRNA (June 15-16 and 23-24,1983).
order of magnitude higher than nitrate-N concentrations Creek. Organic nitrogen was reported to be the major
(Table 1). Organic nitrogen would enter the stream form of nitrogen exported in headwater streams draining
through the same vectors as organic carbon. Organic ni- the Cascade Mountains (Triska et al. 1984). Preliminary
trogen is the major form of nitrogen exported from Kings data suggest that concentrations of organic-N and ni-
I
CO
O
O)
• GRASS/SHRUB
O GALLERY FOREST
22
20
18
16
14
12
10
8
6
0630 1030 1430 1830 2230 0230 0630
TIME
Figure 5.—Changes in phosphate-P concentration (MS/I-) during one storm event in watershed 2D on KPRNA (May 1982).
1
1
«
1
» W -*• ^' •
1 1
till
• til
481
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
trate-N in soil water are similar to those in stream water,
suggesting the contribution of soil water chemistry to
stream water chemistry. In contrast, export from streams
draining in the Hubbard Brook system (Likens et al. 1977)
have higher levels of inorganic nitrogen export relative to
organic nitrogen. The differences between these two eco-
systems support the contention that water quality stand-
ards must be based on local ecosystem processes.
The dynamics of nutrient import, export and storage in
aquatic ecosystems are controlled by many factors, in-
cluding precipitation and runoff patterns, microbial re-
sponse, and invertebrate feeding. Additionally, the sea-
sonality of terrestrial plant uptake affects the movement of
inorganic and organic nutrients through the soil water.
Processes that occur in the headwaters of a watershed
will determine the nature of material imported into lower
reaches, reservoirs and lakes. To adequately address
whether a particular land use is affecting the aquatic eco-
systems draining the landscape type, we must know the
extremes of the pristine system.
Concerns about the effects of artificial organic matter
loading can best be approached after understanding how
the system handles natural loading. Nutrient import from
agricultural sources may affect in-stream processes but
these effects cannot be definitely ascertained without suit-
able baseline information.
Baseline studies must be sufficiently extensive in spa-
tial and temporal scale to allow water quality perspective
to develop for an unmanipulated ecosystem. Only through
acquisition of a large database can we develop an under-
standing of how natural processes influence water quality.
Research must focus both on long-term changes, which
occur gradually over years or decades, as well as short-
term phenomena such as storm events, which can have
long-lasting consequences on stream water quality.
ACKNOWLEDGEMENTS: This work was supported by Contract
E-AC09-76SR00819 between the U.S. Department of Energy
and the University of Georgia's Institute of Ecology. We would
like to acknowledge B. Brock and D. Loring for field collections
of diurnal samples. D. Whittemore provided analyses of phos-
phorus data.
REFERENCES
D'Elia, C.F., P.A. Steudler, and N. Corwin. 1977. Digestion of
total nitrogen in aqueous samples using persulfate digestion.
Limnol. Oceanogr. 22: 760-4.
Gurtz, M.E. et al. 1982. Organic matter loading and processing
in a pristine stream draining a tallgrass prairie/riparian forest
watershed. Contribution no. 230, Kansas Water Resour. Res.
Inst., Manhattan.
Hynes, H.B.N. 1975. The stream and its valley. Verh. Int. Verein.
Limnol. 19:1-15.
Likens, G.E. et al. 1977. Biogeochemistry of a forested ecosys-
tem. Springer-Verlag, New York.
Likens, G.E. and F.H. Bormann. 1974. Linkages between terres-
trial and aquatic ecosystems. BioScience 24: 447-56.
McArthur, J.V. 1984. Interactions of the bacterial assemblages in
a prairie stream with dissolved organic carbon from riparian
vegetation. Ph.D. dissertation, Kansas State Univ., Manhat-
tan.
McArthur, J.V., G.R. Marzolf, and J.E. Urban. 1985. Response of
bacteria isolated from a pristine prairie stream to concentra-
tion and source of soluble organic carbon. Appl. Environ. Mi-
crobiol. 49:238-41.
Merritt, R.W., and K.W. Cummins, eds. 1978. An introduction to
the aquatic insects of North America. Kendall-Hunt, Dubuque,
Iowa.
Omernik, J.M. 1977. Nonpoint source-stream nutrient level rela-
tionships: a nationwide study. EPA-60013-77-105. U.S. Envi-
ron. Prot. Agency, Corvallis, Ore.
Smith, D.L. 1982. Macroinvertebrates and leaf decomposition in
a tallgrass prairie stream. Ph.D. dissertation, Kansas State
Univ., Manhattan.
Tate, C.M. 1985. A study on temporal and spatial variation in
nitrogen concentrations in a tallgrass prairie stream. Ph.D.
dissertation, Kansas State Univ., Manhattan.
Triska, F.J. et al. 1984. Nitrogen budget for a small coniferous
forest stream. Ecol. Monogr. 54:119-40.
Vannote, R.L. et al. 1980. The river continuum concept. Can: J.
Fish. Aquat. Sci. 37: 130-7.
Wallace, J.B., J.R. Webster, and T.F. Cuffney. 1982. Stream de-
tritus dynamics: regulation by invertebrate consumers. Oeco-
logia 53:187-96.
482
-------
IMPLICATIONS OF AIRSHED PROCESSES AND ATMOSPHERIC
DEPOSITION OF NONPOINT POLLUTANTS
ORIEL LOUCKS
Holcomb Research Institute
Butler University
Indianapolis, Indiana
ABSTRACT
Lake managers now routinely use estimates of chemical
loadings from adjacent land surfaces and from ground-
water in evaluating prospective responses to a treatment.
However, large nonpoint sources of pollutants through
atmospheric transport and deposition also need to be
considered. The capacity for watersheds and lakes to
assimilate the acidic component of these inputs varies
widely on a site-by-site basis. Results are presented from
studies of atmospheric chemical inputs and associated
lake chemistries in northern Minnesota and Wisconsin.
Precipitation sulfate input, lake depth, lake-water renewal
time, and bedrock characteristics influence the variability
in lake sulfate and alkalinity.
INTRODUCTION
Evaluating the potential response of lakes to management
now routinely considers nonpoint chemical loadings from
adjacent land and groundwater. However, nonpoint source
loadings also result from atmospheric transport and air-
shed washout through scavenging by rain, fog, and snow.
This paper examines the factors influencing the capacity
of watersheds and lakes to assimilate these nonpoint in-
puts and the circumstances under which lake managers
must take them into consideration.
To begin, we should consider data on the magnitude of
essentially nonpoint source pollutants transported over
the eastern United States in the lower atmosphere. Nu-
merous studies (summarized by OTA, 1984) address one
consequence of these transported pollutants: the altera-
tion of precipitation chemistry over large areas. As with
surface water nonpoint pollutants, the largest sources can
be identified at some distances upwind. The 1980 nitrogen
oxides and sulfur dioxide emissions of three States illus-
trate the burden released to the atmosphere:
Missouri, 0.6 x 106 tons of nitrogen oxides and 1.3 x 106
tons sulfur dioxide,
Illinois, 1.0 x 10" tons nitrogen oxides and 1.5 x 10s tons
sulfur dioxide,
Indiana, 0.8 x 10" tons nitrogen oxides and 2.0 x 10" tons
sulfur dioxide (OTA 1984).
For Indiana, these sulfur dioxide emissions amount to
over 50 tons/square mile/year or 150 Ib/acre, statewide. A
substantial part of these emissions is deposited in the
States of origin. Relatively near the larger sources, the
annual wet plus dry deposition of oxidized sulfur and nitro-
gen compounds (oxides and acids) can exceed 500 Ib/
acre (McFee et al. 1984), scaling down to deposition of 10
to 30 Ib/acre in the northern portions of the Midwest.
In 1978-79, researchers at the University of Wisconsin
and at the Environmental Protection Agency in Duluth be-
gan a series of studies designed to evaluate these atmo-
spheric chemical inputs in the softwater regions of the
northern Midwest (Glass and Loucks, 1980). Plans for
sampling up to 1,000 lake sites in the poorly buffered (few
natural means to offset incoming acidity) regions of north-
eastern Minnesota, northern Wisconsin, and northern
Michigan were developed. By 1983, slightly more than
1,000 lakes and streams had been sampled. Up to 70
watershed properties also had been quantified in over 600
watersheds in the three States (Eilers et al. 1983; Rapp et
al. in press). The goal of this paper Is to review the results
obtained concerning concentrations of nonpoint atmo-
spheric chemical inputs across the region, the differences
in watershed characteristics that lead to pH and alkalinity
changes in lake and stream water quality, and the implica-
tions of acidic inputs and watershed sensitivity for lake
management programs.
BACKGROUND
Watersheds (defined as land and water systems where
inputs, internal processing, and outputs can be quantified,
manipulated, and studied) have been a framework for eco-
system research for nearly 20 years (Likens and Bor-
mann, 1972; Loucks, 1975; Schindler et al. 1976; Likens
et al. 1977; Loucks and Odum, 1978). Most of this re-
search involved intensive studies at a few specific sites to
increase understanding of the flows of materials and the
processes that link land and water ecosystems.
The understanding that various watershed processes
and the wide variability in the physical properties of water-
sheds produce a specific land/lake response system is
now being extended to a larger number (representative
population) of watersheds and lakes within biogeographic
regions (Rapp et al. in press). Studies of a population of
lakes were not feasible until the general principles and
first models of watershed functioning were developed in
the 1970's. Now, with data on variations in chemical in-
puts across these regions, and data on the watershed
characteristics and chemistry of a large population of
lakes, recommendations can be extended to lake man-
agement.
ATMOSPHERIC CHEMICAL INPUTS
ACROSS THE NORTHERN GREAT LAKES
Atmospheric deposition of chemicals has been monitored
at eight sites across the three-State area since 1979
(Glass and Loucks, in review). Table 1 summarizes the
results at the eight sites from December 1980 through
November 1982. Deposition of acidic substances in north-
ern Michigan (Douglas Lake) is less than at Hubbard
Brook, New Hampshire (Glass and Loucks, in review), but
much greater than in the interior of Norway, where moder-
ate effects on lake fisheries are reported (Overrein et al.
1980). The difference in measured acidity in precipitation
from the Fernberg, Minnesota, site is more than twice that
of the Douglas Lake, Michigan, site. The differences be-
tween southwestern Minnesota and south-central Michi-
gan are even larger, but few lakes in those regions are
sensitive to acidic inputs.
Comparative analysis of the modern and historical pre-
cipitation chemistry across the northern part of this three-
483
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
State area (Glass and Loucks, in review) shows that the
input loadings and the gradient (amounts of a substance
carried in solution) have changed considerably since the
first rain-chemistry measurements were made in the area
during the 1950's. The changes are primarily increases in
total nitrate and sulfate in the precipitation. Table 1 shows
large increases in sulfate and nitrate anions (SOi~ and
NO3~) from west to east across the region, with almost a
tripling in the deposition of acidity [H + ] across the three
most northerly sites, Fernberg, Trout Lake, and Douglas
Lake.
RESULTS FROM THE MONITORING OF
LAKE RESPONSE
The 6-year UM-D/ERL-D studies emphasized monitoring
of lake chemistry during the spring and fall turnover per-
iod. A moderately large number of lakes was measured as
a single sampling in each of several years, but a small
number of lakes (12 to 15 each in Minnesota and Wiscon-
sin) were measured during the spring and fall each year.
As of the end of 1983,5 years of data were available in two
of the States, and 2 years in Michigan.
The average pH, alkalinity, and conductivity at 12 Wis-
consin lakes over 3 years show a slight decline in the first
two measures (from 6.09 to 5.87 for pH and from 31.9 to
23.9 neq/l for alkalinity) and the expected rise in conduc-
tivity (from 18.5 to 19.5 /iSiemens/cm for conductivity).
Most of the sites show this pattern in Wisconsin, but few
do in Minnesota, where acidic inputs are lower and back-
ground alkalinities are higher. Research is underway now
to evaluate more fully the statistical significance of these
trends and their implications for lake management, taking
into consideration the possible influence of weather cycles
and the problem of averaging across different types of
lakes.
WATERSHED CHARACTERIZATIONS
ACROSS THE THREE STATES
To evaluate the role of watershed properties in mediating
and neutralizing atmospheric inputs, quantitative data
were obtained for 167 watersheds in Minnesota, 316 in
Wisconsin, and 53 in Michigan. The characterization of
Minnesota and Michigan watersheds was carried out pri-
marily by a group at University of Minnesota-Duluth (UM-
D) (Rapp et al. in press); the Wisconsin watershed data
were developed originally by a group in the Wisconsin
Department of Resources (Eilers et al. 1983) and ex-
panded during 1984 by the UM-D group; and a combined
University of Iowa and UM-D group has developed the
watershed characterization for the upper peninsula of
Michigan.
Various studies (Loucks and Glass, 1984; Loucks et al.
1984; Schnoor et al. 1984) have incorporated a limited set
of the factors and processes considered important in wa-
tershed studies: atmospheric inputs of water and chemi-
cals, evaporation, forest canopy alteration of chemical
concentrations in rainfall, watershed relief and runoff proc-
esses, surface water detention, terrestrial neutralization
processes, and various permeability and below-ground
hydrologic flow parameters. In addition to these factors,
important properties of chemical adsorption and acid neu-
tralization in watersheds are associated with bedrock min-
eralogy, soil chemistry, and flows among various hydro-
logic compartments. In a mathematical model for
predicting aqueous alkalinity in the receiving systems
(lakes) of the study region, Schnoor et al. (1984) proposed
using water volume and alkalinity in each compartment,
along with flow rates, cumulative base cation export, and
a rate constant for acid neutralization by soil or minerals.
When all of these factors are considered as processes
within a watershed, the nine major classes of measurable
properties shown in Table 2, all well known to lake man-
agers, can be recognized as having the potential to influ-
ence the water chemistry of the responding system. Over-
all, data on 3 to 15 measures in each of these 9 categories
were obtained for the watersheds studied in each of the 3
States.
ANALYSIS OF THE WATERSHED/LAKE
RELATIONSHIPS
For lake management, we would like to know which lakes
could show effects from acidic deposition in these north-
ern Great Lakes States and over what time span (ex-
pressed in chemical flushing times of from 0.5 to over 30
years). Ideally, one would like to have data on the re-
sponses of a large population of lakes subjected to acidic
deposition, monitored throughout 3 to 5 decades. Be-
cause such data are not available for a wide range of lake
types, variation in nature (the differences in watershed
acid neutralizing capacity) to quantify the responses of
watershed/lake systems to the nonpoint atmospheric in-
puts can benefit research.
The relationships between transported pollutant deposi-
tion, watershed properties, and lake chemistry have been
evaluated in two ways: first, through analysis of complete
correlation matrices for inputs, watershed factors, and wa-
Table 2.—Watershed factor groups and associated watershed
processes recognized as Important In mediating atmos-
pheric inputs of pollutants (after Rapp et al. In press).
Watershed factor group
1. Atmospheric deposition and
meteorology
2. Catchment size
3. Topographic relief
4. Hydrologic inputs
5. Lake depth and volume
6. Bedrock type
7. Surficial deposit depth and type
8. Land cover type
9. Roads and human development
Watershed processes
Input rates
Averaging responses
Hydrologic flow rates
Flushing rates
Resuspension
Weathering
Neutralization
Remineralization
Sedimentation
Table 1.—Annual Ion deposition (In peq/ha) based on the 2-year period from December 1980 through November 1982, using weekly
composite rain samples from eight NADP precipitation sites (after Glass and Loucks, in review).
Lamberton, MN
Marcell, MN
Fernberg, MN
Spooner, Wl
Trout Lake, Wl
Douglas Lake, Ml
Wellston, Ml
Kellogg, Ml
Precip.
cm.
62
78
70
68
83
72
88
83
Yearly Ion deposition /ieq/ha
H
23
72
85
99
165
264
368
383
S0«
210
220
218
301
313
367
494
529
NO3
136
128
115
158
165
195
279
248
Ca
130
98
74
101
87
100
124
116
Mg
44
35
32
31
28
38
49
45
NH4
216
146
125
220
173
173
291
205
484
-------
tershed output measures (color, in-lake SOl~ and HCO§);
and second, through multiple regression modeling of the
one or two variables in each of the nine classes that mech-
anistic models suggest would have the greatest influence
on resultant lake chemistries.
Using a cluster analysis technique on 275 lake-water-
shed systems, the Wisconsin study (Eilers et al. 1983)
identified lake hydrologic type (seepage, drained, drain-
age, spring-lake, etc.) as by far the dominant watershed
property controlling the prospective response of lakes to
acidic inputs in Wisconsin. The dominance of hydrologic
turnover time over other factors may have been influ-
enced by the comparative homogeneity of the northern
Wisconsin landscape in the study area; soil and bedrock
vary little compared to the high diversity of the Northeast-
ern States. Differences in precipitation inputs across the
population of lakes were not considered at that time, but
will be evaluated in the future.
The influence of chemicals in precipitation, and the
ameliorating influence of other watershed properties on
color, sulfate, and alkalinity were evaluated in Minnesota
using stepwise multiple regression by Rapp et al. (in
press). Here, the widespread occurrence of shallow bed-
rock was expected to yield quite different relationships
from those in Wisconsin. The results in Table 3 show the
watershed components accounting for the greatest varia-
bility in measured color are the surrounding vegetative
characteristics, water renewal time, and evaporative con-
centration. Sulfate concentration in the lake water, how-
EFFECTS ON WATER QUALITY IN THE AGRICULTURAL MIDWEST
ever, is determined primarily by the within-State differ-
ences in atmospheric sulfate deposition (derived from the
gradient in deposition across the lake cluster in Minne-
sota, interpolated from three sites in the surrounding area
and the central Fernberg site). Bedrock type (scale 6 has
some sulfide) and total water renewal time are also impor-
tant in determining sulfate in the headwater lakes. The
adjusted r2 is largest for the headwater lakes (at 0.44) and
is about half of that (0.24) when all watersheds of all sizes
and characteristics are averaged.
Alkalinity is the most difficult of the water chemistry
components to explain; the within-State gradient in atmo-
spheric wet sulfate input is again significant. Water re-
newal time and maximum depth of the lake also contribute
to the variability in alkalinity of these lakes. These results
do not provide causal explanation, but they do identify
variables likely to be important in managing a population
of lakes or streams within a sensitive region.
WATERSHED CHARACTERISTICS AND
THEIR IMPLICATIONS FOR WATER
QUALITY
As is usually the case, some uncertainty remains regard-
ing causality of the results presented, and continuing stud-
ies are planned. Alterations of lake and stream ecosys-
tems are known to become expressed through the
introduction of elevated quantities of chemicals through
atmospheric deposition, and/or through changes in
Table 3.—Factors Identified through stepwise multiple regression T-statistics1 as belonging to the best subset models for
headwater lakes, and all lakes (after Rapp et al. in press).
Headwater
All
Evaporative concentration
Annual runoff
Total renewal time
Maximum depth
%peat
% forest
% marsh
Adjusted r2
N
Evaporative concentration
Wet-deposition sulfate
Total watershed area
Total watershed relief
Total renewal time
Maximum depth
% scale-6 bedrock
% coniferous
% cultural development
Adjusted r2
N
Chlorophyll-a
Annual precipitation
Wet-deposition sulfate
Total watershed relief
Total renewal time
Maximum lake depth
Average bedrock scale
Average soil pH
%peat
% marsh
Adjusted r2
N
Best subsets for color
X
X
.48
52
Best subsets for sulfate
X
X
X
X
X
X
X
X
.43
244
X
X
.44
.24
46
Best subsets for alkalinity
X
X
226
.38
51
X
X
238
.10
1 Numerical values for T-statistics deleted because they are intended as regression coefficients; magnitudes do not correspond
with significance.
485
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
bioavailability and chemical forms induced through
changes in pH. The work of Sollins et al. (1980) and the
studies by the Electric Power Research Institute (see
Chen et al. 1983, and reports that follow) illustrate our
ability to follow acidic substances through the diverse
phase changes of a watershed or stream to the lake. Air-
borne acid-forming compounds deposited with precipita-
tion become part of the hydrocycle and can alter concen-
trations in the watershed, at least at critical times (Loucks
et al. 1984).
To evaluate the potential pH depression of a lake or
stream at a critical time—the period during and following
snowmelt—the total deposition, through both wet and dry
mechanisms in the snow pack, must be summarized. The
relationships to terms in the general atmospheric deposi-
tion models have been evaluated by Ragland and Wilken-
ing (1982), and Samson and Small (1984). The wet-only
measurements at the three National Atmospheric Deposi-
tion Program monitoring sites associated with the three
State clusters (Table 1) were used to compute the annual
wet loadings. A calculated total deposition accumulated
over winter in each watershed can be used as the net
input to the watershed, stream, and lake from snowmelt.
The complex stoichiometry of lakewater or streamwater
for each watershed must be examined carefully. The ionic
composition and changes over time should be studied
using "Stumm" ion balance diagrams (Schnoor and
Stumm, 1984) in conjunction with the relationships given
in the previous sections. In each region, a certain propor-
tion of lakes shows a negative response to the snowmelt
input, with the largest depressions in the Upper Peninsula
of Michigan where SOij~ deposition is largest (Glass and
Loucks, in review).
DISCUSSION AND CONCLUSIONS
Evidence from the studies reported here, and from many
others, is gradually documenting the lake-watershed prop-
erties that determine whether nonpoint pollutants depos-
ited by atmospheric processes are likely to result in a pH
or alkalinity depression in lakes or streams. When com-
bined with computations on hydrology and lakewater re-
newal times, information also is being developed to deter-
mine whether the responses to SO|~ inputs will involve a
serious lag time before effects are expressed. The magni-
tude of SO|~ deposition in the northern Great Lakes
States is emerging as the most important variable deter-
mining lake sulfate and alkalinity, and aspects of water-
shed hydrology, lakewater renewal times, and surrounding
vegetation types are important secondary considerations.
How can lake managers apply these results? Most
probably, in situ management for lake or stream acidifica-
tion effects will be extended soon from direct application
of lime in the lake (which does not control aluminum mobi-
lized in the watershed), to lime application on the adjacent
watershed so as to control the transport of toxic metals.
Over a longer period, we are likely to have risk assess-
ment methods for those systems where the expression of
acidification has a long latency caused by long water re-
newal times. Gradual reduction in loadings combined with
lime treatments may be needed for these systems.
ACKNOWLEDGEMENTS: The author wishes to acknowledge
the fundamental contributions to.this work by Dr. Gary Glass of
the EPA Environmental Research Laboratory, Duluth, Minne-
sota, and the many others at the University of Minnesota-Duluth
who have helped in various aspects of the study.
REFERENCES
Chen, C.W., S.A. Gherini, R.J.M. Hudson, and J.D. Dean. 1983.
The Integrated Lake-Watershed Acidification Study. Vol. 1:
Model Principles and Application Procedures. EPRI EA-3221.
Electric Power Res. Inst., Palo Alto, CA.
Eilers, J.M., G.E. Glass, K.E. Webster, and J.A. Rogalla. 1983.
Hydrologic control of lake susceptibility to acidification. Can.
J. Fish. Aquat. Sci. 40:1896-1904.
Glass, G.E., and O.L. Loucks. 1980. Impacts of Airborne Pollu-
tants on Wilderness Areas along the Minnesota-Ontario Bor-
der. EPA-600/3-80-044, Ecolog. Res. Ser. U.S. Environ. Prot.
Agency, Duluth, MN.
Likens, G.E., and F.H. Bormann. 1972. Nutrient cycling in eco-
systems. Pages 25-67 in J.H. Wiens, ed. Ecosystem Structure
and Function. Proc. Annu. Biology Colloquium, 31. Oregon
State Univ. Press, Corvallis.
Likens, G.E., F.H. Bormann, R.S. Pierce, J.S. Eaton, and N.M.
Johnson. 1977. Biogeochemistry of a Forested Ecosystem.
Springer-Verlag, New York.
Loucks, O.L. 1975. Models linking land-water interactions
around Lake Wingra, Wisconsin. Pages 53-63 in A.D. Hasler
and J. Olson, eds. Land-Water Interactions. Springer-Verlag,
New York.
Loucks, O.L., and W. Odum. 1978. Analysis of the five North
American lake ecosystems. I. A strategy for comparison. Verh.
Int. Verein. Limnol. 20:556-61.
Loucks, O.L. and G.E. Glass. 1984. U.S./Canada aquatic im-
pacts assessment: integration of experimental studies, moni-
toring and modeling of acidic deposition effects. Pages 205-17
in J.L. Schnoor, ed. Modeling of Total Acid Precipitation Im-
pacts. Ann Arbor Sci., Ann Arbor, Ml.
Loucks, O.L., R.W. Miller, and TV. Armentano. 1984. As assess-
ment of aquatic resources altered or at risk from acidic deposi-
tion. Northeastern Environ. Sci. 3:8-23.
McFee, W.W., et al. 1984. Effects on Soil Systems. Pages 2-1
through 2-71 in A.R Altschuler and R.A. Linthurst, eds. The
Acidic Deposition Phenomenon and Its Effects. II. U.S. Envi-
ron. Prot. Agency, Washington, D.C.
Office of Technology Assessment. 1984. Acid Rain and Trans-
ported Air Pollutants: Implications for Public Policy. OTA-0-
204. U.S. Congress, Off. Tech. Assess., Washington, D.C.
Overrein, L.N., H.M. Seip, and A. Tollan. 1980. Acid Precipita-
tion—Effects on Forest and Fish. Final Rep. SNSF proj.
1972-1980. Oslo, Norway.
Ragland, K.W., and K.E. Wilkening. 1982. Relationship between
mesoscale acid precipitation and meteorological factors.
Pages 123-243 in P.M. D'ltri, ed. Acid Precipitation: Effects on
Ecological Systems. Ann Arbor Sci., Ann Arbor, Ml.
Rapp, G. et al. In press. Relationships between water quality
and watershed components for 267 lakes in the Minnesota/
Ontario border area. Submitted to Environ. Geol.
Samson, P.J., and M.J. Small. 1984. Atmospheric trajectory
models for diagnosing the sources of acid precipitation. Pages
1-24 in J.L. Schnoor, ed. Modeling of Total Acid Precipitation
Impacts. Acid Precipitation Ser., Vol. 9, Ann Arbor Sci. Book.
Butterworth Publ., Boston.
Schindler, D.W., R.W. Newbury, K.G. Beaty, and P. Campbell.
1976. Natural water and chemical budgets for a small Pre-
cambrian lake basin in central Canada. J. Fish. Res. Board
Can. 33:2526-43.
Schnoor, J.L., W.D. Palmer, Jr., and G.E. Glass. 1984. Modeling
impacts of acid precipitation for northeastern Minnesota.
Pages 155-173 in J.L. Schnoor, ed. Modeling of Total Acid
Precipitation Impacts. Ann Arbor Sci. Ann Arbor, Ml.
Schnoor, J.L., and W. Stumm. 1984. Acidification of aquatic and
terrestrial systems. In W. Stumm, ed. Chemical Processes in
Lakes. Wiley Interscience, New York.
Sollins, P. et al. 1980. The internal element cycles of an old-
growth Douglas-fir ecosystem in western Oregon. Ecol. Mono-
graphs 50:261-85.
486
-------
Cross Boundary
Nonpoint Source Pollution:
The Implications
GREAT LAKES POLLUTION FROM LAND USE ACTIVITIES
NORM BERG
Soil Conservation Society of America
Ankenny, Iowa
In 1972, the Pollution From Land Use Activities Reference
Group (PLUARG) of the International Joint Commission
(IJC) was established. The purpose was to determine the
levels and causes of any pollution of the Great Lakes Sys-
tem from land use activities. The group was also asked to
recommend appropriate remedial actions to the Govern-
ments of Canada and the United States.
PLUARG reported its findings and recommendations to
the IJC in July 1978. In 1980, the IJC forwarded their
considerations and conclusions to the parties in the Great
Lakes Agreement. Their 18 recommendations basically
adopted the comprehensive Environmental Management
Strategy for the Great Lakes System PLUARG gave IJC.
An Overview of Post-PLUARG Developments by the Non-
point Source Control Task Force of the Water Quality
Board of IJC, was released in 1983. That report stated
that:
PLUARG was a major international cooperative effort un-
dertaken from 1972 to 1978, charged with conducting an
intensive investigation into the pollution of the Great
Lakes System from land use activities. The resulting stud-
ies provided the most exhaustive review conducted up to
that time, and thus remain the most definitive data base
and reference source for any aspect of nonpoint source
pollution in the Great Lakes. The PLUARG final report
contained a comprehensive set of recommendations
which, if implemented, would considerably curtail non-
point sources of pollution. However, despite the magni-
tude of the published scientific output and the submission
of a management-oriented report in 1978, the United
States and Canadian Governments have not yet re-
sponded formally to the PLUARG recommendations. .
In 1982, a report to the Secretary of State by the U.S.
General Accounting Office stated: "The U.S. Government
has not adequately supported or been sufficiently involved
in the water quality activities of the U.S./Canadian Interna-
tional Joint Commission. As a result, the Commission has
had difficulty fulfilling its role as the principal advisor for
water quality matters on the Great Lakes and other bound-
ary waters." GAO recommended actions to improve U.S.
support for the Commission's water quality activities.
In 1984, the IJC Water Quality Board sponsored a Non-
point Source Pollution Workshop, in Windsor, Ontario, to
assess the current status of nonpoint source pollution
problems and control in the Great Lakes basin and to
obtain comments on and provide an opportunity for tech-
nical peer review of the Overview report. Several involved
in PLUARG, including the Co-chairman, Dr. Murray B.
Johnson, and I participated.
The conclusions of the Overview Task Force were dis-
cussed along with programs and practices currently being
carried out by Canada and the United States. The Parties
to the 1978 Great Lakes Water Quality Agreement had
begun to develop management plans in response to the
requirements of the Phosphorus Load Reduction Supple-
ment to Annex 3 of the Great Lakes Water Quality Agree-
ment. Knowledge abut phosphorus transport and
bioavailability, priority area identification, and pesticide us-
age were reviewed, and their impact on nonpoint source
control programs was assessed. In addition, the workshop
was a good international forum for exchange of informa-
tion and experience regarding nonpoint source pollution
control programs and practices: what has and has not
worked.
Dr. Johnson and I were reluctant to reinvolve ourselves
in that exercise. Murray gave his view of the demanding
task of PLUARG, as follows:
As a research limnologist for the Canadian Government, I
was challenged by the terms of reference. The wording
implied authority. The Canada-U.S. Agreement on Great
Lakes Water Quality signed at Ottawa, April 15,1972, by
the President of the United States and by the Prime Min-
ister of Canada said, "I have the honor to inform you that
the Governments—pursuant to Article IX of the Boundary
487
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Waters Treaty of 1909—have agreed to request the IJC to
conduct a study of pollution of the boundary waters of the
Great Lakes System from agricultural, forestry and other
land use activities, in light of provision of Article IV of the
Treaty which provides that the boundary waters and wa-
ters flowing across the boundary shall not be polluted on
either side to the injury of health and property on the other
side." The chance to have my hands on the wheel, have
control, steer a course, be a manager, were prerogatives
of a Chairman, I thought. However, Norm and I learned
much later, to our dismay, that the wheel in our hands
could be turned and turned—BUT IT WAS NOT HOOKED
TO ANYTHING THAT WOULD CHANGE DIRECTION.
We did hear again at Windsor that cost-effective man-
agement practices and implementation programs are
available and have been demonstrated in the basin. Suf-
ficient technical knowledge exists to support implementa-
tion of programs to reduce nonpoint sources of pollution to
the Great Lakes. As a result of extensive review of pro-
grams, practices, and issues surrounding the manage-
ment of nonpoint sources of water pollution, the Task
Force concluded that the basic recommendations devel-
oped by PLUARG—for IJC—remain valid.
This series of concurrent sessions was to address Mak-
ing Decisions About NFS Pollution. International aspects
of the problem complicate the process. The most recent
manifestation involves acid rain, where the decision opts
for more study. Cross-boundary groups are prone to take
this route and rate high in the conduct of that approach.
They were experts at it, and we treasure the experiences
of that IJC assignment. Now, what does that research
reveal that may be of value for the future? If water quality
problems are site-specific, how do local efforts become
part of cohesive, coordinated action?
To fully analyze and apply the benefits of that total expe-
rience to present day problems would require time and
talent beyond my capability. Conditions have changed re-
garding NFS since 1972. The Water Quality Act in the
United States was in the early stages of implementation.
The 208 planning process was being implemented. The
eight Great Lakes States were represented on a basin
commission primarily for data gathering and planning.
Lake Erie was dying, if not already dead, and other Lakes
would suffer a similar fate, unless the conditions causing
pollution were quickly reversed. The rural areas in both
Nations were reluctant to admit that land use activities,
especially agriculture, could cause pollution many miles
from their operations. An early effort of PLUARG was to
determine the state-of-the-art. A detailed study plan em-
phasized four main tasks:
1. Task A: to assess problems, management of pro-
grams, and research, and to attempt to set priorities in
relation to the best information now available on the ef-
fects of land use activities on water quality in the boundary
waters
2. Task B: inventory of land use and land use practices,
with emphasis on certain trends and projections to 1980
and, if possible, to 2020
3. Task C: intensive studies of several small water-
sheds, selected and conducted to permit some extrapola-
tion of data to the entire Great Lakes Basin, and to relate
contamination to water quality at river mouths of the Sys-
tem to specific land uses and practices
4. Task D: diagnosis of degree of impairment of water
quality in the Great Lakes, including assessment of con-
centrations of contaminants of concern in sediment, fish,
and other aquatic resources
The objectives of Task A were to analyze the pollution
problems and the potential of various land uses and to
document the practicality of alternative remedial, or con-
trol measures. To meet these objectives, studies would
assess the state-of-the-art for the following categories:
A1: residential areas
A2: commercial and industrial areas
A3: transportation
A4: extractive areas
A5: pesticides and herbicides
A6: nutrients
A7: erosion and sedimentation
A8: animal wastes
A9: intensive animal feedlots
A10: forestry
A11: recreation land
A12: undeveloped land
A13: liquid waste disposal
A14: solid waste disposal
A15: land fills, including dredging activities
A16: deepwell disposal
A17: management and control of land use/water quality
problems/institutional and legal arrangements
We published in November 1974 in two volumes our
findings of Task A entitled, Management Programs, Ef-
fects of Research and Present Land Use Activities on Wa-
ter Quality of the Great Lakes. This helped PLUARG an-
swer the first question of the Reference: Are boundary
waters of the System being polluted by land drainage (in-
cluding ground and surface runoff and sediments) from
agriculture, forestry, urban and industrial land develop-
ment, recreational and park lands, utility and transporta-
tion systems and natural resources? Table 1 lists param-
eters for which a water quality problem has been identified
in the Great Lakes and those for which no problem has
been identified, but may be a problem in inland or ground
waters.
Although additional work was underway through the
other tasks the affirmative answer to the first query was no
surprise. PLUARG found that the Great Lakes are being
polluted from land drainage sources by phosphorus, sedi-
ments, some industrial compounds, some previously used
pesticides, and, potentially, some heavy metals. Phos-
phorus loads in 1976 exceeded the recommended target
loads for all lakes. We stated that point source control
programs would be sufficient at that time to meet the tar-
get loads only in Lake Superior and Lake Michigan, and
southern Lake Michigan may need further measures.
Toxic substances, such as PCB's, had gained access to
the System from diffuse sources, especially through at-
mospheric deposition.
While in many cases ascribing pollution to any particu-
lar land use was difficult, of note was that the cumulative
effect of a variety of land use activities ultimately contrib-
utes to the pollution of the Great Lakes System. That led
to the second question, "If the answer is in the affirmative,
to what extent, by what causes, and in what localities is
the pollution taking place?"
PLUARG found that the Lakes most affected by phos-
phorus and toxic substances were Erie and Ontario. Many
local problems, including intensive agricultural operations,
were identified as the major diffuse source contributors of
phosphorus. Erosion from land used for crop production
on fine-textured soils and from urbanizing areas were
found to be the main sources of sediment. The most im-
portant land-related factors affecting the magnitude of pol-
lution were found to be soil type, land use intensity, and
materials usage. Northwestern Ohio and southwestern
Ontario were sources of high phosphorus loadings.
The third question was and remains the most difficult to
answer. We were asked, "If the Group should find that
pollution of the character just referred to is taking place,
what remedial measures would, in its judgment, be most
practicable and what would be the probable cost
488
-------
CROSS BOUNDARY NONPOINT SOURCE POLLUTION: THE IMPLICATIONS
Table 1.—Great Lakes water quality pollutants.
Parameters for which a Great Lakes water quality problem has been identified
PROBLEM
SOURCES
POLLUTANT
Phosphorus1
Sediment"
Bacteria of public health concern
PCBs'
Pesticides (past)1
Industrial organics1
Mercury1
Lead1
Lake-
wide
Yes
No
No
Yes
Yes8
Yes
Yes
Potential'
Nearshore
or
Localized
Yes
Yes
Yes
Yes
Yes"
Yes
Yes
Potential'
Land
Runoff
Yes
Yesc
Minor"
Yes
Yes
Yes
Minor
Yes
DIFFUSE
Atmos-
phere
Yes
Negligible
No
Yes
Yes
Yes
Yes
Yes
In-Lake
Sediments
Yes"
Under some
conditions
No
Yes
Yes
Yes
Yes
Yes
POINT
Yes
Negligible
Yes
Yes
No
Yes
Yes
Yes
REMARKS
'Percentage unknown; not considered
significant over annual cycle
bMay contribute to problems other than
water quality (e.g., harbor dredging)
"Including streambank erosion
land runoff is a potential, but minor
source; combined sewer overflows
generally more significant
"Some residual problems exist from past
practices
'Possible methylation to toxic form
Parameters for which no Great Lakes water quality problem has been identified, but which may be a problem In Inland surface waters or ground waters
Nitrogen
Chloride
Pesticides (present)1
Other heavy metals
Asbestos1
Vimenck
Acid precipitation
No
No
No
Potential1
No
No
No"
No"
No
Potential1
Yes
Nom
Yes Yes
Yes Negligible
Yes No
Yes Yes
No ?
. . Kin Data AuailaMa
No Yes
Minor
No
No
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Vac
No
sSome inland groundwater problems
hSome local problems exist in nearshore
areas due to point source
'New pesticides have been found in the
environment; continued monitoring is
required
'See Upper Lakes Reference Group Report
k
mPotential problem for smaller, soft water,
inland lakes
'Sediment per se causes local problems; phosphorus and other sediment-associated contaminants have lakewide dispersion.
thereof?" PLUARG found that the remedies for nonpoint
source pollution would be neither simple nor inexpensive.
NFS are characterized by their wide variety and large
numbers of sources, the seemingly insignificant nature of
their individual contributions, the intermittent nature of
their inputs, the damaging effect of their cumulative im-
pact, the complex set of natural processes acting to mod-
ify them, and the variety of social and economic interac-
tions which affect them.
PLUARG did not favor across the board measures for
nonpoint source pollution control. We recommended a
methodology defining problem areas on a priority basis
and then applying the most practicable control means for
any particular source. We recommended management
plans with four major components:
1. information, education, and technical assistance
2. planning
3. fiscal arrangements
4. regulation
A basic tool for estimating the level and location of man-
agement required in potential pollutant contributing areas
was to be the identification of the most serious hydrologi-
cally active areas (HAA). This was an early concept of
targeting and for equity and public and private costs of
best management practices (BMP's).
The fact is well documented that monitoring for
changes in sediment or nutrient loadings from the imple-
mentation of agricultural control practices is extremely dif-
ficult. Recent evaluations suggest that loading changes of
about 20 percent can be detected, provided good flow and
concentration data are collected and that event sampling
is included. Expert opinion as to the length of time needed
to make definitive judgments range from 5 to 15 years.
To protect the quality of the Great Lakes System the
United States and Canada updated the Water Quality
Agreement of 1972 in 1978. In October of 1983 the United
States and Canada agreed to a supplement to Annex 3 of
the Agreement regarding control of excessive loading of
phosphorus to the Great Lakes System. The mechanism
for preparing the U.S. Load Reduction Plan consists of an
interlake State/Federal Great Lakes Phosphorus Task
Force (GLPTF) and separate State Task Forces (STF). The
STF's are preparing plan elements that will become part
of each State's Water Quality Management Plan. The
GLPTF will integrate the elements into a total U.S. Load
Reduction Plan.
489
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
A better definition of pollution in the Great Lakes is still
required. PLUARG found that traditional yardsticks, such
as water quality objectives or standards insufficiently eval-
uated the impact of nonpoint sources to the System. Fu-
ture studies would be of greater value if they were holistic
in nature. The pages that follow are from the Overview of
Post-PLUARG Developments. Although some progress
has been made on a few of our recommendations since
this report, it will serve the reader as excellent background
to determine whether the Great Lakes ecosystem can, in a
timely manner, maintain its desirable characteristics of
• Diversity,
• Resilience, and
• Stability,
when it is changed, as it has been, by man.
APPENDIX
Nonpoint Source Pollution Abatement in the
Great Lakes Basin: An Overview of Post-
PLUARG Developments ' ,
A Report Submitted by the Nonpoint Source
Control Task Force of the Water Quality Board of
the International Joint Commission
August 1983
Windsor, Ontario
Response to PLUARG Recommendations
Over two years have passed since the PLUARG recom-
mendations were officially transmitted to the Govern-
ments by the International Joint Commission (IJC). The
Parties have so far made no official response to the Inter-
national Joint Commission concerning their positions on
these recommendations. This situation exists despite the
broad based support for the PLUARG recommendations
evident through its own intensive public consultation proc-
ess and further confirmed through the Commission's own
Post-PLUARG hearings.
Likewise, the two Governments have failed to complete
negotiations on Annex 3 of the 1978 Great Lakes Water
Quality Agreement. Confirmation of the target loads for
the lakes and allocation of further phosphorus loading re-
ductions are viewed by this Task Force as being funda-
mental to the resolution of the current impasse on the
PLUARG recommendations.
After a thorough review of the programs and practices
of the Parties, it is the Task Force's position that with the
exception of surveillance, there has been no direct re-
sponse by the Governments. This lack of a direct re-
sponse, while impeding overall program co-ordination and
implementation, has fortunately not prevented govern-
ment agencies and non-governmental groups from under-
taking a number of individual activities. These programs
and activities along with the original PLUARG recommen-
dation which they most closely support are briefly dis-
cussed in this chapter.
RECOMMENDATIONS
1. Development of Management Plans
PLUARG recommends Management Plans, stressing site-
specific approaches, to reduce loadings of phosphorus,
sediments and toxic substances derived from agricultural
and urban areas, be prepared by the appropriate jurisdic-
tions within one year after the International Joint Commis-
sion's recommendations are transmitted to the Govern-
ments. PLUARG further recommends that a mutually
satisfactory schedule for the reduction of nonpoint source
loadings be annexed to the revised Great Lakes Water
Quality Agreement.
Management plans should include:
i) A timetable indicating program priorities for the
implementation of the recommendations;
ii) Agencies responsible for the implementation of
programs designed to satisfy the recommenda-
tions;
Hi) Formal arrangements that have been made to in-
sure inter- and infra-governmental co-operation;
iv) The programs through which the recommenda-
tions will be implemented by federal, state and
provincial levels of government;
v) Sources of funding;
vi) Estimated reduction in loading to be achieved;
vii) Estimated costs of these reductions; and
viii) Provision for public review.
No action to develop comprehensive plans has been
undertaken. In Canada, a number of comprehensive wa-
tershed management studies have been undertaken
which address some of the criteria raised by PLUARG. In
the United States, water quality management plans have
been completed for various states and sub-state areas,
but they are not specifically oriented to reducing loadings
to the Great Lakes except for the Lake Erie Wastewater
Management Study.
2. Planning
PLUARG recommends that Governments make better use
of existing planning mechanisms in implementing nonpoint
source control programs by:
i) Insuring that developments affecting land are
planned to minimize the inputs of pollutants to the
Great Lakes; and
ii) Insuring that planners are aware of and consider
PLUARG findings in the development and review
of land use plans.
In Canada, the Planning Act, the Environmental As-
sessment Act, the Environmental Protection Act (EPA) and
the Federal Environmental Assessment and Review Proc-
ess (EARP) provide a means for addressing nonpoint pol-
lutants during the planning stages of major land develop-
ments. Both the EPA and the EARP, due to their more
restricted application, are not seen as having the potential
to make a major impact on nonpoint source loadings. The
Planning Act, while more all-encompassing, is not actively
used to address such problems. An urban drainage policy
statement is being considered under the Planning Act.
A number of urban municipalities have developed
guidelines and criteria for limiting pollutant loadings dur-
ing construction of new developments. However, the Prov-
ince of Ontario has no uniform policies.
In the United States, regional and statewide water qual-
ity management plans have been developed to address
both point and nonpoint sources of pollution, agricultural
sources in particular. However, they are quite uneven in
the extent they deal with nonpoint sources and none spe-
cifically address loadings to the Great Lakes. The Lake
Erie Wastewater Management Study specifically ad-
dressed lake loadings and stands as the most comprehen-
sive study of agricultural sources in the Great Lakes Ba-
sin.
At the request of the Environmental Protection Agency,
the six Great Lakes States have developed statewide non-
point source control strategies.
3. Fiscal Arrangements
PLUARG recommends that a reveiw of fiscal arrangements
be undertaken to determine whether present arrangements
are adequate to insure effective and rapid implementation
490
-------
CROSS BOUNDARY NONPOINT SOURCE POLLUTION: THE IMPLICATIONS
of programs to control nonpoint pollution. Such a review
should include:
i) Determination of the availability of grants, loans,
tax incentives, cost-sharing arrangements and
other fiscal measures;
ii) Determination of whether or not the terms of fi-
nancial assistance programs are conditional upon
the implementation of nonpoint source remedial
measures; and
Hi) Determination of the extent to which various fi-
nancial assistance programs are conditional upon
the implementation of nonpoint source remedial
measures.
There is no evidence to suggest that there has been an
overview. of Canada's fiscal arrangements concerning
nonpoint pollution control programs. Two provincial inter-
ministerial groups, the Urban Drainage and the Soil Ero-
sion and Sedimentation committees have recently re-
viewed provincial funding of programs and are expected
to make recommendations to the Ontario government in
the near future. Members of the same committees pro-
vided input and consultation to developing the Soil Con-
servation and Environmental Protection Assistance Pro-
gram.
In the United States, no comprehensive review of fiscal
arrangements has occurred; however, several studies
have addressed fiscal problems of individual programs. In
general, United States conservation and environmental
programs are receiving less money. Nonpoint sources
have received a very small share of water quality manage-
ment funds. Soil conservation funding for water quality
purposes has received low priority within the U.S. Depart-
ment of Agriculture, although the shift from structural
measures to tillage practices is providing improved bene-
fits to water quality.
Most states provide substantial annual appropriations to
support local soil and water conservation districts and co-
operative extension programs.
4. Information, Education and Technical Assistance
PLUARG recommends that greater emphasis be given to
the development and implementation of information, edu-
cation and technical assistance programs to meet the
goals of the Great Lakes Water Quality Agreement. This
emphasis should include:
i) Development of broad programs, through school
systems, the media and other public information
sources, describing the origins and impacts of
pollutants on the Great Lakes and alternative
strategies that should be followed by the public
and government agencies to prevent water qual-
ity degradation;
ii) Initiation of more specific programs to improve
the awareness of implementors and those work-
ing in and for government, emphasizing the need
for the further control and abatement of nonpoint
pollution; and
Hi) Strengthening and expanding existing technical
assistance and extension programs dealing with
the protection of water quality, including rural and
urban land management practices.
In Canada, one conservation authority has undertaken
a successful program of information, education and tech-
nical assistance (Upper Thames). A few other authorities
have made some attempts in this area, including pro-
grams aimed at the primary and secondary school level,
providing exhibits at fall fairs and other public events, etc.
The level of effort varies widely among authorities but is
generally a small percentage of their total budgets.
Many county level soil and crop improvement associa-
tions have increased their education efforts on soil conser-
vation matters. The Ontario Ministry of Agriculture and
Food (OMAF) has increased staff available for erosion-
related extension and education purposes. Two films on
soil erosion have been produced and are in great demand
for showing at local meetings.
In the United States, soil conservation is strongly sup-
ported by the field staff of the Soil Conservation Service
(SCS) which provides technical assistance; the field staff
of the Cooperative Extension Service (CES) which pro-
vides education and information; the research segments
of both SCS and CES; cost-sharing funds from the Agri-
cultural and Stabilization and Conservation Service
(ASCS) and other forms of support from various other
USDA organizational units. Very little of this support is
directed specifically toward water quality, however, it has
water quality benefits associated with it.
In addition to soil conservation per se, several major
demonstration programs in the United States and water
quality management planning have greatly increased
knowledge and awareness of nonpoint source pollution.
Special projects have greatly increased the availability of
technical assistance in several regional areas. Several
States and counties have prepared comprehensive con-
servation tillage guides and the State of Ohio holds five to
10 regional conservation tillage workshops each year.
The International Joint Commission, through its Great
Lakes Regional Office, has been disseminating an infor-
mation piece on citizen action for reducing pollution from
land use activities as well as a display about land use
pollution since 1978, and is in the final stages of develop-
ing a slide-tape program from loan distribution to groups.
5. Regulation
PLUARG recommends:
i) That the adequacy of existing and proposed legis-
lation be assessed to insure there is a suitable
legal basis for the enforcement of nonpoint pollu-
tion remedial measures in the event that voluntary
approaches are ineffective; and
ii) That greater emphasis be placed on the preven-
tion aspects of laws and regulations directed to-
ward control of nonpoint pollution.
In Canada, some new regulations are in place to reduce
nonpoint sources of pollution. A few municipalities have
by-laws and guidelines for sediment runoff from construc-
tion sites; and under the Ontario Environmental Assess-
ment Act certain types of development require environ-
mental impact statements. Most conservation authorities
control and inspect development in floodplains and re-
strict filling. The Ontario Waste Management Corporation
(OWMC) is formulating guidelines for industrial waste
management.
OMAF and OWMC are the only agencies with programs
that encompass all of southern Ontario. Each municipality
develops its own runoff control criteria, however, not all
have mapped floodlines and hazard lands and few have
done this for entire watershed. Moreover, many agencies
and types of development are exempt from the Environ-
mental Assessment Act.
Experience has indicated that farmers are more recep-
tive towards the adoption of a nonpoint source manage-
ment program once they are made aware of the advan-
tages to their own operations and the free technical
assistance available.
In urban areas there has been little attempt to promote
policies of controlling pollution at source before it enters
urban runoff.
In the United States, many municipalities have enacted
sediment control and runoff regulations as part of their
subdivision review authority. Statewide sediment control
laws have been passed in several of the Great Lakes
491
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
States but they appear to be having little effect. In the
1983-84 revisions to the Federal Clean Water Act it is
expected that an amendment or amendments regarding
abatement of nonpoint sources of pollution will be devel-
oped.
6. Regional Priorities
i) The water quality conditions within each lake;
ii) The potential contributing areas (PCA) identified
by PLUARG; and
Hi) The most hydrologically active areas (HAA) found
within these potential contributing areas.
Coincidentally, in Canada, most of the work in manag-
ing nonpoint sources has occurred in the Lake Erie Basin.
This is largely because of interest in local water quality
concerns or agricultural production problems, and not an
expressed concern for Great Lakes water quality.
Several agencies have identified priority areas. OMAF
has ranked counties according to the cost of erosion to
agriculture, but has not prioritized its funding accordingly.
Though the Lands Directorate of Environment Canada
has mapped areas prone to erosion and likely to deliver
sediments to waterbodies in southwestern Ontario, no evi-
dence shows that federal priorities or programs have been
influenced.
The Thames River Implementation Committee (TRIG)
study used the mapped priority areas as a basis for guid-
ing implementation of remedial programs. The Grand
River Implementation Study (GRIC) study utilized
PLUARG data in its computer simulations of potential non-
point loadings and embarked on a federally assisted pro-
gram to identify priority management areas within the wa-
tershed.
With the exception of TRIG, GRIC and Environment
Canada, few agencies or studies have utilized the concept
of potential contributing areas. The objective of most
agencies is to meet MOE water quality criteria in streams
under their jurisdictions. Few are concerned with potential
impacts upon the Great Lakes.
In the United States, the demonstration projects of the
Environmental Protection Agency's Great Lakes Demon-
stration Grant Programs have addressed nonpoint source
problems in each of the Great Lakes. EPA has focussed
much of its demonstration grant resource in Lake Erie
Basin where a series of projects and the Corps of Engi-
neers' Lake Erie Wastewater Management Study have
focussed resources on identifying and implementing ef-
fective low cost measures for the control of phosphorus
from nonpoint sources. Focussing the projects in the high
phosphorus clay soils of the western basin was clearly in
response to a water quality priority. However, within the
selected watersheds the emphasis has been on obtaining
successful demonstrations rather than seeking out the
fields with the highest unit loads. The assumption is that
the entire western basin is a hydrologically active area and
that once successfully demonstrated, low cost measures
will be adopted throughout the area.
At the state level, Wisconsin has a well-developed prior-
ity system for selecting its nonpoint source grant projects.
Other Great Lakes States have identified their priority
problem areas as part of their State nonpoint source strat-
egies.
7. Control of Phosphorus
PLUARG recommends that phosphorus loads to the Great
Lakes be reduced by implementation of point and nonpoint
programs necessary to achieve the individual lake target
loads specified by PLUARG.
It is further recommended that additional reductions of
phosphorus to portions of each of the five Great Lakes be
implemented to reduce local nearshore water quality prob-
lems and to prevent future degradation.
While the Governments have moved to meet the phos-
phorus eflluent requirement at sewage treatment plants of
1 mg/L, the target loadings have not been met due to
deficiencies in the nonpoint program. Target loadings set
forth in the 1978 Agreement by the two governments re-
main unconfirmed.
The Toronto Area Watershed Management Study and
the Rondeau Bay Study have both been developed in part
in response to degradation of an important nearshore wa-
ter resource. The extent of support to implement recom-
mendations of these studies is unknown.
In the United States, point source control has made
excellent progress. Nonpoint source controls have also
progressed, particularly in the Lake Erie Basin. Also, the
Water Quality Board and International Joint Commission
are focussing attention on phosphorus control problems in
three Areas of Concern: (geographic area where specific
water quality objectives under the Agreement are violated)
Green Bay, Saginaw Bay and the Maumee River/Western
Lake Erie area.
8. Control of Sediment
PLUARG recommends that erosion and sediment control
programs be improved and expanded to reduce the move-
ment of fine-grained sediment from land surfaces to the
Great Lakes system.
Reductions in soil erosion from cropland and stream-
bank have received the most attention. OMAF's financial
assistance program is designed to reduce erosion on
farmland thereby maximizing net production returns. The
program still lacks a major resource commitment to plan-
ning, technical assistance/demonstration and evaluation
to ensure widespread adoption and implementation in pri-
ority areas over the long-term.
Conservation Services Programs have increased the
amount of effort devoted to erosion control and sedimen-
tation, but most remedial work focusses on the erosion of
streambanks, a relatively minor source of sediments to the
Great Lakes System. Only UTRCA and ERCA have pro-
grams to reduce sedimentation from field erosion. The
UTRCA is also the only conservation authority that con-
ducts most of its remedial measures in priority problem
areas.
There is no evidence to show that a significant reduc-
tion of sediment loadings to the Great Lakes Basin has
been accomplished.
In the United States, the U.S. Department of Agricul-
ture's soil conservation programs continue to operate with
increasing emphasis on control through tillage practices.
The Great Lakes Demonstration Grant Program of EPA
and the Lake Erie Wastewater Management Study of the
Corps of Engineers both stress sediment control as a
means of controlling phosphorus loads to the lakes. Some
sediment control regulations have been adopted by State
and local governments as reported above.
9. Control of Toxic Substances
PLUARG recommends the following actions be taken to
reduce inputs of toxic substances to the Great Lakes:
i) Control of toxic substances at their sources;
ii) Closer co-operation of both countries in the imple-
mentation of toxic substances control legislation
and programs; and
Hi) Proper management and ultimate disposal of
toxic substances presently in use.
Organochlorines migrating from industrial waste sites
are still creating problems. Their regulation will eventually
come under the jurisdiction of the Ontario Waste Manage-
ment Corporation. The OWMC, in conjunction with the
Ministry of the Environment, is starting to embark on a
492
-------
CROSS BOUNDARY NONPOINT SOURCE POLLUTION: THE IMPLICATIONS
program to identify historic and existing waste disposal
sites. OWMC has identified areas suitable for hazardous
waste facilities and a site-specific search is in progress. A
study of the potential in Ontario for reduction, reuse and
recycling of hazardous and other industrial wastes has
been commissioned by OWMC and the Ministry of the
Environment has been active in promoting recycling.
Few joint efforts to assess cumulative and synergistic
effects of contaminants exist. This aspect of toxic sub-
stances is still poorly understood, but both the federal and
provincial governments are conducting research in this
field. Water quality objectives continue to be refined as
impacts upon water quality and aquatic biota are better
understood.
In the United States, many of the most persistent and
bioaccumulative pesticides have been banned from use
and biodegradable alternatives have replaced them. How-
ever, the overall quantity of pesticides in use has steadily
increased.
In the United States legislation enacted since PLUARG
has rapidly changed toxic substances regulation. The
Toxic Substance Control Act addresses the manufacture
and use of compounds, the Resource Conservation and
Recovery Act addresses the transport and disposal of
toxic substances and the so-called Superfund Program
addresses clean up of hazardous waste sites. The com-
bined effect is regulation of virtually every aspect of toxic
substances.
10. Control of Microorganisms
PLUARG recommends that epidemiological evidence be
evaluated to establish applicable microbiological criteria for
body contact recreational use of water receiving runoff
from urban and agricultural sources.
No changes in criteria have been established.
11. Agricultural Land Use
PLUARG recommends that agencies which assist farmers
adopt a general program to help farmers develop and im-
plement water quality plans.
This program should include:
i) A single plan developed for each farm, where
needed;
ii) Consideration of all potential nonpoint source
problems related to agricultural practices, includ-
ing erosion, fertilizer and pesticide use, livestock
operations and drainage; and
Hi) A plan commensurate with the farmers' ability to
sustain an economically viable operation.
None of the agencies mandated to assist farmers have
adopted a program which is directed towards developing
individual farm water quality management plans. Assist-
ance programs are generally offered on a first-come, first-
serve basis and are largely restricted to the provision of
fixed cost-share funds emphasizing the construction of
structural remedial measures.
In the United States, major change is underway in till-
age practices as described elsewhere in this report. The
greatest changes are occurring in the western Lake Erie
basin under the stimulus of changing technology, chang-
ing economic conditions, education and assistance pro-
grams. Some address soil conservation and some (EPA
and COE) address water quality, but are focussing on con-
servation tillage. Similar EPA and USDA projects and pro-
grams operating elsewhere are encouraging tillage prac-
tice changes.
12. Urban Land Use
PLUARG recommends the development of management
plans for controlling urban stormwater runoff. These plans
should include:
i) Proper design of urban stormwater systems in
developing areas such that the natural stream
flow characteristics are maintained; and
ii) Provision for sediment control in developing ar-
eas, and control of toxic substances from com-
mercial and industrial areas.
Because of the expense of up-grading existing systems,
stormwater management plans should deal primarily with
new development. Old development should be improved
only if it is creating severe problems in a localized area.
In Canada, urban sources of nonpoint pollution have
received very little attention. Most provincial and water-
shed agencies addressed problems associated with ex-
cessive stormwater runoff and have taken the position
that urban nonpoint sources of pollution are negligible
compared to agricultural sources. Agencies have tended
to identify phosphorus and sediments as the key prob-
lems, and have ignored compounds such as phenols,
PCBs, mercury and lead which originate almost exclu-
sively from urban areas.
With the assistance of provincial and watershed agen-
cies on urban nonpoint sources of pollution, several mu-
nicipalities have developed comprehensive stormwater
management policies, guidelines and plans. These plans
are designed to minimize flooding, sediment and related
pollutant loads from new developments. However, the lack
of design criteria, inadequate planning tools and limited
surveillance and enforcement, limit the effectiveness of
these initiatives.
In urban areas there has been little attempt to promote
policies of controlling pollution at its source before it en-
ters urban runoff.
The Toronto Area Watershed Management Study is
dealing with urban nonpoint sources of pollution on a
"sewershed" basis. Management plans and^guidelines
will be formulated for each basin and severe problems will
be addressed as they are found.
In the United States, urban land use is the jurisdiction of
local government. A number of municipalities have
passed sediment control ordinances and are conducting
land use planning to protect water quality. The Water Qual-
ity Management Program funded under Section 208 of the
Clean Water Act provided support for major water quality
planning efforts at regional and state levels during the late
1970's. Many of the resulting plans were linked to land
use. The best example of this is in southwestern Wiscon-
sin. There all extensions of sewer service into new areas
must be consistent with the regional land use/water qual-
ity plan on a site-specific basis. Unfortunately, such strong
programs are uncommon.
13. Wetlands and Farmlands
PLUARG recommends the preservation of wetlands, and
the retention for agricultural purposes of those farmlands
which have the least natural limitations for this use.
In Ontario, OMAF recognizes and promotes the value of
preserving prime agricultural land through the use of its
foodland guidelines. The fact that these are only guide-
lines has limited their overall effectiveness in reducing the
loss of prime agricultural land.
Over the past two years the Ontario Ministry of Natural
Resources has been developing a policy statement for
conserving important wetlands. In support of this policy
statement, Environment Canada and the Ontario Ministry
of Natural Resources have jointly developed a wetland
evaluation system to be used to determine the relative
value of wetlands when making land use planning deci-
sions. Environment Canada has also mapped the areas of
wetlands, dating from presettlement time until the present
to determine rate of loss of this important resource. Maps
will be provided to local jurisdictions. A number of wetland
acquisitions have been made but acquisition programs
493
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
are hampered due to lack of fund and long-term manage-
ment. A number of studies directed at improving our un-
derstanding of key wetlands have also been undertaken.
In the United States, the Dredge and Fill permit pro-
gram based upon Section 404 of the Clean Water Act
requires that a permit be obtained from the U.S. Corps of
Engineers before any wetland can be dredged or filled.
Two presidential executive orders are of note: order 11988
addressing flood plain management and order 11990 ad-
dressing the protection of wetlands. Also, the U.S. Depart-
ment of Agriculture has a formal policy: regulations 9500-3
concerning prime agricultural lands, wetlands and flood
plains.*
14. Local Problem Areas
PLUARG recommends that the InternationalJoint Commis-
sion, through the Great Lakes Regional Office, insure that
local levels of government are made aware of the availabil-
ity of PLUARG findings, especially as they relate to local
area problems, to assist them in developing and imple-
menting nonpoint source management programs.
PLUARG data were disseminated to conservation au-
thorities and are available at major libraries. It is question-
able if this information was effectively presented at the
county level and certainly not at the township level. The
IJC could not promote the PLUARG recommendations nor
could it assist directly the local agencies in identifying and
solving nonpoint source problems as such actions are the
responsibility of the signatories to the 1978 Great Lakes
Water Quality Agreement—Article Vl(1e).
15. Review of Implementation
PLUARG recommends:
i) The International Joint Commission insure regular
review of programs undertaken for the implemen-
tation of recommendations from this reference;
and
ii) That nonpoint source interests be represented
during these reviews.
The actions of this Task Force represent the first formal
review by the IJC of the activities of the Governments in
support of the PLUARG recommendations. However, IJC
through its Boards and Windsor Office actively partici-
pated in the Post-PLUARG reviews conducted by the
Great Lakes Basin Commission.
16. Surveillance
PLUARG recommends that tributary monitoring programs
be expanded to improve the accuracy of loading estimates
of sediment, phosphorus, lead and PCBs. Sampling pro-
grams:
i) Should be based on stream response characteris-
tics, with intensive sampling of runoff events,
where necessary; and
ii) Should be expanded to include toxic organic
compounds, toxic metals and other parameters
as may be defined in the future.
Further, the role of atmospheric inputs should be consid-
ered in the evaluation of Great Lakes pollution, with special
consideratiorfgiven to determination of the sources of ma-
jor atmospheric pollutants.
Efforts should be made to improve the co-ordination be-
tween data collection and data user groups, and agree-
ments established regarding data collection standards and
accessibility.
PLUARG further recommends that the adequacy of U.S.
Great Lakes nearshore and offshore water surveillance ef-
forts be examined.
In Ontario, the Saugeen River (L. Huron), Thames River
(L. St. Clair) and the Grand River (L. Erie) are sampled
intensively for a full range of toxic organics and metals.
Atmospheric pollutants are monitored in the Canadian
portion of the Basin at 16 sites for nutrients in major ions
and Cu, Pb, Zn, Fe and Cr. The sampling network has
been expanded since 1978 to include each of the Great
Lakes Basins. Both bulk and wet deposition are moni-
tored. The period of record remains too short to make
loading estimates for the individual lake basins with confi-
dence. Data sets are made available annually to the Inter-
national Joint Commission.
Tributary monitoring data are released in an annual re-
port. The most recently available data—for 1980—pertain-
ing to toxic substances could not be analyzed and inter-
preted for this report due to resource and time limitations.
In the absence of such analysis and interpretation, its sig-
nificance to the health of the Great Lakes ecosystem re-
mains unknown.
In the United States, the Geological Survey (USGS)
maintains an extensive system of stream gauging stations
which record flow levels and some limited water quality
data. Each state conducts water quality monitoring at key
tributary mouths. Traditionally, the states have gathered
monthly grab samples and submitted the data to the Great
Lakes Regional Office of the Commission where annual
loads have been calculated using the Beale ratio estima-
tor. During the past two years additional sampling of high
flow events on key tributaries has been supported by the
EPA Great Lakes National Program Office (GLNPO) in
order to verify the loading estimates. A program of fish
tissue and sediment sampling in the tributary mouth areas
is also being conducted by GLNPO using gas chromatog-
raphy/mass spectroscopy scans in order to locate toxic
contamination problems.
17. Role of the Public
PLUARG recommends that the International Joint Commis-
sion establish a comprehensive public participation pro-
gram at the outset of future references.
No new references have been made to the Commission
since this recommendation was made to the Governments
in 1980.
* 1982-83 Biennial Report of the Dredging Subcommittee of the Water Quality Board
also has a chapter on "Great Lakes Wetlands."
494
-------
IRRIGATION RETURN FLOWS AND SALINITY PROBLEMS
IN THE COLORADO RIVER BASIN
MOHAMED T. EL-ASHRY
World Resources Institute
Washington, D.C.
Soil and water salinity occur in arid regions wherever irri-
gation is practiced. In the United States, an estimated 20
percent of all irrigated land, about 4 million ha (10 million
acres), suffer from salt-caused yield reductions. Though
less serious than pollution from heavy metals or toxic or-
ganic compounds, salinity constitutes the most serious
water quality problem in the western United States.
In the United States, the Colorado River Basin, includ-
ing the Imperial and Coachella Valleys of southern Califor-
nia that receive Colorado River water, contains more ma-
jor salinity problem areas than any other river basin in the
western United States. It is closely followed by the Rio
Grande Basin and the Central Valley of California.
Land degradation and reduced agricultural productivity
are some of the likely downstream effects of high salt
content in irrigation water. For example, in the Mexicalli
Valley land degradation is more severe than in the adjoin-
ing Imperial Valley because: (1) the salinity of the irriga-
tion water is worse where Mexico diverts Colorado River
water, and (2) the Mexicalli Valley does not have the inten-
sive drainage network that the Imperial Valley has.
In all affected river basins, salinity has progressively
increased as the water resources have been developed
and put into use. The water in these rivers becomes in-
creasingly saline from the headwaters to the mouths,
mostly from seepage and return flows from irrigated land.
In the Colorado River, salinity concentrations increase
from less than 50 mg/L in the headwaters to about 900
mg/L at Imperial Dam to about 1,200 mg/L in Mexico. The
heavy salt load of 10 million tons is estimated to cost all
water users, in the United States alone, more than $133
million a year and is projected to more than double to $267
million annually by 2010 if controls are not instituted.
The general objective of irrigation is to provide a suit-
able moisture environment in the soil for plant growth. A
tendency among farmers in dry regions, however, is to
overirrigate. In a typical irrigation field near Grand Junc-
tion, Colorado, 38 percent of the water applied was unnec-
essarily wasted because irrigation was continued for 14
hours after the root zone was filled. The root zone was
filled after 23 hours had elapsed. By stopping then, the
irrigation efficiency could have been raised from 39 per-
cent to 63 percent. On the average, irrigation efficiencies
in the Colorado River Basin are less than 50 percent.
Overirrigation is due to: (1) pricing of water below its
scarcity value, which encourages inefficient water use;
(2) poor conditions of irrigation infrastructure, that lead to
losses through seepage before the water reaches the
crops; (3) irrigators' lack of knowledge of water require-
ments; and (4) inappropriate water policies or provisions
in water laws that serve to discourage conservation.
When irrigation water is applied to the soil surface, wa-
ter is lost in one of three directions: evapotranspiration,
surface runoff, and deep percolation. Deep percolation
losses add large quantities of salts to ground waters and
may add to drainage and downstream salinity problems.
Since all waters used for irrigation contain some dis-
solved salts, repeated application of water to soils will
result in the accumulation of salts (mainly the chlorides,
sulfates, and bicarbonates of calcium, magnesium, so-
dium, and potassium) in the soil profile. To maintain agri-
cultural productivity, these salts must be leached out of the
root zone.
Because of the concentrating effect of consumptive use
of water by crops, and leaching of soil salts, gradual dete-
rioration of water quality in natural aquifers and receiving
streams takes place over a period of time. However, rises
in salinity concentrations can often be dramatic. In 1961,
salinity concentrations in Colorado River water flowing
into Mexico virtually doubled, from 800 mg/L to 1,500 mg/
L. The jump was principally a result of pumping of saline
ground water (6,000 mg/L) from the Wellton-Mohawk Irri-
.gation District in Arizona, coupled with the closure of Glen
Canyon Dam and the reduction in the quantity of water
delivered to Mexico.
Negotiations with Mexico resulted in the signing of a
treaty, Minute 242, in 1974 and the passage by the U.S.
Congress of the Colorado River Salinity Control Act of
1974. The treaty guaranteed that salinity in the water de-
livered to Mexico would not exceed 115 mg/L (± 30 mg/L)
over the annual average at Imperial Dam. This was to be
achieved through the construction of a multimillion dollar
desalting plant near Yuma, Arizona. Similarly, the Colo-
rado River salinity control program for the American
States was to be achieved through the construction of
several Federally funded structural control measures,
mostly for natural point sources of salinity. Such a program
proved acceptable to the States because it interfered little
with irrigators' activities and because it would be financed
by the Federal government. Gradually, however, the em-
phasis has been shifting to the cost-effective on-farm con-
trol measures.
Many technologies and techniques exist at a variety of
costs for water quality management of irrigation return
flows. They involve increase in the efficiency of irrigation
water application (for example, trickle and sprinkler irriga-
tion), reduction in the amount of irrigation water lost from
conveyance systems through seepage (by canal and ditch
lining), disposal of saline return flows, or desalinization of
saline drainage water.
The U.S. Soil Conservation Service has found these on-
farm management practices, including land leveling, crop
management, irrigation scheduling, and proper water
management, to be the most cost-effective measures for
controlling salinity. Yet, only limited enthusiasm exists for
changing on-farm water use practices because of water
management policies and laws, water pricing, and espe-
cially the perception that the actor is not generally the
beneficiary.
Water laws in most States in the West discourage water
conservation, and in many areas irrigators have little in-
centive to conserve because they cannot apply the saved
water to new land. In addition, most water resources in the
United States have been developed and managed by the
public sector and are heavily subsidized. Prices for water
are often based on distribution system operating costs
495
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
rather than on scarcity value or true costs. Water pricing
should be considered as a means to effect efficient water
use. Rate structures can be adjusted to make water quan-
tities exceeding efficient irrigation needs more costly.
Such rate structures can use existing or other appropriate
base rates for efficient irrigation, and excess funds can be
invested in on-farm management measures on a project
or a district scale.
Finally, on-farm salinity control measures for water qual-
ity improvement will not be undertaken by the private
farmer at his own expense in the absence of economic
incentives and disincentives by the Federal and State gov-
ernments. As in the case of many soil and water conserva-
tion issues, the question of who benefits from and who
pays for improved conservation practices is also a valid
one for salinity control.
496
-------
AGRICULTURAL NONPOINT SOURCE POLLUTION IN THE MIDWEST
ROBERT D. WALKER
Cooperative Extension Service
College of Agriculture
University of Illinois
Urbana, Illinois
The area often referred to as the North Central States and
Northern Great Plains includes Ohio, Indiana, Illinois, Mis-
souri, Iowa, Minnesota, Wisconsin, and Michigan, the
North Central States, and Kansas, Nebraska, South Da-
kota, and North Dakota, the Northern Great Plains. These
areas generally have the highest percent of land devoted
to agricultural crops in the United States. Over 60 percent
of the total land is used to grow agricultural crops. The
major crops are corn, soybeans, sorghum grains, and
wheat; corn and soybeans are grown in the North Central
States and wheat and sorghum grains in the Northern
Great Plains States.
Concentrations of orchards on the east side of Lake
Michigan and commercial vegetable production are found
throughout the area with heavy concentrations in Wiscon-
sin, southern Minnesota, and northern Illinois for canning.
Along with concentrated feed grain production, animals
are farmed. Dairy cattle are found throughout the area
with very heavy concentrations in Wisconsin and Minne-
sota. Cattle are fed grain and sold for slaughter through-
out the region, with heavy concentration in western Iowa
and eastern Nebraska. Hogs and poultry are also raised
throughout the area.
Changes in agriculture have been occurring that cause
higher levels of nonpoint pollution. The acres of row crops
have nearly doubled over the past 40 years. This shift
occurred because low-cost nitrogen fertilizer became
available after World War II. It was no longer necessary to
grow legume crops to provide nitrogen for corn. The result
was a shift from grasses and legumes to soybeans on a
large portion of the Corn Belt land.
A change in soil erosion occurred as a result of the shift.
On a typical central Illinois soil with 4 percent slope, we
had annual soil erosion losses of approximately 5-6 tons
per acre with a typical corn-corn-oat-hay rotation in 1940.
In the 1980's, we have annual soil erosion losses of ap-
proximately 15 tons per acre with a typical corn-soybean
rotation using conventional tillage on the same field. Crop
yield increases (3 to 4 times) during this same time period
have helped to reduce soil erosion, but not enough to
make up for the increased erosion resulting from the shift
to more row crops.
The soil erosion problem changes as we go from west to
east in the area. Wind erosion is the dominant problem in
the Western Great Plains, changing to water erosion as
we move eastward into the North Central States. Water
erosion is affected mainly by rainfall amounts, but also by
soil types. Sandy and high organic matter soils are af-
fected more by wind erosion than other soils. Wind ero-
sion reduces air quality, but the blowing soils can also
affect water quality. Both types of erosion degrade envi-
ronmental quality.
Sediment is a major water pollutant, especially through
the central Corn Belt of Indiana, Illinois, and Iowa. Sedi-
ment delivered to streams and lakes is also affected by the
soil type and rainfall. The deep loess soils found along the
Missouri, Mississippi, and Illinois Rivers are highly ero-
sive, resulting in high sedimentation rates in lakes and
rivers. Rainfall intensities also increase as we go from
north to south and west to east. This is reflected in the R
values used in the Universal Soil Loss Equation (USLE)
with R-50 being used in the western edge of North Dakota
to R-250 used in southern Missouri.
Until recently, soil erosion was generally thought to be
associated with reduced crop yields. With the addition of
fertilizers and improved agricultural technology, yields in-
creased. In fact, we have had a three- to fourfold increase
in most of our feed grain crop yields from 1930 to the
present. Since yields were increasing, many landowners
reasoned that soil erosion could not be a problem. It
wasn't untjl the Section 208 studies that we began to get a
clearer picture of the real impact of soil erosion as a water
pollutant.
Sediment is a pollutant when suspended and carried by
water. In addition, sediment reduces water storage capa-
city in lakes and streams as it settles to the bottom. Sedi-
ment may also carry plant nutrients, pesticides, and or-
ganic matter. The amount of plant nutrients reaching
streams can be greatly reduced simply by reducing soil
erosion. Phosphorous is held tightly by the soil clay parti-
cles, and a large portion of the nitrogen is in the soil or-
ganic matter. A soil erosion control program reduces the
loss of both of these nutrients.
A major soil erosion control practice that is now pro-
moted to reduce soil erosion is conservation tillage. Con-
servation tillage systems include those reduced tillage
systems that leave 20- to 30-percent crop residue soil
cover after planting. The systems range from changing
from a moldboard plow to a chisel plow, till-plant or ridge
planting, strip tillage, and no-till planting. The shift to con-
servation tillage has been increasing in the Corn Belt.
Conservation tillage is now used on 34 percent of the
cropland in the North Central and northern Great Plains
States.
The major environmental problem with these systems
has been the increased dependence on pesticides, partic-
ularly herbicides to control weeds as tillage is reduced.
While the systems are effective in reducing soil erosion by
50 to perhaps 80 percent, we do not know the long-term
consequences of using the pesticides. Since the major
increase in pesticides use has been with herbicides, we
have assumed that the tradeoffs have been in favor of
using conservation tillage. Conservation tillage permits us
to continue to grow most crops competitively, while sub-
stantially reducing soil erosion.
Many studies are attempting to find ways of reducing
the amount of pesticides used on our row crops. Other
studies are examining the long-term impacts pesticides
may have on the environment.
I believe that we are beginning to make progress in
reducing soil erosion in the Corn Belt. Iowa provided lead-
ership in starting a State soil erosion control program, but
Minnesota, Wisconsin, Illinois, Missouri, Ohio, and other
States have recently developed programs and are provid-
ing State cost-share money to get soil conservation prac-
tices applied.
The large amount of publicity given to soil erosion has
convinced most landowners that soil erosion is a problem.
The State program has gone far enough to set specific soil
erosion goals as a target for farmers. Generally, this long-
term goal is to reduce soil erosion to the established soil
loss tolerance.
497
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Changes have also occurred with livestock farms affect-
ing nonpoint pollution from livestock waste. We now have
fewer, but larger, livestock farms. Livestock waste washing
into streams adds substantially to the nutrient load of
streams from these large units. In addition, the organic
matter decomposition process uses up oxygen in water,
killing fish and other aquatic life, as well as causing odors
and decreasing esthetic values.
I see only one logical answer to handling livestock
waste: to store the waste until it can be properly applied on
cropland at rates usable by crops. This practice will be-
come more economical as the price of fertilizers continues
to increase. Federal and State livestock waste handling
programs are bringing the livestock waste problem under
control.
Nitrate levels in ground water and streams have in-
creased in some areas as a result of agricultural operation
over the past few years. I am not sure how serious the
problem is or how to reduce the nitrate levels and maintain
a viable agriculture, but in several areas we have large
acreages of corn nitrogen fertilized. Nitrate levels that ex-
ceed public health standards at times during the year are
found in several of the streams used for public water sup-
ply. Also, I understand that nitrate levels in ground water
under irrigation in sections of Nebraska and other areas
have increased above the public health standards.
There is much concern regarding nonpoint sources of
pollution in the northern Lake States and Missouri be-
cause of the recreational uses of lakes and streams. Rec-
reation brings in many dollars in northern Minnesota,
Michigan, Wisconsin, and southern Missouri. These
States are working on the problems.
In summary, the major nonpoint sources of water pollu-
tion from agriculture that I see in the Midwest are: sedi-
ment, livestock waste, plant nutrients, and pesticides.
These materials are impacting water quality in specific
areas. However, the impacts are different for different ar-
eas requiring different programs.
498
-------
Contributed Papers
THE EFFECTS OF CARBONATE GEOLOGY ON URBAN RUNOFF:
WATER QUALITY ASPECTS
JACK D. MILLIGAN
ROGER P. BETSON
Tennessee Valley Authority
Chattanooga, Tennessee
ABSTRACT
A study of urban runoff in four subbasins within a Knox-
ville, Tenn., watershed was conducted as part of EPA's
Nationwide Urban Runoff Program. The four subbasins
represent three urban land-use types, residential, strip
commercial, and central business district. Flow propor-
tional runoff samples, rainfall, and dryfall samples were
collected for water quality analysis. A water quality mass
balance was used to differentiate the effect of different
urban land-use types and carbonate geology on urban
runoff water quality. Rainfall/runoff load ratios illustrate
the effect of carbonate geology on different water quality
constituents within the subbasins and their potential im-
pact on ground water. Correlation coefficients between
input/output ratios of conservative constituents and ante-
cedent dry days were calculated; these further illustrate
the relationship of carbonate geology to urban runoff. The
seasonal nature of the magnitude of urban runoff losses
to the ground water system was evidenced by the reten-
tion of potential runoff mass within the subbasins during
dry weather storms. It was found that urban runoff in
areas underlain by carboante geology may have a signifi-
cant impact on ground water quality.
BACKGROUND
The information presented in this paper is part of a larger
study of urban runoff conducted in Knoxville, TN, as a
component of EPA's Nationwide Urban Runoff Program
(NURP). Only three of the 28 NURP studies addressed
urban runoff impacts on ground water. The authors be-
lieve this is the first study to examine urban runoff in an
area underlain by carbonate geology, where direct access
and/or extremely shallow overburden allows the chemi-
cally unhindered movement of surface runoff to the
ground water system.
The two other NURP studies that addressed ground
water were performed in Long Island, NY, and Fresno, CA.
The thrust of these studies was to evaluate the effective-
ness of recharge devices as a means of managing urban
runoff and their effect on ground water quality. At both
locations, ground water surfaces were at least 20 meters
below the base of the recharge device, allowing most ur-
ban runoff pollutants to be retained by the unconsolidated
material above the water table. The unconsolidated over-
burden in the Knoxville study area is shallow to nonexist-
ent, with numerous outcroppings of the carbonate bed-
rock.
SITE DESCRIPTION, WATER QUALITY:
SAMPLING AND MEASUREMENT
Figure 1 shows the location of the four subbasins within
Knoxville's Second Creek drainage basin that were used
for detailed study of rainfall and runoff. The four subbasins
represent three types of urban land use: two residential
areas, designated R1 and R2 with areas of 0.28 km2 and
0.36 km2, respectively; a strip commercial (SC) area of
0.39 km2; and a central business district (CBD) area of
0.10 km2. Additional information on each of the study ar-
eas is presented in the companion paper by Betson and
Milligan, this vol.
At the runoff exit point from each of the four subbasins,
storm runoff flows were measured and flow proportional
composite water samples collected for water quality analy-
499
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Figure 1.—Subbasins In the Second Creek watershed repre-
senting three urban land use types—residential (R1 and R2),
strip commercial (SC), and central business district (CBD)—
Instrumented for detailed study.
sis. Volumes of wet and dry precipitation were measured
and samples collected at a central location within each
subbasin. Samples were analyzed for chemical constitu-
ents in accordance with methods described in Standard
Methods (1980). A thorough description of the sampling
and measurement methodology is presented in Milligan et
al. (1984).
Betson and Milligan (this vol.) showed that the soluble
carbonate geology within the four water quality study ar-
eas caused a loss of potential storm runoff significantly
above what would be expected in a noncarbonate area.
This loss of runoff as a result of infiltration into the soil/rock
system was found to be variable among the four catch-
ments and depended upon the percentage of impervious
area within each basin. The CBD, for example, had the
lowest runoff losses and contained the highest percentage
of impervious area. The SC, R1, and R2 catchments had
progressively higher amounts of runoff losses and con-
tained respectively lower percentages of impervious area.
This can be shown hierarchically where:
For runoff losses: CBD < SC < R1 < R2
For percent impervious area: CBD > SC > R1 >R2
These findings suggest that concurrent with hydrologi-
cal effects pollutant loading in Knoxville's urban runoff
may also be affected by the carbonate geology of the
area. To evaluate this aspect of urban runoff, a mass bal-
ance was calculated for each runoff event!! The total atmo-
spheric mass input of constituents to each study water-
shed preceding a runoff event was determined from the
dryfall and wetfall data collected during the intervals be-
tween successive runoff events. This total atmospheric
input mass (dryfall plus wetfall) was divided by the total
mass (load) of a constituent measured in the runoff associ-
ated with the preceding atmospheric input .interval.
The resulting storm- input-output pollutant load ratio
permits an assessment of the influence of carbonate geol-
ogy on urban runoff loads and the potential for impacts on
ground water quality. For example, ratios greater than.
unity indicate that atmospheric input exceeds runoff out-
put, suggesting the retention of mass within the drainage
basin or loss through the carbonate geology. Ratios less
than unity indicate that runoff output exceeds atmospheric
input, suggesting a contribution of mass by the drainage
basin to the runoff load, such as from road deposits left by
vehicles and lawn and garden fertilizers.
In Table 1 the mean of the ratios of the atmospheric
input loads to the storm runoff loads is given for all storms
sampled in the CBD, SC, R1, and R2 subbasins. The most
significant aspect of the data in this table is that among
the four subbasins the input-output ratios progressively
increase so that in general: CBD < SC < R1 < R2. This
hierarchy holds for most constituents and suggests a di-
rect causative relationship with the previously established
runoff losses to the ground water and an inverse relation-
ship to the percentage impervious area in the subbasins.
For example, runoff losses are lowest for the CBD catch-
ment because of the high percentage of impervious area,
while runoff losses in the R2 subbasin are highest be-
cause of the lower percentage of impervious area. Conse-
quently, mean input-output load ratios are lowest in the
CBD and highest in the R2 subbasin.
The retention of mass within a basin during a storm
event may be a result of infiltration and subsequent en-
trapment or adsorption in the solids, or it may be the result
of loss to the ground water through surface-connected
solution channels. In the CBD subbasin the high percent-
age of impervious area prevents storm runoff from reach-
ing the soil or subsurface solution channels. Conse-
quently, the input-output ratios are much lower than those
for the R1 and R2 subbasins where access is less re-
strained by large impervious areas.
Whether significant portions of urban runoff pollutants
are lost to the ground water or are merely retained within
the soils of the Second Creek basin is difficult to deter-
mine without directly measuring ground water quality.
However, indirect measurements in this study have pro-
vided an abundance of circumstantial evidence to con-
clude that significant quantities of pollutants in Knoxville's
urban runoff are entering the ground water. This evidence
includes:
1. Mass balance data from Betson, 1976, showed an
accumulation of atmospheric pollutants (input-output load
Table 1.—Mean ratios of atmospheric Input loads (wet and
dry) to storm runoff loads for storms sampled in the
Second Creek basin.
Constituent
CBD
SC
R2
TDS
TSS
TKN
TP
NO2 + NO3
TOC
COD
K
Ca
Mg
Na
Cl
SO4
Fe
Mn
Cu
Zn
Cr
Se
Hg
Ni
Cd
Pb
Al
0.51
0.12
2.0
0.33
2.2
1.2
1.0
0.16
0.09
0.10
0.95
1.0
1.8
0.15
0.31
4.1
0.77
0.21
4.1
10.8
3.8
3.7
0.46
0.27
1.1
0.27
1.7
0.17
19.4
2.0
1.4
0.24
0.06
0.35
0.86
0.79
1.5
0.18
0.23
7.9
0.51
3.4
4.1
59.2
4.1
2.8
0.17
0.34
2.3
0.53
4.3
0.43
12.9
6.8
5.7
0.48
3.2
0.50
4.6
2.7
9.0
0.33
0.25
9.1
2.2
13.7
17.3
28.0
17.3
36.3
0.82
0.34
5.3
1.7
9.8
2.6
23.9
13.6
8.9
1.5
0.42
0.76
9.7
19.9
31.4
1.0
1.5
100.4
4.6
11.8
27.1
85.0
27.1
60.8
1.3
1.3
500
-------
CONTRIBUTED PAPERS
ratios > 1.0) within a Knoxville area watershed (Plantation
Hills) geologically similar to the Second Creek basin.
2. Mass balance data acquired during this study indi-
cated a large portion of urban runoff load was retained
within study watersheds that are underlain by soluble car-
bonate rock and lightly developed.
3. Hydrological modeling (Betson and Milligan, this
vol.) indicated that significant losses of urban runoff to the
ground water occurred above that which would be re-
tained merely as soil moisture.
In addition, further analysis of data from the R2 subba-
sin, where the largest runoff losses would be expected,
showed high positive correlations between the input-out-
put ratios of conservative constituents, such as potas-
sium, sodium, chloride, and sulfate, and antecedent dry
days. This would imply that during the season of high
intensity, short duration rainfall, and longer antecedent dry
periods, the ground water table would be low. Recharge
by urban surface runoff would readily occur, carrying with
it the relatively nonreactive elements. This hypothesis op-
erates following the premise that storms of high intensity
usually are of short duration and low intensity storms are
of long duration. Bearing out this hypothesis, a negative
correlation did exist between intensity and duration (r =
- 0.64) for storms during the study period.
The seasonal variation of urban runoff losses to the
ground water is further evidenced by the change in the
input-output ratios of individual storms during the study
period. For example, historically during January, February,
and March, the greatest amount of precipitation occurs in
the study area, resulting in saturated soil conditions and a
high water table. Under these circumstances the amount
of runoff leaving a watershed, such as the R1 or R2 sub-
basins, represents a higher percentage of the input to the
basin. Consequently, rainfall-runoff load ratios are low.
Conversely, during September, October, and November,
the driest months of the year, soils are usually far below
saturation. The water table then is depressed, resulting in
a smaller percentage of the runoff leaving the basin
through surface runoff and a greater accumulation of pol-
lutants within the basin.
Under these latter circumstances the rainfall-runoff
load ratios are at an annual high. The seasonal fluctuation
in load ratios at the R1 and R2 study sites is shown in
Table 2. The table presents the average rainfall-runoff
load ratios for the 3 months of the year with the greatest
and least precipitation. The data indicate that nearly all
constituents have larger load ratios during the dry months
Table 2.—Average rainfall runoff ratios during wet months
(January-March) and dry months (September-November)
in the residential subbasins, R1 and R2.
R1
R2
Constituent
Wet Dry Wet Dry
Months Months Months Months
TDS
TKN
TP
NO2 + NO3
TOC
COD
Ca
Mg
Fe
Mn
Cu
Zn
Cr
Hg
Cd
Pb
Al
0.5
3.0
0.3
6.7
0.8
0.4
0.6
0.1
0.02
0.1
15.9
0.5
0.3
—
0.7
0.1
0.02
1.9
3.4
0.2
2.4
1.8
1.3
0.3
0.4
0.7
0.3
4.5
3.2
3.9
—
6.5
0.8
0.3
2.0
5.1
0.5
25.5
5.1
1.4
0.07
0.1
0.5
0.4
47.3.
2.1
1.4
8.5
3.0
0.6
1.5
4.1
18.0
10:2
15.3
5.6
6.9
0.1
0.7
1.6
1.7
9.6
4.7
8.8
12.7
20.3
1.2
1.8
and smaller ratios during the wet months. This additional
information provides further evidence that pollutants in
urban runoff from the R1 and R2 subbasins are entering
the ground water in significant quantities.
CONCLUSIONS
The hydrological data collected during this study and the
model simulations performed using the data have shown
with some certainty that the soluble carbonate rock in the
Second Creek basin is a dominating factor relative to the
urban hydrology of the area. Regarding the areas in this
study, a significant quantity of the potential runoff does not
leave the study basin via surface runoff. It drains through
soluble carbonate rock cavities, probably emerging in
springs within the same watershed or perhaps transferring
into an adjacent drainage basin.
The study has shown how the soluble carbonate rock
dominates the urban hydrology in the Second Creek ba-
sin. The chemical constituents associated with the urban
hydrology then could be similarly affected. However, this
may not be the case because physical/chemical proper-
ties of the soil overburden can act as a chemical sink for
urban runoff pollutants, thus reducing their presence in
either the surface runoff or the ground water. Conversely,
equilibrium conditions in the shallow soil overburden
could eventually manifest, facilitating the passage of pollu-
tants to the ground water, while a percentage of all constit-
uents could enter the ground water directly through sur-
face-connected solution channels.
Although it would be necessary to directly measure
ground water by sampling wells to determine the precise
impact of urban runoff pollutants on ground water quality
in karst terrain, several conclusions can be made from this
study regarding the indirect relationship between urban
runoff and ground water:
1. The data indicate that in drainage areas containing
high percentages of impervious area (such as the CBD
and SC subbasins) the urban runoff will generally have a
higher pollutant load in terms of mass per unit area than
basins with a low percentage of impervious area (such as
residential drainages).
2. The ratio of atmospheric mass input to surface mass
output (runoff) is larger for study basins with small percent
impervious areas (such as residential areas) in compari-
son to basins with higher percent impervious areas (such
as high development areas).
3. A positive relationship was found between the input-
output ratios of less reactive pollutants and antecedent
dry days, indicating the transport of these constituents to
the ground water system.
4. The loss of pollutants in urban runoff to the ground
water system is seasonal, with the greatest quantities lost
during dry months, the least during wet months.
In general, indirect evidence indicates that both hydrol-
ogy and pollutants in urban runoff may be profoundly influ-
enced in lightly developed drainages underlain by soluble
carbonate rock.
REFERENCES
Betson, R. 1976. Urban hydrology, a systems study in Knoxville,
Tennessee. Div. Water Manage., Tenn. Valley Author, Knox-
ville, TN.
Betson, R.P., and J.D. Milligan. This vol. The effects of carbon-
ate geology on urban runoff, part I—hydrologic aspects. In
Proc. Perspectives on Nonpoint Source Pollution, a natl. conf.
Kansas City, MO, May 19-22,1985.
Milligan, J.D., E.I. Wallace and R.P. Betson. 1984. The relation-
ship of urban runoff to land use and ground water resources.
TVA/ONRED/AWQ-84/1. Tenn. Valley Author.
Standard Methods for the Examination of Water and Wastewa-
ter. 1981.15th ed. Am. Pub. Health Assn., Washington, DC.
501
-------
USING IN-STREAM MONITORING STATIONS TO EVALUATE
POLLUTION FROM URBAN RUNOFF
LELAND L. HARMS
Department of Civil Engineering
South Dakota School of Mines and Technology
Rapid City, South Dakota
BACKGROUND
A 3-year study of the urban runoff which entered Rapid
Creek was begun in 1980. The Rapid City study was part
of EPA's National Urban Runoff Program (NURP) in which
28 communities across the United States participated.
Rapid Creek, a high-quality mountain stream, enters
Rapid City from the west after flowing through primarily an
undeveloped, forested area. The creek discharge is con-
trolled by a manmade reservoir located approximately 15
miles from Rapid City. No known point sources of pollution
are upstream from Rapid City The creek changes into a
slower moving, prairie stream as it exits onto the plain east
of the city.
Rapid City is located at the foothills to the Black Hills in
western South Dakota and is the county seat for Pen-
nington County. The largest community in western South
Dakota, Rapid City is the business center for much of the
surrounding area. Between 1940 and 1960, Rapid City
experienced a dramatic growth rate resulting in a 1960
population of 42,399, up from 13*844 in 1940. A more
stable, much slower growth occurred in the next two dec-
ades, giving a 1980 census figure of 46,492.
FIELD METHODS
The locations of the six sampling stations are indicated on
Figure 1. Ideally, a sampling station should have been
placed just upstream of all urban development, and just
downstream from the city. Station 1 essentially satisfies
the upstream station, but some urban influence will mani-
fest itself because of an increase in density of dwellings
and upstream highway traffic. It was not hydraulically pos-
sible to establish a sampling station downstream from the
study area before other nonpoint sources of runoff would
enter Rapid Creek. Consequently, a theoretical station,
No. 56, was developed which mathematically establishes
downstream values based on mixing the loads present at
station 5 (the last station on Rapid Creek) and station 6
(the last major urban drainage into Rapid Creek).
Drainage areas ranged in size from 13,650 ha (33,730
acres) at station 1 to 650 ha (1,610 acres) at station 5.
Land usage varied from 96 percent nonurban at station 1
to 19 percent nonurban at station 5. Complete land use
information can be obtained from a document by Harms
etal. (1983).
The majority of the water samples were collected using
automated equipment, but some samples were collected
manually, primarily at the upstream stations. Periodic
base-line samples were taken at normal flow conditions by
manual methods. Even with the automated equipment,
the field crews tried to be on site during the runoff events.
Manually collected water samples were obtained by
wading streams and collecting depth-integrated aliquots,
using between 10 and 15 verticals across the channel. In
the first year of the study, baseline and precipitation runoff
samples were collected directly into new 3.8-L (1-gallon)
plastic milk containers. During 1981 and 1982, samples
were collected in DH-77 depth-integrating samplers. After
collection, the samples were transferred to new plastic
milk containers. All sampling equipment and containers
were rinsed three times with native water prior to taking
the first aliquot. The sample was immediately cooled to
4°C. Date, time, and stage were recorded immediately
before and after sample collection.
The automated sampling equipment, referred to as an
urban hydrology monitoring system by the U.S. Geologi-
cal Survey (USGS), incorporates a microprocessor-based
system control unit to receive an on-site record and to
control the automated water sampling device. The system
would switch on to storm mode when a select stage, cor-
responding to a definite discharge, was reached in-stream
during a storm event. The system would make continuous
recording data (time, date, stage, accumulated rainfall,
and sequential sample number) at a predetermined time
interval ranging from 30 seconds to 1 hour. If a rapid rise
in stage occurred, the timing sequence was overridden
and additional samples were collected. Samples were
pumped into new 3.8-L (1-gallon) plastic containers which
were stored in a refrigerator unit at 4°C, and transported
as soon as possible to the South Dakota School of Mines
Figure 1.—Map of Rapid Creek and major drainage basins
within Rapid City proper; sampling site locations.
Figure 2.—Automated sampling station.
502
-------
CONTRIBUTED PAPERS
and Technology (SDSM&T) environmental engineering
laboratory Under most conditions, samples were deliv-
ered to the lab within 3 hours of collection. A sketch of this
equipment is shown in Figure 2.
LABORATORY METHODS
All sample preservation and preparation were done in the
laboratory. Flow-weighted composites were normally
made for each station using the appropriate aliquot from
each discrete sample. Volumes for each aliquot were de-
termined by calculating the volume under the hydrograph
that each sample represented, baseflow included. A com-
puter program was developed to compute the correct ali-
quots, Harms and Smith (1983).
Composites were produced using a USGS cone splitter
(U.S. Geol. Surv. 1980). The splitter splits any sample into
10 equal aliquots to the nearest 1 percent. The cone split-
ter was evaluated for accuracy prior to use. After some
practice the error averaged about 3 percent, the sum of
the cone splitter and analytical error. After the composite
was properly constituted, the cone splitter was used to
split the composite sample into individual containers for
storage and preservation.
A quality assurance (QA) plan was developed and ap-
proved by EPA. Standards (both known and unknown),
duplicates, blanks, and spikes were routinely analyzed.
Values not satisfying the QA requirements were not en-
tered into the data base.
RESULTS
Interpretation of the results of a field study are only as
good as the samples collected during the investigation.
The error introduced into the data from the sampling pro-
cedure is often overlooked or not evaluated. Data for
events 27 and 30 were sacrificed in an attempt to deter-
mine whether the automatic samplers were collecting rep-
resentative samples. Manual samples were collected at
the same time that the automatic sampler was going
through its sampling sequence by integrating with depth
at various vertical sections along the horizontal cross sec-
tion of the sampling station. All manual samples were col-
lected using standard USGS equipment and procedures.
Both suspended solids (SS), and volatile suspended
solids (VSS) tests were conducted in duplicate on all indi-
vidual samples for both events.
A comparison of the automated versus manually depth-
integrated samples was made using paired t statistics in
both the conventional, normally distributed mode, and
also in the log-normal mode. Results of this comparison
are shown in Table 1. The skewness coefficient was used
as an indication that the data were better evaluated in the
log-normal mode. The data were positively skewed which
indicates that the data points are clustered to the left with
the higher readings being the extreme values to the right
of the distribution.
A critical examination of the sampling approach demon-
strated the importance of the sample size. For example, if
the paired t-test was performed on the suspended solids
data of Site 4 by eliminating one sample in turn, the result-
ing seven tests would not show a significant difference five
times (an acceptance level of 71 percent).
Perhaps a more practical approach would be to evalu-
ate the data based on the difference and the percentage
of difference between the two methods of sample collec-
tion. This scheme is presented in Table 2. Station 3 does
seem to show the effects of the heavier bedload as most
of the depth-integrated samples had higher suspended
solids. Even then, the average difference in concentration
is slightly less than 10 percent.
A summary of the differences in paired samples is given
in Table 3. These data were obtained at each station as
described above for station 3.
It is difficult to compare the automatic samples because
a firm and true value for each sample is not available.
Although unknown, some sort of sampling error is un-
doubtedly associated with the depth-integrated method
being used as the standard in this comparison. In addi-
tion, each laboratory test has its own errors and does not
yield a single true value.
In conclusion, the automatic samplers are apparently
collecting reasonably representative samples at all the lo-
cations. To collect enough samples to confirm this statisti-
cally was beyond the scope of the project. The samples
collected at station 4 would appear to be the most sus-
pect, but the average difference between automated and
depth-integrated samples was only approximately 7 per-
cent.
SNOWMELT RUNOFF
Three snowmelt runoff events were sampled during the
winter of 1981-82. In addition, some in-stream conductiv-
ity measurements were taken on some minor snowmelt
events. Figure 3 shows a typical conductivity pattern and
its relation to discharge. These data are from runoff
caused by the melting of about 41/2 inches of snow, the
cool and cloudy weather making it a slow developing
event. A substantial portion of the pollutants, as evi-
denced by the change in conductivity, were present in the
runoff from the first day although the increase in stream
flow is barely perceptible. It is also interesting that the
Table 1.—Paired student's t Test Results.
Site
No.
3
4
5
6
t, Normal
distr.
n
9
7
7
6
t.05
2.31
2.45
2.45
2.57
SS
-1.34
2.541
0.00
-0.75
VSS
0.00
4.252
1.22
1.00
t, Log-Normal
distr.
SS
-2.801
2.811
0.68
-0.27
VSS
0.01
3.361
1.32
1.39
1A significant difference exists between automated and depth-integrated samples.
2A highly significant difference exists between automated and depth-integrated sam-
ples.
Table 2.—Differences in paired samples for site 3.
SS, mg/L VSS, mg/L
Sample
27-3.1
27-3.2
27-3.3
27-3.4
27-3.5
30-3.1
30-3.2
30-3.3
30-3.4
A'
- 10
- 16
- 3
- 8
- 16
-T 74
-170
43
7
%A
-12.05
-13.91
- 2.68
- 7.34
-12.70
-15.70
-13.82
7.06
1.67
A
0
0
-3
2
0
-1
-4
4
2
%A
0
0
- 23.08
12.50
0
- 1.92
- 3.92
7.69
5.26
| % A | = 9.66%
| % A | = 6.04%
A = Automated sample value - depth-integrated sample value.
A = 100 A/automated sample value.
| % A | = Mean of absolute values of %A.
Table 3.—Summary of differences in paired samples.
Site No. SS, % VSS, %
3
4
5
6
9.66
6.95
6.51
4.43
6.04
4.88
5.57
6.74
503
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
RAPID CREEK AT EAST MAIN STREET
BASELINE CONDUCTIVITY @ 21 CFS
0000
oeoo 1200 ion
2/24/82
oeoo 1200 iaoo
2/25/82
oeoo 1200 iaoo
2/26/82
OeOO 1200 1800
2/27/82
Figure 3.—Conductivity and discharge at station 4 during snowmelt runoff.
maximum temperature for the first day was still below
freezing at -3°C.
The background values for all parameters, as measured
at station 1, were consistently low and nearly constant.
Chlorides, for example, remained less than 4 mg/L at sta-
tion 1 while exceeding 100 mg/L at stations 4 and 5. Data
for sodium show the same trend. Even though the City has
made efforts to limit the salt it places on the streets, the
effects are obvious.
Snowmelt runoff events can result in gross contamina-
tion of the receiving watercourse as evidenced by the SS
concentrations found. For event 25, the only site that did
not exceed 200 mg/L of SS was station 1, upstream from
Rapid City A concentration of 200 mg/L is normally ac-
cepted as an average SS value for untreated domestic
sewage. The municipal wastewater treatment plant is cur-
rently required to reduce SS to a 30 mg/L level prior to
discharge into Rapid Creek.
UNDILUTED EVENT
MEAN CONCENTRATIONS
The undiluted event mean concentration, UEMC, was se-
lected to indicate the quality of runoff entering the stream.
The UEMC was calculated for each parameter from labo-
ratory data for a particular event by subtracting out the
portion associated with the baseline flow, as follows:
QBL CBL + QRO UEMC = Q EMC
and,
UEMC = Q EMC - QBL CBL
where,
QRO
Base flow preceding runoff events, cfs
CBL = Assumed base flow concentration for the param-
eter under consideration. Value is assigned from
baseline quality obtained from samples collected
when runoff was not affecting the quality, mg/L.
QRO = Mean runoff flow, equal to in-stream mean flow
recorded during a runoff event-base flow (Q -
QBL), cfs.
UEMC = Undiluted event mean concentration, a calculated
approximation of the mean concentration of a pa-
rameter in the runoff, mg/L.
Q = Mean in-stream_flow during a runoff event, flow
weighted; i.e., (Q) (Runoff duration) = Volume of
water passing a station during event, cfs.
EMC = Event mean concentration. Obtained by laboratory
testing of a flow weighted composite sample,
mg/L.
A detailed presentation of the UEMC can be found in a
paper by Harms and Smith (1985).
RAINFALL RUNOFF LOADS
Rainfall runoff loads for each event were calculated by
using the UEMC for each parameter. The UEMC was mul-
tiplied by the volume of runoff to give the total load to the
stream in pounds. The amount of material entering Rapid
Creek is dramatically increased as the stream moves
through Rapid City. Table 4 shows these increases as a
percentage of the load entering Rapid City (at station 1).
The higher percentages are associated with those constit-
uents that fluctuate with suspended solids, the most pre-
dominant being total lead. Lead was almost entirely tied to
particulates, and very little dissolved lead was detected.
An important consideration is that these loads are de-
posited in the stream within a relatively short time. Down-
stream from the Rapid City urban area is the municipal
Table 4.—Increase in runoff loads.
Parameter
Ammonia-nitrogen
COD
Chloride
Total Kjeldahl N
Total lead
Nitrate + nitrite
Percent Increase
Sta. 1 to Sta. 56'
633
24,400
12,900
2,530
104,000
2,300
Parameter
Total P
Dissolved P
Sodium
Suspended solids
Volatile SS
Total residue
Percent increase
Sta. 1 to Sta. 561
14,000
800
5,980
63,500
61,400
4,730
Station 56 Is a theoretical site.
504
-------
CONTRIBUTED PAPERS
Parameter
Table 5.—Days needed for effluent from wastewater treatment facility to match urban runoff load.
Urban runoff Time required to generate
load, Ib urban runoff load from effluent, days
COD
median
minimum
maximum
SS
median
minimum
maximum
TKN
median
minimum
maximum
TOTAL P
median
minimum
maximum
15,600
1,480
120,000
205,000
26,700
1,830,000
253
74
4,670
140
16
1,640
4.5
0.4
34.3
117
15
1,045
0.2
0.06
4
0.4
0.04
4.7
Note: Calculations based on an average effluent of 7 MGD, COD = 60 mg/L, SS = 30 mg/L, TKN = 20 mg/L, Total P = 6 mg/L.
wastewater treatment facility. Assuming that the facility
treats its sewage to normal levels, Table 5 gives an indica-
tion of the time required for the effluent to deposit loads
corresponding to the runoff loads for a limited number of
parameters. The predominant problem is consistently sus-
pended matter and those constituents affiliated with sus-
pended material. Organics, based upon chemical oxygen
demand (COD), and nutrients would not appear to cause
excessive degradation of Rapid Creek based upon ac-
ceptable loading rates at the wastewater treatment plant.
However, the suspended solids load can be extreme. It
would take 3 months for the treatment plant to discharge
the same amount of solids that a moderate storm event
would cause to be washed into the creek. It would take
almost 3 years for the effluent to contribute solids equal to
those deposited in Rapid Creek in 1 day by a 6.35-cm (2.5-
inch) rain (Event 21).
From the discussion above, preventive measures ap-
parently should be directed primarily toward reducing the
suspended solids loads. A suspended solids reduction
could also be expected to reduce COD, total Kjeldahl ni-
trogen (TKN), lead, phosphorus, VSS, and total residue
loadings.
WATER QUALITY STANDARDS
The State of South Dakota has water quality standards,
defined according to beneficial use, for every surface wa-
ter within its geographic boundaries. The beneficial uses
for the stretch of Rapid Creek that was studied are listed in
Table 6.
Table 6.—Beneficial uses of Rapid Creek.
Domestic water supply
Cold water permanent fish life propagation
Immersion recreation
Limited contact recreation
Wildlife propagation and stock watering
Irrigation
Table 7 compares the most stringent criteria for the ben-
eficial uses with the observed in-stream values from this
study. As expected, the urban runoff causes the violation
of the in-stream water quality standards on a regular ba-
sis. Standards for SS and fecal coliform are exceeded
during most runoff events, while the standard for ammonia
is exceeded only occasionally. Evaluating the significance
of these violations is difficult. Data which measure the
impact of these levels during dynamic, shock loading situ-
ations are virtually nonexistent.
SUMMARY
Sampling of urban runoff was conducted for 3 years on
five in-stream stations and one major watershed. A theo-
retical station, 56, was used to indicate the in-stream con-
ditions in Rapid Creek below the urban area.
The water quality in Rapid Creek is significantly de-
graded by urban drainage during both snowmelt and rain-
fall runoff events. Most of the water quality parameters
studied increased in concentration at the downstream lo-
cations. The water quality standards, except specific con-
ductance, for the stream in question were violated during
runoff events.
REFERENCES
Harms, L.L., and H. Smith, Jr. 1983. A compositing program for
water quality sampling. Water Eng. and Manage. 130(12): 39-
40.
Harms, L.L., and M. Smith. 1985. Using the Undiluted Event
Mean Concentration to Determine Runoff Loads. Proc. Non-
point Pollut. Abatement Symp. Apr. 23-25, 1985. Marquette
Univ. Milwaukee, Wl.
Harms, L.L., M. Smith, and K. Goddard. 1983. Urban Runoff
Control in Rapid City, South Dakota. Sixth District Council
Local Gov. Rapid City, SD.
U.S. Geological Survey. 1980. Equipment and Supplies—New
Sample Splitter for Water Quality Samples. Tech. Memo. No.
80.17. U.S. Geol. Surv. Reston, VA.
Table 7.—Comparison of water quality criteria and observed values.
Parameter
pH
Ammonia NHs-N, mg/L
Specific conductance, /^mhos/cm
Suspended solids, mg/L
Fecal coliform, No./100 mL
Most stringent
use criteria
6.0 to 8.3
About 0.3
2,500 @ 25°C
30
200
Observed values
in-stream
Min.
6.96
0.03
217
2
. 35
Max.
8.98
0.67
1,010
2,300
68,000
Comment
Violation
Violation
No problem
Violation
Violation
505
-------
STEPHEN F. BLACK
Soil Conservation Service
U.S. Department of Agriculture
Champaign, Illinois
The major resource concerns in Illinois are soil erosion,
water quality, farmland protection, land use changes, wa-
ter supply, flooding, wildlife habitat, and socioeconomics.
In addressing most of these concerns, the hydrologic
area must be considered. In addition, the hydrologic area
is a major factor in addressing other concerns such as
critical soils, mined land, drainage, and irrigation. The
Soil Conservation Service in Illinois is implementing a
pilot program in 14 counties and field offices to deliver
program services by hydrologic areas. This system pro-
vides for developing comprehensive resource plans for
watershed areas within each county, organization of case
files to deliver services by watershed groups, and report-
ing of all accomplishments by each county hydrologic
unit. The system operates within the current county
boundaries with coordination of watershed areas across
county lines as applicable. Resource planning on hydro-
logic areas provides the means for identifying and mea-
suring offsite, as well as onsite, benefits of various con-
servation alternatives. This allows local decisionmakers
to consider and balance the public and private, or offsite
and onsite, benefits of alternative strategies for address-
ing their local resource concerns. Resource plans pre-
pared for hydrologic areas will outline the objectives,
goals, and actions established by the local people. Vari-
ous agency programs will then be used, as applicable, to
address or implement specific parts of a resource plan.
Delivery of services by hydrologic areas will be expanded
to all counties in Illinois over the next 3 to 4 years.
The major resource concerns in Illinois, as identified by
the Resource Conservation Act process, are soil erosion,
water quality, farmland protection, land use changes, wa-
ter supply, flooding, wildlife habitat, and socioeconomic
issues. In addressing most of these concerns, the hydro-
logic area is a highly effective factor. In addition, the hydro-
logic area is a major factor in addressing other concerns,
such as critical soils, mined land, drainage, and irrigation.
The Soil Conservation Service in Illinois is carrying out
a pilot program in 14 counties and field offices to delivery
program services by hydrologic areas. This system pro-
vides for developing comprehensive resource plans for
watershed areas within each county, organizing case files
to deliver services by watershed groups, and reporting all
accomplishments by each county hydrologic unit. Current
political boundaries will not be changed. The system oper-
ates within the current county boundaries with coordina-
tion of watershed areas across county lines as applicable.
The base map for the hydrologic unit system uses the
major river basins of Illinois. Each basin is subdivided into
additional units ranging from 15,000 to 250,000 acres.
This serves as the base map for coordinating all multi-
county watersheds in the State and serves as a starting
point for each county. The size and boundary of each
county subunit is to be determined by the local soil and
water conservation district, in consultation with watershed
sponsors, steering committees, and other local or commu-
nity groups. The following criteria are to be considered in
developing county subunits:
1. Community and landowner priorities and interests.
2. Potential offsite public benefits, such as water sup-
ply, recreation, and so forth.
3. Number and extent of resource concerns being ad-
dressed.
4. Existing or potential project activities.
5. Size of the area. Maximum size not to exceed 40,000
acres.
6. Manageable size—no more than 100 operating
units.
7. Coordination with adjoining counties as to potential
hydrologic unit boundaries.
8. Others as deemed important by local leaders.
One element of a servicing system includes organizing
landowner case files and records by groups. Organizing
by hydrologic units has the following advantages:
1. Resource planning by hydrologic units will require
coordination with individual farm case file conservation
needs and treatment decisions.
2. Mutual interest group planning by hydrologic units
would be facilitated by this grouping.
3. Having case files of adjacent landowners facilitates
progress reporting and additional followup scheduling for
planning, application, and maintenance.
4. Landowners become oriented to the watershed con-
cept and their relationship to the community and public
benefits that will result from conservation treatment.
After the designation of county hydrologic units, the soil
and water conservation district sets priorities of assistance
needs for each unit. These are based upon the number
and degree of resource concerns that need to be ad-
dressed, the potential on and offsite benefits that could
result, the extent of the eroding land needing treatment to
reach the State's Erosion and Sediment Control Guide-
lines, and the readiness of local people to address re-
source concerns.
Priority areas are selected for development of a re-
source plan. This planning process involves the local land-
owners in determining the objectives and goals they want
to attain for the watershed area. Resource planning by
hydrologic areas provides the means for identifying and
measuring offsite, as well as onsite, benefits of various
conservation alternatives. This allows local decisionma-
kers to consider and balance the public and private, or
offsite and onsite, benefits of alternative strategies for ad-
dressing their local resource concerns.
The hydrologic unit concept includes both onsite and
offsite benefits. The progress reporting system will pro-
vide needed data for field office management and should
be developed to provide data on community benefits of
conservation planning and application.
The best technology for measuring some offsite bene-
fits is the use of models. Such models can take into ac-
count before and after erosion rates, sediment delivery
and transport, and pesticide and herbicide loadings. Addi-
tional data need to be developed on the effects of conser-
vation practices and systems on water quality, wildlife hab-
itat, net economic benefits, and so forth. Methods are
needed to include these benefits in resource plans and to
506
-------
measure the benefits as conservation systems are ap-
plied. Effects of individual practices as well as systems
may need to be evaluated in regard to the various onsite
and offsite benefits.
Onsite benefits include maintaining crop productivity by
preventing soil degradation, lessening crop loss resulting
from sedimentation, and reducing land voids caused by
gully or ephemeral erosion. Offsite benefits include recre-
ation uses such as swimming, boating, and fishing and
the associated reduction in these activities caused by im-
paired water use. Other offsite effects are water supply
treatment, loss of water supply storage volume, sediment
in road ditches, flooding, drainage, wildlife, wind erosion,
drifting snow, and aesthetics. Methods of measuring off-
site benefits will need to be developed for field use in
CONTRIBUTED PAPERS
resource planning and in reporting benefits.
A progress reporting system has been established to
capture progress data by each hydrologic unit in each
county. This system uses a two-digit code that identifies
the hydrologic unit within the county. The system is cross
referenced to the national river basin (eight-digit) coding
system. The system permits consolidation of progress
data by any size watershed or river basin unit. The tradi-
tional progress reports by counties will still be available
and will be used for management purposes.
As microcomputers become available in field offices,
progress reporting can be coordinated with automated
case files. In addition, microcomputer modeling and data
base capabilities should enhance the predicting and mea-
suring of both on and offsite benefits.
507
-------
AGRICULTURAL LAND IMPROVEMENT AND WATER QUALITY IN
SOUTH CENTRAL MINNESOTA
HENRY W. QUADE
Department of Biological Sciences
Mankato, Minnesota
INTRODUCTION
Over the past two decades south central Minnesota has
seen a classic confrontation between agricultural and wet-
lands interests, characterized by polarization and single-
issue stances. I propose that a third issue, water quality,
be introduced and suggest that water quality could bring
agricultural and wetlands interests together with a com-
mon purpose, land and water improvement. This paper
will show that geomorphic and water quality studies on
agricultural drainage system in south central Minnesota
demonstrates a potential for moving in this direction. In
fact, legal and legislative imperatives require that water
quality be included as a parameter. Water quality can and
should be the common denominator for all interests in-
volved.
NATURE OF COUNTY DRAINAGE
DITCH SYSTEMS
Extent
In the U.S. Census of Agriculture, 1959, the Department
of Commerce stated that by the end of the 1950's, Minne-
sota had drained 4.7 million ha (11.7 million acres) or 23
percent of its land. This represented 11.5 percent of all the
agricultural land drained in the United States, second only
to Indiana.
In the four south central Minnesota counties, Blue
Earth, Brown, Le Sueur, and Nicollet, we have several
hundred county and judicial ditches. The U.S. Geological
Survey (USGS) listed public drainage by counties in their
1971-1972 drainage survey of Minnesota (Table 1). This
survey based only on responses of county engineers, in-
cluded problems of classification, efficiency, relationship
of private drainage ditches and ditch sheds (U.S. Geol.
Survey, 1971-72).
For purpose of Section 208 planning by the State of
Minnesota, the Department of Agricultural Engineering,
University of Minnesota, surveyed District Conservation-
ists of the Soil Conservation Service (SCS) (Allred and
Geiser, 1978); these figures varied significantly from the
1971-1972 USGS report.
Because of such discrepancies we began an extensive
mapping program in the counties under study. Our results,
similar to earlier USGS findings, raise real questions
about the 1978 survey on which 208 planning was based
(Table 1).
Reasons for Drainage (Legal)
Originally drainage offered a panacea for the problems of
many early Minnesotans, and it was believed that great
benefits could be derived from the reclamation of wet and
overflowed lands. These benefits included: (1) a greater
certainty of a full crop from reduced frost damage; (2) an
increase in yield per acre and a corresponding increase in
market value of land, (3) improvement of highways, (4)
increased profits by freight companies through increased
shipping of agricultural commodities, (5) increased busi-
ness in towns and villages adjacent to reclaimed areas, (6)
improved railroad service because of decreased losses
from floods and softened roadbeds, and (7) improvements
in public health from elimination of disease-breeding
swamps and marshes (Palmer, 1915).
An historic tabulation of reasons for drainage used in
the petitions for Blue Earth County is shown in Table 2
(Quade, 1978). Although these petitions involved much
legal jargon and constitutional definitions that supported
drainage, a trend away from offensive (making land use-
able) to defensive (protecting land and highways) petitions
appeared with time. While most petitions involved several
reasons, soil erosion and water quality were never among
them.
The agricultural benefits derived from drainage were
obvious. Where land could be relieved of water fairly eco-
nomically, such as in the wet prairies, drainage provided
relatively flat and fertile land without the laborious clearing
of trees (Moline, 1969). Moreover, diffused surface waters
were considered "... a common enemy, which an owner,
in the necessary and proper improvement of his land, may
get rid of as best he may .. ." (Pye v. City of Mankato,
1887, 31 N.W. p 863). With these attitudes, drainage was
considered progressive; as a result the early drainage
laws liberally provided a multiplicity of ways to accomplish
wetlands reclamation. Drainage statutes enacted for the
purposes of improving public health proceeded upon the
proposition that wet, low, and marshy lands provide habi-
tat for malaria-carrying mosquitoes, causing a danger to
the health and life of people nearby. (See King, 1980, for a
detailed history of drainage laws in Minnesota).
What Has Been Drained: SCS and Farmer
Perspectives
The findings that 39.9 percent of Blue Earth County was
within artificial public drainage ditchsheds (Table 1), that
only 5.1 percent of the county was surveyed as swamp
and 3.3 percent as lake from the General Land Survey,
and that 12.0 percent of the county is presently in lake-
sheds indicated that wet lands as well as wetlands have
been drained (Dunsmore and Quade, 1979).
Those soils classified as wet and needing drainage for
agriculture comprise 57.9 percent of the county (Jensen,
1981). Now 47.5 percent of these soils are within public
drainage projects, and they comprise 66.6 percent of the
total drained acreage. How much of these "wet" areas
have been privately drained is unknown, but probably sig-
nificant. Peat, only 0.6 percent of the county area, is
drained at 24.2 percent. Jensen's findings indicate that we
will see more drainage in the future since only half of the
Table
County
Blue Earth
Brown
LeSueur
Nicollet
1 .—Extent of public drainage.
1971-72 1978 1979
Dept.
USGS Ag. Eng. M.S.U.
% in public % in public % in public
drainage drainage drainage
50.4
48.2
43.5
59.4
14.9
34.7
14.3
20.2
39.9
45.9
46.7
58.9
508
-------
CASE STUDIES
wet soils in addition to private drainage have been in-
cluded. Diedrick (1981) states that wet soils (wet land),
when adequately drained, are the most productive of
nearly all soils, and further that drained wet soils will yield
10 to 30 more bushels of corn per acre than associated
well-drained soils. In the Midwest wet soils represent
about 29 percent of the cropland and, with their higher
productivity, produce a much higher percent of total crops
grown. Diedrick further states that wet soils are the least
erodible of all farmland (in Minnesota soil loss from llw
capability subclass is 0.93 and from Me is 3.57; Illw is 1.10
and Ille is 6.67 tons per acre). To increase the intensity of
farming on the more sloping soils would require appropri-
ate soil erosion control practices, and the costs must be
shared by the public sector. He concludes that of the 91.8
million acres of prime farmland in the upper Midwest, 33.4
million are capable subclass w soils and, therefore, 36
percent of the prime farmland consists of artificially
drained wet soils. His data, and Jensen's, indicate signifi-
cant amounts of wet soil still to be drained.
From the farmers' economic viewpoint drainage clearly
represents a good investment. Leitch (1981) shows that in
south central Minnesota after all production expenses,
drainage costs, and real estate taxes have been consid-
ered, values of $630 average net return per acre are ex-
pected (drain construction costs capitalized at 12 percent
discount rate). Nonmonetary incentives to drain are also
involved, and Leitch estimates the value of the elimination
of a nuisance wetland to be as much as $30-$60 per acre.
The nuisance factor has increased with the development
of large-scale machinery and larger sized drainage sys-
GEOMORPHOLOGY: IMPLICATIONS
With the departure of the glaciers, the major development
of the landscape was from streams and their branches
developing upslope into the adjacent landscapes. Stream
energy is controlled by the base level of the major
streams, by time, and now by man. The main streams in
Blue Earth County flow within channels developed by gla-
cial torrents. Secondary streams are few. Their drainage
net is weakly developed and all have low gradient, weakly
incised, or unincised channels that developed during ice
wasting. The drainage net development had to be pre-
empted in places by tile outlet ditches and field tiles. De-
pressions are the most common landscape feature. They
identify positions of remnant ice blocks within or under the
wasting ice field. They are expressed as lakes, marshes,
or potholes.
Nearly level land is a hindrance to artificial drainage,
and long ditches of low gradient are required to aid sec-
ondary systems. Public drainage ditch systems (quantita-
tive data on private ditches are lacking) started in the
1890's in Blue Earth County and for the last nine decades
has had total cumulative public drainage of 0.5, 6.4, 20.1,
28.8, 28.9, 29.9, 36.5, 38.3, and 39.9 percent of the sur-
face area of the county Most south central Minnesota
counties have more kilometers of open drainage ditches
than rivers.
The termination of Public Drainage Ditches in Blue
Earth County is shown in Table 3. Fifteen ditches (88.6
percent) terminate in other ditches which then go to rivers,
and two (11.4 percent) terminate in ditches which then go
to lakes.
( '' Table 2.—Reasons for drainage from petitions, Blue Earth County, Minnesota.
Reasons given (by legend number)
Decade
1890-99
1900-09
1910-19
1920-29
1930-39
1940-49
1950-59
1960-69
1970-
Totals
#of
Petitions Hydrologic
2
11
71
19
0
4
8
10
7
132
4
0
% 0
4
% 36
28
% 39
3
% 16
0
% 0
1
% 13
0
% 0
0
°/o 0
36
6
0
0
5
45
68
96
18
95
0
0
3
38
2
20
0
0
96
13
0
0
0
0
0
0
0
0
0
0
0
0
7
70
3
43
10
1
0
0
0
0
1
1
0
0
2
50
4
50
1
10
3
43
11
7
0
0
1
9
22
31
3
16
2
50
1
13
1
10
0
0
30
Agricultural
8
0
0
0
0
0
0
0
0
2
50
1
13
0
0
0
0
3
10
0
0
0
0
0
0
1
5
2
50
2
25
2
20
0
0
7
12
0
0
1
9
40
56
8
42
0
0
2
25
1
10
0
0
52
14
0
0
0
0
7
10
2
11
0
0
0
0
0
0
0
0
9
Highway
5
0
0
0
0
7
10
5
26
1
25
5
63
8
80
5
71
31
9
0
0
1
9
14
20
2
11
0
0
0
0
0
0
0
0
17
Public
Benefit
2
1
50
11
100
71
100
19
100
3
75
7
88
10
100
7
100
129
3
0
0
11
100
70
99
19
100
3
75
7
88
10
100
7
100
127
11
1
50
11
100
70
99
19
100
0
0
1
13
0
0
0
0
102
No.
"O"
0
8
144
31
0
4
7
4
0
No.
"D"
0
0
8
5
0
3
9
16
11
Legend:
1 = Prevent flooding of agriculture land
2 = Promote or improve public health
3 = Be of public utility
4 = Enhance the value of land
5 o Prevent flooding of roads
6 = Reclamation of wet and overflowed lands
7 = Make land useable for cultivation
8 = Make land useable for pasture
9 = Improvement of highways (drain land for construction)
10 * Increase productivity of land
11 = Public welfare and convenience
12 o Make land available for agricultural uses
13 = Prevent flooding of land
14 = Make land valuable for agricultural uses
O = Offensive
6,7,8,9.12
D = Defensive
1,5.13
509
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 3.—Public drainage ditch terminations, Blue Earth County, Minnesota.
Category
No. ditches
Km drained
Percent of drained area
Percent of numbers
Ditch to
river
79
58,792
74.5
74.5
Ditch to
ditch to
river
15
6,848
8.7
14.1
Ditch to
ditch to
lake
2
2,338
3.0
1.9
Ditch to
lake
10
10,521
13.4
9.5
Totals
(106)
(78,499)
(99.6%)
(100.0%)
Table 4.—Erosion control structures in Blue Earth County, Minnesota, by watershed.
WATERSHEDS
CO
^^
£
e
.1
GC
•s
o>
_cr
1
3
Minnesota
Blue Earth
LeSueur
Maple
Watonwan
Big Cobb
Little Cobb
Direct Blue Earth
Minneopa Creek
Morgan Creek
Direct Minnesota
George Lake
Duck Lake
Madison Lake
Severson Lake
TOTALS
453.0
379.2
57.9
76.8
37.5
80.8
33.5
87.7
17.2
9.7
41.0
453.0
138
123
14
20
20
16
7
46
4
1
10
2
1
1
1
143
3.3
3.1
4.1
3.8
1.8
5.1
4.8
1.9
4.3
9.7
4.1
5,667
4,982
667
709
652
684
378
1,891
202
6
477
20.3
6.1
16.2
12.2
5,722
41
40
48
36
33
43
53
41
45
6
48
10.2
6.1
16.2
12.2
99.1
87.1
11.7
12.4
11.4
12.0
6.6
33.1
3.5
.1
8.3
.4
.1
.3
.2
100.1
As large-scale drainage systems increased in the
county, the greater volumes of runoff began to require
controlled management practices to prevent erosion. Ero-
sion control structures (which drop water behind the berm
to prevent gully formation) were introduced by the Gover-
ment with Agricultural Soil and Conservation Service
(ASCS) assitance in 1958. To date, a total of 143 struc-
tures have been built in Blue Earth County (Table 4). Their
primary purpose is to halt progressive gullying or ravine
formation in farmland by controlling surface runoff. Ra-
vines are also a major location of single farmstead dump-
ing and are potentially a significant pollution vector to our
rivers, lakes, and ground water. For a more complete de-
scription of erosion control structures in Blue Earth
County see Davidson (1984).
In our south central Minnesota study (Quade et al. 1980)
found the average of 5.4 km2 (3.34 mi2) of drainage area
for a drainage ditch is significantly higher than reported for
first-order streams throughout the United States. Fasching
(1984) found the entire Minnesota River watershed to av-
erage 9.1 km2 (5.6 mi2) for first-order streams, a measure-
ment characteristic of second-order streams. Leopold,
Wolman, Miller (1964), using a data base of 1,570,000
first-order streams in the United States, found the average
area to be 1.61 km2 (1 mi2). The tile lines function as first-
order streams and the open ditches as second-order and
sometime as third-order streams as defined by Strahler
(Quade et al. 1980). Strahler's (1957) definition of stream
order is: Order 1 is channels without tributaries; order 2 is
channels with only order 1 tributaries, including only the
length segment between the junction upstream of order 1
channels and the junction downstream with another order
2 channel. Drainage density, length of overland flow, and
other quantitative geomorphic parameters support our hy-
pothesis of agricultural drainage mimicing of first- and
second-order systems (Quade et al. 1980).
It has been hypothesized that the glaciated, relatively
immature landscape of Blue Earth County, over a period of
thousands of years, would naturally progress to a state of
physical dissection as seen in the unglaciated portions of
Southeastern Minnesota. However, the active artificial
drainage in Blue Earth County may be rapidly speeding
up the maturation of the landscape. Naturally drained
landscape over an extended period would be the result of
physical gullying and dissection by first-order streams. Ar-
tificial drainage, however, has replaced the gullying and
physical dissection and changed a lake-marsh landscape
to a mature fluvial landscape.
As part of the entire drainage scheme, structures have
functioned to halt the physical dissection of major underfit
river valley walls with the tile lines dissecting the land-
scape underground. Blue Earth County's natural surface
hydrology, then, represents an odd mixture of immaturity
and maturity. The area is characterized by large, underfit
primary river valleys that do not represent well the associ-
ated low stream orders; and yet, the flat topography is well
drained. Erosion control structures and tiling are the an-
swers to the above riddle.
WATER QUALITY IMPACTS
An early study of the causes of eutrophication in Lake
Tetonka, LeSueur County, south central Minnesota had
shown that a major drainage ditch system (LS-C-59),
which entered the Cannon River before entering Lake Te-
tonka, was diluting the nutrient concentrations for all pa-
rameters except nitrate-nitrogen (similar to that of the
river) (Quade et al. 1979).
To examine the output of nutrients from drainage
ditches and natural streams, a series of 17 sampling sta-
tions was used as shown in four counties (Fig. 1). Al-
though an attempt was made to group data by county
510
-------
CASE STUDIES
management, geomorphology, soils, precipitation, and
stream order variation, stream order worked best. Water
quality data were based on 3,700 data points at 17 sites
and were grouped into three seasons for 1979: spring
runoff (Mar. 21-May 30); growing season (June 10-Oct.
11); and fall harvesting and plowing (Oct. 24-Nov. 11).
The major results of the study included the variability of
ditches and streams within the same order, the variability
of the season of maximum loading, and the lack of predict-
ability of the relationship of flow to nutrient concentration
(Quadeetal. 1980).
In Table 5 the relationship of flow to loads of the various
nutrients was not consistent, indicating individuality of nu-
trient concentration among ditches and streams. For ex-
ample, in comparing three second-order ditches with the
same total flow, Br-J-29, B-J-10, and BE-C-5, one can
see that while total phosphate was fairly similar, TKN and
nitrate-nitrogen varied greatly.
Generally when flow was greatest in one of the seasons
the chemical load was also greatest in the same season
(Table 6). Ditch Br-J-30 of Brown County had its major
loading for all parameters in the first season while ditch
Br-J-10's (also Brown County) loading came in the sec-
ond season. Nitrate-nitrogen followed flow closely
whereas orthophosphate showed the greatest variations.
The third-order ditches and all the rivers had major per-
centages of flow in the second season except for the Can-
non River which had 49.8 percent in the first season. Five
of the second-order ditches had major flows in the second
season.
Shanaska Creek, which comes off a lake, was at vari-
ance with the general trend of the largest chemical loads
being in the same season as the largest percentage flows
and showed a load of nitrate-nitrogen 64.1 percent in the
first season, spring runoff.
From the above study, individual drainage systems
show a great deal of variability in nutrient concentrations,
loads, flow, and seasonality of loading. This is also seen in
the maximum/minimum loading ratios where third-order
ditches (ditches fed by other ditches) are very similar to
third-order rivers, while the second-order ditches show
large differences (Quade et al. 1980).
A study done in northern LeSueur County on proposed
Ditch 71 by Larson-Albers (1982) examined the input of
.drainage in a glacial moraine area. The river studied was
not underfit, and the wetlands were riverine and palustrial
near the river. Although the proposed ditching of the river
was halted precluding a direct before and after water qual-
ity comparison, Larson-Albers was able to compare tribu-
tary chemistry from ditches versus wetlands by an analy-
sis of variance.
This study demonstrated some significant differences in
nutrient concentrations among tributaries of a stream in a
small geomorphically homogeneous watershed. A wet-
land-to-ditch continuum could be formed primarily on the
basis of nitrate levels, low in wetlands and high in drain-
age ditches. One wetland in the spring and two in the
summer were distinct from drainage ditches in terms of
their high levels of orthophosphate (PO4-P) and total Kje-
dahl nitrogen (TKN). Land use was shown to be a possible
contributor to water quality with a lower row crop-to-wet-
land ratio matching lower nitrate-nitrogen (NO3-N) and
higher TKN and PO4-P concentrations. This research indi-
cated that the amount of tiling and the position of wetlands
in a watershed may affect water chemistry.
DISCUSSION AND SUMMARY
Geomorphological studies in south central Minnesota indi-
cate that agricultural drainage systems mimic what would
have been done naturally given enough time. Further-
more, drainage systems may save upland soil (by reduc-
ing turbidity going to receiving bodies), reduce biocide
outflow (less soil loss, less piggybacking of biocides), re-
duce phosphate outflow, and usually increase nitrate-ni-
trogen outflow. One has only to look at the highly dis-
Second Order Ditches
9 Third Order D)tches
A Third Order Rivers
O Second Order River
Figure 1 .—Ditch and river water quality study sampling sites.
511
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
Table 5.—Total cumulative load of water quality parameters by site in kg.
Second-order
ditches
Third-order
ditches
Third-order
rivers
Second-order
rivers
Site
N-C-40A
N-C-38A
Br-J-29
Br-C-5
Br-J-30
Br-J-10
BE-C-56
BE-C-5
LS-C-59
LS-C-58
BE-J-48
N-J-1A
N-J-1A*
Br-BE Morgan
Creek
BE-Little Cobb
River
LS-Cannon
River
LS-Shanaska
Creek
Total PO,
No Flow
1,599.0
1,200.3
51.5
8,711.4
1 ,009.3
2,154.8
1,275.4
795.4
326.0
2,980.3
10,973.2
5,533.5
8,299.9
5,985.9
12,394.6
219.3
TKN
No Flow
19,250.6
1 1 ,743.4
634.9
46,608.8
9,437.3
18,666.5
17,730.9
14,626.1
4,448.4
29,922.3
100,510.0
37,361.1
63,328.5
30,630.5
132,032.0
9,961.4
N03
No Flow
193,882.2
116,137.1
12,190.7
302,405.5
185,116.1
158,199.1
76,149.5
80,262.4
50,935.3
588,588.9
1,176,561.5
658,465.7
1,033,460.4
576,006.7
254,467.2
6,753.3
Total
Dissolved
Solid
No Flow
6,614,791.7
3,973,388.3
481,305.0
9,988,864.5
5,822,894.3
5,947,185.2
4,478,515.6
2,629,249.8
1,093,212.5
16,935,025.5
35,325,877.5
21,576,960.0
38,010,640.0
16,288,744.1
20,767,024.7
1,535,491.0
Total
Flow1
0.0
93.9
75.7
5.8
242.0
70.0
107.0
78.3
58.9
17.7
260.8
535.3
280.8
482.5
303.4
528.7
45.2
'in 100,000m3
Table 6.—Seasonal breakdown of water quality parameter loading as percentage of total.
Second Order Ditches
Seasonal loads
Flow
Spring runoff
Growing season
Fall harvest & plowing
Total orthophosphate-phosphorus
Spring runoff
Growing season
Fall harvest & plowing
Total Kjeldahl Nitrogen
Spring runoff
Growing season
Fall harvest & plowing
Nitrate-Nitrogen
Spring runoff
Growing season
Fall harvest & plowing
Total dissolved solids (conductivity)
Spring runoff
Growing season
Fall harvest & plowing
N-C-40A N-C-38A
No Flow
— 23.8
— 64.9
— 11.3
— 21.4
— 67.3
— 11.3
— 22.6
— 68.1
— 9.3
— 23.1
— 64.8
— 12.1
— 19.6
— 74.2
— 6.2
Br-J-29
28.6
38.9
32.5
35.4
24.8
39.8
36.0
39.7
24.3
20.7
43.0
36.2
23.9
61.7
14.4
Br-C-5
37.9
62.1
.0
13.4
86.6
.0
25.8
74.2
0.0
39.6
60.4
.0
31.7
68.3
.0
Br-J-30
81.3
13.2
5.5
90.5
5.6
3.9
86.6
9.3
4.1
73.5
16.5
10.0
69.3
37.9
2.8
Br-J-10
20.0
66.4
13.6
11.2
78.4
10.4
15.6
77.6
7.4
21.4
63.7
14.8
12.7
73.8
13.5
BE-C-56
31.9
65.2
2.8
37.8
61.2
1.1
25.2
74.0
.7
30.5
65.3
4.2
22.2
75.7
2.1
BE-C-5
67.7
31.1
1.2
57.3
41.9
.8
64.6
34.0
1.4
73.7
25.5
.7
60.3
39.7
.0
LS-C-59
64.4
35.6
.0
40.1
59.9
.0
56.3
43.7
.0
78.0
22.0
.0
50.2
49.8
.0
LS-C-58
60.2
32.9
6.9
53.3
40.5
6.1
66.2
28.6
5.2
58.2
33.7
8.1
45.4
46.5
8.1
sected, nonglaciated regions of Southeastern Minnesota
to see what could happen in time.
The SC8 should, in my opinion, classify agricultural
drainage as a conservation measure and thereby include
more water quality features in drainage system design.
During its first 50 years the SCS received criticism for its
agricultural practices, especially drainage. The 1954 Wa-
tershed Protection and Flood Prevention Act, PL. 566,
along with Congressional directives relating to the act, is a
classic case in point (Jahn, 1973). Bill Sayre, an engineer
active in drainage projects in south central Minnesota,
stated that no significant changes in design have been
made since the original systems were constructed, and
only some materials have been changed (1985).
The design of drainage systems must include a water
quality component now because Minnesota is attempting
to recodify the drainage systems and because of the na-
tional plan to implement best management practices
(BMP's).
Minnesota drainage law is contained in Chapter 106,
Chapter 109 dealing with township rights to establish
ditches, and Chapter 112, dealing with watershed dis-
tricts' rights to establish ditches (Stevens, 1985). The
roots of Chapter 106 go back to the mid 1800's. Because
of all the repeals and amendments to the statutes, any
interpretation of the laws has become a judicial nightmare
with objectives at wide variance. Emphasis, in court, has
been shifted almost completely to process rather than to
substance resulting in lack of meaningful or even under-
standable direction. Minnesota State Senator Gary
DeCramer is sponsoring a bill that calls for recodification
of all the legalese in Chapter 106; 12 different words are
used in Chapter 106 for "ditch" (Cross, 1985). This recodi-
fication, although not a revision, would, however, clear up
disputes about what constitutes ditch repairs and ditch
512
-------
CASE STUDIES
improvements. DeCramer feels it is time for the legislature
rather than the courts to address drainage law.
Most important will be the second step involving sub-
stantive changes. Proposals for new drainage and im-
provement of existing drainage in southern Minnesota are
increasingly finding their way to the State Supreme Court.
Judge Lawrence Yetka, with the concurrence of Judges
James Otis, C. Donald Peterson, and Fallen Kelly, de-
clared, "Surely, under the new environmental laws serious
doubt as to the desirability of any general drainage
schemes must exist" (March 11, 1977, Mankato Free
Press).
Examples of substantive issues include the problem of
benefit assessments that have forced farmers to sell their
land for draining wetlands they wanted to maintain. Prob-
lems exist with local datum which often was a spike in a
tree (Maher, 1985). The custodian of the datum base is
often the county engineer but no law assigns responsibil-
ity for datum maintenance and update. The system is not
user friendly. Thusly created controversies in ditch clean-
out and repair cases have resulted in the drainage of
once-protected wetlands and the flooding of lakes.
By far the most inflammatory issue will be that of wet-
land conservation. Joseph Alexander, Commission of the
Minnesota Department of Natural Resources, states that
substantive changes in the drainage laws are necessary
and that these should include a greater sensitivity to the
overall value of our remaining wetlands to be incorporated
into a revised State Drainage Law (Alexander, 1985). Any
substantive changes will involve an extremely rigorous de-
bate, representing a classic example of personal property
right versus environmental rights. The ditch systems are
viewed as a property right paid for and assessed, a "value
added property right" (Maher, 1985). However, the rela-
tionship of drainage to water quality may emerge as a key
factor. The citizens themselves (from our predominantly
agricultural region) are placing water quality as the most
important water problem. In his report, The Citizen and
Water Management: An Atlas of Water Attitudes in South-
ern Minnesota, Moline (1974) investigated the spatial vari-
ation in the perception of water resources and water prob-
lems in south central Minnesota. His study involved both a
stratified (areal) random poll of residents and a select
group poll of individuals concerned with water manage-
ment and planning within the basin. When asked to rank
the three major water resource problems in this area, the
respondents, surprisingly, in both the random poll and the
select group poll, indicated that "water pollution" was
clearly number one with "inadequate water resource plan-
ning" number two. "Wet agricultural fields" was the num-
ber three priority for the stratified random sample poll.
The impact of agricultural practices on water quality has
only recently been examined in some detail. The data
available by the end of the 1960's were extremely limited
(Willrich and Smith, 1970), and data on the impact of
drainage systems on nutrient loading was almost nonex-
istent (Schilfgaarde, 1974). During the 1970's large quan-
tities of data were collected as a result of government
regulation and funding. (Some of this data is ill-based and
suspect, for example, the drainage extent data presented
in this paper.) Recent national nutrient loading data clearly
show, however, that nonpoint sources greatly exceed
point sources for most parameters (exception of BOD5),
and within agricultural cropland, nonpoint exceeds point
in all parameters except perhaps total phosphorus of
rangeland (Duttweiler and Nicholson, 1983). Because of
the relationship of drainage to cropland this indicates the
potential importance of drainage to water quality.
Duttweiler and Nicholson, 1983, state that Section 209
of the Clean Water Act provides a mechanism for consoli-
dating resource management and BMP's and for integrat-
ing water resources development with water pollution con-
trol but that, regrettably, this aspect of the Clean Water Act
has not received much public attention. They further state
that, in the context of managing the hydrologic-ecologic-
agronomic-economic system, BMP's offer a means of ad-
justing the agronomic subsystem that is the source of pol-
lution and, to a lesser extent, the hydrologic subsystem
that regulates the rate of delivery and quantity of pollu-
tants reaching the receiving waters. They then break into
the familiar scenario that BMP's are generally perceived
to be of less benefit to the landowner who applies them
than to the downstream water users or to the general
public who benefit from improved water quality. Baker and
Johnson, 1983, emphasize the lack of information on field-
to-stream transport systems' effect on chemicals and sed-
iment. They state that tile drainage has generally been
considered a production practice rather than a BMP for
environmental protection.
In the summary of the conference on Agricultural Man-
agement and Water Quality at Ames, Iowa, the question
was raised whether, despite the emphasis on erosion con-
trol, reducing soil loss to the T value is a valid water quality
goal or merely a convenient surrogate? Should a BMP
systems approach be for the integrity of stream ecology,
for water quality based on receiver impacts, or for riparian
ecosystems? BMP's should be tailored to each watershed
hydrology by a unified comprehensive approach (Schaller
and Bailey, 1983).
A further potential benefit of redesigning drainage sys-
tems could be greater utilization by the warmwater fish-
eries of south central Minnesota (Peterson, 1985).
Changes in slope of the banks could greatly aid in replac-
ing lost spawning habitat. At present, ephemeral streams
that become channelized (put into a drainage system) are
often downgraded to the lowest possible classification
available (Class 7, Minnesota Pollution Control Agency
(MPCA)). The U.S. Environmental Protection Agency
(U.S. EPA) has recently charged the MPCA to reevaluate
all Class 7 designations in Minnesota over the next 5
years. Most of these are drainage ditches.
It is my contention that BMP's are not yet holistic be-
cause the emphasis is on the source not on the delivery
system or on the receiver impacts. The research I have
reported on from our area in south central Minnesota
shows that the delivery system (agriculture drainage) is
geomorphically sound, cost-effective for the farmer, and
has some real water quality benefits to the receiving body.
These multiple attributes, however, are not recognized or
credited, but, rather drainage progresses in a defensive,
single-issue posture. We need to examine and recognize
carefully the following points:
1. Agriculture drainage is geomorphically sound and
will prevent problems in the future.
2. Drainage to farmers is cost-effective and they rank
water quality as a prime concern.
3. We need to communicate and recognize the value of
wetlands, not just their extent.
Given the three points above we need to design agricul-
tural drainage systems (engineers) that not only get water
off the fields but do so with a delivery system that has the
least negative impact on the water quality of the receiving
body. Our research has shown that drainage ditch sys-
tems act individually and that differences in load and sea-
son of load vary greatly. This change in engineering em-
phasis should not only be applied to new ditches but also
to ditch cleanouts and repairs, where most of southern
Minnesota's legal problems have arisen recently. Finally, it
should be possible to drain wet land without negatively
affecting wetlands; and one could possibly incorporate the
denitrifying ability of wetlands into a holistic drainage sys-
tem.
513
-------
PERSPECTIVES ON NONPOINT SOURCE POLLUTION
ACKNOWLEDGEMENTS: I wish to thank and to credit the fol-
lowing graduate students whose information I have freely used
in this paper: Catherine Larson-Albers, Kent Boyum, Mark
Davidson, Laverne Dunsmore, Paul Fasching, Jim Jensen,
Kevin E. King, Clay Pierce, Ainars Silis, and Bill Thompson. I
also want to thank the Boards of Commissioners of Blue Earth,
LeSueur, Nicollet, and Brown Counties for their help and support
through this investigation. Finally I want to thank Anita Dittrich
and Pat Rosin for typing this manuscript.
REFERENCES
5Alexander, J. 1985. Wetland drainage: time for a new look at an
old law. Minn. Volunteer. 48 (279): 58-62.
Allred, E.R., and K.M. Geiser. 1978. A survey of Agricultural
'irrigation and Drainage Practices. Rep. Minn. Soil Water Con-
serv. Board and Minn. Pollut. Control Agency.
Baker, J.L., and H.P. Johnson. 1983. Evaluating the effective-
ness of BMPs from field studies. Pages 281-304 in F.W.
Schaffer and G.W. Bailey, eds., Agricultural Management and
Water Quality. Iowa State Univ. Press. Ames.
Cross, J. 1985. Recodification Necessary for State's Drainage
Laws. Page 34. Mankato Free Press, Mankato, Wl.
Davidson, M. 1984. The impact of erosion control structures on
the water quality of the Cobb River in Blue Earth County,
Minnesota. Rep. No. 5. Water Resour. Res. Center.
Diedrick, R.T. 1981. The agriculture value of wet soils in the
Upper Midwest. Pages 97-106 in Brandt Richardson, ed., Se-
lected Proc. Midwest Conf. Wetland Values Manage. June
17-19. Minn. Water Plann. Board, St. Paul.
Dunsmore, L., and H.W. Quade. 1979. Public drainage atlas,
Blue Earth County, Minnesota. Limnol. Contribution No. 6.
Dep. Biology, Mankato State Univ. Mankato, Wl.
Duttweiler, D.W., and H.P. Nicholson. 1983. Environmental prob-
lems and issues of agricultural nonpoint source pollution.
Pages 3-16 in F.W. Schaffer and G.W. Bailey, ed. Agricultural
Management and Water Quality. Iowa State Univ. Press,
Ames.
Fasching, P. 1984. Drainage basin characteristics of the Minne-
sota River. J. Minn. Academy Sci. 49 (3): 10-4.
Jahn, L.R. 1973. Watershed program lacks ecological dimen-
sions. Pages 183-95 in C.J. Goldman, J. McEvoy III, and P.J.
Richerson, ed. Environmental Quality and Water Develop-
ment. W. H. Freeman. San Francisco.
Jensen, J.E. 1981. The relationship of artificial drainage to soils
in Blue Earth County, Minnesota. Rep. No. 2. Water Resour.
Res. Center. Univ. Minn.
King, K.E. 1980. A history of drainage law in Minnesota with
special emphasis on the legal status of wet lands. Bull. 106.
Water Resour. Res. Center. Univ. Minn.
Larson-Albers, C.E. 1982. The impact of wetlands and drainage
on water quality in an agricultural watershed in South Central
Minnesota. Rep. No. 3. Water Resour. Res. Center, Univ.
Minn.
Leitch, J.A. 1981. Prairie wetlands allocation: an overview of
landowner alternatives and regional impacts. Pages 467-78
in B. Richardson, ed. Selected Proc. Midwest Conf..Wetland
Values Manage. June 17-19. Minnesota Water Plann. Board.
Leopold, L.B., M.G. Wolman, and J.P. Miller. 1964. Fluvial Proc-
esses in Geomorphology. W.H. Freeman. San Francisco.
Maher, W. 1985. Pers. Commun. (March) Blue Earth County
Board Comm.
Moline, R.T. 1969. The modification of the wet prairie in South-
ern Minnesota. Ph.D. Thesis. Dep. Geography, Univ. Minn.
. 1974. The citizen and water management: an atlas of
water attitudes in Southern Minnesota. OWRR Project B-
0410 Minn. Minneapolis.
Northwest Reporter. 1925. Volume 205. West Publishing Com-
pany, St. Paul, MN.
1887. Pye vs. The City of Mankato. 31: 863.
Palmer, B. 1915. Swamp land drainage with special reference to
Minnesota. Studies Social Sci. No. 5. Univ. Minn.
Peterson, D. 1985. Personal Commun. (March) Area Fish. Su-
pervisor, Waterville.MN.
Quade, H.W. 1978. County drainage ditches in South Central
Minnesota: a unique riparian ecosystem in R.R. Johnson and
J.F. McCormick, tech. coord. Strategies for Protection and
Management of Floodplain Wetlands and Other Riparian Eco-
systems, Proc. Symp. Dec. 11-13, 1978, Callaway Gardens,
GA. Gen. Tech. Rep. Wo-12. Forest Serv. U.S. Dep. Agric.,
Washington, DC.
Quade, H.W., B. Colakovic, B. Hill, and T. Larson. 1979. Limno-
logical investigation of the Cannon River-County Ditch 59 wa-
tershed of Lake Tetonka. Limnol. Contrib. No. 13. Dep. Biol-
ogy. Mankato State Univ.
Quade, H.W, et al. 1980. The Nature and Effects of County
Drainage Ditches in South Central Minnesota. Bull. 105. Wa-
ter Resour. Res. Center. Univ. Minn.
Sayre, B. March, 1985. Pers. Commun. Drainage Eng., Bolten &
Menk Engineering.
Schaller, F.W., and G.W. Bailey. 1983. Agricultural Management
and Water Quality. Iowa State Univ. Press. Ames.
Schilfgaarde, J.V. 1974. Drainage for Agriculture. No. 17 Agron.
Series. Am. Soc. Agron. Inc., Madison, Wis.
Stevens, D. 1985. Pers. Commun. Blue Earth County Board
Commissioners.
Strahler, A.N. 1957. Quantitative analysis of watershed geomor-
phology. Am. Geophys. Union. Trans. 38: 913-20.
United States Department of Commerce. 1959. Drainage of Ag-
ricultural Lands. U.S. Census Agric. Washington, DC.
United States Geological Survey. 1971-1972. Drainage survey
by counties for Minnesota. Unpubl.
Willrich, T.L., and G.E. Smith. 1970. Agricultural Practices and
Water Quality. Iowa State Univ. Press. Ames.
------- |