&EPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington, DC 20460,
EPA/625/R-00/001
July 2000
National Conference on
Tools for Urban Water
Resource Management &
Protection
Proceedings
Chicago, IL
February 7-10, 2000
-------
EPA/625/R-00/001
July 2000
National Conference on Tools for
Urban Water Resource
Management and Protection
Proceedings
February 7-10, 2000
Chicago, IL
Technology Transfer and Support Division
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
Notice
The views expressed in these Proceedings are those of the individual authors and do not necessarily reflect
the views and policies of the U.S. Environmental Protection Agency (EPA). Scientists in EPA's Office of
Research and Development have prepared the EPA sections, and those sections have been reviewed in
accordance with EPA's peer and administrative review policies and approved for presentation and publication.
-------
Preface
A wide array of effective water quality management and protection tools has been developed for urban
environments, but implementation is hindered by a shortage of technology transfer opportunities. Held in
Chicago, Illinois on February 8-10,2000, the National Conference on Tools for Urban Water Resource
Management and Protection was designed to facilitate the educational process and transfer state-of-the-art
information to state, regional, and local urban water quality practitioners.
The Chicago Botanic Garden, which is owned by the Forest Preserve District of Cook County and managed
by the Chicago Horticultural Society, was pleased to coordinate the Office of Wastewater Management and
its Region 5 office, as well as the Northeastern Illinois Planning Commission. The conference was conducted
in cooperation with the Water Environment Federation. Over 450 attendees participated, including
representatives from Australia, Brazil, Canada, Chile, New Zealand, and Turkey.
The timing for this conference coincided well with the U.S. Environmental Protection Agency's release of the
NPDES Storm Water Phase II Final Rule in October 1999. The conference provided participants with
practical, applied information on the most effective tools and technologies for meeting these new NPDES
permit requirements. Program topics were carefully chosen to reflect the Phase II Program's six priorities:
public education, public involvement, detection and elimination of illicit discharges, construction site runoff
control, post-construction storm water management, and pollution prevention for municipal operations.
Two special pre-conference workshops were held on February 7. Better Site Design and Storm Wafer
Management Techniques for Phase // Communitiesexplored the benefits of alternative urban site design
approaches, as well as new advances in storm water management to protect water resources. The workshop
was led by staff from the Center for Watershed Protection. The second pre-conference workshop,
Introduction to Urban TMDLs, examined current and pending requirements for total maximum daily load
(TMDL) programs. Instructors for this workshop were staff from Tetra Tech, Inc. Each of the workshops
attracted over 135 participants.
This Conference Proceedings includes many of the papers presented during the conference. All papers
included were peer reviewed. Additional copies, in either paper or CD-ROM format, are available free of
charge from the U.S. Environmental Protection Agency, telephone 800/490-9198, or visit the web site
.
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting the Nation's land, air, and
water resources. Under a mandate of national environmental laws, the Agency strives to formulate and
implement actions leading to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate, EPA's research program is providing data and
technical support for solving environmental problems today and building a science knowledge base necessary
to manage our ecological resources wisely, understand how pollutants affect our health, and prevent or
reduce environmental risks in the future.
The National Risk Management Research Laboratory is the Agency's center for investigation of technological
and management approaches for reducing risks from threats to human health and the environment. The
focus of the Laboratory's research program is on methods for the prevention and control of pollution to air,
land, water and subsurface resources; protection of water quality in public water systems; remediation of
contaminated sites and ground water; and prevention and control of indoor air pollution. The goal of this
research effort is to catalyze development and implementation of innovative, cost-effective environmental
technologies; develop scientific and engineering information needed by EPA to support regulatory and policy
decisions; and provide technical support and information transfer to ensure effective implementation of
environmental regulations and strategies.
This publication has been produced as part of the Laboratory's strategic long-term research plan. It is
published and made available by EPA's Office of Research and Development to assist the user community
and to link researchers with their clients.
E. Timothy Oppelt, Director
National Risk Management Research Laboratory
IV
-------
Contents
Jordan Cove Urban Watershed National Monitoring Project , 1
Melville P. Cote, Jr.
U.S. Environmental Protection Agency
Off ice of Ecosystem Protection
Boston, Massachusetts
Dr. John Clausen
University of Connecticut
Bruce Morton
Aqua Solutions, LLC
Paul Stacey
Connecticut Department of Environmental Protection
Stan Zaremba
Connecticut Department of Environmental Protection
Sources of Phosphorous in Stormwater and Street Dirt from Two Urban Residential Basins in
Madison, Wisconsin, 1994-95 9
R.J. Waschbusch
U.S. Geological Survey
Middleton, Wisconsin
W.R. Selbig
U.S. Geological Survey
Middleton, Wisconsin
R.T. Bannerman
Wisconsin Department of Natural Resources
Madison, Wisconsin
Using Biological Criteria to Assess and Classify Urban Streams and Develop
Improved Landscape Indicators 32
Chris 0. Yoder and Robert J. Miltner
Ohio EPA, Division of Surface Water
Ecological Assessment Unit
Groveport, Ohio
Dale White
Ohio EPA, Division of Surface Water
Information Resources Management Section
Columbus. Ohio
Getting Past the Obvious 45
Bobbin B. Sotir
Robbin B. Sotir & Associates, Inc.
Marietta, Georgia
-------
Protecting and Enhancing Urban Waters: Using All the Tools Successfully 53
Eric H. Livingston
Stormwater/NPS Management Section
Florida Department of Environmental Protection
Tallahassee, Florida
Structures to Enhance Fish Habitat in Urban Streams 69
Douglas T. Severn
URS Greiner Woodward-Clyde
Seattle, Washington
Lessons Learned About Successfully Using Infiltration Practices , , , 81
Eric H. Livingston
Stormwater/NPS Management Section
Florida Department of Environmental Protection
Tallahassee, Florida
Potential New Tools for the Use of Tracers to Indicate Sources of Contaminants to
Storm Drainage Systems 97
Robert Pitt, Melinda Lalor, Jennifer Harper, and Christy Nix
Department of Civil and Environmental Engineering
The University of Alabama at Birmingham
Birmingham, Alabama
Donald Barbe'
Department of Civil and Environmental Engineering
The University of New Orleans
New Orleans, Louisiana
Elimination of Illicit Connections in Coastal New Hampshire Spurs Cooperation and Controversy 110
Natalie Landry and Robert Livingston
New Hampshire Department of Environmental Services
Concord, New Hampshire
Using Collaborative Problem-Solving to Protect North Carolina's Coastal Resources:
The Experience of the White Oak Advisory Board , , 117
Leon E. Danielson, C. Suzanne Hoover, and Christy A. Perrin
North Carolina State University
Department of Agricultural and Resource Economics
Raleigh, North Carolina
Nancy M. White
North Carolina State University
School of Design
Raleigh, North Carolina
Ron Elmore
North Carolina Department of Transportation
Raleigh, North Carolina
Jennifer L. Platt
Town of Gary
Cat-y, North Carolina
VI
-------
Community Responses to Stormwater Pollution: Case Study Findings with Examples from the Midwest 124
George P. Aponte Clarke
Policy Analyst
Natural Resources Defense Council
New York, New York
Peter H. Lehner
Chief, Environmental Protection Bureau
New York State Attorney General's Office
New York, New York
Diane M. Cameron
President
Cameron Associates
Kensington, Maryland
Andrew G. Frank
Litigation Associate
Paul, Weiss, Rifkind, Wharton, and Garrison
New York, New York
Integrated Urban Stormwater Master Planning 132
Eric Strecker, VP, and Krista Reininga, Sr. Project Engineer
URS Greiner Woodward-Clyde
Portland, Oregon
Conservation Design: Managing Stormwater through Maximizing Preventive Nonstructural Practices 147
Wesley R. Horner
Environmental Management Center
Brandywine Conservancy
Chadds Ford, Pennsylvania
Low-Impact Development Design: A New Paradigm for Stormwater Management Mimicking and
Restoring the Natural Hydrologic Regime ,... 158
Larry S. Coffman
Associate Director
Department of Environmental Resources
Prince George's County, Maryland
A National Menu of BMPs for the Phase II NPDES Storm Water Program 168
James H. Collins
Tetra Tech, Incorporated
Fairfax, Virginia
John A. Kosco, P.E.
USEPA/Office of Wastewater Management
Washington, D.C.
VII
-------
Determining Urban Stormwater BMP Effectiveness 175
Eric Strecker
URS Greiner Woodward Clyde
Portland, Oregon
Marcus M. Quigley
URS Greiner Woodward Clyde
Boston, Massachusetts
Ben R. Urbonas
Urban Drainage and Flood Control District
Denver, Colorado
Texas Nonpoint SourceBOOK is Now On-Line ! 186
John Promise, P.E.
Director of Environmental Resources
North Central Texas Council of Governments, and
Chair, Water Resources Management Committee
American Public Works Association
Keith Kennedy
North Central Texas Council of Governments
Robert W. Brashear, Ph.D.
Camp Dresser & McKee, Inc.
A Comparison of the Long-Term Hydrological Impacts of Urban Renewal versus Urban Sprawl 192
Jon Harbor
Department of Earth and Atmospheric Sciences
Purdue University
West Lafayette, Indiana
Bernie Engel
Department of Agricultural and Biological Engineering
Purdue University
West Lafayette, Indiana
Don Jones
Department of Agricultural and Biological Engineering
Purdue University
West Lafayette, Indiana
Shilpam Pandey
Department of Earth and Atmospheric Sciences
Purdue University
West Lafayette, Indiana
K.J. Lim
Department of Agricultural and Biological Engineering
Purdue University
West Lafayette, Indiana
Suresh Muthukrishnan
Department of Earth and Atmospheric Sciences
Purdue University
West Lafayette, Indiana
Comparative Nutrient Export and Economic Benefits of Conventional and Better Site Design Techniques 198
Jennifer Zielinski, Deb Caracao, and Rich Claytor
Center for Watershed Protection
Ellicott City, Maryland
viii
-------
Predicting Erosion Rates on Construction Sites Using the Universal Soil Loss Equation in
DaneCounty.Wisconsin 206
J.D. Balousek
Erosion Control Engineer
Dane County Land Conservation Department (LCD)
A. Roa-Espinosa
Urban Conservationist
Dane County LCD
G.D. Bubenzer
Professor
University of Wisconsin-Madison
Madison, Wisconsin
Public Involvement Programs That Support Water Quality Management 214
Josephine Powell
Wayne County Department of Environment
Detroit, Michigan
Zachare Ball and Karen Reaume
Environmental Consulting & Technology Inc.
Detroit, Michigan
The Water-Wise Gardener Program: Teaching Nutrient Management to Homeowners 222
Marc T. Aveni
Virginia Cooperative Extension
Manassas, Virginia
Chicago Wilderness; Toward an Urban Conservation Culture , 225
John D. Rogner
U.S. Fish and Wildlife Service
Chicago Field Office
Barrington, Illinois
A Survey of Resident Nutrient Behavior in the Chesapeake Bay Watershed 230
Chris Swann
Center for Watershed Protection
Ellicott City, Maryland
Lawn Care and Water Quality: Finding the Balance 238
Jerry Spetzman
Minnesota Department of Agriculture
St. Paul, Minnesota
San Francisco Bay Area's Pesticide Toxicity Reduction Strategy , , 243
Geoff Brosseau
Bay Area Stormwater Management Agencies Association (BASMAA)
Oakland, California
Administering the NPDES Industrial Storm Water Program at the Municipal Level 249
Michael J. Pronold
Bureau of Environmental Services
Portland, Oregon
-------
Lessons Learned From Three Watershed-Sensitive Development Demonstration Projects In
The Great Lakes Basin 256
Sarah Bennett Nerenberg
The Conservation Fund
Great Lakes Office
Chicago, Illinois
Continuous Deflection Separation (CDS) for Sediment Control in Brevard County, Florida 264
Justin Strynchuk, John Royal, and Gordon England, P.E.
Brevard County Surface Water Improvement
Viera, Florida
Use of Automated Technologies in Watershed Management Planning 272
Lake County Stormwater Management Commission (SMC)
Libertyville, Illinois
Sediment and Runoff Control on Construction Sites Using Four Application Methods of Polyacrylamide Mix ... 278
A. Roa-Espinosa
Urban Conservationist
Dane County Land Conservation Department, and
Assistant Visiting Professor
Biological Systems Engineering
University of Wisconsin-Madison
G.D. Bubenzer
Professor
Biological Systems Engineering
University of Wisconsin-Madison
E.S. Miyashita
Former Project Assistant
Biological Systems Engineering
University of Wisconsin-Madison
Construction Site Planning and Management Tools for Water Quality Protection 284
Thomas Mumley
California Regional Water Quality Control Board
San Francisco Bay Region
Oakland, California
Regulating Sedimentation and Erosion Control into Streams: What Really Works and Why? 291
Seth R. Reice
Department of Biology
University of North Carolina at Chapel Hill
Chapel Hill, North Carolina
JoAnn Carmin
Department of Public Policy
Virginia Polytechnic Institute and State University
Blacksburg, Virginia
-------
Effectiveness in Erosion and Sediment Control: New Initiatives in Indianapolis , 296
James Hayes and Marcia Mathieu
Marion County Soil and Water Conservation District
Indianapolis, Indiana
Greg Lindsey
Center for Urban Policy and the Environment
School of Public and Environmental Affairs
Indiana University
Indianapolis, Indiana
Using Constructed Wetlands to Reduce Nonpoint Source Pollution in Urban Areas 303
Jon Harbor
Department of Earth and Atmospheric Sciences
Purdue University
West Lafayette, Indiana
Susan Tatalovich
FMSM Engineers, Inc.
Columbus, Ohio
Ron Turco
Department of Agronomy
Purdue University
West Lafayette, Indiana
Zac Reicher
Department of Agronomy
Purdue University
West Lafayette, Indiana
Anne Spacie
Department of Forestry and Natural Resources
Purdue University
West Lafayette, Indiana
Vickie Poole
Department of Forestry and Natural Resources
Purdue University
West Lafayette, Indiana
Advanced Identification (ADID) Techniques Used to Protect Wetlands and Aquatic Resources in a
Rapidly Growing County 314
Dennis W. Dreher
Northeastern Illinois Planning Commission
Chicago, Illinois
Local Government Involvement in Mitigation Banking 322
Lisa T. Morales
Wetlands Division
U.S. Environmental Protection Agency
Washington, D.C.
Massachusetts Stormwater Management Policy/Regulations: Development, Implementation, and Refinement . 332
Bethany Eisenberg
Vanasse Hangen Brustlin, Inc.
Watertown, Massachusetts
-------
Implementation of Michigan's Voluntary Stormwater Permit — A Community Perspective 339
Kelly A. Cave
P.E. Director
Watershed Management Division
Wayne County Department of the Environment
Detroit, Michigan
Dale S. Bryson
Camp Dresser & McKee
Detroit Michigan
Kelly C. Kelly
P.E., Project Engineer
Canton Township, Michigan
Jack D. Bails
Vice-President
Public Sector Consultants
Lansing, Michigan
California's Model Urban Runoff Program (MURP): Urban Runoff Programs for Small Municipalities 349
Jennifer Hays
Assistant Engineer
Public Works Department
City of Monterey, California
Cy Oggins
Coastal Program Analyst III
California Coastal Commission
Central Coast District
Santa Cruz, California
New Stormwater Treatment BMPs: Determining Acceptability to Local Implementing Agencies 354
Gary R. Minton
Resource Planning Associates
Seattle, Washington
Paul Bucich
Pierce County Public Works
Tacoma, Washington
Mark Blosser
City of Olympia Public Works
Olympia, Washington
Bill Leif
Snohomish County Surface Water Management
Everett, Washington
Jim Lenhart
Stormwater Management, Inc.
Portland, Oregon
Joseph Simmler
Entrance Engineers
Bellevue, Washington
Steven True
Vortechnics
Federal Way, Washington
XII
-------
By Any Measure 362
Thomas R. Adams, P.E., Vaikko P. Allen, and Andrea Perley
Vottechnics, Inc.
Portland, Maine
Stormceptor Hydrology and Non-Point Source Pollution Removal Estimates 371
G. Bryant
Stormceptor Canada
Etobicoke, Ontario, Canada
R. Grant
New England Pipe
Wauregan, Connecticut
D. Weatherbe
Donald G. Weatherbe Associates
Mississauga, Ontario, Canada
V. Berg
Stormceptor Corporation
Rockville, Maryland
NPDES Phase II Cost Estimates , 383
Andrew J. Reese, P.E.
Ogden Environmental and Energy Services, Inc.
Nashville, Tennessee
The Stormwater Utility Concept in the Next Decade (Forget the Millenium) 397
Hector J. Cyre
President
Water Resource Associates, Inc.
Kirkland, Washington
Are Green Lots Worth More Than Brown Lots? An Economic Incentive for Erosion Control on
Residential Developments 404
Martha Herzog and Jon Harbor
Department of Earth and Atmospheric Sciences
Purdue University
West Lafayette, Indiana
Keith McClintock
Geauga Soil and Water Conservation District
Burton, Ohio
John Law
Indiana Department of Natural Resources and
Saint Joseph Soil and Water Conservation District
South Bend, Indiana
Kara Bennett
Department of Statistics
Purdue University
West Lafayette, Indiana
XIII
-------
Acknowledgments
The success of the conference and the preparation of this document are due largely to the efforts of the
presenters as well as the following individuals:
Conference Planning Committee
Robert J. Kirschner, Conference Coordinator
Chicago Botanic Garden, Glencoe, IL
Thomas E. Davenport, Project Co-Officer
U.S. Environmental Protection Agency, Region 5, Chicago, IL
John A. Kosco, Project Co-Officer
U.S. Environmental Protection Agency, Washington, DC
Deborah S. Caraco, Center for Watershed Protection, Ellicott City, MD
Martin H. Kelly, Southwest Florida Water Management District, Brooksville, FL
Thomas E. Mumley, California Regional Water Quality Control Board, Oakland, CA
Eric W. Strecker, URS Greiner Woodward Clyde, Portland, OR
Peer Reviewers
Donald Brown, U.S. Environmental Protection Agency, Cincinnati, OH
Deborah S. Caraco and Staff, Center for Watershed Protection, Ellicott City, MD
Alan Everson, U.S. Environmental Protection Agency, Cincinnati, OH
Douglas Grosse, U.S. Environmental Protection Agency, Cincinnati, OH
Larry Shepard, U.S. Environmental Protection Agency, Region 7, Kansas City, KS
Steve Walker, Nebraska Department of Natural Resources, Lincoln, NE
Ruth Wallace and Staff, Missouri Department of Natural Resources, Jefferson City, MO
Jennifer Zielinski and Staff, Center for Watershed Protection, Ellicott City, MD
Technical Editing and Publishing
Scott Minamyer, Jean Dye, and Steve Wilson of the U.S. Environmental Protection Agency, Cincinnati, OH.
XIV
-------
Jordan Cove Urban Watershed National Monitoring Project
Melville P. Cote, Jr.
U.S. Environmental Protection Agency
Office of Ecosystem Protection
Boston, Massachusetts
02114
Dr. John Clausen
University of Connecticut
Bruce Morton
Aqua Solutions, LLC
Paul Stacey and Stan Zaremba
Connecticut Department of Environmental Protection
Introduction
Stormwater runoff from urban and urbanizing areas is widely recognized as a major cause of water pollution in the
United States. The impacts of stormwater runoff are threefold: (1) chemically, contaminants deposited on the land are
carried by runoff and infiltration to surface and groundwater; (2) physically, increases in impervious surfaces raise runoff
rates which, in turn, increase mass pollutant loadings and contribute to erosion and sedimentation; and (3) biologically,
the combined chemical and physical alterations of watershed systems degrade aquatic habitat. Research over the past
20 years consistently shows a strong correlation between the imperviousness of a drainage basin and the health of its
receiving waters, with stream health decreasing with increasing impervious coverage of the watershed.' The U.S.
Environmental Protection Agency cites urban runoff as the second leading cause of impairment to estuaries and the fourth
leading cause of impairment to lakes.* Increased runoff rates, and the erosion and sedimentation associated with new
development and construction, also are significant sources of pollution. In the United States, there are an estimated
522,000 construction "starts" each year, with construction activities disturbing an estimated 5 million acres of land
annually.3
Connecticut communities, like those in many urbanized states, are confronted with meeting nonpoint source
management needs that often conflict with traditional subdivision regulations and construction standards. The challenge
of meeting publicsafetyand maintenance requirements in an environmentally sensitive manner is not currently being met,
as evidenced by continued water quality impairments associated with new development. Can impervious surfaces be
reduced, and curbing and storm drains be eliminated in a way that will not raise objections from municipal boards and
commissions? Will homeowners accept cluster housing, natural landscaping, and "greener" home and yard maintenance
practices? Most important, will those modifications make a difference in the quality and quantity of nonpoint source runoff
under widespread application? Answering these and related questions is the objective of the Jordan Cove Urban
Watershed National Monitoring Project.
Project Overview
The primary purpose of the Jordan Cove project is to compare differences in runoff quantity and quality emanating from
traditional and "environmentally sensitive" development sites. The 18-acre "Glen Brook Green" subdivision, located in
the southeastern Connecticut town of Waterford, is being constructed and monitored to make this comparison. The
subdivision is split into two distinct "neighborhoods": one with building lots arranged in a traditional R-20 (half-acre)
-------
zoning pattern (Figure 1); the other, cluster housing with a variety of best management practices (BMPs) incorporated
into the design (Figure 2).
Figure 1. Glen Brook Green "Traditional" Neighborhood.
Figure 2. Glen Brook Green "BMP" Neighborhood.
-------
Stormwater runoff from the traditional section is collected by curbs and catch basins, then piped through astormwater
treatment system before entering Nevins Brook, a tributary of Jordan Brook and, ultimately, Jordan Cove and Long Island
Sound. Homeowners will not be subjected to any enhanced environmental education, or restrictions on how they manage
their properties.
The BMP neighborhood will feature grass swales; roof leader "rain gardens;" shared, permeable driveways; small
building "foot-prints;"deed restrictions on increasing impervious surfaces; "low-mow,""no-mow,"andconservationzones;
a narrower, permeable road surface (interlocking concrete pavement); and a vegetated infiltration basin, or bioretention
area, located inside a "tear-drop" cul de sac. Several different driveway surfaces will be utilized, including interlocking
concrete pavement, gravel, concrete tire strips, and permeable asphalt, and monitored for their relative runoff rates.
Homeowners and town road maintenance crews will be encouraged to adopt pollution prevention techniques, including
controlled fertilizer and pesticide application, pet waste management, street sweeping/vacuuming, and reduced use of
deicing agents.
The BMP neighborhood is expected to generate less stormwater runoff and pollution. Monitoring conducted before,
during and after construction will document actual results. The Jordan Cove project team comprises a true public/private
partnership, with researchers and educators from the University of Connecticut; federal, state, and local government
officials; private consulting firms; and the developer.
National Monitoring Program
The Jordan Cove Urban Watershed National Monitoring Project is funded, in part, through the Connecticut
Department of Environmental Protection (CT DEP) by the U.S. Environmental Protection Agency's (EPA) Section 319
National Monitoring Program (NMP). It is one of 22 such projects nationwide. The Jordan Cove project is the only NMP
project studying the effects of residential subdivision development on runoff quality and quantity, and of BMPs designed
to mitigate those impacts.
The Section 319 NMP was established pursuant to section 319(1) of the federal Clean Water Act (Nonpoint Source
Management Programs - Collection of Information). Section 319(1) states that EPA shall collect information and make
available:
(1) Information concerning the costs and relative efficiencies of best management practices for reducing nonpoint
source pollution.
(2) Data concerning the relationship between water quality and implementation of various management practices
to control nonpoint sources of pollution.
The objectives of the Section 319 NMP are twofold:
(1) To scientifically evaluate the effectiveness of watershed technologies designed to control nonpoint source
pollution.
(2) To improve our understanding of nonpoint source pollution.
To achieve these objectives, the NMP has selected watersheds across the country to be monitored over a 6-to 10-
year period to evaluate how improved land management and the application of BMPs reduce water pollution. The results
from these projects will be used to assist land use and natural resource managers by providing information on the relative
effectiveness of BMPs to control nonpoint source pollution.
Site Selection
In 1993, nonpoint source program staff from EPA and CT DEP, and a University of Connecticut researcher began
efforts to identify a site at which to conduct a nonpoint source monitoring project under the auspices of the NMP. Initial
-------
site selection involved three criteria: (1) an appropriate hydrologic setting, with distinct drainage patterns amenable to
monitoring; (2) a willing land owner or developer who would allow 1-1 1/a years of advance monitoring before beginning
construction; and (3) a municipality willing to adopt innovative site planning and development strategies. Proximity to the
coast was also considered as an important factor because of the need to reduce nonpoint source pollution loads to Long
Island Sound and coastal waters in general.
CT DEP mailed letters soliciting interest to a number of municipalities recognized for either their progressive approach
to land use planning and management, orforexperiencing high development rates. After positive responses from several
municipalities, and numerous field visits, the "Glen Brook Green" site in Waterford was selected in May 1995. The 18-
acre parcel was an active chicken farm, but its owner, who had grown up on the farm, was planning to develop it into a
residential subdivision. The property owner wanted to develop the parcel in an environmentally-sound manner, was
interested in the NMP solicitation, and was willing to be flexible with his construction schedule to facilitate monitoring.
The hydrology of the parcel featured two distinct drainage areas, an ideal setting for the proposed monitoring design.
Poultry houses and several other buildings occupied the area that would become the traditional neighborhood and an old,
partially mined gravel pit dominated the future BMP neighborhood. Soil tests determined that the chicken manure had
not elevated nutrient levels significantly enough to bias the monitoring. The town of Waterford, and its planning officials,
had a reputation as being progressive on land use issues and had served as one of the pilot communities for the
University of Connecticut Cooperative Extension System's Nonpoint Education for Municipal Officials (NEMO) project.
Because waivers from Water-ford's subdivision regulations would be needed to build the BMP neighborhood, the town's
cooperation was critical to the project's implementation.
Planning
Proceeding from a conceptual design to actual construction required a concentrated effort by the project team working
together toward a common goal. Once an acceptable plan was agreed upon by the project team and committed to paper,
the next step was gaining approval from Water-ford's conservation, and planning and zoning commissions. As is typical
of New England town governments, both commissions paid close attention to planning decisions at a series of public
meetings at which many development alternatives were reviewed. Volunteer commissioners and professional staff raised
numerous concerns regarding the health, safety and general welfare of the town residents, and the social economic,
environmental, and political viability of the proposed plan. Among their concerns were road widths for emergency access,
road surface integrity for plowing and de-icing, traffic, drainage, sidewalks, parking, maintenance of common areas, and
responsibility should BMPs fail. The rigorous review was enlightening to the project team and commissioners alike. As
the ongoing dialogue between the various parties led to further planning details and innovative solutions to problems,
enthusiasm and support for the project grew.
After a series of public meetings in late 1996 and early 1997, the project was approved by both commissions.
Technical modifications of existing standards were handled in four ways: as waivers, special design/operation controls,
mitigation, or discretionary actions. Table 1 lists each of these categories with associated comments and concerns
expressed by Water-ford's professional staff and commissions. In the end, it was the willingness of all parties involved
to work in concert, reaching compromises, that allowed this innovative project to advance to the construction phase.
It is a generally accepted axiom that resource-based site planning can help minimize increases in runoff and reduce
the potential for erosion and sedimentation problems typically associated with new development. In this project, goals
identified at the outset are helping to direct the choice of practices and strategies for site development toward those that
will reduce adverse impacts on hydrology and water quality. These goals include: (1) reproducing pre-development
hydrological conditions; (2) confining development and construction activities to the least critical areas; (3) fitting the
development to the terrain; (4) preserving and utilizing the natural drainage system; and (5) creating a desirable living
environment.
-------
Table 1. Technical Modifications of Existing Development Standards.
Considerations
waivers needed
special design/
operational control
mitigation required
discretionary
actions
Traditional Design
specified road surface materials
typical road width = 28 feet,
reduced to 24 feet
curbs and storm drains required
90 ft paved cul-de-sac radius
planning and zoning standards
home owner discretion
home owner discretion
home owner discretion
road runoff piped to storm sewer
creation of 13,400 sq ft wetland at
subdivision entrance
R-20 single-family zoning
open space not contiguous with
all lots
a driveway for each home
BMP/Cluster Design
segmental concrete pavers
(permeable)
reduced road width to 20 feet
for travel lane
no curbs; grassed swales
and sheet flow off road
one way cul de sac design to
reduce road width and
turning radius
bioretention "rain gardens"
vegetative maintenance
pesticide management
domestic animal
management
cluster and zero setback
from lot lines
open space layout
combined driveways
Comments
must be approved by public
works; costs more
must be approved by public
works, fire, and police
turf stone installed to
maintain road edge
integrity; costs less
further reduction in width
and less need for snow
plowing
retains roof runoff on-site
reduces fertilizer use; costs
less
reduces pesticide use:
costs less
reduces pathogen runoff
need to manage storm
water entering the site from
adjacent public road
required to mitigate filling
of 5000 sq ft of wetlands
within subdivision
allows more open space
and natural landscaping
compact housing; natural
landscaping
reduces curb cuts and
impervious surface; cost
less
Monitoring Design
This study is utilizing the "paired-watershed" monitoring design, which requires a minimum of two watersheds (control
and treatment) and two periods of study (calibration and treatment). This approach assumes that there is a quantifiable
relationship between paired water quality data for the two watersheds, and that this relationship is valid until a major
change is made in one of the watersheds. It does not require that the quality and quantity of runoff be statistically the
same for the two watersheds, but that the relationship between the paired observations of water quality and quantity
remains the same over time - except for the influence of the land use changes in the treatment watershed.4
The control watershed accounts for annual and/or seasonal climate variations. During the calibration period, no
changes in land use occur in the watersheds and paired water quality and quantity data are collected to develop a
baseline. The paired data are used to develop regressions for the control and treatment watersheds. The treatment
period begins when changes in land use occur in the treatment watershed. A new regression is developed following the
-------
treatment period. Analysis of variance (ANOVA) is used to test the significance of the regressions in each period.
Analysis of covariance (ANCOVA) is used to test the differences between the two regression slopes and intercepts. The
changes between periods are calculated based on a comparison of predicted values, using the calibration regression
equation, and observed values during the treatment period.5
For the Jordan Cove project, the treatment period will occur in two phases: (1) during construction of the traditional
and BMP neighborhoods; and (2) after construction when the BMPs are in effect. The paired-watershed approach is
being used to measure the differences in water quality and quantity between the treatment areas (traditional and BMP
neighborhoods) and the control area (a nearby 1 0-year old subdivision) caused by construction in the two treatment areas
and the application of BMPs in the BMP neighborhood. Stormwater quality and quantity are measured at the outlets of
each of the two treatment neighborhoods, and the control watershed (Figure 3). Water quality is measured by analyzing
weekly flow-weighted composite samples for total suspended solids (TSS), total phosphorus (TP), total Kjeldahl nitrogen
(TKN), ammonia nitrogen (NH,-N), and nitrate+nitrite nitrogen (NO,-N). Grab samples are analyzed for fecal coliform and
BOD,. Monthly analyses are conducted for copper, lead, and zinc.
Figure 3. Existing residential (control) watershed.
The calibration period began in January 1996, to establish a baseline for future comparisons. Since the treatment
period began in May 1998, runoff monitoring has focused on the effects of construction, and on the relative effectiveness
of standard erosion and sediment control practices in the traditional neighborhood. When construction commences in
the BMP neighborhood, the focus will be on the effects of construction and the relative effectiveness of enhanced erosion
and sediment control practices (e.g., phased grading, stockpile seeding, open space vegetation, cross grading, and
detention swales). Post-construction monitoring is scheduled to begin in 2001 and will continue for 3-5 years.
Supplemental monitoring will be conducted on selected BMPs, including different driveway surfaces and enhanced
turf management in the BMP neighborhood, and a "state-of-the-art" stormwater treatment device in the traditional
neighborhood. This information will be used to evaluate the effectiveness of these specific practices.
-------
Monitoring Results
During the calibration period, 75 runoff events were sampled for the control watershed and 12 runoff events for the
two treatment watersheds, in the treatment period to date, 21 and 20 events were sampled for each treatment watershed,
respectively. Peak discharge values were obtained for nine paired events in the calibration period and 20 pairs for the
treatment period. The total number of samples analyzed was less than the total number of flow observations because
not all the samples contained a sufficient volume for analysis6.
Sampling results to date, as presented in Table 2, indicate that construction of the traditional neighborhood is causing
significant impacts on runoff quality and quantity, including observed increases in mean weekly flow volume (99%), runoff
frequency (from 16% to 95%), and mean weekly peak discharge (79%).7 The conversion of the watershed's topography
from a "knoll" to a "bowl," combined with an increase in impervious surface, appears to have caused a significant change
in hydrologic responses. Concentrations of NO,-N and Pb in runoff also increased. However, increases in the
concentrations of sediment and sediment-associated nutrients, typical of construction sites, did not occur. In fact, TKN
concentrations have declined during construction. It is believed that erosion and sediment controls are responsible for
TSS concentrations remaining constant before and during construction*.
Table 2. Summary of means and percent increases of flow, Qp, nutrient and metal concentrations for the control and traditional watershed in the
calibration and treatment periods.
Calibration Period
Parameter
Control
Traditional
Treatment Period
Control
Traditional
Observed
fms/umak\
Flow
113.84
0.14
107.76
1.94
Qp
0.05
3.00E-04
0.04
1.00E-03
„_ fmn/l ^
i"ig"-;
TSS
NO3-N
NH3
TKN
TP
31
0.5
0.15
1.3
0.159
132
0.3
0.08
4.0
1.009
28
0.4
0.31
1.8
0.127
106
0.8
0.18
2.1
0.481
/nr,/! \
^g/i_;
Cu
Pb
Zn
8
6
58
8
11
65
14
6
79
21
17
126
Predicted
% Change
0.02
99"*
3.00E-04
79***
121
0.3
0.17
4.5
0.758
-15
62"
2
-113"
-58
13
10
98
38
42*
22
* P value < 0.05
" P value < 0.01
. ** P value < 0.001
Coinciding with the increases in pollutant concentration and flow, the mass export of NO.-N and Pb increased as well,
as did the mass exports of TP, TSS, Cu, and Zn. These increases appear to be attributable to increased stormwater
7
-------
runoff volumes. The preliminary results from this study suggest that increased runoff, rather than erosion, is the cause
of increased pollutant export from this construction site. Traditionally, erosion and sediment controls and stormwater
management plans focus on the prevention of sediment and, occasionally, peakflow impacts on downstream areas. The
preservation of pre-development hydrologic conditions within the watershed where construction is occurring is typically
ignored.
Excess runoff, which is the driving force behind nonpoint source pollution, will transport pollutants into waterways and
contribute to their degradation. Preliminary monitoring results demonstrate that erosion and sediment controls can reduce
sediment and sediment-associated pollutants in construction site runoff. However, current erosion and sediment control
practices do not address the increase in runoff from development sites. Consequently, these practices fail at reducing
pollutant loads.9
Next Steps
By the end of 2000, this combination of traditional and "green" designs for residential subdivisions should be fully
constructed. Monitoring of stormwater quality and quantity will be conducted for several years after build-out to determine
the overall efficiency of the design. It should demonstrate that careful planning, landscaping, and use of vegetative BMPs
can help protect and enhance the environment, while addressing other concerns that local planning and zoning
commissions face. Lessons learned from this project have already been, and will continue to be, passed along to other
communities through ongoing technical assistance and training programs administered by the CT DEP, the University
of Connecticut Cooperative Extension System, and other agencies and organizations.
References
1. Arnold, C.L. and C.J. Gibbons. 1996. Impervious Surface Coverage: The Emergence of a Key Environmental
Indicator. American Planning Association Journal. 62:2. Chicago, IL.
2. USEPA. 1996. National Water Quality Inventory. Washington, D.C. 20460
3. US Bureau of the Census. 1996.1992 Census of Construction Industries. Manufacturing and Construction Division.
Washington, D.C. 20460
4. Clausen, J.C. and J. Spooner. 1993. Paired Watershed Study Design. United States Environmental Protection
Agency. USEPA 841 -F-93-009.
5. Engdahl, J. 1999. Impacts of Residential Construction on Water Quality and Quantity in Connecticut. University of
Connecticut. Storrs, CT.
6. Ibid.
7. Ibid.
8. Ibid.
9. Ibid.
-------
Sources of Phosphorus in Stormwater and Street Dirt from Two Urban
Residential Basins in Madison, Wisconsin, 1994-95
R. J. Waschbusch and W. R. Selbig
U.S. Geological Survey
Middleton, Wl
R.T. Bannerman
Wisconsin Department of Natural Resources
Madison, Wl
Abstract
Eutrophication is a common problem for lakes in agricultural and urban areas, such as Lakes Wingra and Mendota
in Madison, Wisconsin. This report describes a study to estimate the sources of phosphorus, a major contributor to
eutrophication, to Lakes Wingra and Mendota from two small urban residential drainage basins. The Monroe Basin
empties into Lake Wingra, and the Harper Basin into Lake Mendota. Phosphorus data were collected from streets, lawns,
roofs, driveways, and parking lots (source areas) within these two basins and were used to estimate loads from each
area. In addition to the samples collected from these source areas, flow-composite samples were collected at monitoring
stations located at the watershed outfalls (storm sewers); discharge and rainfall also were measured. Resulting data were
then used to calibrate the Source Loading and Management Model (SLAMM, version 6.3, copyright 1993, Pitt & Vorhees)
for conditions in the city of Madison and determine within these basins which of the source areas are contributing the most
phosphorus.
Water volumes in the calibrated model were calculated to within 23% and 24% of those measured at the outfalls of
each of the basins. These water volumes were applied to the suspended-solids and phosphorus concentrations that were
used to calibrate SLAMM for suspended-solids and phosphorus loads. Suspended-solids loads were calculated to be
within 4% and 17%, total-phosphorus loads within 24% and 28%, and dissolved- phosphorus loads within 9% and 10%
of those measured at the storm-sewer outfall at the Monroe and Harper basins, respectively.
Lawns and streets are the largest sources of total and dissolved phosphorus in the basins. Their combined
contribution was approximately 80%, with lawns contributing more than the streets. Streets were the largest source of
suspended solids.
Street-dirt samples were collected using industrial vacuum equipment. Leaves in these samples were separated out
and the remaining sediment was sieved into >250 um, 250-63 urn, 63-25 um, <25 urn size fractions and were analyzed
for total phosphorus. Approximately 75% of the sediment mass resides in the >250 um size fractions. Less than 5% of
the mass can be found in the particle sizes less than 63 um. The >250 um size fraction also contributed nearly 50% of
the total-phosphorus mass. The leaf fraction contributed an additional 30%. In each particle size, approximately 25% of
the total-phosphorus mass is derived from leaves or other vegetation.
Introduction
Eutrophication is a common problem for lakes in agricultural and urban areas, such as Lakes Wingra and Mendota
in Madison, Wl. Primary productivity in northern temperate lakes is most often limited by phosphorus (Schindler 1974;
1977). Urban runoff has been noted to contain high phosphorus concentrations (U.S. Environmental Protection Agency,
1983) that may be increasing the eutrophication. The focus of the study described in this report was to estimate the
sources of phosphorus to Lakes Wingraand Mendotafrom two small urban residential watersheds in Madison, Wl (Figure
1). This study was done in cooperation with the city of Madison and the Wisconsin Department of Natural Resources.
-------
EXPLANATION
(1) Monroe Basin
(2; Harper Basin
CJ Lakeland Basin Outfall
Madison /
Figure 1. Lakes Wingra and Mendota in Madison, Wl.
Lake Mendota and Lake Wingra are both part of the Wisconsin Department of Natural Resources (WDNR) Priority
Watershed Program (Betz and others, 1997). State funding is available to help pay for management aimed at reducing
the amounts of phosphorus and other pollutants discharged to the lakes. The goal of the Lake Mendota Priority
Watershed Project is to reduce the frequency of nuisance algae blooms in the lake from one out of every two days to one
out of every five days. To accomplish this goal, it is estimated that a 50% reduction is needed in the amount of
phosphorus entering the lake. To help reach this target, the Nonpoint Source Control Plan for the Lake Mendota Priority
Watershed Project (Betz and others, 1997) set an objective of reducing phosphorus loading to the lake by20% from urban
areas. The remaining 30% reduction is intended to come from rural phosphorus management.
For this study, phosphorus data were collected from five source areas-streets, lawns, roofs, driveways, and parking
lots-within the two drainage basins from urban residential and commercial areas to estimate loads from each source
area (Table 1). Resulting data were used to calibrate the Source Loading and Management Model (SLAMM, version 6.3,
copyright 1993, Pitt & Vorhees) for conditions in the city of Madison and determine which source areas are contributing
the most phosphorus within these basins. The city is planning to use SLAMM to target specific source areas for
management efforts to meet the 20% phosphorus-reduction objective of the priority watershed project and to meet
requirements of its Wisconsin Pollutant Discharge Elimination System stormwater permit.
10
-------
Table 1. Land-use Characteristics of the Monroe Basin, 1994, and Harper Basin, 1995, in Madison, Wl.
[--, source not present]
Monroe Basin Harper Basin
Source area Residential Commercial Residential
Acreage Percent of basin Acreage Percent of basin Acreage Percent of basin
Lawn 119.8 51.5 .. .. 23.6 57.4
Roof 26.5 11.4 2.3 1.0 5.4 13.2
Street 30.5 13.1 1.6 .7 5.3 13.0
Woodlot -- -- .. .. 3.0 7.3
Driveway 10.6 4.6 -- -- 2.1 5.1
Parkland 19.3 8.3 - .. .7 1.7
Sidewalk 12.5 5.4 - .. .7 1.7
Parking lot .4 .2 3.4 1.5 .3 .7
Railroad bed 5.3 2.3
Total 224.9 96.8 7.3 3.2 41.1 100
Stormwater runoff samples from source areas and the basin outfall were collected from a medium-density residential
watershed draining to Lake Wingra from May to November, 1994 and from a medium-density residential watershed
draining to Lake Mendota from June to November 1995. These runoff samples were used to estimate the phosphorus
and suspended-solids load that each of these source areas and basins contributes. In addition, a third basin, the Lakeland
Basin that drains into Lake Monona, was monitored for lawn runoff in 1995. This basin, which encompasses an older
section of Madison, was sampled in an attempt to determine whether any difference exists between this section of the
city (which has older, smaller lawns) and other areas of the city.
Study-Area Description
The Monroe Basin, monitored during 1994, drains into Lake Wingra. The basin is 232.2 acres, of which 224.9 acres
are residential and 7.3 acres are commercial (Figure 2). Lake Wingra has a surface area of 338.9 acres (1.37 km2) and
a drainage area of 3,889 acres (15.74 km*). About 75% of the Lake Wingra drainage basin is urbanized and about 25%
is composed of forest, prairie, and marsh within the University of Wisconsin Arboretum (Oakes and others 1975).
The Harper Basin, monitored during 1995, drains into Lake Mendota. The Harper Basin is 41 .1 acres, all of which
is residential land use (Figure 2). Lake Mendota has a surface area of 9,859 acres (39.9 km2) and has a drainage area
of 138,823 acres (561.8 km2) (Lathrop and others, 1992). Approximately 20% of the Lake Mendota drainage area is
urban, 57% is agricultural, and the remaining 23% is grassland/woodland/marsh/open-water area (Betz and others, 1997).
The Lakeland Basin is approximately 3 miles southeast of the Harper Basin.
In addition to the samples collected from source areas, a monitoring station was located at each basin storm-sewer
outfall to collect flow-composite samples and to monitor discharge and rainfall. The total rainfall amounts for the months
of June through October 1994 at Monroe and July through October 1995 at Harper were 9.24 and 10.67 in., respectively.
These amounts are 64% and 71% of the average from 1961 to 1990 (Brian Hahn, National Weather Service, oral
communication, 1997).
Acknowledgments
We thank all of the volunteers that allowed us to install our sampling equipment in their yards, the City of Madison
Engineering Division and Department of Public Health for their efforts in making this study possible, Jeff Beck of the
USGS for his exceptional field efforts, and Holly Ray and Dr. Bob Pitt at the University of Alabama at Birmingham for
their work analyzing the street-dirt samples. Also, we thank Mary Anne Lowndes of the WDNR and
11
-------
Troy Or
t>.05 0.10 MtlE
r
0.10 KILOMETER
EOTANA
j Residential roof-
Commercial roof
Lawn-
1 Park-
1 fctfdlot-
Sidewalk
Driveway
Parking lot
Railroad'
High-traffic road
Medium-traffic road
Low-traffic road-
with grass ditch
Low-traffic road-
without curb and gutter-
f llJij Low-traffic road-
with curb and gutter
0 0.1 0.2 KILOMETER
Figure 2. Harper and Monroe Basins.
Huron Hill
12
-------
Steve Corsi of the USGS for insightful comments that have greatly improved the report. Lastly, we thank Gail Moede and
Aaron Konkol at the USGS for their help with the report editing and preparation of illustrations.
Data-Collection Equipment and Methods
Runoff samples were collected from each source area by use of sampling equipment slightly modified from that
described by Bannerman and others (1993). Brief descriptions of the sampling equipment follow.
Street samplers. The street samplers were grouted into the street approximately 5 ft from the curb (Figure 3). The
sample bottle was covered with a 6-in. concave polycarbonate cap, set flush with the street surface, with a center drain
hole. The bottle and cap were placed into a 6-in. diameter polyvinyl chloride (PVC) sleeve. Water flowed over the top of
the cap and drained through the center hole into a collection bottle. The drain hole could be constricted by a set screw
that controlled the flow rate into the sample bottle.
3/8-inch thread
for removing cap
Concave •
polycarbonate1
cap
(PVC) coupling
Polycarbonate set •
screw used to adjust1
size of drain hole
1.5-liter sample bottle
&inch-diameter
PVC sleeve
Buick-set grout
Undisturbed asphalt
Asphalt subgrade •
material
Figure 3. Schematic of street samplers.
Driveway samplers. Runoff water from driveways was diverted into a sampler by means of a flat piece of clear
plastic, 1/4 in. high by 1 in. wide by 3 ft long, glued to the surface of the driveway. The sampler consisted of a 1.5-L glass
bottle placed in a 10-in.-diameter protective PVC sleeve set into the ground next to the driveway. A 1/2-in.-diametersilicon
tube carried the runoff through a plastic cap covering the PVC sleeve and into the sampler. During the 1994 field season,
the tubing emptied directly into the sample bottle, causing several sample bottles to overfill. To alleviate this problem, in
1995, the tube emptied onto a polycarbonate cap like those used with the street samplers, so that the volume of water
entering the sample bottle could be controlled.
Lawn samplers. Lawn sample bottles received runoff through two 5-ft pieces of 1/2-in.-diameter PVC pipe placed
flush with the surface of the ground, on a sloping surface, with an angle of about 150 degrees between the two pipes
(Figure 4). Runoff entered the pipes through two slits cut the entire length of pipe. Each pipe was wrapped with fiberglass
13
-------
1/2-inch
WC cap
Wnch-thick 4-1/3-inch
(diameter gray PVC cap
with V-notch cut out to
accommodate the
tygon tube.
Direction of water flow
1/2-inch PVC tubing TOD Vi 6W
wrapped in nylon
screening
Two rows of 2.5-mcn
slots plunge cut
with a table saw to
allow the water to
enter tube.
Tygon tubing
stretched over
the 1/2-inch
PVC tube to
carry water to
the bottle.
4-inch Schedule
40 PVC
iSide Vievx
Figure 4. Schematic of lawn samplers.
screen to prevent insects and large debris from entering. Wooden clothespins with small pieces of nylon rope held the
pipes in place. Water from the pipes flowed into a sampler through a notched cap. The sampler was a 1- qt glass bottle
placed in a 4-in.-diameter protective PVC sleeve. The cap had a notch to accommodate silicon tubing, which ran from
the end of the PVC collector pipes to the sample bottle.
Roof samplers. Roof samplers were designed to divert a small portion of the water in the gutter downspout to a
sample bottle. A 1/4-in.-diameter vinyl tube was attached to the inside of the downspout by means of plastic wire ties.
Each tube went into a 1.5-L glass sample bottle that was placed in a covered 10-in.-diameter protective PVC sleeve.
Because of problems with overfilled sample bottles, the design was changed in the same manner as the driveway
samplers so that the volume of water entering the sample bottle could be controlled.
Parking Lot sampler. The parking lot sampler collected runoff entering a storm-sewer inlet grate. A small portion of
the inlet flow was diverted to a sample bottle by means of a 6-in. trough made of 1/2-in.-diameter PVC pipe cut length-
wise and held in place with stainless steel hose clamps attached to the inlet grating. Water drained from the trough
through a tube to a 2.5-gal glass sample bottle hanging from the inlet grate. No samples were collected from parking lots
during 1995.
14
-------
Basin storm-sewer outfall samplers. An automated sampling station was placed at the storm-sewer outfall of both
the Monroe and Harper Basins. In 1994, water level in the basin storm- sewer outfall pipe was measured with a pressure
transducer as water drained into a detention pond. Velocity was measured with an electromagnetic velocity meter. In
1995, stormwater-runoff volumes were computed using a modified Palmer-Bowlus flume design (Kilpatrick and others,
1985). The water level was measured one pipe diameter (36 in.) upstream from the entrance to the flume using a
pressure transducer connected to a nitrogen bubble system. This water level was used in the following equation to
calculate the total discharge through the flume:
Q = a[Ha /D] "D2 5,
where
Q is discharge, in cubic feet per second,
a is a constant, 3.685,
b is a constant, 1.868,
Ha is the water level above the upstream lip
of the flume, at a distance of one pipe
diameter upstream from the flume
entrance, in feet, and
D is pipe diameter, in feet.
Flow-composite water quality samples were collected using programmable, refrigerated automaticsamplerswith 3/8-
in.-diameter Teflon-lined sample tubing. Rainfall was measured using a tipping-bucket rain gage and was recorded by
a digital data logger.
Stormwater Sample Collection and Processing Protocols
Sample bottles were placed in the source-area samplers as close to the start of each rain event as possible. As the
bottles were being deployed, the sampling equipment was rinsed with deionized water to remove any accumulated surface
dirt. Before the lawn sampler pipes were rinsed, they were cleaned with a small test tube brush. As soon as possible after
runoff had stopped, the sample bottles were collected and the approximate volume of water in each bottle was recorded.
All the bottles from a given source area were composited by pouring water from each bottle into a 5-gal or l-gal
stainless steel, Teflon-coated churn splitter modified from the type described in Ward and Harr (1990). The City of
Madison Department of Public Health Laboratory analyzed a small subsample taken from the churn for suspended solids
and phosphorus.
Street Dirt Collection and Analysis
In addition to stormwater-runoff samples, samples of street dirt were collected with a 9-gal wet-dry shop-vacuum using
a 6-in.-wide wand. A section of the street was vacuumed from curb to curb, 10 times across each of 3 streets in the basin,
similar to the technique described by Pitt (1979) and Bannerman (1983). Monroe Street, Glenway Street, and Seneca
Place/Spring Trail/Huron Hill (the latter three are considered one residential street) were sampled during 1994 (Figure
2). Woodward Drive, Nova Way, and Luster Avenue were sampled during 1995. Woodward Drive did not have curbs, so
the sample was collected by vacuuming between wooden 4-by 4-in. blocks placed at the edge of the asphalt on each side
of the street. During the fall, leaves on the street would often plug the vacuum hose. To alleviate this problem, a 6-in. by
2-ft wooden frame was placed with the 6-in. side abutting the curb. Before vacuuming the inside of the frame, the leaves
inside it were collected by hand and placed in the vacuum collection bag. Then the street was vacuumed in the normal
manner.
The dirt samples were dried at 105°C, sieved into size fractions of >250 jjm, 250-63 pm, 63- 25 pm, and <25 urn and
weighed. The sieved samples were sent to the University of Alabama at Birmingham (UAB) for phosphorus analysis. In
addition to total phosphorus, samples collected from the Monroe Basin were analyzed for percentage of vegetative
material. Two independent methods were used to determine the percentage of vegetative material: thermal
15
-------
chromatography and microscopic examination. In thermal chromatography, dirt samples were placed in ovens at
increasing temperatures and the mass that was lost to incineration was determined after each increase in temperature.
The mass loss was compared to the standard temperatures where various substances like leaves, rubber, and paper
burned off. The sample mass lost at the temperature range corresponding to the leaf standards was assumed to be
vegetation. In microscopic examination, samples of the dirt were compared to microscopic pictures of vegetation. More
details of these methods can be found in Ray (1997).
Sources of Phosphorus
Measured Concentrations in Runoff from Source Areas
A total of 25 runoff events were monitored at each basin. Runoff samples were collected from May to November of
1994 at Monroe and from June to November of 1995 at Harper and Lakeland. Driveway samples collected from the
Monroe Basin were excluded because of problems with the sample bottle overfilling (discussed in the methods section).
Summary statistics are listed in Table 2.
Table 2. Concentrations for Suspended Solids, Total Phosphorus, and Dissolved Phosphorus at the Monroe Basin and Harper Basin, 1994-95
[--, concentrations were not used because of problems with the samplers; -, source area not present in basin; mg/L, milligrams per liter]
Statistic
Lawns
Geometric mean 59
Coeff. of variation .55
Mean 85
Median 75
Geometric mean 0.79
Coeff. of variation .62
Mean 1.03
Median .99
Geometric mean 0.37
Coeff. of variation .62
Mean .52
Median .61
Geometric mean 122
Coeff. of variation .37
Mean 132
Median 154
Geometric mean 1.61
Coeff. of variation 1.12
Mean 2.34
Median 1.54
Geometric mean 0.77
Coeff. of variation 1.51
Mean 1.54
Median .81
Feeder
Street
68
1.17
99
60
0.40
1.24
.75
.31
Collector
street
51
.97
67
46
0.22
1.23
.36
.16
Source area
Arterial Driveways
street
MONROE BASIN
Suspended solids (mg/L)
65
.92
83
64
Total phosphorus (mg/L)
0.18
1.15
.24 ~
.17 ~
Parking
lots
51
1.27
82
44
0.10
1.04
.14
.09
Pitched
roofs
15
.95
85
18
0.07
.76
.09
.06
Flat
roofs
18
1.21
35
20
0.13
.96
.2
.12
0.16
1.72
.40
.14
69
.68
98
88
0.24
.75
.31
.22
0.08
.98
.12
.08
0.05
1.47
.14
.04
Dissolved phosphorus (mg/L)
0.03
1.20
.05 '.'.
.03
HARPER BASIN
Suspended solids (mg/L)
34
.93
57
31
Total phosphorus (mg/L)
0.18
.80
.24
.20
Dissolved phosphorus (mg/L)
0.07
1.0
.11
.07
0.02
1.24
.04
.02
0.02
1.22
.03
.02
17
.96
25
17
0.15
.68
.19
.15
0.08
.83
.11
.07
0.02
1.24
.04
.02
The concentration data from the Monroe and Harper Basins seem to exhibit log-normal distributions that are
consistent with urban-runoff concentration datacollected during the Nationwide Urban Runoff Project (U.S. Environmental
Protection Agency, 1983). In such cases, the geometric mean is a better estimate of the central tendency than the
16
-------
arithmetic mean because the geometric mean gives less weight to extremes (Helsel, 1992). Several of the coefficients
of variation in Table 2 have a value greaterthan 1, indicating substantial variability in concentrations within a source area.
The large variation seen in the source-area concentration data could cast doubt on the predictability of the data. For
lawn runoff, the difference between geometric mean phosphorus concentrations from 1994 to 1995 was greater than a
factor of 2; however, the lawn-runoff data collected from the Lakeland Basin are remarkably similar to data collected at
Harper (Table 3), indicating that the variation in lawn-runoff phosphorus concentrations is not random. Primary sources
of phosphorus, such as tree canopy, also could have a large effect on the source-area concentrations measured between
basins. Figure 5 illustrates a trend between the concentration of phosphorus and the percentage of overhead tree canopy
on streets for the Monroe and Harper Basins. Canopy in the Monroe Basin tends to be less than 35%, whereas the
percentage of canopy in the Harper Basin ranges from 5 to 78%. Variation also could be caused by meteorological
factors like rain depth, intensity, or inter-event period or by seasonal variables.
Roof runoff had the lowest geometric mean concentrations of suspended solids, and lawn runoff had the highest total
and dissolved phosphorus concentrations in both the Monroe and Harper Basins (Figures 6 and 7). In addition, patterns
in geometric mean concentrations between source areas within each basin were similar; however, their magnitudes were
very different. The geometric mean concentration of phosphorus for low-traffic streets in the Monroe Basin were about
twice those at the Harper Basin. Conversely, geometric mean phosphorus concentrations for lawn and roof runoff in the
Harper Basin were more than twice as high as those in the Monroe Basin. The beginning of the sampling periods differed
by one month between basins (Monroe Basin in May and Harper Basin in June), and this difference may have caused
some of the variability.
Concentration results for suspended solids and phosphorus from earlier source-area studies in Madison, Wl,
Marquette, Ml, and Birmingham, AL (Bannerman and others, 1993; Steuer and others, 1997; Pitt and others, 1995), were
compared to the concentration results from this study. Suspended-solids concentrations in street-runoff samples collected
during the other studies were considerably higher than those in samples collected forthis study. Sandier soils are present
in Marquette that could partially account for this difference. Furthermore, some of the same lawns in the Monroe Basin
were monitored for phosphorus concentrations in the previous Madison study (Bannerman and others, 1993), and both
the dissolved and total phosphorus geometric means calculated for that study were more than three times higher than
those in 1994. Because phosphorus concentrations varied highly from Monroe and Harper Basins, did not closely agree
with each other, and did not agree well with previous studies, the geometric mean of the combined data collected at
Monroe and Harper Basins was used for the modeling phase of this study.
Tables. Rainfall Amountsand Intensities and Total-phosphorus Concentrations from Lawn-runoff Samples for Harperand Lakeland Basins, Madison, Wl,
1995
HARPER
LAKELAND
Start of rain Rainfall Intensity Total-phosphorus Rainfall Intensity Total-phosphorus
event
(date)
06/26/95
07/05/95
07/15/95
07/22/95
08/16/95
08/16/95
08/28/95
10/19/95
(inches)
0.26
.36
.50
.79
.61
.38
.80
.32
(in/hr)
0.12
.62
.12
.10
.43
.29
.19
.07
concentration
(mg/L)
10.72
1.32
3.61
1.08
1.82
.60
1.39
2.24
(inches)
0.31
.10
.80
.80
.55
.55
.67
.33
(inhr)
0.17
.16
.10
.09
.94
.49
.15
.06
concentration
(mg/L)
9.05
2.06
2.99
1.35
2.48
.58
1.58
2.59
17
-------
DISSOLVED PHOSPHORUS
0.35
0.9
0.6
0.7
TOTAL PHOSPHORUS
I 0.4
z
tn
o 0.3
i
o
£0.2
0
0
linear regression
R =0.94
10
20
30
40
50
60
70
80
Figure 5. Trend between concentration of phosphorous and percentage of overhead tree canopy on streets for Monroe and Harper Basin.
18
-------
Monroe Basin
DISSOLVED PHOSPHORUS
DISSOLVED PHOSPHORUS
3
2.5
2
1.5
1
0.5
0
I i I I I i I
1 SK 1
Lawns Low-Traffic* Med. -Traffic
0.25
£j 0-2
§ g
g £ 0.15
z 5
UJ <
ll 0.1
o z!
s
^ 0.05
0
1 1 1 1 i i i 1 1
-
-
-
1 1 1 1 1 1 1 1 1
High-Traffic Parking Pitched Fiat
Streets Streets
SOURCE AREA
Streets Iot5 Roofs Roofs
SOURCE AREA
TOTAL PHOSPHORUS
LU
^5
g|
UJ -^
§1
s
s
3.5
3
2.5
2
1.5
1
0.5
n
I I I I I I I
-
-
- -
r ~
-
^
-
Js
« %
'• I I I I I I I
TOTAL PHOSPHORUS
8
Lawns Low-Traffic. Med.-Traffic-
Streets Streets
SOURCE AREA
£^
LlJ
-V t
\ — UJ
z 2
Uj <"
O Q£
8d
*
1.0
1.4 ,
OS-
0.6
0.4
o :
i i i i i i i i i
*
i i i i i i i i i
High-Traffic Parking' Pitched Flat
Streets lots Roofs Roofs
SOURCE AREA
600
500
E£ 400
£ 300
s
| 200
,_j
1 100
0
SUSPENDED SOLIDS
I I I I i I i I I I I I I
-
_
-
« * a « «
i I I i I I I I I i I SS I % \
iauns Low-Traffic* Med-Traffic- High-Traffic Parking Pitched Flat
Streets Streets Streets Iot5 Roofs Roofs
EXPLANATION
Source-area
concentration
S3 Geometric mean
SOURCE AREA
Figure 6. Dissolved phosphorous concentrations in Monroe Basin.
19
-------
DISSOLVED PHOSPHORUS
9
*
Lawns
SOURCE ARM
Harper Basin
0.45
0.4
0.35
0.3
0.25
0.2
0.15
01
0.05
0
DISSOLVED PHOSPHORUS
!
Streets Driveways
SOURCE AREA
Roofs
TOTAL PHOSPHORUS
TOTAL PHOSPHORUS
z 2
UJ <
O fi/
12
10
I
Lawns
SOURCE AKEA
200
oi
UJ
:5 150
§1 i°°
" c£
o§
u d
1 50
2
z
1
0.9
0.6
0.7
0.6
0.5
0.4
0.3
0.2
0.1
SUSPENDED SOLIDS
!
Streets Driveways
SOURCE ARE^A
i
Lawns Streets Driveways
SOURCE AREA
Roofs
Roofs
EXPLANATION
Source-area •
concentration
Geometric mean
Figure 7. Dissolved phosphorous concentrations in Harper Basin
20
-------
Calibration of the Source Loading and Management Model (SLAMM)
A concentration data base to simulate stormwater quality and theoretical runoff coefficients to simulate runoff volumes
is used in the Source Loading and Management Model (SLAMM). Because large amounts of concentration data and
runoff information were collected during this study, it was an opportunity to calibrate the model's concentration data base
and improve the runoff coefficients with data collected from the Monroe and Harper Basins.
Calibrating SLAMM with concentration and water-volume data was a three-step process. First, the runoff volume
generated by each source area was calibrated (a critical step because an accurate water volume is essential for
estimating all pollutant loads). Second, sediment was calibrated because sediment concentrations and loads are used
in SLAMM to estimate phosphorus loads. The final step was to calibrate the model for phosphorus concentrations.
A systematic procedure was used to calibrate suspended-solids and phosphorus concentrations in SLAMM. First,
a mass-balance approach compared total measured loads from source areas summed over 25 events to the loads
measured at the outfall. Monitored loads from source areas were calculated using SLAMM-generated water volumes.
Individual source areas were not equipped to measure runoff volumes during an event. Therefore, the accuracy of source-
area volumes, as assigned by SLAMM, was subject to agreement with the actual volumes measured at the outfall. If the
sum of all source-area volumes closely matched what was measured at the outfall, the individual source-area volumes
assigned by SLAMM were assumed to be correct. Second, source-area concentrations were adjusted to optimize the
mass balance. SLAMM was adjusted after agreement between measured source-area and outfall loads was achieved.
Water-Volume Calibration
Water-runoff volume from each source area for each rain event is calculated with the model. These calculations are
based on the amount of rainfall and a runoff coefficient developed for various rainfall depths for each source area. Source-
area characteristics such as imperviousness, connectedness (amount of impervious area directly connected to the storm
sewer), and infiltration rates on pervious areas were used to develop runoff coefficients (Pitt, 1987). The volumetric dis-
charges for each source area are then summed for each event. The total runoff volume can be decreased in the model
by using control measures, such as infiltration devices.
SLAMM was used to estimate runoff volume for the 25 storm events from all source areas in each basin. The sum
of the volumes from all of these source areas was compared to the volume measured at the basin storm-sewer outfalls
for these 25 events. Initially, the model overpredicted the water volumes measured at Monroe by a total of 55% (over the
entire study period), whereas it underpredicted those measured at Harper by only 2%. To obtain a balance of
overprediction and underprediction between the basins, the runoff coefficients were adjusted. Historically, more
measurements have been made for runoff from impervious surfaces (Pitt, 1987) and more than 50% of the area within
each basin is pervious, mostly because of residential lawns. Therefore, it was decided that the runoff coefficients for
pervious areas were more uncertain and model calibration could benefit from minor adjustments.
Two sets of runoff coefficients are available for pervious areas; one is designed to represent clayey soils, and the
other represents sandy soils. The predicted water volumes mentioned above were determined using the runoff coefficients
for clayey soils (based on soil maps). Changing the pervious classification from clayey to sandy resulted in SLAMM
underprediction of water volumes; approximately a 4% and a42% underprediction at the Monroe and Harper Basin storm-
sewer outfalls, respectively. A much better agreement was achieved at Monroe by assuming that the original soil
classification was incorrect. Sandy and clayey runoff coefficients, available to the model, probably represented two
extremes, and more realistic runoff coefficients fell somewhere between these two coefficients.
Lawn-runoff data collected from Monroe and Harper Basins were used to create runoff coefficients that more
accurately represent the pervious conditions found in Madison. First, the rainfall depth sufficient to initiate runoff in
SLAMM was changed using data on the amount of stormwater in the lawn-sample bottles after each event. For rainfall
amounts less than approximately 0.3 in., the bottles were less than 10% filled. From this observation, 0.3 in. was
established as the minimum precipitation required to initiate runoff. However, the runoff coefficient table for clayey soils
21
-------
used in SLAMM resulted in 10% runoff for a rainfall depth of 0.2 in. Hence, SLAMM was changed to initiate runoff at 0.3
in. rather than 0.2 in. of precipitation.
In addition to the change described above, a trial-and-error approach was used to change the coefficients until
optimum agreement was reached between water volumes predicted in SLAMM and those measured at the Monroe and
Harper storm-sewer outfalls. The resulting coefficients were between those for sandy and clayey soils and were
approximately two-thirds the value for clayey soils. Figure 8 shows how the new "Madison" runoff coefficients compare
to the sandy and clayey coefficients.
0.45
0.4
0.35
0.3
0.25
8 0-2
o
I 0.15
0.1
0.05
0.5
1.5 2 25 3
PRECIPITATION, IN INCHES
3.5
Figure 8. Madison runoff coefficients compared to sandy and clayey coefficients.
Based on the revised Madison runoff coefficients, SLAMM overpredicted storm-sewer outfall volumes at Monroe by
23% and underpredicted Harper storm-sewer outfall volumes by 24% (Table 4). The Madison coefficients also produced
consistent lawn-runoff contributions in both basins, approximately 20% of the total volume. It is expected that the
percentage of lawn contribution would be similar for both basins because they have nearly the same percentage of lawn
area.
22
-------
Table 4. Percentage Difference in Cumulative Modeled Water Volumes Compared with Measured Outfall Water Volumes Using Three Soil Types,
Madison, Wi.
Basin Sandy soils Clayey soils Madison soils
Monroe -4 55 23
Harper -42 -2 -24
Sediment Calibration
Once the runoff volumes were calibrated, SLAMM was used to estimate sediment loads for the 25 events in each
basin. With the exception of streets, a data base of sediment concentrations for each source area is used in SLAMM and
these concentrations are applied to the water volumes to derive a load. Sediment concentrations from streets are
computed by a wash-off function that is related to a street-dirt accumulation rate.
A mass-balance approach was used to test the source-area concentrations within each basin with those measured
at the storm-sewer outfall. Source-area loads were computed by multiplying the water volumes produced from SLAMM
by the concentrations measured at the source areas for each event and then summing these event loads. Sidewalks and
woodlots were two of the larger unmonitored source areas in each basin, accounting for 12% and 1% of the water volume
produced at Monroe and 7% and 2% at Harper, respectively. To add sidewalks to the load estimates, concentrations
measured at driveways were applied to estimates at sidewalks to create a sidewalk load. Woodlot concentrations were
estimated by use of data collected in an undeveloped urban site near Superior, WI (Steuer and others, 1997). The source-
area loads were 39% lower and 60% lower than the measured load at the storm- sewer outfall in the Monroe and Harper
Basins, respectively. This difference between source-area and storm-sewer outfall loads indicates that one or more source
areas within each basin were not effectively monitored.
Streets were the most likely source area to be ineffectively monitored. Street samplers were placed approximately
5 ft away from the curb to prevent gutter flow into the sampler because gutter flow usually contains a mixture of water
from several source areas. Other street studies (Pitt, 1979) have estimated that 90% of the dirt on residential streets in
good condition with little to no parking accumulates within 3 ft of the curb. A larger amount of dirt can sometimes collect
along the curb itself rather than in the driving lane. Some of this dirt could have been deposited on the driving lane, and
turbulence from passing vehicles and wind may have moved it to the curb. Most street dir-t falls within 1 to 2 ft of the curb
if the driving lane is next to the curb (Pitt, 1979). This information suggests that the street samplers in the Monroe and
Harper Basins were too far from the curb (5 ft) to representatively collect the particulate dirt from the streets.
For the reasons previously described, a trial-and-error approach was used to select a street- sediment concentration
that more accurately reflected the street sediment entering the storm sewer. The final suspended-solids concentrations
for streets were increased by a factor of 5. Applying this factor to the simulated street suspended-solids concentration
during each storm event allowed the sum of source-area loads to be within 7% and 9% of the storm- sewer outfall loads
at Monroe and Harper, respectively (Table 5). The geometric means for the revised street suspended-solids
concentrations were 340 and 325 mg/L for low- and high-traffic streets, respectively. These values were within 5% of
those measured at both Marquette, Ml (Steuer and others, 1997), and Madison, WI (Bannerman and others, 1993).
The geometric means of the observed suspended-solids concentrations, excluding streets, for the 25 storm events
at Monroe and Harper Basins were placed into the SLAMM data base (Table 6). The suspended-solids concentrations
for streets were not as easily altered because they are determined by dirt accumulation and wash-off functions in the
model. Entering the geometric means enabled the model to more accurately predict the loads measured at the storm-
sewer outfall. After summing the 25 events, sediment loads predicted by SLAMM were 17% lower at Monroe and 32%
lower at Harper compared to the measured storm-sewer outfall loads.
To improve the match between measured and simulated storm-sewer outfall loads, the delivery coefficients were
removed from SLAMM calculations, essentially assuming 100% delivery from source area to storm-sewer outfall. The
delivery coefficients had been added in a previous calibration study to force a match to the storm-sewer outfall
23
-------
numbers. This adjustment resulted in a 4% and 17% undersimulation between storm-sewer outfall loads of sus-
pended solids calculated by SLAMM and those measured at the storm-sewer outfall for Monroe and Harper,
respectively (Table 5).
Phosphorus Calibration
One objective in calibrating phosphorus concentrations in SLAMM was to ensure that the Monroe and Harper Basins
were accurately represented by the monitored source areas. The sum of source-area loads for total phosphorus for the
25
Table 5. Percentage Difference in Cumulative Source-area Sediment Loads, Before and After Sediment Adjustment, and Modeled and Measured
Sediment Loads at the Basin Outfall, Madison, WI. [Loads computed as suspended solids; %, percent]
Cumulative source area compared
with storm-sewer outfall
Basin Suspended Suspended Modeled
solids before solids after compared with
adjustment adjustment measured
Monroe -39 -7 -4
Harper -60 -9 -17
Table 6. Suspended-solids Concentrations Used to Calibrate the Source Loading and Management Model for Basins in Madison, Wi. [mg/L, milligrams
per liter: —, a series of algorithms and coefficients are used in the model to calculate a suspended-solids concentration for street runoff]
Suspended solids (mg/L)
Source area Residential Commercial
Driveways 34 34
Lawns 84 84
Parking lots 51 51
Streets
Woodlots 15 15
Roofs 16 18
Sidewalks 34 34
storm events was nearly identical to the storm-sewer outfall load (Table 7). The difference was larger for dissolved
phosphorus, but no information was available to determine what adjustments should have been made to reduce the
difference. For this reason, the unadjusted concentrations were entered into the SLAMM data base.
Table?. Percentage Difference Between Cumulative Source Area Versus Outfall Loads and Modeled Results Versus Outfall Loads, after Calibration of
the Source Loading and Management Model for Basins in Madison, Wi. [%, percent]
Cumulative source area Model results after
versus calibration versus storm-
storm-sewer outfall sewer outfall
Basin Phosphorus load
Dissolved Total Dissolved Total
Monroe 39 -1 -9 -24
Harper 35 4 -10 -28
The model simulates total phosphorus loads by adding the dissolved phosphorus and particulate phosphorus loads.
For all source areas except streets, particulate-phosphorus concentrations were calculated using total- and dissolved-
phosphorus and sediment concentrations measured at the Monroe and Harper Basins. To be consistent with calibration
24
-------
procedures, particulate-phosphorus concentration for street runoff was calculated using the adjusted value for sediment
(an increase by a factor of 5). Changing the phosphorus concentrations resulted in SLAMM undersimulation of dissolved
and total phosphorus by 9% and 24% at the Monroe storm-sewer outfall and 10% and 28% at the Harper storm-sewer
outfall, respectively.
The dissolved- and particulate-phosphorus concentrations entered into the SLAMM data base are listed in Tables 8 and
9. Significant changes in dissolved-phos-phorus concentrations (Table 8) were observed for lawns (from 0.22 to 0.53
mg/L), streets (from 0.39 to 0.12 mg/L), woodlots (from 0.25 to 0.01 mg/L), and sidewalks (from 0.60 to 0.07 mg/L). With
the exception of streets, where the particulate-phosphorus concentrations in runoff decreased, particulate-phosphorus
concentrations increased significantly (Table 9).
Table 8. Dissolved-phosphorus Concentrations Used to Calibrate the Source Loading and Management Model for Basins in Madison, Wi.
[mg/L, milligrams per liter]
Dissolved phosphorus (mg/L)
Source-area Residential Commercial
Driveways 0.07 0.07
Lawns .53 .53
Parking lots .02 .02
Streets .12 .03
Woodlots .01 .01
Roofs .04 .02
Sidewalks .07 .07
Table 9. Particulate-phosphorus Concentrations Used to Calibrate the Source Loading and Management Model for Basins in Madison, Wi.
[mg/kg, milligrams per kilogram]
Particulate phosphorus (mg/kg)
Source area Residential Commercial
Driveways 2,649 2,649
Lawns 4,943 4,943
Parking lots 1,467 1,467
Streets 569 409
Woodlots 5,000 5,000
Roofs 3,777 7,946
Sidewalks 2,649 2,649
Distribution of Source-Area Loads
The distribution of suspended-solids and total- and dissolved-phosphorus loads for source areas in the Monroe and
Harper Basins using measured source-area concentrations multiplied by SLAMM-generated water volumes is shown in
Table 10. The distribution of water volumes is nearly identical at Monroe and Harper Basins. The percentage of the total
basin represented by each source area is similar for both basins (Table 1); thus, one should expect to see similar relative
volumes of water calculated from both basins. Streets contributed most of the suspended-solids loads at both Monroe
and Harper Basins, generating 81% and 73%, respectively. Lawns contributed more than 10% of the solids loads at both
basins. The phosphorus loading, however, was quite different. Lawns in the Harper Basin generate more than two-thirds
of the phosphorus loads, whereas phosphorus in the Monroe Basin is more evenly distributed between lawns and streets.
These differences in load distribution are the result of the measured phosphorus concentrations, especially for lawns,
which are much higher for the Harper Basin (Table 2).
25
-------
Table 10. Distribution of Loads Based on Measured Values at Monroe and Harper Basins, Madison, Wi, and Incorporating the Suspended-solids Adjustment
[N/A, source area not present; %, percent abundance; --, value less than 0.5 percent]
Source area
Water
volume'
HARPER
Suspended
solids (%)
Lawns
Streets
Driveways
Sidewalks
Parking lots
Roofs
Parks
Woodlots
Other
Total
21
37
18
7
3
11
3
N/A
311,122
15
732
7
3
--
--
N/A
3,598
Total Dissolved
phosphorus phosphorus
67
14
9
4
1
3
2
N/A
9
71
11
9
3
4
2
N/A
5
MONROE
Water Suspended Total Dissolved
volume (%) solids (%) phosphorus phosphorus
44 45
37 39
5 4
6 4
1
1
7 7
N/A N/A
70 33
20
38
12
14
6
7
3
N/A
1
2,417,341
10
281
2
3
2
2
N/A
..
26,045
'Water volume totals expressed in cubic feet; all other totals expressed in pounds.
2Street-runoff concentrations multiplied by 5.
The suspended-solids load distribution at the Monroe Basin in 1994 is similar to the distribution observed in a 1991
study (Bannerman and others, 1993). Streets contributed 80% of the total basin suspended-solids load in 1991 and 81%
in 1994. Lawns also were comparable, contributing 7% and 10% of the total basin suspended-solids load in 1991 and
1994, respectively. Total and dissolved phosphorus, however, were very different. During the 1991 study, the proportion
of the total-phosphorus load from streets (58%) outweighed that for lawns (14%). The same was true in 1991 for dissolved
phosphorus, where streets produced 46% and lawns22% of the basin load. However, most total- and dissolved-phospho-
rus loading in 1994 was attributed to lawns rather than streets. Streets and lawns, in 1994, generated 37% and 44% of
the total-phosphorus load and 39% and 45% of the dissolved- phosphorus load in the Monroe Basin. The difference in
distributions between the two studies is possibly due to differences in sampling methodology. The street-sampler design
was modified for the 1994 study to eliminate a first-flush effect, where the sample bottle would quickly fill with stormwater
and act as a sediment trap for the remaining duration of the storm event. Also, during the 1994 study, 25 events were
monitored, whereas only 10 events were monitored in 1991. This larger sample size in 1994 improves confidence in the
loading-distribution estimates.
Distribution of suspended-solids, total-phosphorus, and dissolved-phosphorus loads estimated by SLAMM are given
in Table 11. The distribution of loads is consistent with the distribution of measured loads shown in Table 10. For each
constituent, slightly less total load was simulated with SLAMM than calculated using the measured concentrations (other
than suspended solids from streets), yet the distributions of each constituent were similar. Streets and lawns contribute
nearly all of the suspended-solids load for the entire basin. Streets alone contribute more than 75% of the suspended
solids at both Monroe and Harper Basins. Additionally, the significance of lawns as generators of phosphorus is again
noted in SLAMM simulations. Lawns in the Harper Basin contribute 52% and 61% of total and dissolved phosphorus
loads, and lawns in the Monroe Basin contribute 49% and 57%. Streets contribute the second largest phosphorus loads
(about 25%), whereas driveways and sidewalks combined contribute approximately 10%.
The distribution of suspended solids and total and dissolved phosphorus for source areas in the Monroe and Harper
Basins using measured source-area concentrations multiplied by SLAMM-generated water volumes is shown in Table
12. Only loads for the source areas measured are shown in Table 12; concentrations of suspended solids in street runoff
have not been adjusted. These source areas accounted for 82% and 90% of the total water volume from the Monroe and
Harper Basins, respectively.
26
-------
Table 11. Distribution of Loads from Model Simulation Results at Monroe and Harper Basins, Madison, Wi. [N/A, source area not present; %, percent
abundance; -, value less than 0.5 percent]
HARPER
Source area Water Suspended Total Dissolved
volume solids (%) phosphorus phosphorus
52
32
8
3
1
3
2
1
N/A
7
Lawns
Streets
Driveways
Sidewalks
Parking lots
Roofs
Parks
Woodlots
Other
Total
21
37
18
7
3
11
3
N/A
311,122
11
81
4
1
1
1
--
N/A
3,170
MONROE
Water Suspended Total Dissolved
volume (%) solids (%) phosphorus phosphorus
49 57
26 22
5 4
6 5
1
2
8 9
N/A N/A
2 2
60 29
61
24
7
3
2
2
..
N/A
4
20
38
12
14
6
7
3
N/A
1
2,417,341
12
77
3
3
2
1
2
N/A
1
20,814
1'Water volume totals expressed in cubic feet; all other totals expressed in pounds.
Table 12. Distribution of Loads from Monitored Source Areas Only, Based on Unadjusted Concentrations at the Monroe And Harper Basins, Madison, Wi.
[ %, percent abundance; percentage columns may not add up to 100% because of independent rounding]
Source area
Lawns
Streets
Driveways
Parking lots
Roofs
Water
volume
(%)
23
41
20
4
12
HARPER
Suspended
solids
(%)
41
43
10
3
3
Total Dissolved
phosphorus phosphorus
70
20
7
1
3
75
15
6
0
4
Water
volume
(%)
24
46
14
7
9
MONROE
Suspended Total Dissolved
solids phosphorus phosphorus
28
53
9
7
3
56
33
7
2
2
69
21
8
1
The suspended-solids distributions shown in Table 12 differ from Tables 10 and 11, in that the significance of lawns
as a source increases and the significance of streets as a source decreases. Streets are still the largest source of
suspended solids in both basins. The phosphorus distributions also change, but not as much as the suspended solids
because the measured phosphorus concentrations were used in all three tables (Tables 1 0, 1 1, and 12). The significance
of lawns as a source increases slightly, and streets are a slightly larger source in some cases, and in others, are slightly
smaller sources. Results shown in Table 12 indicate that the adjustments made to suspended-solids concentrations in
street runoff do not greatly affect the phosphorus distributions in the basins.
Sediment and Phosphorus Mass in Street-Dirt Samples
Approximately 75% of the total sediment mass in the street-dirt samples originated in the >250 urn size fraction,
whereas the smaller fractions (<63 Lim) made up less than 5%. Material composed of leaves, twigs, and other organic
debris also were measured, contributing, on average, less than 10% of the total sediment mass of the sample (Figure
9).
Like sediment mass, the largest amount of total phosphorus was found in the >250 pm size fraction (nearly
50%) (Figure 9). Combining this size fraction with the leaf fraction, about 80% of the total phosphorus is
accounted for. The contribution of total phosphorus mass decreased as the size fraction decreased. Other
studies have shown that large phosphorus concentrations correspond with small particle sizes because of the
high surface area to mass ratio for small particles (Sartor and Boyd, 1972).
27
-------
<25
25-63 63-250
PARTICLE SIZE, IN MICROMETERS
>250
Leaves
Figure 9. Relations between sediment and total-phosphorus mass from street-dirt samples for five particle-size fractions for basins in Madison, Wl
However, the bulk of the phosphorus phosphorus load results from the greater particle-size fractions (Ray, 1997).
Approximately 25% of the total phosphorus mass in each size fraction can be attributed to leaves (Ray, 1997).
A recent study of particle-size distribution in stormwater at the Monroe storm-sewer outfall demonstrated that most
of the solids are in the particle sizes <63 ?m (Greb and Bannerman, 1997). This distribution is the opposite of the particle-
size distribution observed for the street dirt collected in this study. These results indicate either a loss of the larger
particles somewhere between the street and the outfall or a problem in collecting larger particles in the runoff samples.
Most of the larger particles (>63 ?m) might settle out before reaching the storm-sewer outfall. Street sweeping,
resuspension onto street terraces, and catch basins can remove these particles from the streets before they reach the
storm sewer. Also, the transport of large particles in a storm sewer is not as efficient as the transport of smaller, more
mobile particles. Large sediment particles may become trapped or part of the bedload before reaching the sampler.
Bedload is not sampled efficiently by the automatic samplers described earlier in this report.
Summary and Conclusions
Concentrations of suspended solids, total phosphorus, and dissolved phosphorus were collected from various source
areas at two urban residential basins in Madison, Wl. To represent a range of source-area concentrations for urban
residential basins in Madison, the geometric means of the combined concentration data from the Monroe and Harper
Basins were incorporated into the urban-runoff model, SLAMM.
Source-area suspended solids and phosphorus loads from the Monroe and Harper Basins were determined based
on measured concentrations that were multiplied by water volume estimated by use of SLAMM. Collected data were used
to calibrate and increase confidence in water volumes, suspended solids, and phosphorus source-area loads estimated
by SLAMM. The calibrated model calculated water volumes to within 23% and 24% of those measured at the outfalls of
28
-------
the Monroe and Harper Basins. These calibrated water volumes were then applied to the calibrated suspended-solids
and phosphorus concentrations entered into the SLAMM data bases. Suspended-solids loads were estimated by the
calibrated SLAMM to be within 4% and 17%, total-phosphorus loads within 24% and 28%, and dissolved-phosphorus
loads within 9% and 10% of those measured at the storm-sewer outfall to the Monroe and Harper Basins, respectively.
Streets and lawns are the largest contributors of suspended-solids, total-phosphorus, and dissolved-phosphorus loads
in a residential urban basin. Lawns are the largest contributors of total and dissolved phosphorus; however, streets
contributed nearly 40% of the basin load, as seen in the Monroe Basin. Streets were found to be the largest source of
suspended solids.
There was a large difference between geometric mean concentrations of phosphorus in lawn runoff from 1994 to
1995. Phosphorus data collected from lawns in the Harper and Lakeland Basins during 1995 are remarkably similar, which
suggests that the phosphorus concentration in lawn runoff is affected by some variable or variables that are not yet
understood.
Street-dirt samples indicate that approximately 75% of the sediment mass resides in the >250 pm particle-size
fraction. Less than 5% of the mass can be found in the particle sizes less than 63 pm. The >250 urn particle-size fraction
also contributed nearly 50% of the total- phosphorus mass, and the leaf fraction contributed an additional 30%. In each
particle-size fraction, approximately 25% of the total-phosphorus mass is derived from leaves or other vegetation.
A possible limitation of this study may be that in order for the sum of the source-area loads to match the basin-outfall
loads, it was assumed that the concentrations of suspended solids in street runoff were about 5-times higher than the
concentrations measured. However, the analysis of load distributions based onlyon unadjusted monitoredconcentration
data shows little change in the distributions. In addition, samples from more rain events were collected in this study than
previous source-area studies. Also, improved data-collection equipment were used during this study. Both of these factors
lead to greater confidence in the study results.
Most of the measured suspended-solids concentrations were lower than those measured from previous studies.
However, when comparing concentration results in this study to results from earlier studies, it is important to note that
with the exception of the Marquette study, previous studies used earlier generation source-area sampling equipment. In
Marquette, Ml, the soils are considerably more sandythan those in Madison, which may explain why the suspended-solids
concentrations determined for the Marquette study are higher than those from Madison even though both studies used
the same sample-collection equipment.
The recalibration of the SLAMM model results in an improved model that should more accurately simulate phosphorus
and sediment runoff loads in Wisconsin than the earlier version of the model. The newly created lawn-runoff coefficients
for Madison represent a compromise between the two previous soil-type options available for model input, which probably
represented runoff extremes. The runoff coefficients calculated for Madison should probably be applied to most urban
lawns in Wisconsin unless soils are known to be either sandy or clayey. The phosphorus- and sediment-concentration
data bases created for this study are the largest to date using the most advanced source-area sample collection
technology available.
References
Andren, A.W., and Stolzenburg, T.R., 1979, Atmospheric deposition of lead and phosphorus on the Menomonee River
watershed-Volume 8, The IJC Menomonee River Watershed Study: U.S. Environmental Protection Agency,
EPA-905/4-79-029-H, 98 p.
Bannerman, R.T., Baun, K., Bohn, M., Hughes, P.E., and Graczyk, D.A., 1983, Evaluation of urban nonpoint source
pollution management in Milwaukee County, Wisconsin-Volume I, Urban stormwater characteristics, pollutant sources
and management by street sweeping: U.S. Environmental Protection Agency, Water Planning Division, PB-84-114164,
191 p.
29
-------
Bannerman, R.T., Owens, D.W., Dodds, R.B., and Hornewer, N.J., 1993, Sources of pollutants in Wisconsin stormwater:
Water Science and Technology, v. 28, no. 3-5, p. 241-259.
Betz, CR., Lowndes, M.A., and Porter, S., 1997, Nonpoint source control plan for the Lake Mendota priority watershed
project-Project summary: Wisconsin Department of Natural Resources, WT-481-97, 16 p.
Cullen, R.S., and Huff, D.D., 1972, Determination of land use categories in the Lake Wingra Basin: US-International
Biological Program, Eastern Deciduous Forest Biome Program, Memo Report 72-43, 17 p.
Dong, A., Simsiman, G.V., and Chester, G., 1983, Particle size distribution and phosphorus levels in soils, sediment and
urban dust and dirt samples from the Menomonee River watershed, Wisconsin, USA: Water Research, v. 17, no. 5, p.
569-577.
Greb, S.R., and Bannerman, R.T., 1997, Influence of particle size on wet pond effectiveness: Water Environment
Research, v. 69, no. 6, p. 1134-1 138.
Helsel, D.R., and Hirsch, R.M., 1992, Statistical methods in water resources: Amsterdam and New York, Elsevier, 529
P-
Kilpatrick, FA, Kaehrle, W.R., Hardee, J., Cordes, E.H., and Landers, M.N., 1985, Development and testing of highway
storm-sewer flow measurement and recording system:
U.S. Geological Survey Water-Resources Investigations Report 85-4111, 98 p.
Kunz, K.W., 1980, Atmospheric bulk precipitation in the Great Lakes basin: Environment Canada, Scientific Series 115,
28 p.
Lathrop, R.C., Nehls, S.B., Brynildson, CL, and Plass, K.R., 1992, The fishery of the Yahara Lakes: Wisconsin
Department of Natural Resources, Technical Bulletin 181, 214 p.
Legg, A.D., Bannerman, R.T., and Panuska, J., 1996, Variation in the relation of rainfall to runoff from residential lawns
in Madison, Wisconsin, July and August 1995: U.S. Geological Survey Water-Resources Investigations Report 96-4196,
11 p.
Murphy, T.J., 1974, Sources of phosphorus inputs from the atmosphere and their significance to oligotrophic lakes:
University of Illinois Water Resources Center, Research Report 92, 41 p.
Oakes, E.L., Hendrickson, G.E., and Zuehls, E.E., 1975, Hydrology of the Lake Wingra Basin,
Dane County, Wisconsin: U.S. Geological Survey Water-Resources Investigations 17-75, 31 p.
Pitt, R., 1979, Demonstration of nonpoint pollution abatement through improved street and sewerage cleaning: Cincinnati,
Ohio, U.S. Environmental Protection Agency, EPA/600/2-79/161, 270 p.
-1987, Small storm urban flow and particulate washoff contributions to storm-sewer outfall discharges: University of
Wisconsin-Madison, Dept. of Civil and Environmental Engineering, 513 p.
Pitt, R., Field, R., Lalor, M. and Brown, B. 1995, Urban stormwater toxic pollutants-Assessment, sources, and
treatability: Water Environment Research, v. 67, no. 3, p. 260-275.
Prentki, R.T., Rogers, D.S., Watson, V.J., Weiler, P.R., and Loucks, O.L., 1977, Summary tables of Lake Wingra basin
data: University of Wisconsin-Madison, Institute for Environmental Studies, IES Report 85, 89 p.
Ray, H, 1997, Street dirt as a phosphorus source in urban stormwater: University of Alabama at Birmingham, Department
of Civil Engineering, 125 p.
30
-------
Ryden, J., Syers, J. and Harris, R., 1972, Nutrient enrichment of runoff waters by soils, phase I- Phosphorus enrichment
potential of urban soils in the city of Madison: University of Wisconsin-Madison, Department of Soil Science, Technical
Report OWRR-A-038-WIS, 70 p.
Sartor, J.D., and Boyd, G.B., 1972, Water pollution aspects of street surface contaminants: U.S. Environmental Protection
Agency, EPA-R2-72-081.237 p.
Schindler, D.W., 1974, Eutrophication and recovery in experimental lakes-Implications for lake management: Science,
v. 184, p. 897-899.
-1977, Evolution of phosphorus limitation in lakes: Science, v. 195, p. 260-262.
Steuer, J.J., Selbig, W.R., and Hornewer, N.J., 1997, Sources of contamination in an urban basin in Marquette, Michigan
and an analysis of concentrations, loads, and data quality: U.S. Geological Survey Water-Resources Investigation Report
97-4242, 25 p.
U.S. Environmental Protection Agency, 1983, Results of the Nationwide Urban Runoff Program, volume l-Final report:
Water Planning Division, Washington, D.C., NTIS PB#84-185552, 200 p.
Ward, J.R., and Harr, C.A., eds, 1990, Methods for collection and processing of surface-water and bed-material samples
for physical and chemical analyses: U.S.Geological Survey Open-File Report 90-140, 71 p.
31
-------
Using Biological Criteria to Assess and Classify Urban Streams and Develop
Improved Landscape Indicators
Chris 0. Yoder and Robert J. Miltner
Ohio EPA, Division of Surface Water
Ecological Assessment Unit
Groveport, Ohio 43125
Dale White
Ohio EPA, Division of Surface Water
Information Resources Management Section
Columbus, Ohio 43216-0149
Abstract
This study consisted of a quantitative analysis of the relationship between the Index of Biotic Integrity (IBI), an
indicator of urban land use, and a qualitative analysis of overlying stressors in six of the major metropolitan areas of Ohio.
A database consisting of 267 sampling locations was extracted from the Ohio EPA statewide biological and habitat
database. Most of these sites were sampled between 1990 and 1998 and contained watershed areas less than 50 mi.2,
with most draining less than 20 mi.2. A negative relationship between IBI and urban land use was observed in four of the
six areas, whereas little or no relationship was seen in two areas. For each area, the highest percentage of urban land
use that corresponded to minimum attainment of the applicable warmwater habitat IBI biocriterion ranged from 1%
(Cleveland/Akron) to 12% (Dayton) for the regression line, and 15% (Cleveland/Akron) to 58% (Columbus) as the highest
%urban land use where the IBI biocriterion was attained at any given site. No significant linear relationship was found
in either the Toledo or Youngstown areas, and only a weak relationship was visually apparent for the Toledo streams.
The lack of association was due to the strong presence of overlying stressors (e.g., legacy pollutants, sewage discharges,
combined sewer overflows) that resulted in very low IBI values at sites with lower levels of urbanization. The percentage
of urban land use explained approximately 35% of the variation in IBI scores in the regression model when these impact
types were excluded (compared to 11% when included). The maximum %urban land use that commonly corresponded
to attainment of the warmwater habitat IBI biocriterion based on inspection of the scatter plot was approximately 26%.
Only a very few sites exhibited attainment at urban land uses between 40-60% and none occurred above 60%. These
former sites had either an intact, wooded riparian zone, a continuous influx of groundwater, and/or the relatively recent
onset of urbanization. These results indicate that it might be possible to mitigate the negative effects of urbanization by
preserving or enhancing near and instream habitats, particularly the quality of the riparian buffer zone. The results also
suggest that there is a threshold of watershed urbanization (e.g., >60%) beyond which attainment of warmwater habitat
is unlikely. This threshold is not the same in all watersheds and it can occupy a rather wide range. It is affected by co-
factors such as pollutant loadings, watershed development history, chemical stressors, and watershed scale influences
such as the quality of the riparian buffer and the mosaic of different types of land use. Thus, single-dimension urban land
use indicators, such as watershed imperviousness, are not sufficiently precise or robust as a single indicator of use
attainability. The further development and refinement of multiple indicators of watershed urbanization has merit from a
management and decision-making standpoint. We suggest that co-factors, in addition to more refined urban land use
indicators, be better developed. More precise definitions of different urban land uses are also needed to better
understand and respond to the water quality management challenges posed in existing and developing urban areas.
32
-------
Introduction
The health and well-being of the aquatic biota in surface waters is an important barometer of how effectively we are
achieving the goals of the Clean Water Act (CWA); namely the maintenance and restoration of biological integrity, and
the basic intent of water quality standards. States designate water bodies for beneficial uses (termed designated uses)
that, along with chemical, physical, and biological criteria, assure the protection and restoration of aquatic life,
recreational, and water supply functions and attributes. Biological criteria are the principal tool for determining impairment
of designated aquatic life uses as defined by the Ohio WQS (Ohio Administrative Code 3745-1). As such,
bioassessments play a central role in the Ohio Nonpoint Source Assessment (Ohio EPA 1990; 1991), the biennial Ohio
Water Resource Inventory (305b Report; Ohio EPA 1998), and watershed-specific assessments, of which Ohio EPA
completes between 6 and 12 each year. Biological criteria represent a measurable and tangible goal, against which the
effectiveness of pollution control and other water quality management efforts can be judged. However, biological
assessments must be accompanied by appropriate chemical/physical measures, land use characterization, and pollution
source information necessary to establish linkages between stressors and the biological responses (Yoder and Rankin
1998). Biological criteria in the Ohio WQS also supports the determination of appropriate aquatic life use designations
for individual water bodies, provides for a "reality check" on the application of surrogate indicators, assesses cumulative
impacts, extends anti-degradation concerns to nonpoint sources and habitat influences, defines high quality waters, and
serves as a meaningful indicator in the management of regulatory programs for environmental results. This provides a
means to incorporate the broader concept of water resource integrity (Karr et al. 1986) in policy and planning while
preserving the appropriate roles of the traditional chemical/physical and toxicological approaches developed over the past
three decades.
We, and others at Ohio EPA, have previously described the status of Ohio's streams and rivers as affected by
watershed urbanization (Yoder et al. 1999; Yoder and Rankin 1997; Yoder 1995). Small watersheds are especially
impacted, as illustrated by Yoder and Rankin (1997), where no headwater streams in established urban settings
throughout Ohio attained the minimum CWA benchmark use designation of warmwater habitat. This finding has led to
the perception that the impairment of beneficial aquatic life uses in these small watersheds is intractable, at least within
the constraints of current land use policies, restoration technologies, and funding levels. Together, these factors present
potentially significant barriers to the objective of fully restoring degraded watersheds or upgrading urban streams that are
presently designated for less than fishable and swimmable uses.
Headwater streams are critical to watershed functioning in that they serve as the principal interface between runoff
from land use and receiving streams. The ability of a headwater stream to physically filter and biologically assimilate the
primary and secondary effects of pollutants is a function of habitat quality and the structure of the biological system. A
healthy headwater stream ecosystem is characterized by good habitat and a well balanced assemblage of aquatic
organismsand plants, one which processes external inputs in a manner which promotes high quality downstream exports.
These exports include good quality water and high value biomass, both of which positively impact the ability of
downstream waters to deliver quality goods and services (e.g., water supply, recreation, waste assimilation, water
retention, ecological values). A degraded headwater stream ecosystem is characterized by poor habitat and an
assemblage of aquatic organisms and plants that processes external inputs in a manner which promotes low quality
downstream exports. Thus in this latter scenario, the effects from urban runoff can accumulate in a downstream direction
and adversely affect water quality and ecosystem goods and services in larger water bodies. In Ohio, more than 78%
of stream miles drain less than 20 mi.2 and are classified as headwater streams. While these may individually seem less
significant than larger water bodies, they are collectively the most numerous and perhaps important stream type. In many
ways, and in a collective sense, headwater streams are analogous to the capillaries of the human circulatory system
where essential product transport and waste assimilation functions are accomplished. Certainly the finding that a high
proportion of headwater streams fail to meet CWA goals in Ohio urban areas translates to the potential for undesirable
impacts in downstream waters and obvious consequences for the overall health of the "patient".
There is concern that the attainment of CWA goal uses (e.g., warmwater habitat in Ohio) within small urban
watersheds may be precluded by the legacy of urbanization. If this is true, how is this determined and what are the
protection endpoints to guide water quality management? Federal water quality standards regulations (40CFR, Part 131)
33
-------
allow for the establishment of a use that is less than the CWA fishable and swimmable goals when it is precluded by the
following:
1) the degraded conditions are naturally occurring;
2) restoring the degraded conditions would result in widespread adverse socioeconomic impacts;
3) the degraded conditions are irretrievable and human induced.
Such uses are established on a waterbody-specific basis and are supported by a use attainability analysis. In Ohio,
such analyses are routinely conducted as a result of the five-year basin approach to monitoring, assessment, and water
quality management. One purpose of this paper is to advance the development of the tools and indicators needed to
make use attainability decisions in urban watersheds.
Ohio EPA routinely conducts biological and water quality surveys, or "biosurveys", on a systematic basis statewide.
A biosurvey is an interdisciplinary monitoring effort coordinated on a waterbody-specific or watershed scale. Such efforts
may be relatively simple, focusing on one or two small streams, one or two principal stressors, and a handful of sampling
sites; or much more complex, including entire drainage basins, multiple and overlapping stressors, and tens of sites. Each
year, Ohio EPA conducts biosurveys in 1 O-l 5 different study areas with an aggregate total of 350-450 sampling sites.
Biological, chemical, and physical monitoring and assessment techniques are employed in biosurveys in order to meet
three major objectives: 1) determine the extent to which use designations assigned in the Ohio Water Quality Standards
(WQS) are either attained or not attained; 2) determine if use designations assigned to a given water body are appropriate
and attainable; and 3) determine if any changes in key ambient biological, chemical, or physical indicators have taken
place over time, particularly before and after the implementation of point source pollution controls or best management
practices. The data gathered by a biosurvey is processed, evaluated, and synthesized in a biological and water quality
report. The findings and conclusions of each biological and water quality study may factor into regulatory actions taken
by Ohio EPA and are incorporated into Water Quality Permit Support Documents (WQPSDs), State Water Quality
Management Plans, the Ohio Nonpoint Source Assessment, and the Ohio Water Resource Inventory (305[b] Report).
In 1990, the Ohio EPA initiated an organized, sequential approach to monitoring and assessment, termed the Five-
Year Basin Approach. One of the principal objectives of this new approach was to better coordinate the collection of
ambient monitoring data so that information and reports would be available in time to support water quality management
activities such as the reissuance of NPDES permits and periodic revision of the Ohio Water Quality Standards (WQS).
Ohio EPA's approach to surface water monitoring and water quality management via the Five-Year Basin Approach
essentially serves as an environmental feedback process taking "cues" from environmental indicators to effect needed
changes or adjustments within water quality management. The environmental indicators used in this process are
categorized as stressor, exposure, and response indicators (Yoder and Rankin 1998). Stressor indicators generally
include activities that impact, but which may or may not degrade the environment. This includes point and nonpoint
source loadings, land use changes, and other broad-scale influences that generally result from anthropogenicactivities.
Exposure indicators include chemical-specific, whole effluent toxicity, tissue residues, and biomarkers, each of which
suggests or provides evidence of biological exposure to stressor agents. Responseindicators include the direct measures
of the status of use designations. For aquatic life uses, the community and population response parameters that are
represented by the biological indices that comprise Ohio EPA's biological criteria are the principal response indicators.
Previously, our analyses examined the water quality and biological assessment database from watersheds in and
near existing and developing urban and suburban areas of Ohio. Yoder and Rankin (1997) compiled their analyses based
on sampling conducted at more than 100 stream sampling locations. Yoder et al. (1999) examined more detailed land
use and stressor relationships with the Index of Biotic Integrity (IBI), based on fish assemblage data, and the Invertebrate
Community Index (ICI), based on macroinvertebrate assemblage data, within two major Ohio urban areas
(Akron/Cleveland and Columbus). This study consisted of a quantitative analysis of the relationship between the IBI, an
indicator of urban land use, and a qualitative analysis of other stressors influencing this relationship using available data
from all six of the major metropolitan areas within Ohio. The importance of understanding these relationships is
heightened by contemporary water quality management issues such as combined sewers and stormwater permitting.
One challenge we faced was in attempting to separate the influences of these multiple stressors on aquatic life attainment
34
-------
status. Could we sufficiently understand the baseline influence of urbanization apart from these other and better
understood stressors?
The principal analysis conducted in this study examined the relationship between urban land cover and the IBI, both
visually and by statistical analysis. Some goals were to determine the extent to which biological performance (as
expressed by the IBI) was correlated with urban land use, thresholds at which this occurred, and the overlying effects of
other stressors.
Met hods
A database consisting of 267 sampling locations from the six major metropolitan areas of Ohio was extracted from
the Ohio EPA statewide biological and habitat database. Most of these sites were sampled between 1990 and 1998 and
contained watershed areas less than 50 mi.2, with most draining less than 20 mi.2. As such, the database represents a
collection of discrete watershed units where land uses may have a significant effect on the composition and quality of the
instream habitat and biological communities. Urban land use effects have been much more apparent in these smaller
watersheds as evidenced by the higher proportion of impaired stream miles compared to larger streams and rivers in Ohio
(Yoder 1995; Yoder and Rankin 1997).
Fish communities were sampled using generator-powered, pulsed D.C. electrofishing units and a standardized
methodology (Ohio EPA 1987a,b, 1989a,b; Yoder and Smith 1999). Fish community attributes were collectively
expressed by the Index of Biotic Integrity (IBI; Karr 1981; Karr et al. 1986), as modified for Ohio streams and rivers (Yoder
and Rankin 1995; Ohio EPA 1987b, 1989b). Habitat was assessed at all fish sampling locations using the Qualitative
Habitat Evaluation Index (QHEI; Rankin 1989, 1995). The QHEI is a qualitative, visual assessment of the functional
aspects of stream macrohabitats (e.g., amount and type of cover, substrate quality and condition, riparian quality and
width, siltation, channel morphology, etc.). Ohio EPA also collected macroinvertebrate assemblage data at some of these
sites, but it was not included in this study because of the partial coverage and the extensive use of the qualitative method
was not always compatible with regression analysis. Some of the analyses in our earlier studies (Yoder et al. 1999)
included macroinvertebrate data.
The urban land use indicator was derived from Landsat Thematic Mapper satellite imagery of land cover classification
(September 1994) provided by the Ohio Department of Natural Resources. The percentage of land use in the urban
classification was calculated for the subwatershed upstream from each fish sampling location to the boundary of the
watershed. Because many of the sites included in the statewide data set are subjected to a variety of stressors, each
site was qualitatively classified by predominant impact type. Impact types included least impacted sites, estate sites (i.e.,
subwatersheds with large lot sizes or green space provided by parks), sites reflecting gross instream habitat alterations
(i.e., channel modifications or impoundment), sites impacted directly by discharges from combined sewer overflows
(CSOs), sites impacted by wastewater treatment plant discharges, sites impacted by instream sewer line placement and
construction (Cincinnati area only), sites with evidence of impacts by legacy pollutants, or sites affected by general
urbanization only. This latter category included urban land uses not containing any of the other impact types and usually
consisted of residential development.
Results
The relationship between the IBI and urban land use was initially characterized by regressing IBI scores against
percent urban land use (log,, transformed) and QHEI scores using a database of 267 sites for all of the six major
metropolitan areas of Ohio (Figure 1). Diagnostic plots (e.g., residuals, normal probability) indicated nonconstancyof error
variance. To provide insights into whether the results varied substantially between each metropolitan area, scatter plots
of the relationship between urban land use and IBI in each of the six metro areas were also made (Figure 2). A negative
relationship between IBI and urban land use was observed in four of the six areas, whereas little or no relationship was
seen in two areas. For each area, the highest percentage of urban land use that corresponded to minimum attainment
of the WWH IBI biocriterion was determined by inspection of the scatter plot and the intersection of the regression line
and the WWH IBI biocriterion were determined (Figure 2). This ranged from 1% (Cleveland/Akron) to 12% (Dayton) for
the regression line, and 15% (Cleveland/Akron) to 58% (Columbus) as the highest %urban land use where WWH was
attained in each area at a given sampling location. No significant linear relationship was found in either the Toledo or
35
-------
CO
IMPACT TYPES
EWH Biocriterion
(IB I = 501
WWH
Biocriterion
/(IBI = 40)
• Habitat
^ Background
a Urban
0 WWTP
* Legacy
± CSO/SSO
Q Estate
^ Sewer Line
•95%
4J* _ al •§ «U- — ^ _X ' '
£ S - »»o - " --—' Jl ' ~ - • • ' — "• * . 95%
IS.TI-' • -r o • "- ""*-*».^ » o ' ' " - ,- 75%
. JB"»--. * f " ""v * ' 1
V O l*-^p A
• T •np "••-. o o """--•.
• AV GO "*••-...
GO
OMB • • •
T •• HT mm
O OT Ai *
O A «
• •""••-.. .A
_ ' • .,
•SB T •• BBT mm mm ••-...A
TB> • O O OT Ai * A B, B
A3BB* AlfT B
L TB B A \ <3k A* * B A ^
A * 0» * (* "^
3 °0 "
A A A
AA A
T A
10.0 20.0 30.0 40.0 50.0 60.0
Percent Urban Landuse
70.0
80.0
Figure 1. Scatter plot of Index of Biotic Integrity (IBI) scores against percentage of watershed upstream from the site in urban land use at 267 small (<50
mi.2) sampling sites.
Youngstown areas, and only a weak relationship was visually apparent for the Toledo streams. All of the sites in the
Youngstown area were impaired, and so severely that no land use relationship was evident (Figure 2). In the Toledo area,
the highest urban land use corresponding to WWH attainment was 28%. However, the WWH IBI biocriterion in the
Huron/Erie Lake Plain ecoregion is the lowest in the state and almost all of the small streams in the Toledo area have
been channel modified to some degree. It was apparent that the lack of a stronger association between IBI and urban
land use was due to overlying stressors (e.g., legacy pollutants, WWTPs, CSOs/SSOs), particularly those that resulted
in very low IBI values at sites with low levels of urbanization. While some threshold relationships were evident in these
results, the resulting variability in IBI scores led to only weak or non-existent linear relationships.
Some of the impact types had a strong effect on the IBI regardless of the effect of urban land use. The IBI results
were examined by impact type across all six metro areas (Figure 3). The legacy, CSO/SSO, habitat, and WWTP impact
types had the strongest negative effects on the IBI, respectively, and this was independent of the urban land use indicator.
While these impact types are common to urban areas, they were removed from the remaining statistical analyses
(elimination of these impact types reduced the sample size to 123 sites) to better develop the IBI/urban land use
relationship. The entire Toledo and Youngstown datasets were also removed since they are comprised entirely of these
impact types. This resulted in a better regression model fit, and diagnostics consistent with regression model assumptions
(Neter et al. 1990). This also allowed us to discern the threshold of urbanization at which WWH attainment is lost with
greater precision and in the absence of potentially confounding impacts, which was a major objective of our study. The
relationship between different levels of urbanization and biotic integrity was further quantified with an analysis of variance
model where quartiles of percent urban land use determined factor level (e.g., all sites within the 1 st quartile of percent
urban land use were coded as factor level 1). Similarly, an analysis of covariance model using QHEI as the covariate
was employed to test for further refinements. Multiple comparisons of factor level mean differences were made using
Tukey's method (Neter et al., 1990).
36
-------
50
IBI
10
Regression Line
Intersects W W H
(11%)
Maximum Observed
%Urban That
Meets WWH
127%)
Cincinnati Area 3(ream 4
. i* i
1H--I--
60
50 -
40 -
IBI
30
20
10 100
Percent Urban Landuse
Maximum Observed
Regression Line %Urban That
Intersects WWH Meets WWH
(3%) 158%!.
'. Columbus Area ^Streams
I
I 10 100
Percent Urban Landuse
Maximum Observed
%Urban That
Meets WWH
(28%)
4O
IBI
30
20
10
Toledo Area Streams
0.1 1 10 100
Percent Urban Landuse
Maximum Observed
Regression Line %Urban That
Intersects WWH Meets WWH
(1%) (35%)
10
T^
O.I I 10
Percent Urban Landuse
50
40
IBI
30
IBI
50
40
I
30
20
10
t Youngstown Area Streams
Regression Line
Intersects WWH
(12%)
Maximum Observed
%Urban That
Meets WWH
(22%)
Dayton Area Stffam s
1 10
Percent U rban Landuse
Maximum Observed
%Urban That
Meets WWH
(0»)
r- i i i i i i i| i i i .
1 10 100
Percent Urban Landuse
Figure 2. Scatter plots of Index of Biotic Integrity (IBI) scores against percentage of watershed upstream from the site in urban land use at small stream
(<50 mi.2) sampling sites in six of the major metropolitan areas of Ohio. Predominant impact types are indicated for each site (see Figure 1) along with the
regression line. The warmwater habitat and exceptional warmwater habitat biological criteria for the IBI are also indicated.
37
-------
60
IBI
1
I
1
I Maximum %Urban
O ° i That Meets WWH
O c| (exploding outliers)
cc^o
o
o
o o o o o
i i i iiiiiTTii iii i ii\\iiirniii \iirn
0 10 20 30 40 50 60 70 80
Percent Urban Landuse
IBI
60
50
40
30
20
12
I
022.6 [
BO O30.2CJ18.4
MS, O O O ' /
¥XJ O ' ./
JljCL <$ir} 03 '^
i 0 00§ 0 « ',
-0 (P O i
- OO O i
a o i CD
- 0 O 0(33
o o
oa
I
-Avg. Age = 28.2 ' o
-Med. Age = 28.5 ]
i i i i i i i i i i i i i i i i i
Maximum %Urban
That Meets WWH
/- (excluding outliers)
019.8
O 37.8 "iiiji • O'"29.9 " — '{'WHV
nn- °17'5 •"'""
°8 Q
QD CD
O. O O
00 R °
u O
O
o
I I I I I I I I I I I I I I I I I I I I
_
_
-
_
-
_
_
-
—
-
-
-
1 1
0 10 20 30 40 50 60 70 80
Percent Urban Landuse
Figure 3. Box-and-whisker plots of Index of Biotic Integrity (IBI) scores by each of the major impact types used in Figures 1 and 2. The warmwater
habitat and exceptional warmwater habitat biological criteria for the IBI are indicated.
38
-------
The percentage of urban land use explained approximately 35% of the variation in IBI scores in the regression model
when the other impact types were excluded. Local habitat quality (as measured by the QHEI) explained an additional
7% of the variation (Table 1). The ANOVA model showed that there were significant differences in mean IBI scores
between quartile level of percent urbanization, with sites exceeding 29 % urban land cover having lower IBI scores on
average than sites with less urban land cover (Figure 4). Sites characterized by less than 4% urban land cover had
higher IBI scores than sites with urban land use exceeding 15% The ANCOVA model provided a slightly better fit, but
the additional variation explained was marginal (Table 2), and the results of pairwise comparisons were similar between
models (Figure 4).
Table 1. Regression Results for the Model Ibl = Log10 (Percent Urban Land Use+l)+qhei for All Sites and the Removal of Selected Impact Types.
Effect Coefficient SE t P(2 Tail) Adj. R-Squared
All Sites
CONSTANT
Urban
QHEI
Impact Types Removed
CONSTANT
Urban
QHEI
21 .9333
-6.8323
0.2676
32.4069
-11.1496
0.2390
2.9450
1.1370
0.0418
4.2184
1.3102
0.0605
7.4477
-6.0092
6.4102
7.6822
-8.5096
3.9493
0.0000
0.0000
0.0001
0.0000
0.0000
0.0001
0.1179
0.2388
0.3500
0.4199
CD 60
oZ 50
O
UJ
O
O
CO
40
30
O 20
X
LJU -10
Q '^
EWH Statewide
Biocriterion
WWH Ecoregion
Biocriteria
Minimum IBI = 12
n = 8 n = 45 n = 1B n = 94 n = 41 n = 26 n = 37 n=11
UJ
o>
o
CO
CD
a)
(/>
O
Figure 4. Distributions of Index of Biotic Integrity (IBI) from small streams (<50 mi.2) in the six major metropolitan areas of Ohio plotted by quartiles of
percent of urbanization upstream from sampling locations. Horizontal lines spanning adjacent box plots indicate similar means. Levels of percent of
urbanization corresponding to the 25th, 50th and 75th percentile are indicated. The shaded areas indicate the applicable warmwater habitat biological
criterion and the range of insignificant departure for the IBI.
39
-------
Table 2. Analysis of Variance Results for the Anova Model, and the Ancova Model Using Qhei as a Covariate.
ANOVA
Source
Urban
Error
ANCOVA
Source
Urban
QHEI
Error
Sum-of-Squares
4248.53
6129.10
Sum-of-Squares
4020.66
640.71
5488.39
df
3
119
df
3
1
118
Mean-Square
1416.18
51.51
Mean-Square
1340.22
640.71
46.51
F -ratio
27.50
F-ratio
28.81
13.78
P
0.0000
P
0.0000
0.0003
R-Squared
0.4094
0.4711
In an attempt to better visualize where attainment of warmwater habitat occurs along the urban land use gradient,
the IBI results were plotted against percent of urban land use for all sites used in this study and with the other impact
types excluded (Figure 5). The elimination of the other impact types provided for a more precise statistical relationship
between urban land use and the IBI (i.e., lower error of regression estimates). For example, the R2 was higher with the
removal of the other impact types and the slope of the regression was steeper, both of which suggest a more meaningful
relationship between the IBI and urban land use (Table 1). However, the percent of urban land use that corresponded
to attainment of the warmwater habitat IBI biocriterion based on inspection of the scatter plot was the same
(approximately 26%) in both plots.
m
50
40
30
20
10
W W H Biocriteria
(IBI = 40)
O
- • ,O
\
1 st
<4.2%
2nd
4.3-I 4.6%
3rd
14.7-29.3%
4th
>29.3%
Quartile Level of Percent Urbanization
Figure 5. Scatter plots of Index of Biotic Integrity (IBI) scores against percentage of watershed upstream from the site in urban land use at small stream
(<50 mi.2) sampling sites in six of the major metropolitan areas of Ohio for all sites (upper) and a subset with non-urban impact types removed (lower). The
age of the urbanized area is indicated for selected sites and the mean and median age for entire dataset. The warmwater habitat biological criterion and
the range of insignificant departure for the IBI are indicated.
40
-------
Also apparent in both plots was the occurrence of "outliers" where IBI scores above the warmwater habitat biocriterion
occurred at sites with 40% to 60% urban land use. These sites had either an intact, wooded riparian zone, a continuous
influx of groundwater, and/or the relatively recent onset of urbanization. Intact riparian buffers can mitigate the effects
of urban land use up to a point (Steedman 1988; Horner et al. 1997) and local hydrology can strongly influence the quality
of the fish assemblage (Poff and Allen 1995). The three sites with the relatively recent onset of urbanization (all <20
years) may not yet have accrued the types of negative effects that are readily apparent in some of the older urbanized
areas of Ohio.
Discussion
Threshold levels of urbanization beyond which biological communities are likely to be impaired have previously been
identified in the range of 8% to 20% impervious cover within a watershed (Schuler 1994). Our previous analyses (Yoder
et al. 1999) produced results of approximately 8% and 33% urban land use cover for the Cuyahoga River basin and
Columbus area streams, as identified by analysis of variance. We also concluded that the threshold level identified by
regression for the Cuyahoga River basin was lowered by the presence of other stressors (e.g., CSOs, point sources,
legacy pollutants). The elimination of those sites impacted by these other stressorsfrom the regression analysis resulted
in a higher threshold of urbanization. Our expanded study seemed to confirm this phenomenon, as the elimination of the
other impact types helped clarify the urban land use/IBI relationship in a broader array of urban influenced streams
throughout Ohio (Figure 5). The upper threshold of urbanization which corresponded to a loss of warmwater habitat
attainment was in the 25-30% range. However, our results show that non-attainment also occurs at lower thresholds of
urbanization (Figure 5) due primarily to the co-occurrence of other stressors. This makes both the linear and visual
derivation of sufficiently precise indicator thresholds such as percentage of impervious surfaces more difficult.
In terms of understanding the potential effect of urbanization on aquatic life use attainment, the most meaningful
results of our analyses are the upper thresholds at which attainment of CWA goal uses are mostly lost (e.g., 25%) and
that beyond which it never occurs (>60%). Only a very few sites exhibited full attainment of the warmwater habitat
biocriteria at urban land uses between 40-60% (Figure 5). A closer examination of these sites and the watersheds
showed the presence of high quality riparian zones, an influx of flow augmenting groundwater, and/or development of the
urban land use occurring within the past 20 years. For the latter, we hypothesized that the full effect of negative impacts
in an urban setting may take time to accumulate and may not be immediately manifest in the form of instream
impairments. This could account for the higher-than-expected urban land use (i.e., 40-60%) correlating with full
attainment of the biocriteria. If this is true, then we might expect these sites to exhibit declines in IBI scores over the next
one or two decades. It also suggests that newly urbanizing watersheds should be developed with an emphasis on
determining which attributes (e.g., riparian zones, wetlands, flow regime) need to be maintained and preserved in order
to protect and maintain instream habitat and biological quality.
The results of this study indicate that it might be possible to partially mitigate the negative effects of urbanization by
preserving or enhancing near and instream habitats, particularly the quality of the riparian buffer zone. The "outlier" sites
that exhibited full attainment of the warmwater habitat biocriteria had more extensive and higher quality riparian zones
and good to excellent instream habitat quality. Some streams were nestled in small valleys which were not amenable
to development and the accompanying encroachment of urban land uses. This generally agrees with the findings of
Steedman (1988) who demonstrated a co-relationship between riparian zone quality and land use in terms of how each
affected the fish communities and IBI values of Toronto area streams. Horner et al. (1997) also found that the negative
effects of urban land use were mitigated by riparian protection and other management interventions. However, in both
studies the quality and extent of the riparian zone ceased to be effective above 45-60% impervious land cover, which
generally corresponds to the thresholds identified by our study. Until we better understand the effect of the "age" of the
urban effect, it seems prudent to advocate policies that preserve existing riparian zones rather than responding with post-
urbanization retrofits.
Yoder et al. (1999) discussed the implications of their findings on the designation of aquatic life uses in state water
quality standards, particularly to the use attainability analysis process. Uses designated for specific water bodies are done
so with the expectation that the criteria associated with the use are reasonably attainable. If CWA goal uses (e.g.,
warmwater habitat in Ohio) are found to be unattainable, then lower quality uses may be established and assigned on
a case-by-case basis (40CFR, Part 131 .10[g]). Recently, the imperviousness of the watershed has been suggested as
41
-------
an indicator that is correlated with use attainability. If the frequently cited threshold of 25% impermeability is used,
streams in watersheds with greater than this value could be considered unlikely to ever attain a beneficial use regardless
of site- and reach-specific factors. This assumes that the negative effects of urbanization cannot be remediated, which
has yet to be extensively tested. However, the results of our study suggest that there is a threshold of watershed
urbanization (e.g., >60%) beyond which attainment of the WWH use becomes increasingly unlikely, at least as affected
by contemporary practices. This threshold is not the same in all watersheds, as evidenced by the results from the six
Ohio metropolitan areas, and it can occupy a rather wide range. In addition, co-factors such as pollutant loadings,
watershed development history, chemical stressors, and watershed scale influences such as the quality of the riparian
buffer and the mosaic of different types of land use, also act singly and in combination to determine to the resultant
biological quality in the receiving streams. Thus, single dimensional urban land use indicators, such as watershed
imperviousness, is not sufficiently precise or reliable as a single indicator of use attainability.
The further development and refinement of multiple indicators of watershed urbanization has merit from a
management and decision-making standpoint. Because of the many co-factors involved (e.g., water quality, habitat
quality, hydrologic regime, etc.), some of which are controllable and amenable to reasonable remediation, this will be a
complex undertaking. We suggest that these co-factors, in addition to more refined urban land use indicators, be
developed and tested using datasets from broad geographic areas spanning the extremes of the urbanization gradient.
Urban land use and its analogs (e.g., % imperviousness) are coarse approximations of the cumulative effect of all
negative influences within a watershed. Thus co-factors and more precise definitions of different urban land uses need
to be defined in order to better understand and respond to the water quality management challenges posed in existing
and developing urban areas.
A management outgrowth of such an effort could be the development of an urban stream habitat use designation.
Yoder et al. (1999) previously indicated where the biological criteria for this potential new use designation might occur
compared to the already existing hierarchy of aquatic life uses in the Ohio WQS (Figure 6). This designated use would
Max.
Quality Gradient of Aquatic Life Uses and Narrative
Descriptions of Biological Community Condition
Exceptional
Warmwater
Habitat (EWH)
Index
Value
(IBI.ICI)
Min.
Warmwater
Habitat (VWVH)
Modified
Warmwater
Habitat (MWH)
Limited
Resource
Waters
Exceptional"
"Very Good"
"Poor"
'"Very Poor"
LOW-
• BIOLOGICAL INTEGRITY^
HIGH
Figure 6. Relationship between the tiered aquatic life uses in the Ohio WQS and narrative evaluations of biological community performance and how this
corresponds to a qualitative scale of biological integrity and of the biological indices that comprise the Ohio biological criteria. The position of a potential
new Urban Stream Habitat (USH) use designation is indicated (after Yoder et al. 1999).
42
-------
satisfy the desire to afford urban streams the maximum protection practicable, while recognizing the inherent limitations
that the irretrievable effects of urbanization may impose on stream quality. In the meantime, simplistic regulatory and
management approaches should be avoided, particularly in those watersheds where uncertainty about the attainability
of CWA goal uses (i.e., WWH and higher) exists. For example, a single indicator of urban development (e.g., proportion
of impermeable surfaces) is alone insufficient to drive this process. We envision that more refined, multiple indicators
of urban development will provide the necessary sophistication to more appropriately define when this less than CWA
goal use should be applied. In the meantime, management strategies such as the nine minimum controls for CSOs seem
reasonable analogies for the management of urban watersheds and stormwater runoff. However, proceeding beyond
such minimum requirements with long-term remediation plans should be done with deference to the use attainability
issues and with the aid of sufficiently robust before-and-after biological and water quality assessments.
References
Horner, R.R., D.B. Booth, A. Azous, and C.W. May. 1997. Watershed determinants of ecosystem functioning, pp. 251-
274. In C. Roesner (ed.). Effects of Watershed Development and Management on Aquatic Ecosystems, American
Society of Civil Engineers, New York, NY.
Karr, J. R. 1981. Assessment of biotic integrity using fish communities. Fisheries 6(6): 21-27.
Karr, J. R., K. D. Fausch, P. L. Angermier, P. R. Yant, and I. J. Schlosser. 1986. Assessing biological integrity in running
waters: a method and its rationale. Illinois Natural History Survey Special Publication 5: 28 pp.
Neter J., Wasserman W., and Kutner M. H. 1990. Applied Linear Statistical Models: Regression, Analysis of Variance,
and Experimental Designs. Irwin, Homewood, Illinois.
Ohio Environmental Protection Agency. 1998. 1998 Ohio water resource inventory - volume I addendum, summary,
status, and trends. Rankin, E.T., Yoder, C.O., and Mishne, D.A. (eds.), Ohio EPA Tech. Bull. MAS/1998-6-1, Division
of Surface Water, Columbus, Ohio. 190 pp.
Ohio Environmental Protection Agency. 1990. Ohio's nonpoint source pollution assessment. Division of Water Quality
Planning and Assessment.. Columbus, Ohio.
Ohio Environmental Protection Agency. 1991. 1991 Ohio nonpoint source assessment. Ohio EPA. Division of Water
Quality Planning and Assessment.. Columbus, Ohio.
Ohio Environmental Protection Agency. 1989a. Biological criteria for the protection of aquatic life, volume III:
standardized biological field sampling and laboratory methods for assessing fish and macroinvertebrate communities,
Division of Water Quality Monitoring and Assessment, Columbus, Ohio.
Ohio Environmental Protection Agency. 1989b. Addendum to biological criteria for the protection of aquatic life, volume
II: users manual for biological field assessment of Ohio surface waters, Division of Water Quality Planning and
Assessment, Surface Water Section, Columbus, Ohio.
Ohio Environmental Protection Agency. 1987a. Biological criteria for the protection of aquatic life: Volume I. The role
of biological data in water quality assessment. Division of Water Quality Monitoring and Assessment, Surface Water
Section, Columbus, Ohio.
Ohio Environmental Protection Agency. 1987b. Biological criteria for the protection of aquatic life: volume II. users
manual for biological field assessment of Ohio surface waters, Division of Water Quality Monitoring and Assessment,
Surface Water Section, Columbus, Ohio.
Poff, N.L. And J.D. Allen. 1995. Functional organization of stream fish assemblages in relation to hydrological variability.
Ecology 76(2): 606-627.
43
-------
Rankin, E. T. 1995. The use of habitat assessments in water resource management programs, pp. 181-208. in W. Davis
and T. Simon (eds.). Biological Assessment and Criteria: Tools for Water Resource Planning and Decision Making.
Lewis Publishers, Boca Raton, FL.
Rankin, E. T. 1989. The qualitative habitat evaluation index (QHEI), rationale, methods, and application, Ohio EPA,
Division of Water Quality Planning and Assessment, Ecological Assessment Section, Columbus, Ohio.
Schuler, T. R. 1994. The importance of imperviousness. Watershed Protection Techniques 1 : 100 - 111.
Steedman, R.J. 1988. Modification and assessment of an index of biotic integrity to quantify stream quality in southern
Ontario. Can. J. Fish. Aq. Sci. 45: 492-501.
Yoder, C.O. 1995. Incorporating ecological concepts and biological criteria in the assessment and management of urban
nonpoint source pollution, pp. 183-197. in D. Murray (ed.). National Conference on Urban Runoff Management:
Enhancing Urban Watershed Management at the Local, County, and State Levels. EPA/625/R-95/003.
Yoder, C.O. and E.T. Rankin. 1998. The role of biological indicators in a state water quality management process. J.
Env. Mon. Assess. 51(1-2): 61-88.
Yoder, C.O. and E.T. Rankin. 1997. Assessing the condition and status of aquatic life designated uses in urban and
suburban watersheds, pp. 201-227. in Roesner, L.A. (ed.). Effects of Watershed Development and Management on
Aquatic Ecosystems, American Society of Civil Engineers, New York, NY.
Yoder, C.O. and E.T. Rankin. 1995. Biological criteria program development and implementation in Ohio, pp. 109-144.
in W. Davis and T. Simon (eds.). Biological Assessment and Criteria: Tools for Water Resource Planning and Decision
Making. Lewis Publishers, Boca Raton, FL.
Yoder, C.O. and M.A. Smith. 1999. Using fish assemblages in a state biological assessment and criteria program:
essential concepts and considerations, pp. 17-56. in T. Simon (ed.). Assessing the Sustainability and Biological Integrity
of Water Resources Using Fish Communities. CRC Press, Boca Raton, FL.
Yoder, C.O., Miltner, R.J., and D. White. 1999. Assessing the status of aquatic life designated uses in urban and
suburban watersheds, pp. 16-28. in Everson, A. et a/, (eds.). National Conference on Retrofit Opportunities for Water
Resource Protection in Urban Environments, Chicago, IL. EPA/625/R-99/002.
44
-------
Getting Past the Obvious
Robbin B. Sotir, President
Bobbin B. Sotir & Associates, Inc.
Marietta, GA 30064
Abstract
Long Leaf Creek is located within an urbanized watershed along coastal North Carolina. The specific stream reach
addressed is located in a residential subdivision. Conditions had dramatically changed there due to the continued
development of the watershed. The stream had deepened and widened as result of increased runoff and high
concentration events, including hurricanes. This increased the loss of aesthetic value, riparian corridor vegetation, and
aquatic and terrestrial habitats. Water quality was also degraded.
Before they could decide how best to control flooding and stabilize and restore Long Leaf Creek to a naturally
functioning channel within its changed watershed conditions, citizens had to be educated about natural stabilization and
restoration technologies and specificmethods that would work, including conventional options. Soil bioengineering was
agreed upon, with numerous modifications to meet specific needs.
This paper is presented from both the client's and consultant's perspectives. It identifies what worked, what did not
work, and what was necessary to improve the process for successful, long-term results. We present the lessons learned
from criteria issue development and understanding, educational process alternatives preparation, design, construction,
and project results since construction.
Paper
Long Leaf Creek is located within an urbanized watershed along coastal North Carolina. The specific stream reach
addressed is approximately 2000 feet in length and is located in a residential subdivision. It is a highly sensitive project
with a variety of multi-objective goals specific to its location and function and typical to urbanizing areas. The watershed
includes residential, office, institutional, and commercial properties, including 25 homes that line the creek in this area.
Residents living along the creek described the former Long Leaf, as a small, picturesque stream that pleasantly flowed
through their neighborhood—a stream that could be jumped across. It was enjoyed by many people. The conditions had
dramatically changed, however, due to the continued development of the watershed, especially a new road and large
shopping center immediately upstream. Residents have seen their stream deepen by almost ten feet and widen by 40
feet in areas, a result of increased runoff and high concentration events, including hurricanes. In many areas, the banks
were vertical. Large, woody debris filled much of the channel, and many people were now using the "ditch" as a yard and
construction waste dump (see Figure 1). This has resulted in increased the loss of aesthetic value, riparian corridor
vegetation, and aquatic and terrestrial habitats. Water quality has also been degraded.
Many people had already lost property due to stream widening and were unwilling to lose more. Flooding was a major
problem in the downstream end, while erosion was occurring throughout the project reach. The City of Wilmington was
interested in exploring a natural approach to solving the problem. After assessing the site and conditions, and listening
to the residents= concerns and desired solutions, it was clear that a strong, continual working relationship had to be
formed with the neighborhood to ensure project success.
45
-------
Figure 1. Pre-Construction Conditions
46
-------
Before they could decide how best to control flooding and stabilize and restore Long Leaf Creek to a naturally
functioning channel within its changed watershed conditions, citizens had to be educated about natural stabilization and
restoration technologies and the specific methods that would work, including comparative conventional options. To "get
past the obvious," it was clear that almost everyone would have to give up some land and existing trees to solve their
continued land loss and flooding problems and to improve the environmental and aesthetic values of Long Leaf Creek.
How much land and how many trees they would lose would ultimately depend on their selected restoration alternative.
A matrix was developed using critical issues and matching these to possible alternatives (Table 1). Soil bioengineering
was agreed upon, with numerous modifications to meet specific needs.
Bobbin B. Sotir & Associates, Inc., (RBSA), served as the soil bioengineering consultant to the prime, the Kimley-
Horn's interdisciplinary team, developed the geotechnical design and hydraulic eff iciencies of a soil bioengineering solution
to address the desired goals and critical engineering, environmental, and aesthetic issues.
Alternatives were compared with such critical issues as erosion control, streambank stabilization, safer and healthier
environment, flood control, timely project completion, environmental and aesthetic improvement, property loss
minimization, hydraulic efficiency, and cost feasibility.
After an initial investigation, an alternative analysis was produced in the summer of 1997. This alternative analysis
explored numerous approaches to solving each of the project goals, with cost and risk factors assigned to each
alternative. Several alternatives were considered, such as box culverts, 3:1 (horizontal : vertical) grassed slopes, 2:1
riprap rock, 2:1 concrete lining, and soil bioengineered slope systems. With input from the residents and permit
authorities, the City selected the soil bioengineering approach and commissioned a design team to produce plan and
specification documents, including construction cost estimates.
The selected systems employed the use of live fascines, brushlayer/live fascines, joint planting and vegetated
geogrids (see Figures 2 through 5).
The majority of the improvement was done using vegetated geogrid, due to its soil reinforcing capabilities and ability
to reduce land losses (see Figure 6).
Pre-bid, preconstruction, and permit application services were provided to support the project. Construction was
completed by the spring of 1999.
The project has performed well from a biological perspective. Willow, baccharis, and myrtle installed as cuttings in
the lower layer had a survival rate of approximately 80%. The rooted stock installed in the upper two layers comprised
of spirea bush, Carolina allspice, serviceberry, and viburnum, were less successful, with a survival rate of approximately
60% due to an insect infestation (see Figure 7). Hydraulically, we have had some bed scour, accompanied by toe
erosion.
The survival rate of the rooted stock would have been higher had the watering maintenance program been followed.
It is also possible that the insect infestation would not have occurred if the plants had been kept healthier by better
maintenance practices. The rooted plants will be replaced by the contractor under the maintenance agreement. The
contractor is also responsible for taking care of the insect infestation. The bed scour in the upper level caused by
Hurricane Floyd is being handled with check dams to stop the bank from lowering and to control the toe scour.
47
-------
Table 1. Long Leaf Hills/Hewletts Creek Alternatives and Critical Issues
Alternatives
Critical Intermediate 3:1 Side Slopes 2:1 Side Slopes 2:1 Side Slopes Reinforced Soil
Issues Action Grass Lining' Riprap Rock' with Concrete Lining* Box Concrete Bioengineering
Stop Erosion • • • n/a •
& Stabilize
Banks
Clean Out • • • • • •
Trash &
Debris, Remove
Fallen Trees
Safer & • •
Healthier
Area
Control • • • •
Flooding
Timely • • • • •
Project
Completion
Environmental • • •
Improvement
Aesthetically • n/a •
Enhancing
Meets Hydraulic • • n/a •
COE and •
Environmental
Permits
Approval
Probability
Minimize •
Property
Loss
Preliminary $250,000 $640,000 $900,000
Cost Estimate to to to
Range $400,000 $800,000 $1,400,000
$785,000 $1,750,000 $1 ,000,000
to to to
$1,200,000 $2,300,000 $1,300,000
'Does not address geotechnical issues of sandy bank material stability and major land loss requirements.
*Does not address increased safety concerns or reduction in property values.
48
-------
Figure 2. Live Fascine.
Figure 3. Brushlayer/Live Fascine.
LIVE CUT
BRANCHES
LIVE STAKE
JUTVE COT BRANCHES
EXCAVATED TERRACE '
LIVE FASCINE
49
-------
EXISTING
RIPRAP
EXISTING
LIVE STAKE
Figure 4. Joint Planting.
HLL MATERIAL
LIVE BRANCHES
EXISTING
- COMPETENT FOUNDATION -
Figure 5. Vegetated Geogri
50
-------
Figure 6. During Construction.
Figure 7. Three Months After Construction.
51
-------
The project is functioning well from the bank stability and flood control aspects and the stream is operating within the
parameters of the new watershed conditions. It is aesthetically attractive and, over time, should develop some ecological
diversity. In summary, it is clear that the soil bioengineering approach is succeeding. The most important lessons learned
were as follows:
. Learn more about the bed conditions in areas that have had high deposits of mobile materials
. Employ sophisticated grade control structures
• Ensure installation procedures are followed correctly and that materials are not changed
. Keep tabs on the contractor's maintenance schedule
• There is no substitute for communication and cooperation
Reference
Sotir, R.B.; "Brushing Up On Erosion," American Cities and Counties, February, 1998.
52
-------
Protecting and Enhancing Urban Waters:
Using All the Tools Successfully
Eric H. Livingston
Stormwater/NPS Management Section
Florida Department of Environmental Protection
Tallahassee, Florida 32399-2400
Abstract
Reducing the hydrologic effects and pollutant loadings from urban drainage systems and restoring aquatic habitats
to improve the health of our aquatic ecosystems presents many unique challenges. These challenges can be categorized
as technical, institutional, financial, and cultural. This paper will examine each of the challenges and the tools that have
and are being developed to overcome them. A case study on how the tools are being used in Florida to enhance the
health of the Tampa Bay aquatic ecosystem will be presented.
Introduction
Research conducted in Florida during the late 1970s characterized stormwater pollutants, provided cost and benefit
information on many types of stormwater treatment practices, and determined the importance of stormwater discharges
as a major source of pollution. As a result, in 1979, the Florida Environmental Regulation Commission adopted the state's
first stormwater treatment requirements. In 1982, the state's stormwater rule was fully adopted, requiring all new
development and redevelopment projects to include site appropriate BMPs to treat stormwater. This technology-based
program establishes a performance standard of removing at least 80% of the average annual post-development loading
of total suspended solids for stormwater discharged to most waters. Stormwater discharges to the state's most pristine
waters, known as Outstanding Florida Waters, are required to reduce pollutant loading by 95%.
Florida's stormwater treatment program, in combination with the state's wetlands protection, land acquisition, and
growth management programs, has greatly minimized the effects of Florida's rapid growth on its water bodies. However,
land uses and hydrologic alterations that occurred before the mid-l 980s has continued to adversely affect the state's
vulnerable and valuable aquatic ecosystems. Accordingly, the focus of Florida's watershed management program shifted
to cleaning up "older sources" such as existing land uses, whether urban or agricultural, and to integrating program
components to eliminate duplication and improve efficiency and effectiveness. This has led to greater emphasis on more
holistic approaches to address cumulative effects of land use activities within a watershed and to a greater emphasis on
regional structural controls and the purchase or restoration of environmentally sensitive lands. The key institutional
components of this watershed approach have been described in detail (WMI, 1997; Livingston et al., 1995).
Development of Florida's Watershed Assessment Tools
Florida's Water Implementation Rule, Chapter 62-40, F.A.C., establishes a performance standard for reducing, on
a watershed basis, the pollutant loading from older stormwater systems. The goal is to protect, maintain or restore the
beneficial uses of the receiving water body. The amount of needed pollutant load reduction is known as a "Pollutant Load
Reduction Goal or PLRG". The rule further specifies that PLRGs be established as part of the states's priority watershed
program, Surface Water Improvement and Management (SWIM), which is implemented by the state's five water
management districts. Consequently, stormwater PLRGs have been established for several water bodies leading to the
development and implementation of watershed plans to reduce pollutants from urban and agricultural stormwater
discharges. For example, farms within the Everglades Agricultural Area have implemented BMPs to reduce phosphorus
loadings by 40%. Additionally, the federal and state government, SFWMD, and landowners within the EAA are sharing
the cost of constructing tens of thousands of acres of wetlands (Stormwater Treatment Areas) to provide for additional
reduction of phosphorus.
53
-------
Having a sound institutional framework, however, is only one of the tools needed to successfully reduce stormwater
pollutant loadings from existing land uses. Equally important are funding and public education to promote the cultural
change necessary to reduce "Pointless Personal Pollution". A final cornerstone is good science -to establish ecologically
meaningful watershed management goals, evaluate the effectiveness of BMPs and management programs, and assess
the cumulative effects of wet weather discharges.
Unlike traditional point sources of pollution, the effluent quality and environmental effects of stormwater and other
nonpoint sources of pollution are highly variable because of their intermittent, diffuse, land use-specific nature. Of
particular environmental concern is the cumulative impact on a water body from the numerous stormwater/nonpoint
sources within a watershed. Consequently, traditional water quality monitoring and management efforts used for point
discharges generally suffer from several deficiencies when trying to understand and manage stormwater/NPS pollution.
These deficiencies include difficulty in:
1. Assessing intermittent, shock loadings of pollutants.
2. Assessing cumulative impacts of multiple sources.
3. Comparing water bodies and establishing priorities for management actions.
4. Assessing hydrologic, geomorphological, and habitat alterations within a watershed.
5. Distinguishing actual or potential problems from perceived problems.
6. Discriminating anthropogenic loadings from natural watershed loadings of metals and nutrients.
7. Establishing cost-effective ways to assess pollution sources and trends on a watershed basis.
To overcome these problems, the Florida Department of Environmental Protection (DEP) has developed cost-effective
sediment and biological monitoring tools that are much better suited for assessing cumulative effects than traditional water
chemistry monitoring (Livingston et. al, 1995; McCarron et al., 1997). Most stormwater pollutants accumulate overtime
in sediments, not the water column. Therefore, the sediments and the organisms that reside in them offer an in-situ
monitoring opportunity to determine the cumulative effects of watershed stormwater/NPS pollution sources on aquatic
systems or to evaluate the effectiveness of management programs.
Sediment quality is a sensitive indicator of overall environmental quality. Sediments influence the environmental fate
of many toxic and bioaccumulative substances in aquatic ecosystems (Long and Morgan, 1990). Sediments tend to
integrate contaminant concentrations over time and may represent long-term sources of contamination. Specifically,
sediment quality is important because many toxic contaminants found in only trace amounts in water can accumulate to
elevated levels in sediments. Sediment-associated contaminants can also directly affect benthic and other
sediment-associated organisms since sediments provide benthic and pelagic communities suitable habitats for essential
biological processes (e.g. spawning, incubation, rearing, etc.).
Sediments provide an essential link between chemical and biological processes. By understanding this link,
environmental scientists can develop assessment tools and conduct monitoring programs to more rapidly and accurately
evaluate the health of aquatic systems (Pardue et al., 1992). Therefore, sediment quality data provide essential
information for evaluating ambient environmental quality conditions in water bodies. Additionally, information about the
amount and quality of sediments within stormwater systems, stormsewers and other stormwater conveyances can help
trace pollution sources, prioritize areas for implementing control measures, and help to assure proper disposal of
accumulated sediments.
Inclusion of biological community monitoring allows a more holistic, systems approach that greatly enhances surface
water quality assessment and management (Yoder, 1989; Yoder and Rankin, 1997). In particular, it allows assessment
of the degradation of habitat (e.g., channel and bank erosion) and siltation within water bodies, neither of which are
54
-------
detected by water chemistry sampling, and both of which are typically associated with wet weather discharges. While
chemical data reflect short-term conditions that exist when a particular sample is collected, biological communities
accurately indicate overall environmental health because they continuously inhabit receiving waters where they integrate
a variety of environmental influences - chemical, physical and biological.
Biological assessment involves an integrated analysis of functional and structural components of aquaticcommunities
(Karr and Dudley, 1981; Karr, 1991). Bioassessments are best used to detect aquatic life impairments and assess their
relative severity. Once an impairment is detected, additional chemical and biological toxicity testing can identify the
causative agent and its source. Both biological and chemical methods play critical roles in successful pollution control and
environmental management programs. They are complementary, not mutually exclusive, approaches that enhance overall
program effectiveness.
A fundamental part of bioassessments is "metrics" (Karr and Chu, 1998). Just as a doctor uses metrics such as blood
pressure and heart rate to assess human health, biological metrics allow the ecologist to use meaningful indicator
attributes to assess the status of communities in response to perturbation. The definition of a metric is a characteristic
of the biota that changes in some predictable way with increased human influence (Barbour et al., in review). By using
multiple metrics to assess biological condition, the information available about the elements and processes of aquatic
communities is maximized. The validity of an integrated assessment using multiple metrics is supported by the use of
measurements of biological attributes firmly rooted in sound ecological principles (Fausch et al. 1990; Lyons 1992).
In 1983, the DEP began developing assessment tools that could be used to assess stormwater and NPS effects and
the effectiveness of management programs, practices, and activities. The first efforts focused on estuarine sediment
assessment tools (FDER, 1988; Schropp et al., 1989,1990; MacDonald, 1994). In 1989, efforts began to modify national
bioassessment protocols (EPA, 1989) to develop quantitative bioassessment protocols for Florida (Griffith et al., 1994;
Barbour et al., 1996, FDEP, 1996). The tools that have been developed and that are under development (noted by a *)
include:
A. Sediment assessment tools
1. Standardized sediment collection and analysis protocols.
2. Estuarine normalization of metal concentrations to aluminum concentrations ratio.
3. Estuarine sediment quality assessment guidelines
4. Freshwater normalization of metal concentrations to aluminum concentrations ratio(*).
5. Freshwater sediment quality assessment guidelines(*).
B. Bioassessment tools:
1. Stream Condition Index
2. Lake Condition Index (* - nearly completed)
3. Wetland bioassessment methods (* - in early stages of development)
4. Canal bioassessment methods (* - in middle stages of development)
5. Estuarine bioassessment methods (* - in early stages of development)
Using the Tools for Watershed Management
Sediment assessment, together with watershed characterization, future land use plans, stormwater master plans,
and mapping of pollution sources, can be used to screen watersheds and sub-basins to determine potential "hot spots".
55
-------
Bioassessment and water chemistry sampling can then be done to assess the actual health of the aquatic system in these
locations. An important component of the bioassessment is habitat quality, especially since urbanization often leads to
dramatic changes in stream hydrology, geomorphology, riparian zones, habitats, and ultimately biological communities.
Possible outcomes of the bioassessment are: (1) no biological effects; (2) effects due to habitat degradation; (3) effects
due to sediment or water quality; or (4) effects due to a combination of sediment, water quality and habitat degradation.
Bioasssessments also allow the establishment of an ecologically-based aquatic resource goal, rather than one based
solely on traditional water chemistry standards to which the applicability to wet weather discharges is questionable. Once
an aquatic resource goal is established, then relationships between the needs of the biological community and the
chemical, sediment, and habitat influences on the community can be established. The next step is to then use all of this
information to develop a watershed management plan that includes specific projects, schedules, and funding sources that
will ultimately result in the acheivement of the desired aquatic ecological goal. Much of this effort can be coordinated and
implemented using the rotating basin approach that many states are now using in combination with the NPDES
stormwater permitting program.
Putting the Puzzle Pieces Together: Tampa Bay Case Study
Puzzle Piece 1: The Assessment- As part of the development of the Tampa Bay SWIM Plan (SWFWMD, 1992) and
the Tampa Bay NEP Comprehensive Conservation and Management Plan (TBNEP, 1996), existing environmental
information was assessed to determine the ecological health of the bay system. Major findings of these assessments are
summarized below.
Habitats
1. Since 1950, about half of the bay's natural shoreline and 40% of its seagrasses have been destroyed.
2. In 1950, the bay's shallow shelf supported about 40,000 acres of seagrasses. By 1982, only 21,600 acres
remained and virtually all of Hillsborough Bay's 2,700 acres were gone. Seagrass decline is due to dredging and
filling for residential development, turbidity caused by dredging of the main shipping channel, and reduced light
penetration caused by shading by algae fueled by excess nutrient discharges.
3. Since the early 1900s approximately 13,200 acres of bay bottom (3.6% of the bay's surface area) were filled, with
over 90% of the activity occurring along the bay's shallow shelf where seagrasses once thrived. The surface area
of Hillsborough Bay has been reduced by 14%.
4. Upgrading sewage plants in the 1980s to provide advanced wastewater treatment reduced nitrogen loadings,
leading to a decline in phytoplankton, an increase in water clarity, and greater light penetration. Consequently,
between 1982 and 1992 seagrass coverage increased by about 4,000 acres (18.5%) raising the bay's total
acreage to over 25,600 acres.
5. About 43% (9,700 acres) of Tampa Bay's original saltwater wetlands were lost between 1950 and 1990, primarily
because of dredging and filling for waterfront development. However, as many as 5,900 acres of new wetlands
formed along causeways and other emergent land created by dredged spoil material during this period. Recent
estimates of wetland habitat in Tampa Bay indicate that about 18,000 acres of mangroves and saltmarsh remain
but many thousands of acres are damaged by invasion by exotic plants such as Brazilian pepper.
Fish and Wildlife
1. Between 1966 and 1990, the harvest of 11 commercial species offish declined by 24%, primarily because of
smaller catches of mullet and sea trout. Each of these species is dependent on seagrass habitats.
2. Harvest of spotted sea trout declined by 86% between 1950 and 1990, from 487,000 pounds to 67,000 pounds.
Similarly, red drum harvests plummeted from 80,000 pounds in 1950 to 15,000 pounds in 1986.
56
-------
3. Tampa Bay's once-thriving shellfish industry has virtually collapsed, except for bait shrimping. Harvests of clams
or oysters are restricted or prohibited throughout the bay because of high bacterial levels associated with
stormwater discharges and septic tanks. The bay scallop, a highly pollution sensitive organism, all but
disappeared from the bay in the 1960s.
Water and Sediment Quality
1. While water quality has improved over the past ten years, primarily as a result of better wastewater treatment,
water clarity, nutrients, and toxics continue to be a problem.
2. Because of natural circulation and flushing from the Gulf of Mexico water clarity is greatest in the lower part of
Tampa Bay (2.5 m), and naturally decreases moving up the bay, dropping to an average of 2 meters (6.6 feet)
in Middle Tampa Bay and Old Tampa Bay. The lowest average water clarity is in Hillsborough Bay (1.5 m) which
has poor circulation and receives a larger share of nutrients and sediments from major rivers.
3. Excessive amounts of nitrogen continue to accelerate algal growth which subsequently reduces light penetration
to seagrasses and contributes to oxygen depletion. The bay's total annual nitrogen load was estimated to be 2.5
times greater in 1976 than the load computed for 1985 to 1991 (Figure 1).
4. Recent studies by NOAA, in cooperation with FDEP, provide an excellent overview of the levels and distribution
of toxics in bay sediments (Long, et. al. 1991, 1994; FDEP, 1994). Compared to other urban estuaries, Tampa
Bay has low-to-moderate levels of most toxic parameters. Contamination is centered around large urban centers,
ports and marinas, and concentrations generally decrease from the top of the bay toward the Gulf of Mexico.
5. Generally, the highest levels of sediment toxic contamination occur in Hillsborough Bay, the bay's most
industrialized area and home to the state's busiest port. Upper Hillsborough Bay has the highest levels of
cadmium, copper, mercury, zinc, and lead, as well as the pesticide DDT. Concentrations in sediments at a site
in northern Hillsborough Bay were the highest of any toxic pollutant measured in Tampa Bay. Two other bays with
heavily urbanized watersheds, Boca Ciega Bay and Bayboro Harbor, also can be considered as hot spots of toxic
contamination.
6. Figure 2 shows sites in Tampa Bay where concentrations of toxic contaminants in sediments exceeded Florida's
Probable Effects Level (PEL) and No Observable Effects Level (NOEL). Sites above the PEL indicate a high
probability for biological impact to marine organisms while those above the NOEL are considered "at risk" to
biological impact (MacDonald, 1994).
Puzzle Piece 2: The Goal: A critical component of watershed management is using biological living resources as a
measure of a water body's health, with far less emphasis on traditional laboratory-based water quality standards. This
approach addresses critical ecological effects that are not seen by water chemistry standards, allows greater flexibility
to achieve the desired ecological goals, and provides taxpayers with a better benchmark to judge the return on their
expenditures. Through the SWIM Plan and the Tampa Bay NEP CCMP, the primary overall goals have been established
for the restoration and protection of Tampa Bay:
1. To reverse the environmental degradation of the Tampa Bay estuarine system.
2. To optimize water quality and other habitat values, thereby promoting the sustained existence or re-establishment
of thriving, integrated, biological communities.
3. To ensure the maintenance of a productive, balanced ecosystem complimentary with human needs and uses of
the resources.
57
-------
Figure 1. Tampa Bay Total Nitrogen Loading and Sources
Hilltborough Bay
121 tons/
S§322 tons/ 17.3%
236 tons/ 12.7%
195 tons/ 10.5%
Total 1,856 tons
Old Tampa Bay
172 tons/ 34.5%
250 tons/ 50.1%
77 tons/ 15.4%
ton/ 0%
Total 499 tana
Middle Tampa Bay
340 tons/ 46.2%
339 tons/ 46.1%
26 tons/ 3.7%
ton/ 0%
Total 70S ton*
Boca Clega Bay
: 83 tons/ 43.2%
: 106 tons/ 55.2%
3 tons/ 1.6%
Total 192 tons
Middle Tampa Bay
^J BaeaCiega
Boy
• 10 tons/ 29.4%
Li 20 tons/ 58.6%
H 4 ton*/11
Total 34 ton*
Lower Tampa Bay
1 ton/ 0.3%
302 tons/ 82.7%
36 tons/ 9.9%
26 ton*/ 7.1%
ton/ 0%
Total 365 tona
tmw Tampa Bay ;
r *•?. "
• Stormwater Runoff
U Atmospnaric Deposition
Industrial & Municipal Point Sources
Fertilizer Material Losses
Springs & Grouodwater
364 tons/ 63.3%
53 tons^ 8.7%
170 tons/ 26.0%
Total 607 tons
Figure 1. Tampa Bay Total Nitrogen Loading and Sources.
58
-------
Figure 2. Sediment Toxic Hot Spots in Tampa Bay
Figure 2 shows sites in Tampa Bay where concentrations of toxic
contaminants in sediments have exceeded Florida's Probable Effects
Level (PEL) and No Observable Effects Level (NOEL) for biological
impact. Sites registering above the PEL indicate that some biological
impact to marine organisms is likely. Sites registering above the NOEL
are "at risk" to biological impact.
= -A-
1
A
Tampa -^j
St. Petersburg
Tampa
4
«
A
*%••
, A 1
Sites exceeding
Probable Effects Level
Sites exceeding No
Observable Effects
Level
Figure 2. Sediment Toxic Hot Spots in Tampa Bay.
59
-------
To achieve these overall goals the following specific goals have been established:
1. The overall goal is to restore seagrasses to 1950s levels. This will lead to restoration of commercially important
species such as the bay scallop, mullet, sea trout, and red drum.
2. To restore seagrasses to 14,000 acres of the bay. The ability of seagrasses to recolonize the bay depends on
the amount of sunlight the grass species require, as well as shading factors such as the amount of drift macro-
algae and attached algal growth on grass blades. For most seagrasses in the bay, an estimated 20% to 25% of
the light striking the bay's surface must penetrate to target depths to allow seagrass regrowth. Reducing nitrogen
loadings will reduce chlorophyll a concentrations thereby increasing the depth of sunlight penetration.
3. As many as 12,000 acres of seagrass can be recovered by maintaining recent water quality conditions. This will
require local communities to reduce their nitrogen loadings to the bay by about 10% by the year 2010 to
compensate for increases in nitrogen loadings associated with the watershed's population growth.
4. A coastal habitat master plan has been developed for the watershed that will help to coordinate and prioritize
existing state, regional, and local restoration programs. The long term goal is to recover 1,800 acres of low-
salinity tidal marshes while maintaining and enhancing salt marshes and mangroves at existing levels. A minimum
goal is to restore 100 acres of tidal marsh habitat every five years.
5. Reduce sediment toxicity to minimize risks to marine life and humans. Using three tests -evaluation of sediment
chemistry, sediment toxicity, and benthic community health - bay sediments will be characterized and prioritized
for management.
6. Reduce bacterial contamination to levels safe for swimming and shellfish harvesting.
Puzzle Piece 3: The PLRG: To achieve the ecological goal, nitrogen loadings would need to be held to those
occurring in 1992-94 meaning that the 17 tons of nitrogen loading that would accompany projected growth within the
watershed would have to be compensated. Reduction of nitrogen would reduce chlorophyll a levels which would increase
how deep the minimum levels of light needed for seagrass growth would penetrate the water column.
Puzzle Piece 4: Quantifying Pollution Sources: Before load reductions can be achieved, it is essential that the sources
and relative contribution of the sources be quantified.
Figure 3 summarizes sources of nonpoint nitrogen loadings to Tampa Bay.
1. Stormwater runoff from the Tampa Bay watershed contributes about 47% of the bay's total annual nitrogen load
with urban runoff accounting for about 16%, or 680 tons. Residential areas, the watershed's largest land use, is
responsible for over half of the nitrogen loading while commercial/industrial sites account for about 20%.
2. About 28% of the bay's total nitrogen loadings, or 1,200 tons, are from atmospheric pollutants falling directly on
the water. An additional 7,500 tons fall in the watershed, although no one can determine how much enters the
bay via stormwater. EPA estimates that as much as 67% of the bay's total nitrogen load may be from the
atmosphere.
3. Stationary sources, primarily power plants, contribute an estimated 50% of the anthropogenic NO, emissions as
compared to 35% from motor vehicles.
4. Domestic wastewater discharges still discharge about 8% (340 tons) of the bay's total annual nitrogen loadings,
even though all plants provide AWT. Hillsborough Bay receives about two-thirds of the cumulative nitrogen load
from the 36 billion gallons of effluent discharged to Tampa Bay each day.
5. Industrial wastewater discharges, primarily fertilizer manufacturing and shipping facilities, are responsible for
about 6% of the bay's total annual nitrogen loadings.
60
-------
Figure 3. Total Nitrogen Loadings
Per-Acre Nitrogen Loadings
from Non-Point Sources
Ruidential
Commercial Industrial/Institutional
Mining
Range and Pasture
Intensive Agriculture
Undeveloped Land
% Loading
11
5
4
13
6
8
* Watershed
15.5
6.4
3.2
20.4
6.5
39.93
Yield Iba/oc/yr
4.52
5.26
4.97
2.81
5.63
1.15
Total nitrogen Loadings to Tampa Bay (1882-1894 average)
6%
Industrial
Wastewater
26%
Atmospheric
Deposition
8%
Municipal
Wastewater
6%
Accidental
Fertilizer Losses
4%
Groundwater
4%
Mining
47%
Stormwater
Runoff
5%
Commercial/
Industrial Runoff
8%
Undeveloped
Land
6%
Intensive
Agriculture
13%
Pasture/Range
Lands
Figure 3. Total Nitrogen Loadings.
6. Septic tanks, which serve about 20% of the watershed's population, are another important source of nitrogen and
pathogen loadings, especially in some areas such as Allen's Creek and tributaries to McKay Bay.
7. Another 7% of the bay's total nitrogen loadings had been attributed to losses of fertilizer during ship loading and
en route to port. However, this figure has declined substantially since 1991 as source control BMPs were
implemented at the port.
61
-------
8. More than 60% of the bay's annual loadings of chromium, zinc, mercury and lead, as well as significant amounts
of petroleum hydrocarbons and pesticides are conveyed by stormwater.
9. Atmospheric deposition also is a major source of toxic substances accounting for 44% of the bay's total cadmium
loading and about 17% of the copper and lead loadings. PAHs also enter the bay from the atmosphere.
10. Industrial and domestic point sources also contribute about 30% of the bay's total loadings of aresenic, cadmium,
chromium, and copper.
Puzzle Piece 5: The Watershed Management Plan: The state's Surface Water Improvement and Management
(SWIM) program was established by the legislature in 1987. Tampa Bay was named in the SWIM Act as a priority
waterbody within the Southwest Florida Water Management District and a SWIM Plan was adopted in 1992. In 1990,
Tampa Bay was adopted into the National Estuary Program by EPA leading to the development of Charting the Course,
a Comprehensive Conservation and Management Plan (CCMP) for the bay. Community participation was an essential
component of the development of both watershed plans. In particular, many of the agencies, citizen groups, and others
long active in the restoration and management of Tampa Bay participated in the development of the CCMP. The CCMP
built on many of the region's ongoing environmental programs, from land acquisition to urban stormwater retrofitting to
habitat restoration. It also identified where unneccesary duplication existed in current environmental programs and
provided recommendations to ensure that limited public funds are spent in the most environmentally effective manner.
Puzzle Piece 6:Setting Priorities:!'o assure that limited financial resources were used judiciously, restoration projects
and programs were prioritized and targeted to specific subbasins. This was done by combining GIS analysis with
watershed modeling and characterization with the results of the sediment, biological, habitat, and chemical assessments.
For example, an essential early activity of the SWIM program was a watershed-wide assessment of pollution sources,
especially stormwater, to identify "hot spots" - subbasins with high stormwater loadings - and to prioritize urban
stormwater retrofitting projects (SWFWMD, 1990). Similarly, priority habitat restoration sites were also selected as were
projects to reduce overall nitrogen loadings to the bay, including those from atmospheric deposition.
fuzz/e Piece 7: Action Hans: A successful watershed management plan must include a specific set of actions that
will be taken within a specified time. Charting the Courseincludes 41 specific actions that are needed to achieve the plan's
goals. These include the construction of numerous urban stormwater treatment and habitat restoration projects that have
been built, are underway, or are planned in priority subbasins (Figure 4). Since vacant land in the highly urbanized area
is scarce or extremely expensive, many of these projects are being conducted on existing public lands providing multiple
benefits including regional stormwater management, open space, and recreation. Public education is a frequent
component of these projects with the placement of signs depicting the effects of urbanization, the importance of wetlands
and riparian vegetation, and the need for stormwater treatment and habitat restoration.
Programs and actions that rely upon nonstructural BMPs are also being used to reduce "Pointless Personal Pollution"
at its source and increase the effectiveness of existing programs. For example, since surveys showed that up to 70% of
the stormwater BMPs serving new development are not being properly maintained, assuring their long-term operation
and maintenance can greatly reduce stormwater pollution. Maintenance and operation of BMPs typically is the legal
responsibility of private land owners and property owner associations. Unfortunately, DEP, SWFWMD, and local
governments do not have enough staff to conduct regular inspections. To improve this deficiency, DEP is implementing,
in cooperation with local governments and the WMDs, a training and certification program for public and private sector
individuals involved in erosion, sediment, and stormwater inspections. Local governments also are encouraged to
implement Stormwater Operating Permit systems which require an annual inspection and certification that the stormwater
system has been maintained and is properly operating. As an economic incentive, some local stormwater utilities provide
credits for individuals served by a properly maintained and operating system. Additionally, Hillsborough County has
implemented the "Adopt a Pond" program to help educate stormwater system owners on how to maintain their systems.
Other nonstructural efforts include assisting businesses in developing and implementing pollution prevention plans
and the continued implementation of the Florida Yards and Neighborhoods Program. This program is being expanded
to help develop model landscaping guidelines for commercial landscapes, and promote the incorporation of FYN
62
-------
Figure 4. Active SWIM Habitat Restoration and Stormwater
Retrofitting Projects
Figure 4. Active SWIM Habitat Restoration and Stormwater Retrofitting Projects.
landscaping guidelines into local government site review processes for new development. The region's continuing rapid
growth provides opportunities through local government comprehensive plans and land development regulations to
promote compact development and to reduce impervious surfaces, especially parking lots at commercial developments.
Puzzle Piece 8: Assuring Implementation: An important aspect of any watershed management program is an
institutional framework that assures that all of the partners will implement their responsibilities. In 1998, an Interlocal
Agreement was signed by the Tampa Bay NEP's local government and regulatory implementation partners. The
agreement requires the partners to submit detailed plans describing how they will fulfill their responsibitlities for bay
restoration and protection. Additionally, all of the local governments within the watershed have been issued, either as
individual permittees or co-permittees, NDPES muncipal Stormwater permits. These permits include specific requirements
that are identified in the CCMP's action plans.
63
-------
Active Habitat Restoration Projects
Active Stormwater Retrofitting Projects
1. Lowry Park
2. NE McKay Bay
3. Delaney Creek
4. Simmons Park
5. Hendry Fill
6. Peanut Lake
7. Bayshore Blvd.
8. Gandy Park
9. Cabbage Head Bayou
10. Boca Ciega
11. Cargil S. parcel
12. Mangrove Bay
13. Cockroach Bay
14. Little Bayou
15. MacDillAFB
16. Picnic Island
1. Lowry Park
2. Horizon Park
3. Old Coachman
4. S. Pasadena
5. Jungle Lake
6. Pinellas Park
7. N. Redington
8. EMS Site
9. St. Pete/Clearwater Airport
10. Brushy Creek
11. Safety Harbor
12. 102nd Avenue
13. 94th Avenue
14. Lake Carroll
15. Delaney Creek
16. Haynsworth
17. 141st Avenue
Puzzle Piece 9: Funding Implementation: Partners in the implementation of the CCMP include EPA, Florida DEP,
SWFWMD, local cities and counties, and the private sector. Each of the partners has made a substantial commitment
of financial resources since 1995 to accomplish the desired aquatic ecological goals. Primary funding sources have
included the P2000 (state and local land acquisition funds), the SWIM Program (state and SWFWMD funds), SWFWMD
Basin Boards, the private sector, and local stormwater utility fees. In many cases, funding for projects is from a
combination of sources that often allow the leveraging of other funds needing nonfederal matching funds such as those
from the Section 319 nonpoint source implementation grant program.
Costs associated with the individual actions presented in the Tampa Bay SWIM Plan and the NEP CCMP are
considerable. However, these should not automatically be construed as requirements for new sources of revenues, since
some of these initiatives can and are being accomplished with existing resources or by redirecting current funding
allocations to better address the bay's needs. A number of actions seek to improve coordination, cooperation, and
planning among state and local governments, and the private sector. These may actually result in cost savings for
currently funded activities.
A 1994 survey by the Tampa Bay NEP attempted to quantify how much money is spent to manage and monitor bay
quality and administer environmental programs. Based on FY94-95 budgets, the study indicates that over $260 million
is spent annually by federal, state, and local agencies on the restoration and management of Tampa Bay. As seen in
Figure 5, the largest par-t of the funds (65% or $170 million) are spent on wastewater collection, treatment, and reuse.
Figure 5. Funds Spent
e.r%
Habitat Restoration
4 Msnag«m*nt
3.7*
Land Acquisition
S.2%
RsguMlon t
Enforctmsnt
0.5%
Public Avnrann»
O.D%
MnHnlMntlon
Figure 5. Funds Spent.
64
-------
Approximately $35 million (13%) is spent by local governments and the SWFWMD on stormwater management.
Habitat restoration and land acquisition, two long favored and implemented environmental programs in the Tampa Bay
region, account for over $27 million in expenditures.
Puzzle Piece 10: Monitoring Implementation: Since the adoption of Charting the Course in 1996, numerous projects
and programs have been implemented. The Tampa Bay NEP recently issued the first official progress report on
implementation (TBNEP, 1999). It shows that the program's partners are on or ahead of schedule in achieving most of
the priority goals for bay improvement. Highlights of implementation activities that occurred between 1995-99 include:
. Goal: Recover an additional 12,350 acres of seagrassover 1992 levels, while preserving the bay's existing 25,600
acres.
Status: Since 1988, seagrass acreage is increasing at about 500 acres per year meaning the goal will be
reached in 25 years.
• Goal: Restoring and protecting bay habitats
Status: A total of 250 acres of low-salinity habitat will be restored in all bay segments, exceeding the five year
target by 150 acres. Additionally, a total of 1,340 acres of mangrove and saltmarsh habitat have been
restored. All 28 priority sites identified in the habitat master plan have been given the highest priority for
acquisition under the state's land-buying programs.
. Goal: "Hold the line" at nitrogen loadings estimated from 1992-94.
Status: When fully implemented, the 105 projects constructed, underway, or planned will reduce the amount of
nitrogen entering the bay by an average of 134 tons per year, exceeding the target by 60%.
In addition to progress on the above goals, very good progress has been made on other goals. These include:
. The goal of protecting the endangered manatee population in Tampa Bay. The Manatee Awareness Coalition has
implemented "Manatee Watch" where trained volunteers help educate and encourage boaters to go slow in waters
frequented by manatees.
. The goal of returning bay scallop to Tampa Bay. Stocking programs are adding bay scallops and citizen volunteers
are measuring the effectiveness of these efforts through the Great Bay Scallop Search.
• The goal of reducing atmospheric deposition into and onto Tampa Bay. To better understand the linkage between
air pollution and water quality, eight research and monitoring programs addressing atmospheric deposition are
underway.
. The goal of making Tampa Bay safe for shellfish harvesting and swimming. As part of the "Healthy Beaches
Project", research is underway to identify and test better indicators of microbial contamination, the prevalence of
the indicators at bay and gulf beaches, and probable sources of the contamination.
. The goal of providing flows necessary to support plant and animal communities below the dam on the Hillsborough
River. In February 1999, the SWFWMD approved a draft minimum flow rule for the Hillsborough River that hopefully
will provide a basis for resolving conflicts over competing uses of the river.
. The goal of developing a long term dredge material management plan for Tampa Bay. A Dredged Material Advisory
Committee is being organized in partnership with the U. S. Army Corps of Engineers.
Discussion and Recommendations
Florida has established a wide variety of laws, regulations and programs at the state, regional and local level to
protect, manage and restore the state's incredibly valuable yet vulnerable natural resources, especially its water
65
-------
resources. There is no doubt that these programs have been effective in helping to reduce adverse impacts on natural
resources resulting from the state's rapid and continuing growth over the past twenty years. However, even with the
implementation of these programs, many of Florida's natural resources have been severely strained or degraded. Some
of these adverse effects can be attributed to activities that occurred before the implementation of modern watershed
management programs such as the channelization of the Kissimmee River and the creation of the vast drainage canal
network south of Lake Okeechobee both of which are contributing to the decline of Lake Okeechobee, the Everglades
and Florida Bay. Other adverse impacts, though, are directly related to the state's rapid growth and development during
the last twenty years. These include water supply problems, water quality problems, declining habitat and impacts on
endangered species such as the manatee and the Florida panther.
Why are these adverse impacts still occurring given the wide range of watershed management programs that have
been implemented in Florida? What could be done to reduce these effects and possibly restore already degraded areas?
The continuing evolution of Florida's land and water management programs into a more holistic approach which seek to
manage cumulative effects can help to overcome many of the current program deficiencies. Cooperative efforts and
partnerships, together with citizen education and involvement to improve the stewardship ethic of all Floridians is
essential. With increased support and participation by all Floridians, the effectiveness of the state's programs can be
improved helping to assure that our natural resources will be able to be enjoyed by future generations.
References
Barbour, M.T., J. Gerritsen, and J.S. White. 1996. Development of a Stream Condition Index (SCI) for Florida. Prepared
for Stormwater/NPS Management Section, FDEP, Tallahassee, Florida.
Barbour, M.T., J.B. Stribling, and J.R. Karr, (in press). The Multimetric Approach for Establishing Biocriteria and
Measuring Biological Condition. |n W. Davis, T. Simon (eds.), Biological Assessment and Criteria: Tools for Water
Resource Planning and Decision Making. Lewis Publishers.
Barbour, M.T., J.B. Stribling, and J.R. Karr, (in review). Biological Criteria: Technical Guidance for Streams. U.S.
Environmental Protection Agency, Office of Science and Technology, Health and Ecological Criteria Division, Washington,
DC.
EPA. 1989. Rapid bioassessment protocols for use in streams and rivers: Benthic macroinvertebrates and fish.
EPA/444/4-89/001. Washington D.C.
Fausch, K.D., J. Lyons, J.R. Karr, P.L. Angermeier. 1990. Fish communities as indicators of environmental degradation.
Am. Soc. Symp. 8:123-44.
FDER. 1988. A guide to the interpretation of metal concentrations in estuarine sediments. Florida Dept. of Env.
Regulation. Office of Coastal Zone Management. Tallahassee, Fl.
FDEP. 1994. Florida coastal sediment contaminants atlas and technical volume. Florida Dept. of Environmental
Protection. Sediment Research Group. Tallahassee, Fl.
FDEP. 1994. Florida Coastal Sediment Contamination Atlas. Office of Water Policy, Tallahassee, Florida.
FDEP. 1996. Standard Operating Procedures for Biological Assessments. Bureau of Laboratories, Tallahassee, Florida.
Griffith, G, J. M. Omernik, C. Rohmand, S. Pierson. 1994. Florida regionalization project. USEPA Environmental Research
Laboratory. Corvallis, Oregon. Final Report Prepared for Fla. Dept. Environ. Protection. Tallahassee, Fl.
Karr, J.R. 1991. Biological integrity: A long-neglected aspect of water resource management. Ecological Applications 1:66-
84.
Karr, J. and D. Dudley. 1981. Ecological perspective on water quality goals. Environ. Mgmt. 5(1): 55-68.
66
-------
Karr, J.R., and E.W. Chu. 1998. Restoring Life in Running Waters: Better Biological Monitoring. Island Press, Washington,
D.C.
Livingston, E.H. 1995. The evolution of Florida's stormwater/watershed management program. In Proceedings of the
National Conference on Urban Runoff Management: Enhancing Urban Watershed Management at the Local, County, and
State Levels. EPA 625/R-95/003. Cincinnati, Ohio.
Livingston, E.H., E. McCarron, T. Seal, and G. Sloane. 1995. Use of sediment and biological monitoring. In Stormwater
NPDES Related Monitoring Needs, Proceedings of an Engineering Foundation Conference. ASCE, New York.
Long, E.R. and Morgan, L.G. 1990. The potential for biological effects of sediment-sorbed contaminants tested in the
National Status and Trends Program. NOAA Tech. Memo NOS OMA 52.
Long, E. R., D. MacDonald, and C. Cairncross. 1991. Status and Trends of Toxicants and the Potential for Their Biological
Effects in Tampa Bay, Florida. NOAA Technical Memorandum NOS OMA 58.
Long, E. R., D.A. Wolfe, R.S. Carr, K.J. Scott, G.B. Thursby, H.L. Windom, R. Lee, F.D. Calder, G.M. Sloane, and T.
Seal. 1994. Magnitude and extent of sediment toxicity in Tampa Bay, Florida. NOAA Technical Memorandum NOS ORCA
78.
Lyons, J. 1992. Using the index of biotic integrity (IBI) to measure environmental quality in warmwater streams of
Wisconsin. General Technical Report, NC-I 49. U.S. Department of Agriculture, Forest Service. St. Paul, Mn.
MacDonald, D. 1993. Development of an approach to the assessment of sediment quality in Florida coastal waters. Final
report submitted to FDER/SRG. Tallahassee, Fl.
MacDonald, D. 1994. Approach to the Assessment of Sediment Quality in Florida Coastal Waters. Final report submitted
to the Florida Department of Environmental Protection, Tallahassee, Florida.
McCarron, M. E., E. H. Livingston, and R. Frydenborg. 1997. Using bioassessments to evaluate cumulative effects. In
Effects of Watershed Development and Management on Aquatic Systems, Proceedings of an Engineering Foundation
Conference. ASCE, New York.
Pardue, J., R. DeLaune, and W. Patrick, Jr. 1992. Metal to aluminum correlation in Louisiana coastal wetlands:
Identification of elevated metal concentrations. J. Environmental Quality 21: 539-545.
Schropp, S., F.Calder, L. Burney, and H. Windom. 1989. A practical approach for assessing metals contamination in
coastal sediments - An example in Tampa Bay. In:
Proceedings of the Sixth Symposium on Coastal and Ocean Management. July 11-1 4, 1989. Published by ASCE.
Charleston, S.C.
Schropp, S., F. Lewis, H. Windom, J. Ryan, F. Calder, and L. Burney. 1990. Interpretation of metal concentrations in
estuarine sediments of Florida using aluminum as a reference element. Esturaries 13(3): 227-235.
Southwest Florida Water Management District. 1990. Urban stormwater analysis and Improvements for the Tampa Bay
watershed. Brooksville, Florida.
Southwest Florida Water Management District. 1992. Tampa Bay Surface Water Improvement and Management Plan.
Brooksville, Florida.
Tampa Bay National Estuary Program. 1996. Charting the Course for Tampa Bay: Final Comprehensive Conservation
and Management Plan, St. Petersburg, Florida.
Tampa Bay National Estuary Program. 1999. Bay Guardian: News of the Tampa Bay National Estuary Program. Summer
1999. St. Petersburg, Florida.
67
-------
Yoder, C. 0.1989. The development and use of biological criteria for Ohio surface waters. Pp 139-1 46. In: Water Quality
Standards for the 21 st Century.
Yoder, C. 0. and E. T. Rankin. 1997. Assessing the condition and status of aquatic life designated uses in urban and
suburban watersheds. In Effects of Watershed Development and Management on Aquatic Systems, Proceedings of an
Engineering Foundation Conference. ASCE, New York.
68
-------
Urban Stream Structure And Selection Of Structures To Build Habitat
To Support Wild Fish Populations
Douglas T. Severn
URS Greiner Woodward-Clyde
Seattle, Washington
Abstract
This paper gives a brief overview of the physical impacts of urbanization on streams and examines the selection of
in-stream methods, tools, and devices for stabilizing streams and creating habitat to support native fish species. Although
the paper discusses salmonid species in the Pacific Northwest in particular, the methodologies and tools employed to
evaluate and support fish habitat can be generally applied to streams and watersheds in other regions.
The effects of urbanization, such as decreased pervious area and vegetative cover and increased stormwater runoff
and erosion, destabilize watersheds and streambeds and destroy aquatic habitats. Stable stream environments are
necessary if biological systems including fish and their supporting food web are to flourish. Changes to urban streams
and watersheds may be so significant, however, that decades may pass before they reach stability. Even then, the
resulting Astable environment© might not provide the type of habitat needed to support species from the natural
environments.
The evaluation of channel erosion and sedimentation in urban streams provides one measure to assess the relative
stability of streams and, thus, their ability to support fish and amphibian species. For most streams, an evaluation of
relative streambed stability can be completed through a visual examination of streambed morphology and minimal
supporting calculations. Tools for performing these analyses will be presented in this paper.
Rehabilitation of streams in urban and heavily logged watersheds requires establishing a stream structure that will
maintain streambed stability and create the different types of habitats needed to support desired fish species. One size
or type of in-stream device cannot meet all stabilization and habitat requirements. The selection of devices should
correspond to the relative stability of the individual stream reach. Devices for maintaining streambed stability and creating
habitat, as well as the procedures for selecting them, will be discussed in this paper.
Introduction
Salmon populations in the Pacific Northwest are dwindling. One significant cause of this reduction in population is
the destruction of small stream (1s1 to 4th order) habitats.
In order to respond to the destruction of fish habitat in small streams, we must have an understanding of watershed
processes and natural stream morphology. Although this paper concerns western Washington streams, which are
surrounded by heavy forests and fed by rain and groundwater, the general principles for stabilizing streams discussed
here can be applied to most natural small stream systems.
Natural Stream Morphology
Streambed gradients gradually decrease from the upper reaches of a watershed to the outlet of a stream because
flow rates increase in the lower parts of the watershed (Leopold, Wolman, and Miller, 1964).
The streambed gradients create different types of fish habitat features (Rosgen, 1994). Figure 1 shows the
relationship of habitat type to streambed slope in western Washington streams. Pool/drop habitat is dominant in reaches
with smaller flows and steeper valley gradients. The pools are formed by large organic debris or rock formations. As the
stream flow rates increase, valley gradients tend to decrease and the streams are dominated by pool/riffle habitats.
69
-------
POOL RIFFLE
Stop* Range2
TYPICAL SPECIES
Chum
PLANEBED
GLIDES
POOL/DP
PCQL/RIFFLE
TRANSITION
Slope Ranga =2-4%
TYPICAL SPECIES
Goto
POOL DROP
Slope Range =4-10%
TYPICAL SPECIES
Qfturo (tKtwm ftM Qrfef (Hfe ftafe)
Figure 1. Relationship of Stream Habitat to Streambed Slope in Western Washington Streams.
Pools and riffles form alternately on the outside of stream bends. These alternating pools and riffles are present in
practically all perennial channels. In straight or meandering streams, pools and riffles generally form every 5 to 7 channel
widths. As the stream widths increase, however, the number of pools decrease (Leopold, Wolman, Miller, 1964).
In each non-rigid, natural stream, a dominant channel is formed by the stream=s dominant discharge. This dominant
discharge channel is a component of most fish habitats. The dominant discharge has a recurrence period of
approximately once each 1.5 years in natural systems (Simons, Senturk, 1991). In urban systems, the bank-full discharge
has a recurrence period of about one year (MacCrae, 1996).
Western Washington Natural Stream Characteristics The small streams of western Washington are fed by rain and
groundwater and are found in steep-sided canyons with fairly straight valley bottoms. In their natural (forested) state,
small streams have a slightly meandering, low-flow channel in a narrow valley bottom. The vegetation along the stream
banks is often dense and provides shade, channel stability, and cover. Debris jams are common and act to slow stream
flows during storm events. Western Washington soils are products of glacial activity and consist of smooth cobbles and
stones, as well as fine materials. Clayey bank materials, heavy root structures along the banks, and steady base flows
create channels with nearly vertical sides and small widths (3 to 6 feet wide, and sometimes as small as 1 foot wide)
relative to the width of the valley bottom. Typical old-growth forest streams have low nutrient levels and low annual
sediment yields.
70
-------
Most natural Pacific Northwest streams can be described as sediment starved. Natural watersheds are heavily
forested and act as a sponge for rainfall. In natural watersheds, the storm runoff response is slow and the flow rates are
low. Because the ratio of the 1.5-year storm to the base flow is quite low (often less than 5), these streams have small
width to depth ratios, with most of the dominant discharge channel acting as an aquatic habitat channel. The width of the
dominant discharge channel is coincident with the width of the aquatic habitat channel. The aquatic habitat channel is
the normally wetted, low-flow part of the streambed (Seattle, 1997). Most small, natural Western Washington streams
are dominated by pools and drops formed by large organic debris (Gustav, Severn, Washington, 1993)
The aquatic biological community in western Washington may include as many as 250 plant and animal species. The
aquatic biological community depends on a stable aquatic habitat associated with old growth, coniferous watersheds.
Much of this aquatic community functions as a food web, with fish populations representing the mega fauna. Some
salmon and trout species are the top aquatic predators (Severn, Washington, 1996).
Salmon and trout in western Washington evolved to take advantage of these regional stream conditions. Figure 1
lists different species found in the various regions of the watershed. More athletic fish species like coho, steelhead, and
cutthroat trout occupy the upper reaches (steeper gradient) of the watershed. The young coho and steelhead reside in
the stream for a year before migrating to saltwater. Cutthroat may reside in fresh water for two years before migrating
to saltwater. Less athletic species, such as chum and pink salmon, can not migrate through the steeper gradients of the
upper watershed to spawn. The fry of these species occur in the lower regions of the watershed and migrate to saltwater
shortly after emergence. Young salmon, as well as young and adult trout, will utilize any part of the watershed that meets
their habitat requirements. For example, if fish habitat in the upper reaches of a watershed is unsuitable, young coho may
seek winter refuge in the lower regions of a watershed (Severn, Washington, 1996).
Five general categories of habitat occur in natural western Washington streams (Severn, Washington, 1996):
. Estuaries/Deltas
. Passage
. Refuge
. Rearing
. Spawning and Incubation
Pacific Northwest fish derive most of their food from organisms (benthos) that live in, or on, the substrate of the
stream. Most food production occurs in the same stream areas that provide spawning and incubation habitat for fish. An
annual surplus of approximately 10 pounds of biomass is required to support one pound of fish in the stream (Washington,
1999).
Assessing Stream Deterioration
Visual inspection of stream morphology can provide rapid and relatively accurate assessments of astream=s ability
to support fish populations. Practical experience working in western Washington streams has shown that visual
streambed assessments correlate well with benthic sampling (Seattle, 1999). Although benthic sampling is necessary,
visual inspections can reduce the amount of benthicsampling required when quickassessments are needed orextensive
benthic testing is cost-prohibitive.
Natural streams are generally non-rigid. Their cross-sections vary with changes in flow rates and yearly rainfall
volumes. Stream systems will generally aggrade during low flow periods, and degrade during high flow periods. To
assess the deterioration of non-rigid streams, it is necessary to understand the following three concepts:
*Sediment transport
71
-------
*Effects of urbanization on stream stability
*Effects of urbanization on fish habitat
Sediment Transport Concepts Understanding the dynamics of sediment transport is useful for predicting hydraulic
equilibrium conditions in a stream. Any stream will respond to imposed changes. Six basic relationships exist between
discharge levels and channel form, regardless of stream size (Simons and Senturk, 1991):
Depth of flow in the dominant discharge channel is directly proportional to discharge.
. Width of the dominant discharge channel is directly proportional to water discharge and sediment discharge.
. Dominant discharge channel shape is directly related to sediment discharge.
. Channel gradient is inversely proportional to water discharge and directly proportional to sediment discharge and
grain size.
. Sinuosity is directly proportional to valley gradient and inversely proportional to sediment discharge (larger valley
gradient causes greater meander, larger sediment discharge causes less meander).
. Transport of bed materials is directly related to flow velocity and concentration of fine material, and inversely
proportional to the fall diameter of the bed material (greater depths and higher velocities cause larger bed load
volume in transport, sediments shaped like kites fall slower than round-shaped sediments).
A stable channel exists when a stream has the bed slope and cross-section which allow its channel to transport water
and sediment from upstream without aggradation, deposition, or streambank erosion (Simons and Senturk, 1991).
When natural flow rates are exceeded, sedimentation and erosion can be a dominant limiting factor for fish
populations. The exaggerated volumes and rates of stormwater runoff in urban areas increase both the rate of erosion
and volume of sediments generated from upland and riparian areas in the watershed. Soil erosion can lead to excess
streambed erosion and sedimentation and destroy redds, fish rearing habitats, and food production areas.
Effects of Urbanization on Stream Stability Urbanization permanently alters the hydrologic balance within stream in
the following ways:
. Total water passing through urban streams increases.
. Stormwater runoff rates and volumes increase.
. Increased impervious surface areas prevent groundwater recharge; as a result, base flow rates during summer and
fall are often less than natural flow rates were.
. Increased stormwater runoff causes erosion and transports significant amounts of sediments and pollutants,
including oil, grease, and polluted fine sediments from streets and parking lots, into urban streams,
In urban areas, the ratio of the 1 -year storm to the stream=s base flow (dominant discharge) is large, sometimes
greater than 100. In many watersheds, terrestrial sediment volumes are dramatically increased by urbanization. Excess
sediment increases the width of the dominant discharge channel.
Stable urban streambeds have 182% gradients, compared to the 2810% gradients that support anadromous species in
natural watersheds. To reach a stable gradient, the streambed can lower several feet, causing significant bed load
sediments from shallow landslides. Measurements taken from several western Washington streams show that
streambeds will flatten from a 4% gradient to a 1% gradient as a result of urbanization. A change in streambed gradient
from 4% to 1% over the distance of 1,000 feet can result in streambed erosion and an elevation difference of 30 feet at
the upper end of the reach. During the transition from steep to flat gradients, fish habitat is in a perpetual state of change
(Severn, Washington, 1996). Unfortunately, it can take decades before stability is again reached.
72
-------
In unstable streams, braiding occurs at crossover points between bends where stream gradients are steep. At normal
stage, a braided section has a divided flow with small, mid-channel bars and a single large channel composed of
subordinate channels. The base flow channel often changes location within the bottom of the dominant discharge channel
(Severn, Washington, 1996).
The erosion of the channel bottom along the basin (often called head-cutting) indicates a readjustment of the basin's
gradient, the stream discharge, and the sediment load. (Simons and SentCirk, 1991).
Effects of Urbanization on Habitat Stable habitat conditions within the channel have stringent requirements in urban
streams, including a sediment-starved condition and minimal movement of spawning-sized gravel (3-inch and smaller)
during most storm runoff events.
In urban streams, the benthic system can be limited both by erosion (which provides conditions of constant change) and
by sedimentation (which smothers redds and food production areas, and fills rearing habitats with silt). In addition,
streams in urban or deforested watersheds experience significant habitat loss and are unable to support the biological
diversity that fish species depend upon. In contrast to natural watersheds, where 250 plant and animal species may
comprise the aquatic habitat, urban watersheds may have fewer than 50 plant and animal species.
Living systems do not adapt to constantly changing environmental conditions. The changes in aquatic habitats
caused by urbanization decrease food production and destroy spawning and incubation areas (Bell, 1990). As flow rates
and volumes increase, streambeds become unstable. When streambeds become unstable, the aquatic habitat channel
may retain a small width to depth ratio, but it will be substantially less than the width of the dominant discharge channel.
In addition, streambed instability causes a constant shifting of the aquatic habitat channel and this limits development of
the benthic community and destroys redds.
As an urban stream approaches stability, the resulting aquatic habitat channel will be too wide, shallow, and
homogeneous to support fish populations. Streams naturally deposit bed load on the inside of bends and form point bars.
Because natural sinuosity is low in western Washington streams, point bars form infrequently or incompletely leaving a
wide, shallow, cross section during base flows. Under these conditions, the flow depths of most urban streams are
insufficient to submerge returning adult fish. Because small-grained sediments settle as flow rates decrease, redds and
food production areas are smothered with silt and pool habitats are filled with sediment.
Perhaps the greatest general impact is the permanent loss of habitat types that sustain coho and steelhead
populations. These species prefer pool/drop habitat. Coho, in particular, require quiet pools (Seattle, 1997). Examination
of Figure 1 shows that as the streambed gradient lessons, the habitat type shifts from pool/drop to pool/riffle. Pool/riffle
habitat does not provide sufficient pool depth for normal fish rearing or urban storm refuge. Because large storms
frequently occur after the fry emerge from the streambed gravels, the need for urban storm refuge habitat is critical.
Juvenile fish cannot maintain their position in high velocity reaches. In fact, normal urban storm flows often wash juvenile
fish into larger bodies of water (salt water or streams) where they cannot survive.
Tools and Methodologies
The methodologies available to assess existing stream conditions and predict future conditions include the use of visual
streambed assessments and analytical tools such as simple hydraulic mathematics.
Streambed Assessment
General indicators of habitat degradation in urban streams are visually apparent and include the following elements:
. Dominant discharge channel wider than in natural conditions
. Reduced pool frequency and less diverse habitat
. Increased sediment from terrestrial sources
73
-------
. Reduced large woody debris
. Drastically reduced aquatic community diversity
Both natural and urban streams fluctuate between stability, degrading, or aggrading. Compared to natural streams,
however, streams in urban watersheds exhibit extreme traits of aggradation, degradation, or instability. Urban stream
deterioration indicators differ according to the condition of the specific reach (stable, degrading, or aggrading). In a
degrading streambed, there is a progressive lowering of the channel due to scour. In an aggrading streambed, there is
a progressive buildup or raising of the channel due to sediment deposition. Both degradation and aggradation are
indicators that a change in the stream=s discharge and sediment load is taking place (Simons and Senturk, 1991).
Urban stream deterioration indicators differ according to the condition of the specific reach. Table 1 (the following
bulleted paragraphs) describes the deterioration indicators in a stable, degrading, or aggrading stream.
Table 1. Urban Stream Deterioration Assessment Indicators
Stable B If an urban stream reach is stable and terrestrial sediment loads are low, the reach may be able to support species that reside temporarily in the
stream before moving to salt water. The wide, shallow channel of a stable stream provides little protection from predation, however, and also lacks resting
pools. The following are indicators of stable urban streams:
. Apparent changes in channel shape and configuration afler large storms are small.
. Width of aquatic habitat channel coincident with width of dominant discharge channel B shallow flow depth, prevents fish passage.
. Head-cutting and nick points are absent or nearly absent.
. Substrate stability:
. Periphyton stays on streambed after significant storms B streambed is stable.
. Streambed gravel lack periphyton B gravel is being moved during significant runoff and replaced when the storm flow recedes,
Small-grained sediments settle when storm flows recede, smothering redds and food production areas.
Pools (on-stream or off-stream) that can retain newly hatched fry are generally not present.
Degrading B In a degrading urban stream the dominant discharge channel is as wide as in a stable stream, but base flow rarely covers the bottom of the
channel except at crossover points between bends. When streams begin to unravel due to degradation, the effects do not appear instantly. Head-cuts
move through a stream until it reaches a vertical drop. When enough head-cuts accumulate, the vertical drop will be undercut, releasing large amounts
of bed load type sediments. Occasional pools will develop that may support anadromous fish, however, they are often inaccessible.
. Streambed gravel sizes are larger than stable sections of the stream, but are mostly bare of periphyton or other aquatic growth.
. Channel braiding occurs at crossovers between bends.
. The substrate of the base flow channel is not coated with periphyton.
. The base flow channel and dominant discharge channel lack large woody debris.
. Large woody debris that spans the banks of the dominant discharge channel may indicate a recent streambed elevation and may illustrate the
amount of degradation that has occurred.
. Stream banks are bare and often nearly vertical.
Aggrading B Aggrading stream reaches may be visually similar to stable reaches. Generally, aggrading streams will have a flatter streambed gradient
and accumulate more fine sediment. Disturbing the bed of an aggrading reach usually results in long periods of murky water flow. Aggradation will occur
locally in pools, which reduces habitat value, but has less impact on habitat than an entire aggrading stream reach would.
. Head-cuts and nick points do not exist.
. Deltas may be visible at the top of the reach.
. Periphyton covers the substrate.
. Large woody debris in streambed is partially buried.
. Pool/riffle and pool/drop habitat can occur as isolated conditions.
. Substrate surface aravel sizes are small.
The visual indicators described in Table 1 (combined with assessments of fish passage problems and periodic benthic
population checks) can be used to describe the potential habitat capacity for a stream reach or, collectively, for an entire
stream. Needless to say, habitats for fish species that reside in the stream for one year or longer must be able to support
the full life cycle of the species. In addition, fish need to have full access to these habitats.
A methodology based on visual streambed assessment was developed for rapid stream assessments in Longfellow
Creek and Pipers Creek in Seattle, Washington (it was also used within six watersheds in Snohomish County,
Washington). The streambed assessments provided an accurate measurement of the ability of the watershed or stream
in question to support fish. An example of the summary rating for Longfellow Creek is shown in Table 2.
74
-------
Table 2. Summary Rating for Longfellow Creek
Creek Map Bank Sediments Other
Segment Description Sheet Habitat' Erosion' Sources* Pollutants* Total
1 Pipe from outfall to Andover Street 1,2,3 5 3 3 2 13
2 Open channel from Andover Street to 4,5 10 1 1 2 14
Genessee Street
3 Open channel from Genessee Street to 5,6,7 10 2 2 2 16
confluence of unnamed tributary in West
Seattle Golf Course
4 Unnamed tributary in West Seattle Golf 6,7 6 2 2 212
Course
5 Open channel from confluence with West 7,8,9 91 1 213
Seattle Golf Course tributary to Brandon
Street
6 Open channel from Brandon Street to Findlay 9 9 2 2 215
Street. Contains confluence of Juneau Street
bypass via "biochannel."
7 Open channel from Findlay Street to Juneau 9,10 822 2 14
Street. Also contains piped high flow bypass
starting at Juneau Street and rejoining Creek
in Segment 6.
8 Open channel from Juneau Street to Graham 10,11 11 2 3 2 18
Street
9 Open channel from Graham Street to Willow 11,12 8 2 1 2 13
Street
10 Open channel from Willow Street to Myrtle 12,13 822 2 14
Street
11 Piped channel Webster Basin; open channel 13,14 722 2 13
to Holden Street. Contains "K-Mart bypass."
12 Open channel from Holden Street to Thistle 14,15,16 821 2 13
Street
13 Pipe from Thistle Street to head of basin at 16 6 3 1 2 12
Roxbury Street
Notes:
Ranking Codes:
1 = Poor condition
2 = Moderate
3 = Relatively good
• B The value in the "habitat" column is the result of another ranking process. The total "habitat" volume for each Creek segment is
transferred into this table to complete the ranking process. The habitat rank has a range of O-18 which is developed from evaluating bed
erosion, fine sediment accumulation, gravels (clean/stable), benthic (quantity, quality), habitat structure, and riparian vegetation. All other
columns in this table are ranked from 1 (poor condition) to 3 (relatively good condition).
Analytical Tools Simple hydraulic mathematical tools can be applied to analyze existing conditions and are
needed to check passage conditions for channel rehabilitation projects. In addition to depth and velocity, the amount
of turbulence in a stream has significant impact on the amount of sediment that can be moved. Greater turbulence
increases the amount and size of sediments that can be moved. Maximum turbulence occurs when the Froude
number is equal to 1 .0. The Froude number is defined by Equation 1:
FN= V/(gY)
1/2
(Equation 1)
Where
FN = Froude number
V = average velocity in the cross-section
g = acceleration force due to gravity
Y = the hydraulic depth, which is the cross-sectional area divided by the top width
75
-------
The Froude number should not exceed 0.8 for large storms (like the 25 or 100-year event) except at channel drops.
This criterion is the same for rigid, grass, and dirt lined channels in most stormwater manuals. For frequently occurring
storm flows (1 to 2 year events), the Froude number has to be much less in order to meet the criteria for larger flows.
Spawning-size gravel will generally be stable if this Froude number criterion is met.
In the author's experience, backwater analysis (HEC-2 or HEC-RAS) may not be strictly applicable to analyze natural
streams, but is a useful tool to analyze deteriorated channels and to model proposed improvements for stream
rehabilitation. For stream rehabilitation backwater analysis is done for large and small events (base flow, 1 -year flow, and
one point in between) to ensure passage of juveniles throughout their life-cycle habitat within the stream. To provide
accurate results, more cross-sections are needed to conduct backwater analysis on non-rigid, urban streams compared
to conventional backwater analysis.
Selecting Rehabilitation Devices
Unless excess stormwater from both frequent and rare storm runoff events can be eliminated, it is the author=s
opinion that some form of structural intervention is needed to create fish habitat in urban streams. In western
Washington, streams become unstable and significant fish habitat is lost when the impervious area reaches 10 to 15
percent (Booth, 1996). The dominant discharge channel will respond to hydrologicchanges (Simons and Senturk, 1991).
As a result, modifying how land is urbanized can reduce the effects of urbanization, but it will not obviate the hydraulic
impact on the dominant discharge channel. Stream rehabilitation measures will still work best where there are fewer
disturbances to the watershed and where wide riparian corridors are maintained.
The goal of urban stream rehabilitation is to stabilize the streambed with devices that also create a habitat for fish
populations. The stream must develop sufficient food mass and diversity to support desired fish species. Quality salmon
and trout habitat can exist in urban streams when hydraulic/habitat criteria are met, the streambed is stable, and the base
flow channel is confined (Severn and Washington, 1996).
Establishing stable streams in urban watersheds is often a moving target. As more urbanization occurs, hydrologic
and biological changes accumulate. The extent that the dominant discharge channel spreads is a direct function of the
amount of pavement in a watershed.
The New Urban Stream. Restoring an urban stream to pre-development conditions is not possible (National Research
Council, 1992). It is the author's opinion that too many stream rehabilitation projects emphasize stream bank rehabilitation
rather than focusing on the root causes of stream habitat destruction. Often, that root cause is streambed instability,
which is a natural response to increased flow rates and volumes in the stream.
To maintain pre-development species in urban and deforested streams, a "new urban stream" is needed that can
provide a variety of fish habitats including pools. Without intervention, the urban streams will convert pool/drop habitat
into pool/riffle habitat, eliminating the diversity of habitat required to support a variety of fish species (Severn and
Washington, 1996).
The goal is to stabilize the stream and return it to its original sediment-starved condition. While bed load and
suspended sediment are readily available, the object is to sculpt sediment deposition to form bars and banks and confine
the aquatic habitat channel within a single location. Concentrating flows in the aquatic habitat channel helps keep the
substrate size optimal and clears the stream of fines.
Urban stream rehabilitation must focus on historic conditions that can be recreated, rather than on the conditions
that cannot be meaningfully restored. Except for the following three exceptions, historic environmental conditions can
be recreated in urban streams:
. The width of the dominant discharge channel will always be greater in urban streams.
. The banks of the aquatic habitat channel cannot be coincident with the dominant discharge channel (in pool/riffle
habitat).
. High flood flows, deep flow depths, and large velocities are more frequent.
76
-------
In form, the resulting channels resemble snowmelt type streams, where flow rates significantly exceed base flows
for several weeks each year. Most urban streams lack large woody debris that can be incorporated into the stream's
structure when a reach attains stability. Most reaches will take decades to reach stability. Just because the channel will
not look "natural", doesn't mean that we will have failed or that anadromous species cannot be supported. Many eastern
Washington snowmelt type streams support anadromous species.
A rehabilitated stream has five primary needs:
. A dominant discharge channel sized to carry the 1 -yearto 1 S-year storm (depending on the degree of urbanization)
at full bank. It is important to recognize this need because the stream will reshape the dominant discharge channel
and may undo much of the rehabilitation effort.
. Within the dominant discharge channel, hydraulic conditions must provide biologic and stream stability (keep most
of the spawning sized gravel and rock from moving during frequent storms).
. Within the dominant discharge channel, habitat must be provided for the entire life cycle of the desired species.
. Because fine-grained sediment falls out last and needs to be kept in transport, the base flow channel has to be
narrow, deep, and stable.
. Stream banks need to be stable to support riparian vegetation. Bio-stabilization techniques can help reduce the
width of the dominant discharge channel.
Ideally, long reaches of unstable streams will be stabilized. Near the spanning structures, aggradation will replace
degradation (as long as a sediment supply is available). If only short reaches are stabilized, large storms will deposit
substantial sediments within the stabilized reach, particularly at the upstream end of the reach. The sediment sizes most
likely to accumulate during the stabilization of a reach are the larger sizes that are moved as bed load. Once stable
streambed gradients are attained, the amount of sediment that can move and cause aggradation is finite. After a few
larger storms, bed load movement will be minimal.
Rehabilitation Devices A variety of devices may be used to confine the base flow channel and provide streambed
stability. The author has experience with several types of bed control structures, glides, lunkers, and confining devices.
In small streams, these devices may be a well-placed piece of timber, a boulder, or randomly placed stones and rootwads
(to increase roughness). In larger streams, stabilization and confining devices are much more complex.
Selection and siting of devices is dependent on the stream condition in a specific area and the type of habitat that
is to be provided. It is important to realize that one type of structure is not suitable for all applications. Selection of a
stream rehabilitation device depends on the type of habitat needed and the device's perceived hydraulic attributes. Both
non-rigid channel design and biologic skills are required to be successful.
This section will address issues related to the selection of three types of structures: timber stepdown structures,
boulder bed control structures, and deflectors.
Timber Stepdown stepdown. Sometimes referred to as a "K-structure". A form of timber is shown in Figure 2. The
logs, which are set at 45 degrees to the channel, are called weir logs and create pools during high storm flows. The weir
logs need to have a steep pitch (the bank end higher than the center of the stream). In dense, urban western Washington
watersheds, storms with high yearly return frequencies produce flows 20 to 50 times the base flow. Timber stepdown
structures form large, quiet pools during storms, allowing newly emerged fish juveniles to find refuge. In
77
-------
Figure 2. Timber Stepdown K-Structure (shown with weir logs).
addition, the K-type stepdown creates rollers. Rollers assist fish passage upstream for less athletic adults and juveniles
during higher flow rates (Kerr Wood Leidal Associates, LTD, 1980).
Common variations of the K-structure include a straight timber stepdown (no weir logs) and a vortex structure (with
weir logs). In creating a diverse habitat environment, both types can be used.
While the straight timberstepdown does not form an upstream pool, the substrate above the stepdown is turned-over,
even for frequently occurring storm flows. The Washington Department of Fisheries and Wildlife has successfully installed
many straight timber stepdowns in logged watersheds and those with minimum to moderate urbanization. For lower flow
rates, the substrate above the stepdown remains stable and provides excellent spawning habitat.
The vortex type timber stepdown does not form storm refuge pools as well as the k-structure and the log configuration
stymies formation of a roller. The structure can help develop confined low flows, however, and creates good fish-rearing
habitat for many northwest species (not Coho).
Boulder Bed Control Structure. Figure 3 shows a boulder bed control structure. Like the K-structure, the boulders
that form the structure are on the upstream side, but these boulders could also be on the downstream side to form a
vortex-type structure. In the author's experience, changing the configuration and the angles of the boulders provides
slightly different habitat characters, all of which are acceptable. While wood is usually preferred in small streams, boulder
bed control structures are flexible and can adjust to channel degradation in unstable streams. Boulder bed control
structures are most beneficial at the downstream end of a reach where no streambed control is established in the reach
below.
Deflectors. Deflectors can be made of either wood or boulders. Figure 4 shows a timber deflector with a bank log on the
opposite bank. Point bars form on the insides of bends. Deflectors should be installed on the insides of bends to help
build larger point bars and to confine the base flow channel. The reach in Figure 4 is nearly straight (sinuosity about 1 .0).
To develop point bars in straight reaches, several deflectors may be needed and could be on one side of the channel,
or on alternating sides. The reach shown in Figure 4 has a large bed load, and point bar formation would have occurred
if the deflector installations were correctly located and implemented. Without modification, however, the deflector shown
in Figure 4 will not form a point bar.
78
-------
Figure 3. Boulder Bed Control Structure.
3-^
Figure 4. Deflector.
Conclusions
Streambed assessment based on sediment transport principles can be a useful tool to rapidly determine a stream's
capability to support fish populations. Standard engineering analytical tools can be used with streambed assessment to
support hydraulic design for stream rehabilitation projects. Although the stream will not have the same appearance as
a natural stream, stream rehabilitation can be successful and urban streams can support anadromous fish populations
in western Washington. Hydraulic design is needed to develop the dominant discharge channel and properly place
structures to attain the desired habitat conditions. Selection of the rehabilitation devices must consider design needs of
non-rigid channels, as well as habitat requirements. Finally, one size or type of habitat rehabilitation device cannot serve
79
-------
all purposes. Selection of habitat rehabilitation devices should achieve specific habitat requirements and provide habitat
diversity.
References
Bell, Milo C., Fisheries Handbook of Engineering Requirements and Biological Criteria, US Army Corps of Engineers,
Portland, Oregon, January, 1990.
Booth, Derek B.; Montgomery, David R.; Bethel, John; "Large Woody Debris in Urban Streams of the Pacific Northwest;"
Proceedings of an Engineering Foundation Conference, Edited by Larry Roesner, Snowbird, Utah, August 1996.
Gustav, Richard S.; Severn, Douglas T.; Washington, Percy M.; "If You Build It They Will Come: A Maintenance
Approach to Restoring Fish Habitat in Urban Streams Heavily Influenced by Stormwater Flows;" Proceedings of the Water
Environment Federation 66th Annual Conference & Exposition, Anaheim California, October 1993.
Kerr Wood Leidal Associates LTD., D.B. Lister & Associates LTD, Stream Enhancement Guide, Province of British
Columbia, Vancouver, British Columbia, 1980.
Leopold, Luna B., Wolman, Gordon M., Miller, John P., Fluvial Processes in Geomorphology, W. H. Freeman and
Company, San Francisco and London, 1964.
MacCrae, C. R., "Experience From Morphological Research on Canadian Streams: Is Control of the Two-Year Frequency
Runoff Event the Best Basis for Stream Channel Protection?" Proceedings of an Engineering Foundation Conference,
Edited by Larry Roesner, Snowbird, Utah, August 1996.
National Research Council, Restoration of Aquatic Ecosystems, Washington, D.C., 1992.
Rosgen, David L., "A Classification of Natural Rivers,"Catena, ElsiverScience, B.V., Amsterdam, The Netherlands, 1994.
Seattle, City of, Longfellow Creek Habitat Restoration Master Plan, Seattle, Washington, 1999.
Seattle, City of, Pipers Creek Rehabilitation, Erosion and Sedimentation Management Program and Design Manual, 1997.
Simons, Daryl B. and Senturk, Fuat, Sediment Transport Technology, Water and Sediment Dynamics, Water Resources
Publications, Littleton, Colorado, 1991.
Severn, Douglas T. and Washington, Percy M., "Effects of Urban Growth on Stream Habitat," Proceedings of an
Engineering Foundation Conference, Edited by Larry Roesner, Snowbird, Utah, August 1996.
Washington, Percy M., Conservation, Seattle, Washington, August 1, 1999.
80
-------
Lessons Learned About Successfully Using Infiltration Practices
Eric H. Livingston
Stormwater/NPS Management Section
Florida Department of Environmental Protection
2600 Blair Stone Road (MS3570)
Tallahassee, Florida 32399-2400
Abstract
Infiltration practices are one of the most valuable urban stormwater BMPs because they not only help to reduce
stormwater pollutants but, more importantly, help to reduce stormwater volume. Unfortunately, infiltration practices have
gotten a bad reputation over the past 20 years because of their potential to fail. This paper will review the successes and
failures of the use of infiltration practices in the United States. It will summarize the lessons that have been learned about
the use of infiltration practices. This will include recommendations on when they should be used and how, and
recommendations on when they should not be used. Finally, the paper will discuss what can be done to reduce stormwater
volume when infiltration practices cannot be used successfully.
Introduction
To achieve the desired objectives of flood control, water quality protection, erosion control, improved aesthetics, and
recreation, a stormwater management system must be an integral part of the site planning for every development site.
Although the basic principles of stormwater management remain the same, each individual site and each specific project
presents unique challenges, obstacles, and opportunities. The many variations in climate, soils, geology, groundwater,
topography, vegetation, and planned land use require site-specific design.
The natural attributes of each site will strongly influence the type and configuration of the stormwater system. These
features will suggest which particular combination of Best Management Practices (BMPs) can be successfully integrated
into an effective system. Whenever site conditions allow, the stormwater management system should be designed to
achieve maximum on-site storage of runoff by incorporating infiltration practices throughout the remaining natural and
landscaped areas of a site. A stormwater management system should be viewed as a "treatment train" in which the
individual BMPs are the cars. Generally, the more BMPs that are incorporated into the system, the better the performance
of the treatment train. Inclusion of infiltrative practices as one of the cars should be a primary goal of stormwater system
designers, even on sites where detention practices will be the primary BMP.
Infiltration practices (or retention practices) retain stormwater on-site, allowing it to soak into the ground or evaporate. There
are a number of different infiltration practices that have been widely used throughout the United States, including basins,
trenches, dry wells, pervious pavements, and swales. Often infiltration practices include vegetation with a wide variety of
trees and shrubs. In 1987, Prince George's County, Maryland, began evolving this type of infiltration practice into
"bioretention," which is a BMP that uses the soil matrix and vegetation to sequester pollutants within the terrestrial
environment (PGC, 1993, 1997)
Infiltration practices are one of the few BMPs that can help to assure that all four stormwater characteristics (volume,
rate, timing, and pollutant load) after development closely approximate the conditions which occurred before development.
That is because infiltration practices help to maintain pre-development site perviousness and vegetative cover, thereby
reducing stormwater volume and discharge rate, which further promotes infiltration and filtering of the runoff.
The benefits of infiltration include reducing stormwater volume and peak runoff rate; recharging groundwater, which
helps to replenish wetlands, creeks, rivers, lakes, and estuaries; augmenting base flow in streams, especially during low
flow times; settling and filtering of pollutants as they move through the system's vegetation and surficial soils; lowering the
probability of downstream flooding, stream erosion, and sedimentation; and providing water for other beneficial uses.
81
-------
Another benefit of infiltration practices is their ability to serve multiple uses since they are temporary storage basins.
Recreational areas (e.g., ballfields, tennis courts, volleyball courts), greenbelt areas, neighborhood parks, and even parking
facilities provide excellent settings for the temporary storage of stormwater. Such areas are not usually in use during
periods of precipitation and the ponding of stormwater for short durations does not seriously impede their primary functions.
Longevity of Infiltration Systems
One of the problems with infiltration BMPs that has been consistently identified, either quantitatively or qualitatively,
is their high rate of failure. Maryland's Stormwater Program produced one of the most comprehensive quantitative reviews
of the longevity of infiltration systems (Pensyl and Clement, 1987; Lindsey et al, 1992). This information is summarized
in Table 1, where it can be seen that the overall condition and functioning of infiltration systems declined over time. In
1986, about two-thirds of all surveyed facilities were functioning as designed, while in 1990, only about half were. Only 42%
of the facilities were functioning as designed in both 1986 and 1990, while about 27% were not functioning as designed in
both years. About 24% of the systems were functioning in 1986, but not in 1990; while only 7% of those not working in
1986 were working in 1990. Maintenance was needed at more facilities in 1990 (66%) than in 1986 (45%). Additionally,
38% of facilities that needed maintenance in 1986, still needed maintenance in 1990, while 32% of the facilities that did
not need maintenance in 1986, did need it in 1990. Only 10% of the systems that needed maintenance in 1986 did not need
maintenance in 1990. These data suggest that little effort is expended on maintaining the operational capabilities of
stormwater management systems.
Additional quantitative information on the success and failure of infiltration systems was collected in the Puget Sound,
Washington, area (Klochak, 1992; Gaus, 1993; Hilding, 1993; Jacobson and Horner, 1993). Of 23 infiltration basins
evaluated, 12 did not comply with the region's guidelines for either infiltration rate or time for the basin to recover its storage
volume. Interestingly, the authors found no relationship between lack of basin maintenance and failure, with examples of
basins with and without maintenance that did not function properly. Some basins were functioning properly even though
they had never been maintained, while 43% of the 23 basins had been scarified to enhance performance.
The above data, when combined with qualitative information from Florida and Delaware (Baldwin, 1999, personal
communication), seem to indicate that infiltration basin failures are associated with:
1. Inaccurate estimation of infiltration rates
2. Inaccurate estimation of the seasonal high water table
3. Excessive compaction during the construction process
4. Excessive sediment loadings either from improper erosion and sediment control during the construction process
or a lack of pretreatment BMPs
5. Lack of maintenance
Factors Influencing Successful Use of Infiltration Systems
Factors that influence the successful use of any stormwater BMP can be categorized as institutional, technical, and
implementational. This section of the paper will examine the components of each of these categories that must be included
in a stormwater program if the causes of infiltration system failure are to be minimized.
Institutional Components
The "BMP Golden Rule" states that stormwater BMPs should never be required unless the stormwater program includes
components that will assure that the BMPs are correctly designed, approved, constructed, inspected, maintained,
82
-------
Table 1. Results of Maryland Infiltration Practices Surveys
Basins
# facilities
(% of total)
1986
63
30%
1990
48
27%
Trenches
1986
94
45%
1990
88
30%
Dry Wells
1986
30
14%
1990
25
14%
Perv. Paving
1986
14
7%
1990
13
7%
Veg. Swale
1986
6
3%
1990
3
2%
Facility Evaluations
Functioning
OM Needed
30
48%
41
65%
18
38%
39
81%
75
80%
28
30%
47
53%
64
73%
23
77%
9
30%
18
72%
4
16%
7
50%
10
71%
2
15%
8
62%
3
50%
6
100%
2
67%
3
67%
Performance and Maintenance Criteria
Buffer strip
inadequate
Stabilization
needed
Sediment
entry
Inappropriate
ponding
No observed
well
20 4 65
32% 8A 69%
12 23 11
19% 48% 12%
24 28 32
38% 58% 34%
41 25 25
65% 52% 27%
na na 45
45%
35
39%
13
15%
58
66%
20
23%
58
56%
24 0 14 3
80% 100% 23%
13 11
3% 12% 7% 8%
02 99
8% 64% 64%
93 74
30% 12% 50% 31%
47 10 11
13% 28% 71% 85%
1 0
17%
3 1
50% 33%
4 1
67% 33%
4 0
67%
na na
and operated (Livingston, 1997). Specifically, the program must have stormwater treatment and management goals,
performance standards, education, and an institutional framework that includes plan approval, inspections during and after
construction, legal operation and maintenance entity requirements, effective compliance mechanisms, and dedicated
funding mechanisms.
83
-------
Program goals: Experience has shown that stormwater programs need to have multiple objectives that are important
to the general public in order to gain the support of the community. Typically, these will include flood protection, erosion
and sediment control during construction, water quality and habitat protection, and open space and recreation. Infiltration
systems can help achieve all of these goals.
Performance standards: Nearly all stormwater treatment programs in the United States are technology-based, relying
upon a performance standard (minimum level of treatment) and design criteria for various BMPs that assure that they
achieve the desired treatment goal. A review of 32 stormwater programs around the country showed that the most common
performance standard is removing at least 80% of the average annual loading of total suspended solids (WMI, 1997a).
Some programs require higher levels of treatment for stormwater discharges to sensitive waters, such as Florida's
requirement that discharges to Outstanding Florida Waters remove at least 95% of the average annual pollutant load.
Technology-based performance standards such as these provide water quality goals for nonpoint sources that create equity
with the minimum treatment requirements for domestic wastewater point sources (Livingston, 1988).
Institutional framework: The stormwater program must have a strong institutional framework that assures that all BMPs
are (1) properly designed, (2) reviewed and approved, (3) inspected during and after construction, and (4) operated and
maintained. The components of this institutional framework are set forth in Figure 1. One of the most important components
especially for infiltration practices, is a feedback mechanism among system inspectors, plan reviewers, and designers
about what is working and what is not. This information can then be used to revise the design criteria for infiltration BMPs
and improve their potential for long-term success.
Technical Components:
Successful implementation of any BMP depends on a thorough understanding of the factors that determine the BMP's
treatment effectiveness, a strong scientific basis for the BMP's design criteria, and an understanding of the site conditions
that are required or that limit the utility of a specific BMP. Infiltration practices are also commonly called retention practices
because they retain the runoff on-site and are designed to infiltrate a design volume (treatment volume) of stormwater.
Factors that influence the treatment effectiveness and feasibility of infiltration practices include (1) precipitation patterns,
(2) whether the system is designed as an on-line or off-line system, (3) whether pretreatment via the BMP treatment train
is provided, and (4) site characterisitcs such as land use, soil type, geology, water table elevation, topography, and
vegetation.
Infiltration areas, especially off-line ones, can be incorporated easily into landscaping or open space areas of a site.
These can include natural or excavated grassed or landscaped depressions and recreational areas. Parking lots, with their
landscape islands, offer an excellent opportunity for the use of this concept since even the infiltration of a quarter inch of
runoff will greatly reduce sediments, metals, oils and greases. Placing storm sewer inlets within recessed parking lot
landscape areas, raising the inlet a few inches above the bottom, and using curb cuts to allow runoff to enter this area
represents a highly effective treatment train. If site conditions prevent the exclusive use of infiltration, then off-line retention
areas should be used as pretreatment practices in a stormwater treatment train. This is especially true if detention lakes
are the primary component of the stormwater system and the lakes are intended to serve as a focal point of the
development.
Pollutant removal efficiency factors: Average annual pollutant removal efficiency is calculated considering the annual mass
of pollutants available for discharge and the annual mass removed. The primary removal mechanism for infiltration practices
is the volume of stormwater that is infiltrated, since this eliminates the discharge of stormwater and its associated
pollutants. In addition, the system's vegetation and the surficial soils play an important role in binding and transforming
pollutants as the water infiltrates. As with any type of stormwater management practice, actual field efficiency will depend
on a large number of factors. For infiltration practices, such factors include:
1. Long-term precipitation characteristics; such as mean number of storms per year, their intensity and volume, and
average inter-event time.
84
-------
Fig yre 1, Storm wife r Program Institutional Framework Components
Inspections
Erosion/sediment controls
Storm water system construction
Storm water system operation
Plan review and approval
Site plans
Erosion control plans
Storm water plans
Feedback evaluation process
Storm water system operation/maintenance
Public facilities
Private facilities
Adopt a pond program
Education programs
General public
Elected officials
School curriculum
Designers
Structural BMPs
Design criteria
Re s ea rch/perfo rma n ce
Proper construction
Proper operation
Maintenance
Performance standards
Peak discharge rate
Volume
Treatment
Nonstmctural BMPs
Site planning
Source controls
Land acquisition
Street sweeping
Local Tand use plan
Administration
Lead agency?
Separate agency?
Staffing
Engineers
Inspectors
Planners
Scientists
Maintenance
Clerical
Program Evaluation
Citizen surveys
Bldg, community surveys
BMP monitoring
Water body monitoring
Developers
Builders
Practitioners
Inspectors
Comp F ia nce/enfo rcerrta nt
Stop-wo/k orders
Fines
Civil or criminal
Storm water Retrofitting
Watershed goal
Targ eting/pri o ri tizati on
Capital improvements
Regional BMPs
Storm water master planning Watershed planning
Integration wfth other federal, state, regional, and local programs
Adopt prog ram lawsfregulations Adopt starmwater utility ordinance/fees
Govt. Roles
Responsibilities
85
-------
2. The occurrence of first flush, which is related to the amount of directly connected impervious area; type of
stormwater conveyance system; and the pollutant of interest
3. Whether the system is an "on-line" or an "off-line" design
4. Cumulatively, the above three factors determine the minimum treatment volume and maximum storage recovery
time
The U. S. Weather Bureau has measured weather statistics at many locations around the country. Long-term
precipitation records, including such information as day and duration of event, intensity, volume, etc., are available from
either the Federal government or private vendors. Statistical analysis of these records can develop probability frequencies
for storm characteristics such as the mean storm volume and the mean inter-event period between storms.
"First flush" describes the washing action that stormwater has on accumulated pollutants in the watershed. In the early
stages of runoff, the land surfaces, especially impervious ones like streets and parking areas, are flushed clean by the
stormwater. This flushing creates a shock loading of pollutants. However, the occurrence and prevalence of first flush
depends largely on precipitation patterns, the degree of imperviousness of the contributing drainage area, the size of the
drainage area, and the type of stormwater conveyance system. Florida studies have determined that for highly impervious
urban land uses with drainage areas under 100 acres, there is a first flush for many pollutants, especially particulates
(Yousef et al., 1985; Miller, 1985). In areas such as Oregon and Washington, however, where rainfall consists of low
intensity, long-duration "events,"first flush is not very prevalent.
On-line stormwater practices store runoff temporarily before most of the volume is discharged to surface waters. These
systems capture all of the runoff from a design storm. This mixes all stormwater within the system, thereby masking first
flush and reducing pollutant removal. They primarily provide flood control benefits, with water quality benefits usually
secondary—although on-line wet detention systems provide both.
Off-line practices are designed to divert the more polluted first flush stormwater for water quality treatment, isolating
it from the remaining stormwater that is managed for flood control. In infiltration systems, the diverted first flush is not
discharged to surface waters, but is stored until it is gradually removed by infiltration, evaporation, and evapotranspiration.
Vegetation, such as grass in the bottom and sides of infiltration areas, helps to trap stormwater pollutants and reduce the
potential for transfer of these pollutants to groundwater. Off-line retention practices are the most effective for water quality
enhancement of stormwater.
Since an off-line retention area primarily provides for stormwater treatment, it must be combined with other BMPs for
flood protection to form a comprehensive stormwater management system. Figure 2 is a schematic of an off-line system,
commonly referred to as a "dual pond system." In these systems, a smart weir directs the first flush stormwater into the
infiltration area until it is filled, with the remaining runoff being routed to the detention facility for flood control.
A more recent investigation of the influence of long-term rainfall characteristics on the efficiency of retention practices
included inter-event dry periods, leading to the development of diversion volume curves for inter-event dry periods of varying
length (Wanielista et al., 1991 a). Figure 3 shows an example diversion volume curve for the Orlando area. It is important
to note that first flush is not considered in these curves. If a first flush effect does exist, the design curves would be
conservative in that the percent treatment efficiency of the infiltration system would increase. Furthermore, these curves
are based on precipitation-interevent-frequency (PIF) curves that also include consideration of the probability that a storm
greater than the design storm will occur. The PIF analysis looked at exceedance probabilities storms with a return period
of 2, 3, 4, or 6 months, representing a chance that the storm will exceed the design volume 6, 4, 3, or 2 times per year.
Other useful products from this project are Rate-Efficiency-Volume charts and curves that can be used to design wet
detention ponds that reuse runoff and thereby help to balance pre- and post-development volume (Livingston et al., 1999).
86
-------
GH-Une Tiwafmenf System
DfflfSiSM'l Ct-vil t
"Smart in*'
K
Dntot!:••, Pjri
"Smart" Box Bchemntle
OutlBt 5VrjOfUK3
or PIB-PK*.
Figure 2. Schematic of an Off-line Stormwater System.
C.S 1
Dvcia'on Volunc
4 faur/E 24ta.rLE
Figure 3. Diversion Volume Curve for Orlando, Florida.
87
-------
Site Characteristics: The suitability of a site for using infiltration practices will depend on a careful evaluation of the site's
natural attributes. Proposed infiltration areas should be evaluated for feasibility on any particular site or project by
examining the following:
SOILS - Must be suitable for infiltration. Nationally, most states recommend that soil textures should not have more
than 30% clay content or 40% silt content. Most importantly, they need to be able to percolate the diverted volume to
infiltrate within 72 hours, or within 24-36 hours for infiltration areas that are planted with grasses. Therefore, soils that have
been classified by the NRCS as HSG A are recommended for infiltration practices, although they can be successfully used
with HSG B soil types.
INFILTRATION RATES - In recent years, the minimum permeability rate recommended for infiltration practices has
been raised by implementing agencies. Shaver (1986) recommended a minimum rate of 0.25 inches per hour, but
Maryland's regulations now recommend 0.52 inches per hour. One of the most difficult aspects of designing infiltration
practices is obtaining reliable information about the actual infiltration rate of the soil where the practice will be constructed.
Unfortunately, such information is not easily obtainable. Avellaneda (1985) conducted 20 hydrologic studies of vegetated
swales constructed on sandy soils with a water table at least one foot below the bottom during dry conditions. Infiltration
rates were measured using laboratory permeability tests, double ring infiltrometers, and field mass balance experiments.
The field mass balance method measured a minimum infiltration rate of 2-3 inches/hour. This measured rate was much less
than lab permeabilities, rates measured by double ring infiltrometer tests (5-20 in/hr), or rates published in the Detailed Soil
Survey.
The following should be considered for determining the infiltration rate for retention practices:
1. Since the infiltration rate is the key to designing any retention practices, conservative estimates should be used,
and safety factors incorporated into the design to ensure that the design volume will actually be percolated into the
soil and not discharged downstream.
2. It is important that on-site infiltration measurements be taken at the locations where retention practices will be
located. More importantly, since soil characteristics and infiltration rate change with depth, it is crucial that the
measurements be made at the depth of the design elevation of the bottom of the retention practice.
3. Infiltration rates should be determined by mass balance field tests if possible. They provide the most realistic,
accurate estimate of the percolation rate. If field tests are not possible, infiltrometer tests should be used, with lab
permeability tests a third option. In either of these latter two tests, the design infiltration rate should be half of the
lowest measured rate. As a last resort, information from Detailed Soil Surveys can be used to estimated the
infiltration rate. However, the lowest rate should be used-as should a safety factor of two.
A recent assessment of infiltration practices in Carroll County, Maryland, quantified the infiltration rates for six basins
and six trenches of differing ages (Nelson et al., 1999). They found that 64% of the systems had an average infiltration
rate below the state's minimum recommended rate. However, 70% of the practices were still recovering their storage
volume within the required 72 hours. Interestingly, for some facilities (mainly trenches), the infiltration rate met or exceeded
the minimum state rate for a large percentage of the volume of water infiltrated, while the remaining water persisted for much
longer periods of time before infiltrating. This may indicate that (1) the infiltration rate is related to the hydraulic head, where
the higher depth of the stormwater in the BMP creates a higher pressure pushing water into the ground, or (2) the bottoms
of the systems accumulate fines that impede percolation, while the sides of the systems are still infiltrating runoff rapidly.
WATER TABLE - The seasonal high water table should be at least three feet beneath the bottom of the infiltration area
to assure that stormwater pollutants are removed by the vegetation, soil, and microbes before contacting the groundwater.
Jacobson and Horner (1993) recommend a minimum of five feet if the seasonal high water table cannot be estimated
accurately. When considering the groundwater elevation, it is important to remember that the retention area can cause a
mounding effect on the water table, thereby raising it above the predevelopment level. The Southwest Florida Water
Management District (SWFWMD) has developed a model that can be used to more accurately determine the seasonal high
water table and the effects of mounding (SWFWMD, 1998).
88
-------
GEOLOGY - Bedrock should be at least three feet beneath the bottom of the infiltration area. In those parts of the
country where limestone is at or near the land surface, special precautions must be taken when using infiltration practices.
The potential for groundwater contamination in such areas is quite high, especially in "Karst Sensitive Areas" (KSA) where
sinkhole formation is common. In KSAs, solution pipe sinkholes may form in the bottom of infiltration areas creating a
direct conduit for stormwater pollutants to enter the groundwater. Solution pipes often open in the bottom of retention areas
because the natural soil plug capping the solution pipe is thinned by partial excavation to create the retention area and
because the stormwater creates a hydraulic pressure which can wash out the plug.
In KSAs, a site-specific hydrogeologic investigation should be undertaken that includes geologic borings wherever
infiltration areas are proposed and mapping limerock outcroppings and sinkholes on site. Infiltration systems in KSAs
should (1) include several small off-site areas, (2) use swale conveyances for pretreatment, (3) be as shallow as possible,
(4) be vegetated with a permanent cover such as sodded grasses, and (5) have flat bottoms to keep the stormwater spread
out across the entire area.
TOPOGRAPHY - Infiltration practices should not be located on areas with slopes over 20% to minimize the chance
of downstream water seepage from the subgrade. Sloping sites often require extensive cut and fill operations. Infiltration
practices should not be sited on fill material, since fill areas are very susceptible to slope failure, especially when the
interface of the fill/natural soil becomes saturated.
VEGETATION - To reduce the potential for stormwater pollutants to enter groundwater, and to help maintain the soil's
capacity to absorb water, infiltration practices should be vegetated with appropriate native vegetation, especially grasses.
However, this type of vegetation cannot tolerate long-term inundation, so the retention area must be capable of infiltrating
all of its runoff within a relatively short time period (i.e., 24 to 36 hours). The design of "bioretention" systems incorporates
soils and vegetation that are proficient in trapping stormwater pollutants within them and takes advantage of microbial
processes that help transform and trap pollutants in the terrestrial environment.
SET BACKS - Infiltration areas should be located at least 100 feet from any water supply well and at least 12 feet down-
gradient from any building foundations. Additionally, they should be set back at least 50 feet from on-site wastewater
systems, especially drain fields.
LAND USE RESTRICTIONS - Certain infiltration practices can only be applied to particular land uses. For example,
some sites are so small or intensively developed that space is insufficient for surficial practices (e.g., retention basin), but
they may allow for infiltration or exfiltration trenches if pretreatment can be provided. A concern with any infiltration practice
is the potential for hazardous or toxic wastes to enter the system and migrate into the groundwater. Land uses where such
substances are used should implement comprehensive pollution prevention, spill containment, and emergency response
plans that will prevent dangerous materials from getting into the infiltration system.
POTENTIAL FOR GROUNDWATER POLLUTION - A possible concern about infiltration practices is whether they
simply are transferring the stormwater pollution problem from surface waters to groundwaters. Stormwater pollutants,
especially heavy metals, quickly bind to soil particles and vegetation is effective in filtering pollutants, thereby minimizing
the risk of groundwater contamination (Harper, 1985; Yousef et al, 1985b). However, groundwater beneath swales, and
retention areas located in highly sandy soils with low organic content, did show elevated levels of heavy metals down to
depths of 20 feet (Harper, 1988).
Design Criteria
Once all of the above factors have been quantified using state, regional, or local data as appropriate, specific design
criteria can be established. Table 2 summarizes the design criteria for infiltration systems set forth in Florida's stormwater
regulations. St. John's River Water Management District (SJRWMD, 1992).
89
-------
Table 2. Florida's Design Criteria for Infiltration Practices
BMP 80% Treatment Effectiveness Diversion Volume
Swales Infiltrate 80% of the runoff from a 3-yr, 1-hr storm (2.5 inches)
Retention - Off-line Infiltrate the larger of 0.5 inches of runoff or 1.25" X % impervious
Retention - On-line Infiltrate an additional 0.5 inches of runoff
DESIGN FACTOR CRITERIA
Soil type HSG A or B with < 30% clay or < 40% silt/clay
Treatment volume recovery time 72 hours, 24 to 36 hours if grassed
Water table or bedrock At least 3 feet beneath bottom after mounding
Topography On slopes < 20%, not on fill soils
Vegetation Recommended to reduce potential for groundwater pollution and to maintain soil permeability
Land use May not be appropriate at sites where hazardous materials spills may occur
Swales: Traditionally, swales are used primarily for stormwater conveyance and, as such, are considered an on-line
practice. The removal of stormwater pollutants by swales can occur by infiltration or vegetative filtration. Investigations
in Florida (Yousef et al., 1985a; Harper, 1985) have concluded that swale treatment efficiency largely depends on the
volume of stormwater that can be infiltrated through the filtering vegetation and into the soil.
Avellaneda (1985) developed the following equation for a triangular-shaped swale to estimate the length of swale
necessary to infiltrate the design stormwater treatment volume:
5/8 c 3/16
n3/8i
where:
L = swale length (m) n = Mannings roughness coefficient
Q = average runoff flow rate (m3/S) i = infiltration rate (cm/hr)
S = longitudinal slope (m/m)
K = constant which is a function of side slope that varies from 4,722 to 10,516
For most residential, commercial, and highway projects, the length of swales necessary to percolate the stormwater
needed to achieve the 80% performance standard was found to be excessive, or at least twice the distance available.
Thus, some type of swale block (berm) or on-line detention/retention may be more helpful. Swales make excellent
pretreatment practices by providing for the infiltration of some stormwater and for some vegetative filtration. By using a
raised storm sewer inlet, swales can provide water quality enhancement via retention and still serve as effective
conveyances for flood protection. Swales can incorporate retention by using swale blocks, small check dams, or elevated
driveway culverts to create storage; thereby reducing runoff velocity, reducing erosion, and promoting infiltration. In
highway designs for high speed situations, safety must be considered; thus, a maximum depth of water equal to 1.5 feet
and flow line slopes on the berms of 1 vertical/20 horizontal are recommended. Along lower speed highways or in some
residential/commercial urban settings, steeperflow line berm slopes (1 on 6) are acceptable (Wanielista et.al., 1986).
Unlike Florida, investigations in Washington State (Horner, 1988; WPCD, 1992) indicate that swales can also act as
a biofilter, with removal of particulate pollutants without infiltration of stormwater. The following recommendations were made
to improve their water quality benefits:
1. Maximum design velocity should not exceed 27 cm per second.
90
-------
2. A hydraulic residence time of at least 9 minutes is recommended for removal of about 80% of the total suspended
solids. Longer residence times will provide higher removal effectiveness.
3. Swale width should be limited to 6 to 8 feet, unless special measures are provided to assure a level swale bottom,
uniform flow spreading, and management of flows to prevent formation of low-flow channels.
4. Swale slopes should be between 2 and 4%.
5. Water depth should be limited to no greater than one half the height of the grass, up to a maximum of 3 inches of
water depth.
6. Swale length will be a function of the hydraulic residence time, swale width, and stormwater volume and velocity.
Implementation Components
Even if effective design criteria have been established for an infiltration system, the design has been reviewed and
approved, and the institutional framework to assure performance has been established, an infiltration system may still not
work correctly. In fact, assessments of the success or failure of infiltration systems have determined that poor construction
is a major factor in system failure (Pensyl and Clement, 1987; Lindsey et al., 1991). We will discuss five considerations
that are essential to proper implementation of infiltration practices including (1) education, (2) erosion and sediment control,
(3) construction, (4) inspections, and (5) maintenance.
Education: The stormwater program needs to include an extensive education program that targets BMP designers,
plan reviewers, inspectors, contractors, and maintenance personnel. Each of these practitioners is an important part of
the stormwater team. They must each understand their role in BMP design and implementation, as well as the technical
factors discussed above. Additionally, a communication mechanism needs to be established among all of these
practitioners so that in-the-field knowledge of what works, and what does not work, is transferred back to all other team
members. With respect to BMP installation, the plan reviewers and inspectors should meet with the project engineers and
contractors on-site to review the site plan, construction sequencing, erosion and sediment control plan and details, and the
infiltration system's detailed standards and specifications.
Erosion and sediment control: Infiltration practices must be protected from sediment loadings, especially during the
project's construction phases. Infiltration practices should never be used as part of the erosion and sediment control
system, nor should they be put into operation until all contributing drainage areas are fully stabilized. Although sediment
loads drop sharply after construction is completed, gradual clogging of infiltration practices can still occur. Pre-treatment
practices such as swale conveyances or vegetated buffer strips can help to filter out sediments and extend the life of
retention practices. Do not forget the treatment train concept.
Construction considerations: To prevent clogging of infiltration areas, special precautions must be taken during the
entire construction phase of a project to prevent reduction of the system's infiltration capacity. In particular, two areas need
to be stressed, including preventing sedimentation during construction and preventing compaction of the soil. Areas that
are selected for infiltration use should be well marked during site surveying and protected during construction. Specifical
construction recommendations are as follows (WMI, 1997b):
1. If possible, schedule construction so it does not occur during the rainy season but does occur during the vegetation
growing season. For example, in Auckland, New Zealand, construction sites are shut down during the winter when
long, prolonged rains make erosion and sediment controls ineffective and when vegetation does not
grow well. Unfortunately, in the United States, these seasons often overlap and the economics of development
dictate the time frame for starting construction.
2. Before the development site is graded, areas planned for use as infiltration systems should be well marked during
site surveying, and all traffic and heavy equipment kept away from the area to prevent compacting the underlying
soils.
91
-------
3. Construction should be overseen by someone who is trained and experienced in the installation of infiltration
practices, and who is knowledgeable about their purpose and operation.
4. The design team should inspect the exposed soil after excavation to confirm that soil conditions are as expected
and are suitable for the approved design. If they are not, work should not proceed and the situation should be
analyzed to determine whether or not design or construction changes should be made to the approved design.
5. Construction of the infiltration system should not begin until after the site has been completely stabilized. If this
is not possible, then:
a. Diversion berms should be placed around the perimeter of the infiltration area during all phases of construction
to divert runoff and sediment away from it.
b. Sediment and erosion control plans for the site should be oriented to keep sediment and runoff completely away
from the area.
c. The facility should not be excavated to final grade until after the contributing drainage area is stablilized. Leave
at least two feet of native soil during the initial excavation.
6. Infiltration areas should never be used as a temporary sediment basins during the construction phase.
Unfortunately, it is common for infiltration areas, especially basins, to be used as a sediment trap, with initial
excavation to within two feet of the final design elevation of the basin floor. If the facility is to be used during
construction, this soil can be removed in layers as it clogs. Once construction is completed, sediment that
accumulated during the construction phase can then be removed when the basin undergoes final excavation to its
design elevation. However, recent experience indicates that even with this type of construction practice, infiltration
areas used as sediment traps have a higher rate of failure.
7. Infiltration areas/basins should be excavated using light earth-moving equipment with tracks or over-sized tires.
Normal rubber tires should be avoided since they compact the subsoil and reduce its infiltration capabilities. For
the same reason, the use of bulldozers or front end loaders should be avoided.
8. During construction, place excavated material at least 10-15 feet away from the infiltration area.
9. Since some compaction of the underlying soils is still likely to occur during excavation, the floor of the basin should
be deeply tilled with a rotary tiller or disc harrow at the end of the excavation process.
10. Rock used in infiltration or exfiltration trenches should be washed clean of sediments. Rock should be placed in
lifts and compacted after each lift.
11. Trenches should be clear of any protruding objects and carefully inspected before installing geotextile fabrics. The
fabrics should have the proper permeability and be installed with at least a 12-inch overlap, in a shingle-like manner
12. Trenches should be covered and not put into operation until the contributing drainage area is completely stabilized
and all pretreatment BMPs installed.
13. Pervious asphalt and concrete should be installed only by certified personnel who are specifically trained in their
batching, pouring, and finishing.
14. The basin should be stabilized with vegetation within a week after construction. Use of low maintenance, rapidly
germinating native grasses are recommended. The condition of the newly established vegetation should be
checked several times over the first two months, and any necessary remedial actions taken (e.g., reseeding,
fertilization, and irrigation).
92
-------
Inspections: Like all stormwater treatment practices, infiltration systems need to be inspected during construction and
on a regular basis after construction. Inspections during construction are needed to assure that the infiltration system is
built in accordance with the approved design and standards and specifications. Five inspections are recommended: (1)
pre-construction, (2) during excavation, (3) during construction of the embankment (if applicable), (4) after final excavation,
and (5) after construction is completed. During this final inspection, the inspector should have a copy of the "As-built or
record drawings." In addition, infiltration systems should be inspected semi-annually after construction (before and after
wet seasons) to ensure that they continue to function. Two site trips are recommended: one during or immediately after
a rainfall, so that conditions during operation can be observed, and a second from 24 to 72 hours after the rainfall, to
determine whether the system is recovering its storage volume as designed. Inspection forms are highly recommended.
Examples of inspection forms (WMI, 1997b) can be downloaded from the EPA web site located at
http://www.epa.gov/owow/NPS/orderform.html.
Maintenance: All infiltration practices will require regular and non-routine maintenance to maintain their ability to
infiltrate stormwater. The frequency and need for maintenance will depend primarily on the loading of particulates and
whether pre-treatment practices have been employed. Routine maintenance includes revegetating eroding areas, removing
materials that accumulate in pretreatment BMPs, and removing materials from inlets and outlets. Non-routine, restorative
maintenance activities should be conducted whenever inspections reveal that stormwater remains in the system beyond
the designed time. These may include structural repairs to the inlets or outlets and restoration of the infiltration capability
of the system.
Additional Concerns and Recommendations
Concerns with Pervious Pavement: Local land development codes typically specify the type of material for a parking
lot (i.e., paved, grass, gravel) and determine the number and size of parking spaces within a parking lot. These
requirements should be reviewed carefully to ensure that they are necessary (is paving really required in every case) and
that the number of spaces is related to actual traffic demands. After these requirements have been reviewed and verified,
the use of pervious pavement within a parking lot should be examined. Pervious pavement materials include pervious
asphalt, pervious concrete, turf blocks, and even Geoweb covered with sod.
Overall, experiences with pervious pavements have not been very good. Pervious pavements have been prone to
clogging. Causes include poor erosion and sediment control during construction, unstabilized drainage areas after
construction, improper mixing and finishing of the pavement, and poor maintenance. However, field investigations of
pervious concrete that has been in use for up to 15 years in Florida indicate that these parking lots can continue to infiltrate
rainfall and runoff if they were installed and maintained properly (FCMA, undated). Pervious concrete not only helps reduce
site imperviousness, but also reduces hydroplaning and road noise.
Recommendations: Specific recommendations and other important information about infiltration systems that will help
increase their successful implementation are summarized in Table 3. This table includes essential information about the
advantages, disadvantages, maintenance, and other aspects of successfully using infiltration practice. To improve
evaluation of site conditions for the suitability of infiltration practices, Jacobson and Horner (1993) recommended a quan
titative rating system. The factors used in the system included: (1) soil till layer (presence and location), (2) location of
seasonal high water table, (3) removal efficiency of the pretreatment BMPs, (4) degree of siltation protection, (5) soil type,
and (6) infiltration rate. Different degrees of acceptability are possible: (1) disqualifying (characteristics that eliminate design
or location from further consideration), (2) passable (characteristics that allow consideration but not ideal), and (3) ideal
(optimum characteristics for design or siting of facility). Table 4 illustrates the proposed rating system factors.
93
-------
Table 3. Additional Information About Infiltration Systems to Enhance Successful Implementation
Infiltration
Bmp Type
Surface Basin
(Typically recessed
areas or, in Mid-Atlantic
States, rock filled)
Infiltration Trench
(Typically a rock filled
trench)
Exfiltration Trench
(Typically a perforated
pipe with a gravel
envelope)
Advantages
•Integrate into land-
scaping, open space,
parking lot islands
•Use for recreation
•Easier inspection and
maintenance
•Can be used where
land or space is limited
•Can be used where
land or space is
limited
Disadvantages
•Land area required
•Potential mosquito
problem if not designed
or maintained properly
•Easily clogged
•Difficult to unclog
•Difficult to monitor
performance
•Easily clogged
•Difficult to unclog
•Difficult to monitor
performance
Maintenance
•Vegetated basins should be
mowed and clippings removed
•Remove sediments when dry
and cracked
•Non-vegetated basins require
annual disking
•Removing sediments that
accumulate in rocks
•Remove materials that enter
pipe
•High pressure wash perforated
pipe
Comments
•Can serve larger drainage
3:1 or flatter side slopes, flat
bottom
•Bottom and side slope vege-
tation recommended
•Pretreatment essential
•Use observation well
•Keep covered until drainage
area stabilized
•Pretreatment essential
•Source controls useful
•Geotextile is limiting infiltra-
tion factor
Pervious Pavement
Swale (Typically
a shallow, grassed
conveyance system)
•Reduces impervious-
ness
•Reduces hydroplaning
and highway noise
•Higher recharge rates
•Can be incorporated
into site 's landscaping/
open areas
•Great car in BMP Train
•Easily clogged
•Lack of trained
practitioners
•Anaerobic conditions
may develop in soils
•Not for flood control
•May "disappear" from
residential back yards
•May become depository
fnr traQh varH waQtAQ
•Removing sediments that
Accumulate in rocks, replace
rocks
•Regular vacuum street
sweeping
•High pressure cleaning
•Drilling holes to restore
infiltration
•Replacement
•Mow and remove grass
clippings
•Hydroscope accumulated
sediments and resod
•Inf. rate 0.5'Vhr if use sides
and bottom
•Inf. rate 1"/hr if use bottom
•Proper batching and place-
ment is crucial
•Education programs needed
for practitioners
•Post signs to inform users
and keep dir and mud out
•Wet swales (wetland plants)
work great too
•Use swale blocks, raised
driveway culverts to retain
Table 4. Possible Rating System to Evaluate the Suitability of Infiltration BMPs
Disqualifying Passable
Ideal
Factor
Soil till layer
Characteristic
Impenetrable, thick layer
near surface
Characteristic
Layer present but at >5' depth, or
easily penetrable
Characteristic
No till layer present
Seasonal High Water
Table
Close to surface, within 5'
At intermediate depth, at least 5' below
BMP bottom
Very deep, well below
BMP bottom
Pretreatment
None provided
Some,minimum 50% TSS removal
Pretreatment provides >80%
TSS removal
Siltation
Protection
None provided
Any silt or construction sediment
removed before final BMP construction
Fully protected from silt during
construction
Soils
Saturated or with >30% clay
or >40% silt/clay content
Coarse, highly infiltrative soil that can be
modified to produce proper inf. rate
Loam or loamy sand
Infiltration Rate
<0.57hr
>2.5"/hr but with very deep water table
or modified to slower rate
0.5 to 2.57hr
94
-------
References
Avellaneda, Eduardo. Hydrologic Design of Swales. M.S. Thesis, University of Central Florida, Orlando, FL December
1985.
Florida Concrete Manufacturers Association. Portland Cement Pervious Pavement Manual. FCPMA, 649 Vassar Street,
Orlando, FL, 32804. No date.
Gaus, J. J. Soils of Stormwater Infiltration Basins in the Puget Sound Region: Trace Metal Form and Concentration and
Comparison to Washington State Department of Ecology Guidelines. M. S. Thesis. College of Forest Resources, U. of
Washington, Seattle, WA 1993.
Harper, Harvey H. Effects of Stormwater Management Systems on Ground Water Quality. Final Report, Contract WM-190,
Submitted to Florida Dept. of Environmental Regulation, Tallahassee, FL 1988.
Harper, Harvey H. Fate of Heavy Metals from Highway Runoff in Stormwater Management Systems. Ph.D. Dissertation,
University of Central Florida, Orlando, FL 1985.
Hilding, K. A Survey of Infiltration Basins in the Puget Sound Area. Non-thesis paper, Dept. of Biological and Agricultural
Engineering, University of California, Davis, CA 1993.
Horner, R.R. Biofiltration Systems for Storm Runoff Water Quality Control. Final Report Submitted to City of Seattle, WA
1988.
Jacobson, M.A. and R.R. Horner. Summary and Conclusions from Applied Research on Infiltration Basins and
Recommendations for Modifying the Washington Department of Ecology Stormwater Management Manual for the Puget
Sound Basin. Submitted to Urban Nonpoint Management Unit, WDOE, Seattle, WA 1993
Klochak, J.R. An Investigation of the Effectiveness of Infiltration Systems in Treating Urban Runoff. M.S.E. Thesis,
Department of Civil Engineering, University of Washington, Seattle, WA 1992.
Lindsey, G., L. Roberts and W. Page. Inspection and Maintenance of Infiltration Facilities. J. Soil Water Conserv. 47 (6):
481-186. 1992.
Livingston, E.H. State Perspectives on Water Quality Criteria. In: Design of Urban Runoff Controls, an Engineering
Foundation Conference, Potosi, MO 1988.
Livingston, E.H. Successful Stormwater Management: Selecting and Putting the Puzzle Pieces Together. In: Proc. Future
Directions for Australian Soil and Water Management, September 9-12, 1997, Brisbane, Australia 1997.
Livingston, E.H. M.P. Wanielista, J. Bradner. Stormwater Reuse: An Added Benefit of Wet Detention Systems. In:
Proceedings of the First South Pacific Conference on Comprehensive Stormwater and Aquatic Ecosystem Management,
Auckland, New Zealand, 1999.
Miller, R.A. Percentage Entrainment of Constituent Loads in Urban Runoff, South Florida. USGS WRI 84-4329. 1985.
Nelson, S., T. Devilbiss, and E. Bolton. Assessment of Stormwater Management Infiltration Practices in Carroll County,
MD Final Report. March 15, 1999.
Pensyl, L.K. and P.F. Clement. Results of the State of Maryland Infiltration Practices Survey. Presented at the State of
Maryland Sediment and Stormwater Management Conference, Washington College, Chestertown, MD August 1987.
Prince George's County, Maryland. Bioretention Manual. Dept. of Env. Resources. 1993.
95
-------
Prince George's County, Maryland. Low-Impact Design Manual. Dept. of Env. Res. 1997.
St. Johns River Water Management District. Rule 40C-42 and associated Applicant's Handbook. Palatka, FL 1992.
Shaver, H. Earl. Infiltration as a Stormwater Management Component. In: Proceedings of the Urban Runoff Technology,
Engineering Foundation Conference, Henniker, NH 1986,
Southwest Florida Water Management District. Methods for Identifying Soils and Determining Seasonal High Ground Water
Elevations. Handout Literature for Technical Training Workshops. Brooksville, FL. 1998.
U.S. EPA, "Results of the Nationwide Urban Runoff Program: Volume 2: Final Report", Washington, DC, 1983.
Wanielista, M.P. Quality Considerations in the Design of Holding Ponds, Stormwater Retention/Detention Basins Seminar,
University of Central Florida, Orlando, FL 1977.
Wanielista, M.P., Y.A. Yousef, L. M. Van DeGroaff and S.H. Rehmann-Koo. Best Management Practices - Enhanced
Erosion and Sediment Control Using Swale Blocks. Final Report, FL-ER-35-87, Submitted to Florida Dept. of Transportation,
Tallahassee, FL 1986b.
Wanielista, M.P., M.J. Gauthier, and D.L. Evans. Design and Performance of Exfiltration Systems. Final Report, Contract
C3331, Submitted to the Florida Dept. of Transportation, Tallahassee, FL 1991 a.
Wanielista, M.P., Y.A. Yousef, G.M. Harper, T.R. Lineback, and L. Dansereau. Precipitation, Inter-Event Dry Periods, and
Reuse Design Curves for Selected Areas of Florida. Final Report, Contract WM- Submitted to the Florida Dept.
Environmental Regulation, Tallahassee, FL 1991b.
Water Pollution Control Dept. Biofiltration Swale Performance, Recommendations, and Design Considerations. Publication
657, Seattle, WA 1992.
Watershed Management Institute, Inc. Institutional Aspects of Urban Runoff Management: A Guide for Program
Development and Implementation. Cooperative Agreement with the Nonpoint Source Control Branch, U.S. EPA,
Washington, D.C. 1997a.
Watershed Management Institute, Inc. Operation, Maintenance, and Management of Completed Stormwater Management
Systems. Cooperative Agreement with the Nonpoint Source Control Branch, U.S. EPA, Washington, D.C. 1997b.
Yousef, Y.A., M.P. Wanielista, H.H. Harper, D.B. Pearce and R.D. Tolbert. Best Management Practices - Removal of
Highway Contaminants by Roadside Swales. Final Report, Submitted to the Florida Dept. of Transportation, Tallahassee,
FL 1985a.
Yousef, Y.A., M.P Wanielista, and H. Harper. Fate of Pollutants in Retention/Detention Ponds. In: Stormwater
Management: An Update. Publication 85-1, University of Central Florida, Orlando, FL. 1985b.
96
-------
Potential New Tools for the Use of Tracers to Indicate Sources of
Contaminants to Storm Drainage Systems
Robert Pitt, Melinda Lalor, Jennifer Harper, and Christy Nix
Department of Civil and Environmental Engineering
The University of Alabama at Birmingham
Donald Barbe'
Department of Civil and Environmental Engineering
The University of New Orleans
Abstract
This paper is a description of previously developed methods used to identify sources of contaminants in storm
drainage systems, plus a review of emerging techniques that may also be useful. The original methods, along with
selected new procedures, were tested using almost 700 stormwater samples collected from telecommunication manholes
from throughout the U.S. About 10% of the samples were estimated to be contaminated with sanitary sewage, using these
methods, similar to what is expected for most stormwater systems. The original methods are still recommended as the
most useful procedure for identifying contamination of storm drainage systems, with the possible addition of specific tests
for E. co//and enterococci and UV absorbance at 228 nm. Most newly emerging methods require exotic equipment and
unusual expertise and are therefore not very available, especially at low cost and with fast turn-around times for the
analyses. These emerging methods may therefore be more useful for special research projects than for routine screening
of storm drainage systems.
Introduction
Urban stormwater runoff includes waters from many other sources which find their way into storm drainage systems,
besides from precipitation. There are cases where pollutant levels in storm drainage are much higher than they would
otherwise be because of excessive amounts of contaminants that are introduced into the storm drainage system by
various non-stormwater discharges. Additionally, baseflows (during dry weather) are also common in storm drainage
systems. Dry-weather flows and wet-weather flows have been monitored during numerous urban runoff studies. These
studies have found that discharges observed at outfalls during dry weather were significantly different from wet-weather
discharges and may account for the majority of the annual discharges for some pollutants of concern from the storm
drainage system.
There have been numerous methods used to investigate inappropriate discharges to storm drainage systems. Pitt,
ef a/. (1993) and Lalor (1994) reviewed many of these procedures and developed a system that municipalities could use
for screening outfalls in residential and commercial areas. They are currently updating these earlier methods under
funding from the U.S. EPA and the University of New Orleans. In these areas, sewage contamination, along with low-rate
discharges from small businesses (especially laundries, vehicle repair shops, plating shops, etc.) are of primary concern.
One of the earliest methods used to identify sewage contamination utilized the ratio of fecal coliform to fecal strep.
bacteria. This method is still in use, but unfortunately has proven inaccurate in most urban stormwater applications The
following discussion reviews the methodology developed by Pitt, etal. (1993) and Lalor (1994), and some new
approaches that were investigated.
Use of Tracers to Identify Sources of Contamination in Urban Drainage Systems
This research investigated inappropriate discharges into storm drainage systems. It was of most concern to identify toxic
or pathogenic sources of water, typically raw sewage or industrial wastewaters, that were being discharged accidentally
into the storm drainage system.
97
-------
Investigations designed to determine the contribution of urban stormwater runoff to receiving water quality problems
have led to a continuing interest in inappropriate connections to storm drainage systems. Urban stormwater runoff is
traditionally defined as that portion of precipitation which drains from city surfaces and flows via natural or man-made
drainage systems into receiving waters. In fact, urban stormwater runoff also includes waters from many other sources
which find their way into storm drainage systems. Sources of some of this water can be identified and accounted for by
examining current National Pollutant Discharge Elimination System (NPDES) permit records for permitted industrial
wastewaters that can be legally discharged to the storm drainage system. However, most of the water comes from other
sources, including illicit and/or inappropriate entries to the storm drainage system. These entries can account for a
significant amount of the pollutants discharged from storm sewerage systems (Pitt and McLean 1986).
Permits for municipal separate storm sewers include a requirement to effectively prohibit problematic non-stormwater
discharges, thereby placing emphasis on the elimination of inappropriate connections to urban storm drains. Section
122.26 (d)(l )(iv)(D) of the rule specifically requires an initial screening program to provide means for detecting high levels
of pollutants in dry weatherflows, which should serve as an indicator of illicit connections to the storm sewers. To facilitate
the application of this rule, the EPA's Office of Research and Development's Storm and Combined Sewer Pollution
Control Program and the Environmental Engineering & Technology Demonstration Branch, along with the Office of
Water's Nonpoint Source Branch, supported research for the investigation of inappropriate entries to storm drainage
systems (Pitt, et a/. 1993). The approach presented in this research was based on the identification and quantification
of clean baseflow and the contaminated components during dry weather. If the relative amounts of these components
are known, the importance of the dry weather discharge can be determined.
The ideal tracer to identify major flow sources should have the following characteristics:
. Significant difference in concentrations between possible pollutant sources;
. Small variations in concentrations within each likely pollutant source category;
. A conservative behavior (i.e., no significant concentration change due to physical, chemical or
biological processes); and,
. Ease of measurement with adequate detection limits, good sensitivity and repeatability.
In order to identify tracers meeting the above criteria, literature characterizing potential inappropriate entries into storm
drainage systems was examined. Several case studies that identified procedures used by individual municipalities or
regional agencies were also examined.
Parameters Suitable for Indicators of Contamination by Sanitary Sewage
Tracer Characteristics of Local Source Flows. Table 1 is a summary of tracer parameter measurements for
Birmingham, Alabama. This table is a summary of the "library" that describes the tracer conditions for each potential
source category. The important information shown on this table includes the median and coefficient of variation (COV)
values for each tracer parameter for each source category. Appropriate tracers are characterized by having significantly
different concentrations in flow categories that need to be distinguished. In addition, effective tracers also need low COV
values within each flow category. The study indicated that the COV values were quite low for each category, with the
exception of chlorine, which had much greater COV values. Chlorine is therefore not recommended as a quantitative
tracer to estimate the flow components. Similar data must be collected in each community where these procedures are
to be used. Recommended field observations include color, odor, clarity, presence of floatables and deposits, and rate
of flow, in addition to the selected chemical measurements.
Simple Data Evaluation Methods to Indicate Sources of Contamination
Indicators Implying Contamination. Indicators of contamination (negative indicators) are clearly apparent visual or
physical parameters indicating obvious problems and are readily observable at the outfall during the field screening
98
-------
Table 1. Tracer Concentrations Found in Birmingham, Alabama, Waters (Mean, Standard Deviation and Coefficient of Variation)
Fluorescence
(% scale)
Potassium
(mg/L)
Ammonia
(mg/L)
Ammonia/Potassium
(ratio)
Fluoride
(mg/L)
Toxicity
(% light decrease after
25 minutes, I26)
Surfactants
(mg/LasMBAS)
Hardness
(mg/L)
PH
(pH units)
Color
(color units)
Chlorine
(mg/L)
Specific conductivity
(pS/cm)
Number of samples
Spring
water
6.8
2.9
0.43
0.73
0.070
0.10
0.009
0.016
1.7
0.011
0.02
2.0
0.031
0.027
0.87
<5
n/a
n/a
<0.5
n/a
n/a
240
7.8
0.03
7.0
0.05
0.01
<1
n/a
n/a
0.003
0.005
1.6
300
12
0.04
10
Treated
potable water
4.6
0.35
0.08
1.6
0.059
0.04
0.028
0.006
0.23
0.018
0.006
0.35
0.97
0.014
0.02
47
20
0.44
<0.5
n/a
n/a
49
1.4
0.03
6.9
0.29
0.04
cl
n/a
n/a
0.88
0.60
0.68
110
1.1
0.01
10
Laundry
wastewater
1020
125
0.12
3.5
0.38
0.11
0.82
0.12
0.14
0.24
0.050
0.21
33
13
0.38
99.9
<1
n/a
27
6.7
0.25
14
8.0
0.57
9.1
0.35
0.04
47
12
0.27
0.40
0.10
0.26
560
120
0.21
10
Sanitary
wastewater
250
50
0.20
6.0
1.4
0.23
10
3.3
0.34
1.7
0.52
0.31
0.77
0.17
0.23
43
26
0.59
1.5
1.2
0.82
140
15
0.11
7.1
0.13
0.02
38
21
0.55
0.014
0.020
1.4
420
55
0.13
36
Septic
tank
effluent
430
100
0.23
20
9.5
0.47
90
40
0.44
5.2
3.7
0.71
0.99
0.33
0.33
99.9
<1
n/a
3.1
4.8
1.5
235
150
0.64
6.8
0.34
0.05
59
25
0.41
0.013
0.013
1.0
430
311
0.72
9
Car wash
water
1200
130
0.11
43
16
0.37
0.24
0.066
0.28
0.006
0.005
0.86
12
2.4
0.20
99.9
<1
n/a
49
5.1
0.11
160
9.2
0.06
6.7
0.22
0.03
220
78
0.35
0.070
0.080
1.1
485
29
0.06
10
Radiator
water
22,000
950
0.04
2800
375
0.13
0.03
0.01
0.3
0.011
0.011
1.0
150
24
0.16
99.9
<1
n/a
15
1.6
0.11
50
1.5
0.03
7.0
0.39
0.06
3000
44
0.02
0.03
0.016
0.52
3300
700
0.22
10
activities. These observations are very important during the field survey because they are the simplest method of
identifying grossly contaminated dry-weather outfall flows. The direct examination of outfall characteristics for unusual
conditions of flow, odor, color, turbidity, floatables, deposits/stains, vegetation conditions, and damage to drainage
structures is therefore an important part of these investigations. Table 2 presents a summary of these indicators, along
with narratives of the descriptors to be selected in the field.
99
-------
Table 2. Interpretations of Physical Observation Parameters and Likely Associated Flow Sources
Odor - Most strong odors, especially gasoline, oils, and solvents, are likely associated with high responses on the toxicity screening test. Typical obvious
odors include: gasoline, oil, sanitary wastewater, industrial chemicals, decomposing organic wastes, etc.
sewage: smell associated with stale sanitary wastewater, especially in pools near outfall.
sulfur ("rotten eggs"): industries that discharge sulfide compounds or organics (meat packers, canneries, dairies, etc.).
oil and gas: petroleum refineries or many facilities associated with vehicle maintenance or petroleum product storage.
rancid-sour, food preparation facilities (restaurants, hotels, etc.).
Color - Important indicatorof inappropriate industrial sources. Industrial dry-weather discharges may be of any color, but darkcolors, such as brown, gray,
or black, are usually of most common.
yellow: chemical plants, textile and tanning plants.
brown: meat packers, printing plants, metal works, stone and concrete, fertilizers, and petroleum refining facilities.
green: chemical plants, textile facilities.
red: meat packers.
gray: dairies.
Turbidity - Often affected by the degree of gross contamination. Dry-weather industrial flows with moderate turbidity can be cloudy, while highly turbid
flows can be opaque. High turbidity is often a characteristic of undiluted dry-weather industrial discharges.
cloudy: sanitary wastewater, concrete or stone operations, fertilizer facilities, automotive dealers.
opaque: food processors, lumber mills, metal operations, pigment plants.
Floatable Matter - A contaminated flow may contain floating solids or liquids directly related to industrial or sanitary wastewater pollution. Floatables of
industrial origin may include animal fats, spoiled food, oils, solvents, sawdust, foams, packing materials, or fuel.
oil sheen: petroleum refineries or storage facilities and vehicle service facilities.
sewage: sanitary wastewater.
Deposits and Stains- Refer to any type of coating near the outfall and are usually of a dark color. Deposits and stains often will contain fragments of
floatable substances. These situations are illustrated by the grayish-black deposits that contain fragments of animal flesh and hair which often are
produced by leather tanneries, or the white crystalline powder which commonly coats outfalls due to nitrogenous fertilizer wastes.
sediment construction site erosion.
oily petroleum refineries or storage facilities and vehicle service facilities.
Vegetation -Vegetation surrounding an outfall may show the effects of industrial pollutants. Decaying organic materials coming from various food product
wastes would cause an increase in plant life, while the dischargeof chemical dyes and inorganicpigmentsfrom textile mills could noticeably decrease
vegetation. It is important not to confuse the adverse effects of high stormwater flows on vegetation with highly toxic dry-weather intermittent flows.
excessive growth: food product facilities.
inhibited growth: high stormwater flows, beverage facilities, printing plants, metal product facilities, drug manufacturing, petroleum facilities, vehicle
service facilities and automobile dealers.
Damage to Outfall Structures - Another readily visible indication of industrial contamination. Cracking, deterioration, and spelling of concrete or peeling
of surface paint, occurring at an outfall are usually caused by severely contaminated discharges, usually of industrial origin. These contaminants are
usuallyveryacidicor basic in nature. Primary metal industries have a strong potential forcausing outfall structural damage because their batch dumps
are highly acidic. Poor construction, hydraulic scour, and old age may also adversely affect the condition of the outfall structure.
concrete cracking: industrial flows
concrete spalling: industrial flows
peeling paint industrial flows
metal corrosion: industrial flows
Correlation tests were conducted to identify relationships between outfalls that were known to have severe
contamination problems and the negative indicators (Lalor 1994). Pearson correlation tests indicated that high turbidity
and obvious odors appeared to be the most useful physical indicators of contamination when contamination was defined
by toxicity and the presence of detergents. High turbidity was noted in 74% of the contaminated source flow samples, This
represented a 26% false negative rate (indication of no contamination when contamination actually exists), if one relied
on turbidity alone as an indicator of contamination. High turbidity was noted in only 5% of the uncontaminated source flow
samples. This represents the rate of false positives (indication of contamination when none actually exists) when relying
on turbidity alone. Noticeable odor was indicated in 67% of flow samples from contaminated sources, but in none of the
flow samples from uncontaminated sources. This translates to 37% false negatives, but no false positives. Obvious odors
identified included gasoline, oil, sewage, industrial chemicals or detergents, decomposing organic wastes, etc.
100
-------
False negatives are more of a concern than a reasonable number of false positives when working with a screening
methodology. Screening methodologies are used to direct further, more detailed investigations. False positives would be
discarded after further investigation. However, a false negative during a screening investigation results in the dismissal
of a problem outfall for at least the near future. Missed contributors to stream contamination may result in unsatisfactory
in-stream results following the application of costly corrective measures elsewhere.
The method of using physical characteristics to indicate contamination in outfall flows does not allow quantifiable
estimates of the flow components and, if used alone, will likely result in many incorrect determinations, especially false
negatives. These simple characteristics are most useful for identifying gross contamination: only the most significantly
contaminated outfalls and drainage areas would therefore be recognized using this method.
Detergents as Indicators of Contamination. Results from the Mann-Whitney U tests (Lalor 1994) indicated that
samples from any of the dry-weather flow sources could be correctly classified as clean or contaminated based only on
the measured value of any one of the following parameters: detergents, color, or conductivity. Color and high conductivity
were present in samples from clean sources as well as contaminated sources, but their levels of occurrence were
significantly different between the two groups. If samples from only one source were expected to make up outfall flows,
the level of color or conductivity could be used to distinguish contaminated outfalls from clean outfalls. However, since
multi-source flows occur, measured levels of color or conductivity could fall within acceptable levels because of dilution,
even though a contaminating source was contributing to the flow. Detergents, on the other hand, can be used to
distinguish between clean and contaminated outfalls simply by their presence or absence, using a detection limit of 0.06
mg/L All samples analyzed from contaminated sources contained detergents in excess of this amount (with the exception
of three septage samples collected from homes discharging only toilet flushing water). No clean source samples were
found to contain detergents. Contaminated sources would be detected in mixtures with uncontaminated waters if they
made up at least 10% of the mixture.
Flow Chart for Most Significant Flow Component Identification. A further refinement is the flow chart shown on Figure
1. This flow chart describes an analysis strategy which may be used to identify the major component of dry-weather flow
samples in residential and commercial areas. This method does not attempt to distinguish among all potential sources
of dry-weather flows identified earlier, but rather the following four major groups of flow are identified: (1) tap waters
(including domestic tap water, irrigation water and rinse water), (2) natural waters (spring water and shallow ground
water), (3) sanitary wastewaters (sanitary sewage and septic tank discharge), and (4) wash waters (commercial laundry
waters, commercial car wash waters, radiator flushing wastes, and plating bath wastewaters). The use of this method
would not only allow outfall flows to be categorized as contaminated or uncontaminated, but would allow outfalls carrying
sanitary wastewaters to be identified. These outfalls could then receive highest priority for further investigation leading
to source control. This flow chart was designed for use in residential and/or commercial areas only.
In residential and/or commercial areas, all outfalls should be located and examined. The first indicator is the presence
or absence of dry-weather flow. If no dry-weather flow exists at an outfall, then indications of intermittent flows must be
investigated, Specifically, stains, deposits, odors, unusual stream-side vegetation conditions, and damage to outfall
structures can all indicate intermittent non-stormwater flows. However, frequent visits to outfalls over long time periods,
or the use of other monitoring techniques, may be needed to confirm that only stormwater flows occur. If intermittent flow
is not indicated, then the outfall probably does not have a contaminated non-stormwater source. The other points on the
flow chart serve to indicate if a major contaminating source is present, or if the water is uncontaminated. Component
contributions cannot be quantified using this method, and only the "most contaminated" type of source present will be
identified.
If dry-weather flow exists at an outfall, the flow should be sampled and tested for detergents. If detergents are not
present, the flow is probably from a non-contaminated non-stormwater source. The lower limit of detection for detergent
should be about 0.06 mg/L.
101
-------
Start
Residential or
Commercial
^
•j
*
Land Use
In Area
fe
Industrial or
Industrial/
Commercial
Sanitary
Wastewater or
Washwater
Not a
Contaminated
Non-Stormwatrr
Source
Likely
Sanitary
Wastewater
Source
Likely
Wastewater
Source
No
fc
ifylng
(Lalor 1994).
Natural
Water
Source
Flow Chart Methodology for Idrntifylng
Most Significant Flow Component (Lalor 1994).
Figure 1. Simple Flow Chart Method to Identify Significant Contaminating Sources
102
-------
If detergents are not present, fluoride levels can be used to distinguish between flows with treated watersources and
flows with natural sources in communities where water supplies are fluoridated and natural fluoride levels are low. In the
absence of detergents, high fluoride levels would indicate a potable water line leak, irrigation water, or wash/rinse water.
Low fluoride levels would indicate waters originating from springs or shallow groundwater. Based on the flow source
samples tested in this research (Table 1), fluoride levels above 0.13 mg/L would most likely indicate that a tap water
source was contributing to the dry-weather flow in the Birmingham, Alabama, study area.
If detergents are present, the flow is probably from a contaminated non-stormwater source, as indicated on Table
1. The ratio of ammonia to potassium can be used to indicate whether or not the source is sanitary wastewater.
Ammonia/potassium ratios greater than 0.60 would indicate likelysanitarywastewater contamination. Ammonia/potassium
ratios were above 0.9 for all septage and sewage samples collected in Birmingham (values ranged from 0.97 to 15.37,
averaging 2.55). Ammonia/potassium ratios for all othersamplescontaining detergents were below 0.7, ranging from 0.00
to 0.65, averaging 0.11. One radiator waste sample had an ammonia/potassium ratio of 0.65.
Non-contaminated samples collected in Birmingham had ammonia/potassium ratios ranging from 0.00 to 0.41, with
a mean value of 0.06 and a median value of 0.03. Using the mean values for non-contaminated samples (0.06) and
sanitary wastewaters (2.55), flows comprised of mixtures containing at least 25% sanitary wastes with the remainder of
the flow from uncontaminated sources would likely be identified as sanitary wastewaters using this method. Flows
containing smaller percent contributions from sanitary wastewaters might be identified as having a wash water source,
but would not be identified as uncontaminated.
Emerging Tools for Identifying Sources of Discharges
Coprostanol and Other Fecal Sterol Compounds Utilized as Tracers of Contamination by Sanitary Sewage. A more
likely indicator of human wastes than fecal coliforms and other "indicator" bacteria may be the use of certain molecular
markers, specifically the fecal sterols, such as coprostanol and epicoprostanol (Eaganhouse, eta/, 1988). However, these
compounds are also discharged by other carnivores in a drainage (especially dogs). A number of research projects have
used these compounds to investigate the presence of sanitary sewage contamination. The most successful application
may be associated with sediment analyses instead of water analyses. As an example, water analyses of coprostanol are
difficult due to the typically very low concentrations found, although the concentrations in many sediments are quite high
and much easier to quantify. Unfortunately, the long persistence of these compounds in the environment easily confuses
recent contamination with historical or intermittent contamination.
Particulates and sediments collected from coastal areas in Spain and Cuba receiving municipal sewage loads were
analyzed by Grimalt, ef a/. (1990) to determine the utility of coprostanol as a chemical marker of sewage contamination.
Coprostanol can not by itself be attributed to fecal matter inputs. However, relative contributions of steroid components
can be a useful indicator. When the relative concentrations of coprostanol and coprostanone are higher than their 5?
epimers, or more realistically, other sterol components of background or natural occurrence, it can provide useful
information.
Sediment cores from Santa Monica Basin, CA, and effluent from two local municipal wastewater discharges were
analyzed by Venkatesan and Kaplan (1990) for coprostanol to determine the degree of sewage addition to sediment.
Coprostanols were distributed throughout the basin sediments in association with fine particles. Some stations contained
elevated levels, either due to their proximity to outfalls or because of preferential advection of fine-grained sediments. A
noted decline of coprostanols relative to total sterols from outfalls seaward indicated dilution of sewage by biogenic
sterols.
Other chemical compounds have been utilized for sewage tracer work. Saturated hydrocarbons with 16-1 8 carbons,
and saturated hydrocarbons with 16-21 carbons, in addition to coprostanol, were chosen as markers for sewage in water,
particulate, and sediment samples near the Cocoa, FL, domestic wastewater treatment plant (Holm and Windsor, 1990).
The concentration of the markers was highest at points close to the outfall pipe and diminished with distance. However
the concentration of Cl 6-C21 compounds was high at a site 800 m from the outfall indicating that these compounds were
103
-------
unsuitable markers for locating areas exposed to the sewage plume. The concentrations for the other markers were very
low at this station.
The range of concentrations of coprostanol found in sediments and mussels of Venice; Italy, were reported by
Sherwin, ef al, (1993). Raw sewage is still discharged directly into the Venice lagoon. Coprostanol concentrations were
determined in sediment and mussel samples from the lagoon using gas chromatography/mass spectroscopy. Samples
were collected in interior canals and compared to open-bay concentrations. Sediment concentrations ranged from 0.2-41 .0
ug/g (dry weight). Interior canal sediment samples averaged 16 ug/g compared to 2 |jg/g found in open bay sediment
samples. Total coprostanol concentrations in mussels ranged from 80 to 620 ng/g (wet weight). No mussels were found
in the four most polluted interior canal sites.
Nichols, etal. (1996) also examined coprostanol in stormwater and the sea-surface microlayer to distinguish human
versus nonhuman sources of contamination. Other steroid compounds in sewage effluent were investigated by Routledge,
ef al. (1998) and Desbrow, ef al. (1998) who both examined estrogenic chemicals. The most common found were 173-
Estradiol and estrone which were detected at concentrations in the tens of nanograms per liter range. These were
identified as estrogenic through a toxicity identification and evaluation approach, where sequential separations and
analyses identified the sample fractions causing estrogenic activity using a yeast-based estrogen screen. GC/MS was
then used to identify the specific compounds.
Estimating Potential Sanitary Sewage Discharges into Storm Drainage and Receiving Wafers using Detergent Tracer
Compounds. As described above, detergent measurements (using methylene blue active substance, MBAS, test
methods) were the most successful individual tracer to indicate contaminated water in storm sewerage dry-weather flows.
Unfortunately, the MBAS method uses hazardous chloroform for an extraction step. Different detergent components,
especially linearalkylbenzene sulphonates (LAS) and linear alkylbenzenes (LAB), have also been tried to indicatesewage
dispersal patterns in receiving waters. Boron, a major historical ingredient of laundry chemicals, can also potentially be
used. Boron has the great advantage of being relatively easy to analyze using portable field test kits, while LAS requires
chromatographic equipment. LAS can be measured using HPLC with fluorescent detection, after solid phase extraction,
to very low levels. Fujita, ef a/. (1998) developed an efficient enzyme-linked immunosorbent assay (ELISA) for detecting
LAS at levels from 20 to 500 ng/L.
LAS from synthetic surfactants (Terzic and Ahel 1993) which degrade rapidly, as well as nonionic detergents (Zoller,
ef a/. 1991) which do not degrade rapidly, have been utilized as sanitary sewage markers. LAS was quickly dispersed
from wastewater outfalls except in areas where wind was calm. In these areas LAS concentrations increased in
freshwater but were unaffected in saline water. After time, the lower alkyl groups were mostly found, possibly as a result
of degradation or settling of longer alkyl chain compounds with sediments. Chung, ef a/ (1995) also describe the
distribution and fate of LAS in an urban stream in Korea. They examined different LAS compounds having carbon ratios
of Cl 2 and Cl 3 compared to Cl 0 and Cl 1, plus ratios of phosphates to MBAS and the internal to external isomer ratio
(I/E) as part of their research. Gonalez-Mazo, etal. (1998) examined LAS in the Bay of Cadiz off the southwest of Spain.
They found that LAS degrades rapidly (Fujita, ef a/., 1998, found that complete biodegradation of LAS requires several
days), and is also strongly sorbed to particulates. In areas close to shore and near the untreated wastewater discharges,
there is significant vertical stratification of LAS: the top 3 to 5 mm of water had LAS concentrations about 100 times
greater than found at 0.5 m.
Zeng and Vista (1997) and Zeng, ef al, (1997) describe a study off of San Diego where LAB was measured, along
with polycyclic aromatic hydrocarbons (PAHs) and aliphatic hydrocarbons (AHs) to indicate the relative pollutant
contributions of wastewater from sanitary sewage, nonpoint sources, and hydrocarbon combustion sources. They
developed and tested several indicator ratios (alkyl homologue distributions and parent compound distributions) and
examined the ratio of various PAHs (such as phenanthrene to anthracene, methylphenanthrene to phenanthrene,
fluoranthene to pyrene, and benzo(a)anthracene to chrysene) as tools for distinguishing these sources. They concluded
that LABs are useful tracers of domestic waste inputs to the environment due to their limited sources. They also describe
the use of the internal to external isomer ratio (I/E) to indicate the amount of biodegradation that may have occurred to
the LABs. They observed concentrations of total LABs in sewage effluent of about 3 ug/L, although previous researchers
have seen concentrations of about 150 ug/L in sewage effluent from the same area.
104
-------
The fluorescent properties of detergents have also been used as a tracer by investigating the fluorescent whitening
agents (FWAs), as described by Poiger, ef a/. (1996) and Kramer, ef a/. (1996). HPLC with fluorescence detection was
used in these studies to quantify very low concentrations of FWAs. The two most frequently used FWAs in household
detergents (DSBP and DAS 1) were found at 7 to 21 ug/L in primary sewage effluent and at 3 to 9 ug/L in secondary
effluent. Raw sewage contains about 10 to 20 ug/L FWAs. The removal mechanisms in sewage treatment processes is
by adsorption to activated sludge. The type of FWAs varies from laundry applications to textile finishing and paper
production, making it possible to identify sewage sources. The FWAs were found in river water at 0.04 to 0.6 ug/L. The
FWAs are not easily biodegradable but they are readily photodegraded. Photodegradation rates have been reported to
be about 7% for DSBP and 71% for DAS 1 in river water exposed to natural sunlight, after one hour exposure.
Subsequent photodegradation is quite slow.
Other Compounds Found in Sanitary Sewage that may be used for Identifying Contamination by Sewage. Halling-
Sorensen, eta/. (1998) detected numerous pharmaceutical substances in sewage effluents and in receiving waters. Their
work addressed human health concerns of these low level compounds that can enter downstream drinking water supplies.
However, the information can also be possibly used to help identify sewage contamination. Most of the research has
focused on clofibric acid, a chemical used in cholesterol lowering drugs. It has been found in concentrations ranging from
10 to 165 ng/L in Berlin drinking water samples. Other drugs commonly found include aspirin, caffeine, and ibuprofen.
Current FDA guidance mandates that the maximum concentration of a substance or its active metabolites at the point
of entry into the aquatic environment be less than 1 ug/L (Hun 1998).
Caffeine has been used as an indicator of sewage contamination by several investigators (Shuman and Strand 1996).
The King County, WA, Water Quality Assessment Project is examining the impacts of CSOs on the Duwamish River and
Elliott Bay. They are using both caffeine (representing dissolved CSO constituents) and coprostanol (representing
particulate bound CSO constituents), in conjunction with heavy metals and conventional analyses, to help determine the
contribution of CSOs to the river. The caffeine is unique to sewage, while coprostanol is from both humans and
carnivorous animals and is therefore also in stormwater. They sampled upstream of all CSOs, but with some stormwater
influences, 100 m upstream of the primary CSO discharge (but downstream of other CSOs), within the primary CSO
discharge line, and 100 m downriver of the CSO discharge location. The relationship between caffeine and coprostanol
was fairly consistent for the four sites (coprostanol was about 0.5 to 1.5 ug/L higher than caffeine). Similar patterns were
found among metals; chromium was always the lowest and zinc was the highest. King Co. is also using clean transported
mussels placed in the Duwamish River to measure the bioconcentration potential of metal and organic toxicants and the
effects of the CSOs on mussel growth rates (after 6 weekexposure periods). Paired reference locations are available near
the areas of deployment, but outside the areas of immediate CSO influence. US Water News (1998) also described a
study in Boston Harbor that found caffeine at levels of about 7 ug/L in the harbor water. The caffeine content of regular
coffee is about 700 mg/L, in contrast.
Kratch (1997) summarized several investigations on cataloging the DMA of £ co/i to identify their source in water.
This rapidly emerging technique seems to have great promise in addressing a number of nonpoint source water pollution
issues. The procedure, developed at the Virginia Polytechnic Institute and State University, has been used in Chesapeake
Bay. In one example, it was possible to identify a large wild animal population as the source of fecal coliform
contamination of a shellfish bed, instead of suspected failing septic tanks. DMA patterns in fecal coliforms vary among
animals and birds, and it is relatively easy to distinguish between human and non-human sources of the bacteria.
However, some wild animals have DMA patterns that are not easily distinguishable. Some researchers question the value
of £ coll DMA fingerprinting believing that there is little direct relationship between £ co/i and human pathogens.
However, this method should be useful to identify the presence of sewage contamination in stormwater or in a receiving
water.
One application of the technique, as described by Krane, ef a/ (1999) of Wright State University, used randomly
amplified polymorphic DMA polymerase chain reaction (RAPD-PCR) generated profiles of naturally occurring crayfish.
They found that changes in the underlying genetic diversity of these populations were significantly correlated with the
extent to which they have been exposed to anthropogenic stressors. They concluded that this rapid and relatively simple
technique can be used to develop a sensitive means of directly assessing the impact of stressors upon ecosystems.
These Wright State University researchers have also used the RAPD-PCR techniques on populations of snails, pill bugs,
105
-------
violets, spiders, earthworms, herring, and some benthic macroinvertebrates, finding relatively few obstacles in its use for
different organisms. As noted above, other researchers have used DMA profiling techniques to identify sources of £ coli
bacteria found in coastal waterways. It is possible that these techniques can be expanded to enable rapid detection of
many different types of pathogens in receiving waters, and the most likely sources of these pathogens.
Other Tracer Methods for Identifying Sources of Water. Stable isotopes had been recommended as an efficient
indicator of illicit connections to storm sewerage. A demonstration was conducted in Detroit as part of the Rouge River
project to identify sources of dry weather flows in storm sewerage (Sangal, ef a/. 1996). Naturally occurring stable
isotopes of oxygen and hydrogen can be used to identify waters originating from different geographical sources (especially
along a north-south gradient). Ma and Spalding (1996) discuss this approach by using stable isotopes to investigate
recharge of groundwaters by surface waters. During water vapor transport from equatorial source regions to higher
latitudes, depletion of heavy isotopes occurs with rain. Deviation from a standard relationship between deuterium and 180
for a specific area indicates that the water has undergone additional evaporation. The ratio is also affected by seasonal
changes. As discussed by Ma and Spalding (1996), the Platte River water is normally derived in part from snowmelt from
the Rocky Mountains, while the groundwater in parts of Nebraska is mainly contributed from the Gulf air stream. The
origins of these waters are sufficiently different and allow good measurements of the recharge rate of the surface water
to the groundwater. In Detroit, Sangal, ef a/. (1996) used differences in origin between the domestic water supply, local
surface waters, and the local groundwater to identify potential sanitary sewage contributions to the separate storm
sewerage. Rieley, eta/, (1997) used stable isotopes of carbon in marine organisms to distinguish the primary source of
carbon being consumed (sewage sludge vs. natural carbon sources) in two deep sea sewage sludge disposal areas.
Stable isotope analyses would not be able to distinguish between sanitary sewage, industrial discharges, washwaters,
and domestic water, as they all have the same origin, nor would it be possible to distinguish sewage from local
groundwaters if the domestic water supply was from the same local aquifer. This method works best for situations where
the water supply is from a distant source and where separation of waters into separate flow components is not needed.
It may be an excellent tool to study the effects of deep well injection of stormwater on deep aquifers having distant
recharge sources (such as in the Phoenix area). Few laboratories can analyze for these stable isotopes, requiring shipping
and a long wait for the analytical results. Sangal, etal. (1995) used Geochron Laboratories, in Cambridge, Massachusetts.
Dating of sediments using 137Cs was described by Davis, et a/. (1997). Arsenic contaminated sediments in the
Hylebos Waterway in Tacoma, WA, could have originated from numerous sources, including a pesticide manufacturing
facility, a rock-wool plant, steel slags, powdered metal plant, shipbuilding facilities, marinas and arsenic boat paints, and
the Tacoma Smelter. Dating the sediments, combined with knowing the history of potential discharges and conducting
optical and electron microscopic studies of the sediments, was found to be a powerful tool to differentiate between the
different metal sources to the sediments.
Conclusions
Recent tests examined several potential tracer parameters during a project characterizing stormwater that had
collected in telecommunication manholes, funded by Tecordia (previously Bellcore), AT&T, and eight regional telephone
companies throughout the country (Pitt and Clark 1999). Numerous conventional constituents, plus major ions, and
toxicants were measured, along with candidate tracers to indicate sewage contamination of this water. Boron, caffeine,
coprostanol, £ coll, enterococci, fluorescence (using specific wavelengths for detergents), and a simpler test for
detergents were evaluated, along with the use of fluoride, ammonia, potassium, and obvious odors and color. About 700
water samples were evaluated for all of these parameters, with the exception of bacteria and boron (about 250 samples),
and only infrequent samples were analyzed for fluorescence. Coprostanol was found in about 25% of the water samples
(and in about 75% of the 350 sediment samples analyzed). Caffeine was found in very few samples, while elevated E.
co//and enterococci (using IDEXX tests) were observed in about 10% of the samples. Strong sewage odors in water and
sediment samples were also detected in about 10% of the samples. Detergents and fluoride (at >0.3 mg/L) were found
in about 40% of the samples and are expected to have been contaminated with industrial activities (lubricants and
cleansers) and not sewerage. Overall, about 10% of the samples were therefore expected to have been contaminated
with sanitary sewage, about the same rate previously estimated for stormwater systems.
106
-------
Additional laboratory tests, funded by the University of New Orleans and the EPA, were conducted using many
sewage and laundry detergent samples and found that the boron test was a poor indicator of sewage, possibly due to
changes in formulations in modern laundry detergents. Other laboratory tests found that fluorescence was an excellent
indicator of sewage, especially when using specialized "detergent whitener" filter sets, but was not very repeatable. We
also examined several UV absorbance wavelengths as sewage indicators and found excellent correlations with 228 nm,
a wavelength having very little background absorbance in local spring waters, but with a strong response factor with
increasing strengths of sewage. We recommend that our originally developed and tested protocol still be used as the most
efficient routine indicator of sewage contamination of stormwater drainage systems, with the possible addition of specific
E. co//and enterococci measurements and UV absorbance at 228 nm. The numerous exotic tests requiring specialized
instrumentation and expertise reviewed in this paper do not appear to warrant their expense and long analytical turn-
around times, except in specialized research situations.
References
Chung, K.H., K.S. Ro, and S.U. Hong. "Syntheticdetergent in an urban stream in Korea." Proceedings of the 65th Water
Environment Federation Technical Exhibition and Conference, Miami Beach, Florida. Volume 4, Surface Water Quality
and Ecology, pp. 429 - 434. Alexandria, VA. 1995.
Davis, A., P. DeCurnou, and I.E. Eary. "Discriminating between sources of arsenic in the sediments of a tidal waterway,
Tacoma, Washington." Environmental Science & Technology. Vol. 31, no. 7, pg. 1985. 1997.
Desbrow, C., E.J. Routledge, G.C. Brighty, J.P. Sumpter, and M. Waldock. "Identification of estrogenicchemicals in STW
effluent. 1. Chemical fractionation and in vitro biological screening." Environmental Science & Technology. Vol. 32, no.
11, pg. 1549. 1998.
Eganhouse, R.P., D.P. Olaguer, B.R. Gould and C.S. Phinney. "Use of Molecular Markers for the Detection of Municipal
Sewage Sludge at Sea." Marine Environmental Research. Volume 25, No.1. pp. I-22. 1988.
Fujita, M., M. Ike, Y. Goda, S. Fujimoto, Y. Toyoda, and K-l. Miyagawa. "An enzyme-linked immunosorbent assay for
detection of linear alkylbenzene sulfonate: development and field studies." Environmental Science & Technology. Vol.
32, no. 8, pp. 1143-1 146. 1998.
Gonalez-Mazo, E., J.M. Forja, and A. Gomez-Parra. "Fate and distribution of linear alkylbenzene sulfonates in the littoral
environment." Environmental Science & Technology. Vol. 32, no. 11, pp. 1636-1 641. 1998.
Grimalt, J. 0., ef a/. "Assessment of fecal sterols and ketones as indicators of urban sewage inputs to coastal waters."
Env. Science & Technology, Vol. 24, no. 3, p357. March 1990.
Halling-Sorensen, B., S. Nors Nielsen, P.F. Lanzky, F. Ingerslev, H.C. Molten Lutzhoft, andS.E. Jorgensen. "Occurrence,
fate and effects of pharmaceutical substances in the environment -A review." Chemosphere. Vol. 36, no. 2, pp. 357 -
393.1998.
Holm, S.E., J.G. Windsor. "Exposure assessment of sewage treatment plant effluent by a selected chemical marker
method." Archives Env. Contam. & Tox. Vol. 19, no. 5, p674. Sep-Ott 1990.
Hun, T. "NewsWatch: Water Quality, Studies indicate drugs in water may come from effluent discharges." Water
Environment & Technology, pp. 17 - 22. July 1998.
Kramer, J.B., S. Canonica, J. Hoige, and J. Kaschig. "Degradation of fluorescent whitening agents in sunlit natural
waters." Environmental Science & Technology. Vol. 30, no. 7, pp. 2227 - 2234. 1996.
Krane, D. E., D. C. Sternberg, and G.A. Burton. "Randomly amplified polymorphic DMA profile-based measures of genetic
diversity in crayfish correlated with environmental impacts." Environmental Toxicology and Chemistry. Vol. 18, no. 3, pp.
504 - 508. March 1999.
107
-------
Kratch, K. "NewsWatch, Water Quality: Cataloging DMA of E. coli sources pinpoints contamination causes." Water
Environment & Technology. Vol. 9, no. 8, pp. 24 - 26. August 1997.
Lalor, M. An Assessment of Non-Stormwater Discharges toStorm Drainage Systemsin Residential and Commercial Land
Use Areas. Ph.D. Dissertation. Department of Civil and Environmental Engineering. Vanderbilt University. 1994.
Ma, L. and R.F. Spalding. "Stable isotopes characterization of the impacts of artificial ground water recharge." Water
Resources Bulletin. Vol. 32, no. 6, pp. 1273 - 1282. December 1996.
Nichols, P.O., R. Leeming. M.S. Rayner, and V. Latham. "Use of capillary gas-chromatography for measuring fecal-
derived sterols application to stormwater, the sea-surface microlayer, beach greases, regional studies, and distinguishing
algal blooms and human and nonhuman sources of sewage pollution." Journal of Chromatography. Vol. 733, no. 1 - 2,
pp. 497-509. May 10, 1996.
Poiger, T., J.A. Field, T.M. Field, and W. Giger. "Occurrence of fluorescent whitening agents in sewage and river
determined by solid-phase extraction and high-performance liquid Chromatography." Environmental Science &
Technology. Vol. 30, no. 7, pp. 2220 - 2226. 1996.
Pitt, R. and S. Clark. Communication Manhole Water Study: Characteristics of Water Found in Communications
Manholes. Final Draft. Office of Water, U.S. Environmental Protection Agency. Washington, D.C. July 1999.
Pitt, R. and J. McLean. Humber River Pilot Watershed Project, Ontario Ministry of the Environment, Toronto, Canada.
483 pgs. June 1986.
Pitt, R., M. Lalor, R. Field, D.D. Adrian, and D.Barbe'. A User's Guide forthe Assessment ofNon-Stormwater Discharges
into Separate Storm Drainage Systems. Jointly published by the Center of Environmental Research Information, US EPA,
and the Urban Waste Management & Research Center (UWM&RC). EPA/600/R-92/238. PB93-131472. Cincinnati, Ohio.
January 1993.
Rieley, G., C.L. VanDiver, and G. Eglinton. "Fatty acids as sensitive tracers of sewage sludge carbon in a deep-sea
ecosystem." Environmental Science & Technology. Vol. 31, no. 4, pg. 1018. 1997.
Routledge, E.J., D. Sheahan, C. Desbrow, CC. Brighty, M. Waldock, and J.P. Sumpter. "Identification of estrogenic
chemicals in STW effluent. 2. In vivo responses in trout and roach." Environmental Science & Technology. Vol. 32, no.
11, pg. 1559. 1998.
Sangal, S., P.K. Aggarwal, and D. Tuomari. "Identification of illicit connections in storm sewers: An innovative approach
using stable isotopes." Rouge River Studies. Detroit, Ml. 1996.
Sherwin, M.R., et al. "Coprostanol (5?-cholestan-3?-ol) in lagoonal sediments and mussels of Venice, Italy." Mar. Potlut.
Bull. Vol. 26, no. 9, p501. September 1993.
Shuman, R. and J. Strand. "King County water quality assessment: CSO discharges, biological impacts being assessed."
Wet Weatherx, Water Environment Research Foundation. Fairfax, VA. Vol. 1, no. 3, pp. 10 - 14. Fall 1996.
Terzic, S., and M. Ahel. "Determination of alkylbenzene sulphonates in the Krka River Estuary." Bull. Environ. Contam.
Toxicol. Vol. 50, no. 2, p241. February 1993.
US Water News. "Scientists say Boston Harbor water full of caffeine." Pg. 3. July 1998.
Venkatesan, M.I. and I.R.Kaplan. "Sedimentary coprostanol as an index of sewage addition in Santa Monica Basin,
Southern California." Env. Science & Technology, Vol. 24, no. 2, p208. February 1990.
108
-------
Zeng, E. and C.L. Vista. "Organic pollutants in the coastal environment off San Diego, California. 1. Source identification
and assessment by compositional indices of polycyclic aromatic hydrocarbons." Environmental Toxicology and Chemistry,
Vol. 16, no. 2, pp. 179- 188. 1997.
Zeng, E., A.M. Khan, and K. Iran. "Organic pollutants in the coastal environment off San Diego, California. 2. Using linear
alkylbenzenes to trace sewage-derived organic materials." Environmental Toxicologyand Chemistry. Vol. 16, no. 2, pp.
196-201. 1997.
Zoller, U., et al. "Nonionic detergents as tracers of groundwater pollution caused by municipal sewage." Chemistry for
the Protection of the Environment (Plenum). p225. 1991.
109
-------
Elimination of Illicit Connections in Coastal New Hampshire Spurs Cooperation
and Controversy
Natalie Landry and Robert Livingston
New Hampshire Department of Environmental Services
Concord, New Hampshire
Introduction
Discharging stormwater runoff into our waterways has long been an accepted practice. In theory, storm drainage
pipes should only discharge during and after storm events unless the source is groundwater or surface water piped
underground. Therefore, the dry weather discharge should be relatively free of contaminants. However, many
communities across the country are finding out this is not always true. Some cities and towns are discovering illegal
connections of residential and commercial sewer lines to storm water collection systems. Illicit connections have been
identified by the New Hampshire Department of Environmental Services (DES) as the point source of high fecal coliform
levels in the New Hampshire coastal basin (Jones, 1995). These illegal connections pose a health riskto those recreating
in the coastal waters and have forced the closure of shellfish growing waters to harvesting.
Goals of the Coastal Investigations Programs
Determining the extent of dry weather contamination in storm drainage systems is the first step an investigator should
take when researching stormwater pollution. Dry weather flows in storm drainage systems are often the result of
groundwater infiltration, but can also result from inappropriate connections from commercial, industrial, or residential
buildings. In 1996, the New Hampshire DES published the Coastal Basin Nonpoint Source Pollution Assessment and
Abatement Plan (NHDES, 1996) that directed coastal investigations of each community's storm drainage system during
dry weather. This decision to conduct dry weather investigations in the coast was made after 300 illicit connections were
identified in the northern New Hampshire city of Berlin. State environmental officials were convinced that illicit connections
were always present in storm drainage systems that were once considered a pollution threat only during wet weather.
DES initiated a multi-year effort that focused on identifying and abating the sources of the bacterial violations found
in the state's coastal waters with the goal of opening shellfish growing waters during dry weather (Landry, 1997). About
the same time, the New Hampshire Estuaries Project (NHEP) began a three-year process of developing a comprehensive
management plan aimed at restoring, protecting and enhancing the water quality and living resources of the state's
estuaries. The major goal of the NHEP was to address the sources of pollution currently impacting the estuaries and
prevent future problems through effective land use planning and shoreline protection of the coastal resources (NHEP,
1996). To accomplish this goal, part of the NHEP strategy was to locate and remediate the sources of the water quality
violations, primarily bacterial violations, found in the estuaries and coastal waters (Landry, 1997). DES and NHEP
combined resources and developed an investigation strategy with the overall goal of improving and protecting estuarine
water quality.
Specific Program Objectives
The main objectives of the investigation strategy were to identify inappropriate connections in the storm drainage
systems of urban, coastal communities and to eliminate the illicit connections through the available means, which include
voluntary compliance and enforcement.
A Brief Look at Coastal New Hampshire
The eighteen miles of New Hampshire coastline do not begin to tell the story of the state's abundant marine
resources. The relatively modest coastline is only a small part of the coastal basin. The estuarine resources include the
Great Bay Estuary and seven associated tidal rivers, Hampton Harbor, and Rye Harbor. These waters are used by
residents and many visitors for swimming, boat touring, shellfish harvesting, surfing, and angling. Forty-two communities
110
-------
comprise the coastal basin watershed, with a population density just under 300 residents/mile2(Jones, In Review). The
upper watershed is generally undeveloped and forested while the more urban centers are situated in the lower watershed
as the rivers approach the coast.
Ten wastewater treatment facilities are situated on the tributaries of Great Bay and Hampton Harbor and two facilities
discharge directly into the Atlantic Ocean. Coastal communities are working diligently to upgrade the wastewater
treatment facilities and sewage collection systems. Inflow/infiltration problems and undersized pump stations plague the
treatment facilities and have resulted in financial hardshipsforaffected municipalities. Shellfish growing waters have been
temporarily closed after heavy rainstorms when bacteria levels rise due to sewage by-passes. Sewage is a well-
recognized threat to the marine environment because it often contains harmful chemicals, disease-causing bacteria and
viruses, dissolved material and solid matter. Pathogens can cause avariety of illnesses and humans are exposed to these
organisms through contaminated water, shellfish, and fish (Sea Grant, 1999).
Investigating Illicit Connections
Recently, more and more watershed studies are investigating inappropriate discharges in storm drainage systems.
This pollution source originates from an identifiable point and flows through the storm drainage system to the outfall pipe.
For example, instead of connecting to the sewer system, a direct connection of sewer service discharges into the storm
drainage system. Other inappropriate sources include floor drains and laundry pipes. These inappropriate connections
are also referred to as illicit or illegal cross connections. The health threat and the potential to interfere with stormwater
contamination assessments elevate illicit connections to priority status for watershed managers to investigate.
Pitt et al. (1993), in cooperation with the Center of Environmental Research Information, U.S. Environmental
Protection Agency, published a user's guide for conducting investigations of illicit connections. Several of the methods
suggested in this guide were implemented during the New Hampshire coastal investigations. Detailed surveys to
determine the extent of contamination through specific water quality monitoring and careful observation of storm drainage
outfalls are recommended for each type of land use in the watershed. Pitt recommends an initial phase of investigative
protocol that includes the initial mapping and field surveys. The initial activities are followed by more detailed watershed
surveys to locate and correct the sources of the contamination in the identified problem areas. After corrective action has
been taken, repeated outfall field surveys are required to ensure that the outfalls remain uncontaminated.
111
-------
Surveys of Storm Drainage Systems
Over the course of the investigations, several methods were used, ranging from the initial screening process of
surveying storm drainage discharges to dye testing the indoor plumbing of suspected sources. Steps between the initial
survey and the final determination of the source, included analyzing the discharge for water quality; visual and odor
observations at outfalls, manholes, and catchbasins; smoke testing; and video inspection of the storm drainage and sewer
systems.
Tidal rivers and coastal waters were divided into study sections by community. The urban, downtown centers of these
communities were targeted based on the existence of the storm drainage infrastructure. The investigators compiled maps
and as-built drawings of the storm drainage and sewer infrastructure. If the maps were inaccurate, insufficient, or
unavailable, information on the storm drainage system was developed based on field investigations by the staff, typically
with the assistance of public works employees.
Communities with maps based in a geographic information system (CIS) saved staff time and were generally more
accurate than record drawings that are not updated regularly (Landry, 1997). Tuomari (1996) applied the Rouge
Watershed Geographic Information System to the Wayne County Illicit Connections Detection Program and concluded
that the new GIS strategy eliminated the need to use maps and graphics from disparate reports and sources, significantly
reducing the time and effort once spent on research, field data acquisition, and interpretation.
Beginning in the summer of 1996, the coastal shorelines were surveyed at low tide, on foot or by canoe, depending
on access, for potential pollution sources. All pipes, seeps, streams, and swales with flow were sampled for bacteria. In
addition, temperature was measured and observations relating to the condition of the pipe (stained or structurally
damaged), odor, evidence of untreated wastewater (toilet paper, etc.), turbidity, color, debris, estimated flow, and any
other observations were noted. Dry pipes were rechecked on several occasions for intermittent flow. Evidence indicating
the presence of wastewater and/or elevated bacteria levels prompted further investigation of these locations.
Upstream catchbasins and manholes associated with the outfall pipes that were identified in the screening process
previously described, were surveyed for evidence of wastewater and sampled for bacteria. Smoke testing (using non-toxic
smoke blown into catch basins) was then used to identify buildings connected to the storm drainage system by canvassing
the neighborhood for vents emitting smoke. Final confirmation of an illicit connection from the buildings that emitted smoke
was accomplished with dye testing of indoor plumbing and observing the storm drainage and sewer systems for the
presence/absence of the dye.
Feeder streams were surveyed for outfall pipes with dry weather flow. Other potential bacteriological sources (e.g.,
pigeon roosting sites on bridges) were bracketed with water quality sampling stations. Where contaminated seeps and
swales were suspected, the drainage area was surveyed for potential sources such as broken sewer mains.
Water Quality Results
Bacteria data (1997/98) from outfall pipes with confirmed cross connections ranged from 1,700 - >1,000,000 E. coli
counts/I 00 ml during dry weather in Dover, New Hampshire. Many outfall pipes with cross connections had a gray biomat
comprised of filamentous bacteria coating the inside of the pipe and, often, the rocks or sediment below. These biomats
were used as a wastewater indicator based on the presence of these mats at more than 50% of the outfalls with confirmed
cross connections.
Dr. Stephen Jones of the University of New Hampshire Jackson Estuarine Laboratory conducted a twelve-month
study that examined the significance of all flow coming from urban storm drainage systems in the downtown Dover
watershed of the Cocheco River (Jones, 1998). Jones found that storm drains were consistent sources of relatively high
concentrations of bacterial indicators and pathogens at concentrations that exceeded state standards for recreational and
shellfish-growing waters during both dry and wet weather.
Flow from a damaged stormwater outfall pipe was determined to have a geometric mean £. coli concentration of
1,047,199cfu/100 ml and a dissolved inorganic nitrogen (DIN) concentration of 22.4 mg/l.The data were brought to the
attention of DES and an investigation revealed a cross connection from a commercial building. Dr. Jones continued to
monitor the quality of flow after the cross connection was eliminated and the results show a significant decline in both
bacteria and DIN geometric means. The post-repair results were 93 E. coli cfu/100 ml and 7.2 DIN mg/l.
112
-------
Although no public health problems were known to have occurred as a result of exposure to bacterial pathogens in
the Cocheco River, the contamination may be a significant contribution to the fecal-borne bacteria that are presently the
reason for closing the area's shellfish growing waters in New Hampshire and restricting harvests in Maine (Jones, 1998).
Remedial Actions
Once confirmed, illicit connections in coastal New Hampshire have been eliminated in different ways. The most
desired course of action from the regulatory perspective is voluntary compliance and many fixes have been accomplished
through this process. Economics and prioritization of the many demands on public works departments sometimes compel
the state and federal environmental agencies to initiate regulatory action to eliminate raw wastewater discharges into
surface waters. Lawsuits, although not common, have been filed against municipalities after cross connections were
discovered.
Voluntary Compliance: Town of Exeter Case Study
In 1994, researchers at the UNH Jackson Estuarine Laboratory (Jones and Langan, 1995) reported elevated dry-
weather bacterial levels collected in Morris Brook, a tributary to the Squamscott River in Exeter, New Hampshire. In 1996,
DES collected bacteria samples at various locations on Norris Brook (NHDES, 1997) and found relatively low E. coil
concentrations of <150 counts/100 ml. In 1998, an Exeter official urged DES to investigate the watershed for
contamination based on the 1994 data that showed a fecal coliform concentration of 600 counts/I 00 ml. In April of 1998,
DES and a town official conducted a survey of the lower watershed and discovered a storm drainage outfall discharging
a large volume of flow even though the weather had been dry. Upon closer inspection, toilet paper was observed in the
outfall pipe and the immediate area.
The town public works department was notified of the survey results and, following reminders by DES, began
investigations to determine the sources of untreated wastewater. Progress was slowed because several of the residences
were rental properties which involved contacting the owner, who was in some cases from out-of-state, and gaining
permission to access the building for dye testing. By November, the town reported that a few of the cross connections
still remained. DES considered enforcement action and an administrative fine but did not take that action to maintain the
spirit of cooperation. In January of 1999, the town reported that the owner of the last remaining property to be dye tested
was not responding to requests for access. More prodding by DES followed and in February 1999, DES received
notification that the cross connections were eliminated. A follow up inspection confirmed the absence of untreated
wastewater in the storm drainage outfall.
A lesson learned from this experience is that a persistent, local advocate is often the key to maintaining attention on
a local water quality problem. In addition, local advocates, whether a conservation commissioner, selectmen, or citizen,
often have detailed knowledge of the complaint and the locale, which provides valuable and time-saving information to
the state investigators.
Time and resource demands on local officials as well as state investigators can cause this process to be distressingly
slow. Budgeting for 2-3 cross connection investigations and fixes per year is recommended at approximately $6,000 per
fix, to help alleviate the unexpected financial burdens on urban communities when illicit connections are found.
Bacteria alone should not be the determining factor of the presence or absence of an illicit connection for a variety
of reasons. Chlorine or other toxins in untreated wastewater may depress bacteria levels and bacteria lack conservative
behavior, which deem it a poor indicator (Pitt, 1993). Investigators have found that a careful and thorough outfall survey
is usually more informative than just collecting water samples.
Enforcement Action: City of Newmarket Case Study
Enforcement is another tool available to DES to achieve compliance. For example, setting timetables for compliance
milestones in a legal document is a method that, while typically thought of as a burden to a community, may actually
provide the impetus for action in a positive way. Public works departments of New Hampshire coastal communities are
not equipped with large, discretionary budgets to address unplanned remediation of illicit connections. When faced with
this dilemma, an enforcement action against the community provides public works departments with the validation to
support a request for additional funding from the officials who approve the allocation of funds.
113
-------
As the mill Town of Newmarket, New Hampshire, developed over the years, a small watercourse named Moonlight
Brook was built over and culverted in the center of downtown. In 1996, DES investigated Moonlight Brook based on
historic elevated bacteria levels. The DES investigation revealed dry weather E. coll concentrations as high as 41,600
counts/I 00ml in the brook. DES encouraged the town to initiate dye testing of the structures in the vicinity of Moonlight
Brook but, at that time, the town was reluctant to allocate staff and funding for clean up efforts (NHDES, 1997).
An administrative order was issued by the US Environmental Protection Agency for various violations of permitted
effluent limitations in October, 1997 and included a requirement that the town eliminate the raw sewage discharges from
the storm sewer system (USEPA Docket No. 97-78). The order required a plan and schedule for eliminating any pollutants
discharging during dry weather. The order also specified sampling of each active dry weather discharge that remained
following elimination of the illicit connections to the system identified by the town's fieldwork.
In response to the order, the town hired an environmental consultant to address the problems at the wastewater
treatment facility and the illicit connections. During the summer of 1997, the consultant and the town performed a dye
study of the subdrainage area that the town suspected was the likely source of bacterial contamination identified at the
discharge. The dye study resulted in the identification of a total of four untreated discharges to the storm drainage system
from three properties. A subsequent video inspection of the sewer lines adjacent to these properties revealed that the
sewer service connections from these properties might have been installed at the time the original sewer was constructed.
The consultant then concluded that this would indicate the sources of sewage discharging to the storm drainage are
broken sewer service connections rather than direct connections. The town stated that the remedial work would be
completed by June 1998 (Plante, 1998).
Another storm discharge pipe servicing this area was separated in 1985 and 1986, at which time dye testing was
performed to identify sanitary services that were connected to the sewer. The consultant determined that it would be
unlikely that direct sanitary service connections to the storm drain were present in this area, however, broken service
connections could result in sewage entering the storm drain culvert along Main Street. A dye study was planned for May
1998.
A total of 59 properties were included in the dye-testing program. Four of the properties were confirmed to be cross-
connected to the storm drainage system. Two of the four were the result of direct connections of sewer laterals to the
drainage system. The remaining two were a result of exfiltration from the sewer lateral through the ground to the drain
line (Town of Newmarket, 1999). The town reports a 90% reduction in the E. co//counts following the elimination of the
illicit discharges.
Legal Action: City of Dover Case Study
In the 1970's, the City of Dover, New Hampshire, constructed a new sewage collection system and treatment facility.
In 1997, DES investigators began surveying the storm drainage outfalls for contamination. Around this same time,
University of New Hampshire researcher Dr. Stephen Jones initiated a study in Dover to determine the significance of
flow (both dry and wet weather) coming from urban storm drainage systems (Jones, 1998). Jones identified a source of
bacterial contamination to be a cross connection later confirmed by DES and the City of Dover Public Works Department.
The city fixed the illicit connection by connecting the sanitary service into the sewer main, at no charge to the building
owner, while noting substantial flow from this service due to a hair salon in the building.
After learning about the existence of the cross connection, the building owners made an unsuccessful request to the
city for an abatement of the sewer fees they had paid since 1981 and initiated legal proceedings. The city alleged that
the case law mandated a decision in its favor and filed a motion for summary judgement (Strafford Superior Court, Order
#98-C-207). In a responding order from the judge, the case law was said to illustrate that the Court had considered a
variety of factors in related cases including (1) whether the new and old system were integral to one another, (2) whether
the benefit provided to the plaintiff under the new and old systems was comparable, and (3) whether the property owner
had access to the new system. The motion for summary judgment was denied because the Court found that these were
issues for a jury and that summary judgment at that stage would be premature.
A trial date was set. One week before the trial was to occur, the two parties settled out of court. The terms of the
settlement were confidential. If the property owners were successful in seeking a tax abatement and damages for unjust
enrichment, implied contract, and negligent misrepresentation, as sought, the pollution investigations could have been
114
-------
in jeopardy of becoming ineffective. Such a precedent could have led to other similar suits and would effectively remove
the incentive for municipalities to be proactive in fixing cross connections.
Conclusions
The Department of Environmental Services, in conjunction with the New Hampshire Estuaries Project, has
systematically identified illicit connections in the urban communities of coastal New Hampshire. Applying both voluntary
compliance and enforcement has resulted in the removal of cross connections to the storm drainage systems and a
decrease in the contamination reaching the coastal surface waters. DES is currently monitoring the shellfish growing
waters to determine the extent of water quality improvement resulting from the removal of illicit connections.
References
Hamilton R. Krans, and Allan B. Krans v. City of Dover, No. 98-C-207 (Strafford, SS. N.H. Dec. 28,1998) (order denying
defendant's motion for summary judgment.)
Jones, S.H. and R. Langan. 1995. Assessment of Nonpoint Source Pollution in Tributaries to the Great Bay Estuary. Final
Report. New Hampshire Office of State Planning/Coastal Program. Concord, New Hampshire.
Jones, Stephen, Ph.D. In Review. A Technical Characterization of Estuarine and Coastal New Hampshire. New
Hampshire Estuaries Project, Office of State Planning, Portsmouth, New Hampshire.
Jones, Stephen, Ph.D. 1998. Stormwater contamination of New Hampshire Coastal Surface Waters. Final Report
Submitted to the New Hampshire Coastal Program/Office of State Planning, Concord, New Hampshire.
Jones, Stephen, Ph.D. 1999. Public Health Significance of Stormwater-Borne Microorganisms. Final Report to the New
Hampshire Department of Environmental Services, Concord, New Hampshire.
Landry, Natalie. 1997. An Investigations of Water Quality in New Hampshire Estuaries. Final Report to the New
Hampshire Estuaries Project. New Hampshire Department of Environmental Services, Concord, New Hampshire.
New Hampshire Department of Environmental Services (NHDES). 1996.1996 Coastal Basin Nonpoint Source Pollution
Assessment and Abatement Plan. Concord, New Hampshire.
New Hampshire Department of Environmental Services (NHDES). 1997. 1996 Nonpoint Source Coastal Assessment
Report. Concord, New Hampshire.
New Hampshire Estuaries Project (NHEP). 1996. New Hampshire Estuaries Project National Estuary Program: EPA/State
Management Conference Agreement. New Hampshire Office of State Planning, Concord, New Hampshire.
Pitt, Robert, Melinda Lalor, Richard Field, Donald Dean Adrian, and Donald Barbe. 1993. Investigation of Inappropriate
Pollutant Entries into Storm Drainage Systems, A Users Guide. United States Environmental Protection Agency.
EPA/600/r-92/238.
Plante, Thomas R., P.E. 1998. Report to the US Environmental Protection Agency Regarding Administrative Order No.
97-87, Item 6-Dry Weather Discharges From Stormwater System. New Hampshire Department of Environmental
Services, National Point Discharge Elimination System files, Concord, New Hampshire.
Sea Grant. 1999. Sewage: Wastewater Major Pollutant Challenge. Sea Grant National Media Relations Office, National
Sea Grant College Program, http://www.seagrantnews.org/news/sewage.html.
Town of Newmarket. 1999. Storm Drain/Sewer Cross Connections, Identification and Correction. Final report to the New
Hampshire Estuaries Project. Prepared by Underwood Engineers, Portsmouth, New Hampshire.
Tuomari, Dean. 1996. Rouge River Watershed Illicit Sewer Connection Detection Program: A GIS Application. Watershed
96. Wayne County Department of Environment, Detroit, Michigan.
115
-------
United States Environmental Protection Agency. 1997. Administrative Order No. 97-78, Findings of Violation and Order
for Compliance In the Matter of Newmarket, New Hampshire, NPDES Permit No. NH01 00196.
116
-------
Using Collaborative Problem-Solving to Protect North Carolina's Coastal
Resources: The Experience of the White Oak River Advisory Board*
Leon E. Danielson, C. Suzanne Hoover, Christy A. Perrin
North Carolina State University
Department of Agricultural and Resource Economics
Raleigh, NC
Nancy M. White
North Carolina State University
School of Design
Raleigh, NC
Ron Elmore
North Carolina Department of Transportation
Raleigh, NC
Jennifer L. Platt
Town of Gary
Gary, NC
Introduction
In North Carolina coastal estuarine systems, land use change has been implicated as a significant cause of water
quality impairment (NC Department of Environment and Natural Resources, 1997; White, et al., 1998). Such development
processes change surface hydrology, pollutant delivery, and, as a consequence, adjacent water quality. Decisions
regarding placement, density, and type of development are controlled by policy implementation at the local level.
Furthermore, while the degree of impact may vary with each location, it is the cumulative effects throughout a watershed
that can be most damaging to water quality. Hence, there is a need to develop and enact policy locally, but on a multi-
jurisdictional, watershed basis.
Increasingly, local communities and governments are showing interest in playing a role in developing and
implementing solutions to water quality problems (NC Department of Environment and Natural Resources, 1997).
However, logistical complications arise upon implementation of this concept. First, a mechanism for effectively involving
local citizen stakeholders in the policymaking process may not exist and/or is difficult to establish (Danielson, 1998).
Second, technical data needed to address local issues and concerns are often not readily available, or are in a form not
easily understood. Third, programs for addressing water quality problems on a watershed-basis may not exist, suggesting
a need to develop, coordinate, and deliver multi-jurisdictional education on water quality issues and policy alternatives.
Through a project entitled Watershed Education for Communities and Local Officials (WECO), the North Carolina
Cooperative Extension Service has worked with a number of state and federal agencies, along with citizens and local
governments within a coastal watershed to address these needs.
The goal of this project is to improve water quality in all of the White Oak River Watershed through involvement and
education of citizens and government officials who live and work in the watershed. The project's main thrusts are : 1) the
delivery of technical information and educational material on water quality, management strategies, and policy options
that support watershed-based planning; 2) the empowerment of local citizens by facilitating collaborative partnerships
between communities, local officials, and state agencies within the watershed; and 3) the facilitation of the development
• Funding for this project is provided by USDA-CSREES under project number 97-EWQI-1-0150.
117
-------
of local stakeholder - driven policy recommendations for the entire watershed. This paper discusses the application of
these concepts to an issue of critical importance to local citizens.
Background
The White Oak River watershed is one of four rivers in the White Oak River Basin (Figure 1). It is 48 miles long and
encompasses 320 mi2 . The watershed begins in freshwater creeks and swamps of Jones County, NC, and contains
portions of three other counties—Craven, Onslow, and Carteret. Along its route to Bogue Sound and the Atlantic Ocean,
the river traverses between 30 ft. banks, which are relics of ancient dune ridges. This river is home to five threatened
or endangered organisms, including alligators; loggerhead, green, and leatherback turtles, and the Croatan crayfish. The
river and its estuarine waters have extensive primary nursery waters and provide habitat for several anadromous species--
herring, shad, striped bass, and sturgeon. The majority of the river is classified as SA, or saltwater suitable for
commercial shellfish harvesting.
White Oak River Watershed
North Carolina
•aven County
Figure 1. General map of White Oak River Watershed in North Carolina.
The White Oak River watershed has six major land cover/land use classes with wetlands encompassing the largest
single type at 52% of the total. Forests are the second largest land cover type constituting the majority of the headwaters
in the Croatan National and Hoffman State Forests (22%). A very small portion of the watershed is urban (2%) and
agricultural (11%) (NC Department of Environment and Natural Resources, 1997).
Despite the low level of urbanization, the North Carolina Division of Water Quality's basinwide management plan notes
an increase in shellfish closures in the river (North Carolina Division of Environmental and Natural Resources, 1997). At
state-sponsored public meetings, over 100 citizens expressed concern and called for more public education on water
quality.
118
-------
Recognizing the interest of their constituencies in water quality education, local NC Cooperative Extension Service
leaders assembled a project team involving members from the North Carolina Division of Water Quality, North Carolina
Division of Coastal Management, North Carolina Division of Environmental Health - Shellfish Sanitation Branch, North
Carolina Cooperative Extension, and 25 citizens who comprise the stakeholder-based Advisory Board for the White Oak
River Watershed. This group includes crop farmers, livestock farmers, fisherpersons, developers, foresters, tourism
directors, teachers, scientists, and local government officials from the watershed (see Table 1). The citizen advisory
board is the decision-making entity. The government agency representatives and Cooperative Extension personnel
function as support staff to the Board. Support staff provide resources, perform research and reviews, make reports,
serve as technical advisors, and provide formal facilitation and consensus-building services.
The White Oak Advisory Board's Primary Issue of Concern
The Board began meeting in August of 1996. Their first task was to prioritize water quality issues upon which to focus
their efforts, Board members expressed concern that past bridge and road construction across the mouth of the river had
contributed to a decline in water quality. Furthermore, this road, Highway 24, was slated for expansion, and they were
concerned that this would exacerbate the problems. The Board acknowledged the need for expansion of the road, but
recognized a unique opportunity to mitigate its impact if they could move quickly to work with the North Carolina
Department of Transportation (NC-DOT)
At the time that the Board was convened and identified the highway and its expansion as an issue, NC-DOT was in
the process of conducting an Environmental Assessment of the project and were anticipating a Finding of No Significant
Impact. During a meeting between the Extension Project Team and NC-DOT, DOT representatives were made aware
of the Board's concerns and expressed an interest in working with the Board to address those concerns. However, timing
was an issue because in several months, NC-DOT was planning right of way acquisition to begin the expansion project.
Because of the urgency of the matter, the Board resolved to meet twice monthly and work to develop their comments and
recommendations.
Technical Information Gathered by the Board
In response to the Board's inquiry, the Project Team reviewed, summarized, and presented scientificstudies that had
been conducted on the river that related to sedimentation and flow patterns in the river and the possible effects of highway
construction over the mouth of the river. Results from the following four studies were especially useful in understanding
the science behind this policy issue.
Table 1. White Oak River Watershed Advisory Board - Stakeholder Composition.
Stakeholder Groups
Fishing, Commercial
Fishing, Recreational
Real Estate or Development
Environment/ Conservation
Farming, Crop
Farming, Livestock
Forestry, Private
Business & Industry
Local Government
Academia/ Public Schools
Travel & Tourism
NC Shellfish Sanitation
Soil & Water Cons.
Public Forestry
|Totals
Carteret
12
Jones
5
Onslow
5
At Large
3
Total
4
1
1
1
2
1
1
3
3
4
1
1
2
25
119
-------
One study (Martens and Goldhaber, 1978) determined that the metabolic pathways by which bacteria degrades
organic matter in the sediments differ depending on whether the overlying aquatic environment is salty or fresh.
Chemistry analyses done on soil cores taken at various locations in the river found framboidal pyrite at upstream samples,
which indicated that saltwater wedges had previously penetrated further upstream than current patterns in soil chemistry
showed. These results provided evidence to the Board that saltwater flows in the river had changed over time.
Adams, Benniger, Hosier, Overton, and Reed (1982) studied water circulation and sedimentation patterns in the White
Oak Estuary and found that sedimentation in the estuary varies from 0.3 cm/yr to 5 cm/yr. approximating the annual rate
of submergence along the Atlantic coast. Their study confirmed for the Board that the system is a flood tide dominated
system with sediment transport primarily occurring during storms with strong on-shore winds. The study also noted that
the construction of the Intracoastal Waterway (ICWW) in 1930-32 in conjunction with the construction of Highway 24 in
1933 altered channel flow from one channel (adjacent to Muggins Island) to another (adjacent to the mainland near
Highway 24). Spoil deposition from dredging operations may also be responsible for decreased channel flow in the west
channel of the Inlet. The authors noted no evidence of a declining fishery based on the fact that it was comparable to
other fisheries in the area and in line with historical production rates for the estuary. This study also quantified the extent
of fill and alteration to the estuary caused by the original construction of the ICWW and the road in 1932 and 1933,
respectively. Historical maps, when compared to current data, showed that two inlets were closed and that more than
80% of the river was obstructed by these projects.
Benniger and Martens (1983) investigated the age and the sources of organic matter in the estuary. This study
characterized the organic matter degradation rates, which is important in understanding the estuary's capability to process
organic inputs. The researchers determined that the upstream organic matter inputs were primarily terrestrial and the
downstream organic inputs were primarily marine. However, they found that microbial processes acted preferentially to
remove recently produced organic matter. This implies that recently produced or partially treated organic matter could
substantially increase sediment oxygen demand and the rate of nutrient regeneration. This would increase the
vulnerability of the estuaries to anoxia and algal blooms.
Kelley, Martens, and Chanton (1990), by collecting and analyzing sediment cores, characterized the relative
remineralization rates of sedimentary carbon for the fresh and saltwater environments in the river. They found that the
upstream environment, which is dominated by terrestrial inputs and the process of methane reduction, remineralized at
a rate three times faster than the downstream site, which is dominated by marine inputs and uses sulfate reduction as
the energy pathway for organic matter remineralization. As indicated in the previously described paper by Martens and
Goldhaber (1978), saltwater circulation patterns, as well as freshwater inputs, appear to have changed such that the
estuarine ecology has shifted towards a more freshwater system. Since freshwater facilitates rapid remineralization of
organic carbons, this, over time, can reduce the river's buffering capacity and result in nutrient enrichment.
Initial Conclusions of Board
Based on these and other related studies, the Board concluded the following:
. Salt wedges that used to extend upstream have not occurred in recent history.
. Organic inputs upstream are from terrestrial sources and downstream are from marine sources.
. Salinity regimes in the river are highly variable seasonally and spatially.
. Salinity helps buffer the river from nutrient inputs.
. Sedimentation at the mouth of the estuary was considered normal for coastal estuary systems.
. There was no evidence to support a perceived decline in the fishery.
. Increased fresh water inputs from the expanded impervious surface area related to the highway expansion may
have a negative impact.
120
-------
. Higher salinity reduces concentrations of fecal coliform bacteria.
. Ditching and other means of moving water faster off the land causes problems with increased freshwater to the river
as well as increased bacterial contamination in shellfish beds.
. There is a significant shellfish resource at the mouth of the river that has historically remained open.
Next, the Advisory Board convened a panel of specialists to discuss this information and potential mitigation
strategies. The panel participants included:
. Dr. Larry K. Benninger, Geologist, University of North Carolina at Chapel Hill (UNC-CH);
. Archie Hankins, Biologist, NC-DOT;
. Tom Jarrett, Hydraulic Modeling, United States Army Corps of Engineers (USCOE);
. Dr. Chris Martens, Marine Sciences, UNC-CH;
. Dr. Paul Hosier, Biologist, UNC-CH;
. Dr. Rick Leuttich, Sedimentary Geologist, UNC-CH; and
. Howard Varnam, Hydrologist, USCOE.
The panel reviewed the scientific information presented to the Board, and they agreed that there has been an impact
on the circulation and flushing of the White Oak River since the construction of the causeway and the ICWW, but
quantification of those effects would require intensive modeling that would take a minimum of 1.5 to 2 years. The panel
felt that any action to increase circulation and salt water inputs to the river would have an overall positive effect on water
quality. However, the best manner in which to accomplish those goals and the particular effects on fisheries,
sedimentation, or other water resource values would be difficult without modeling studies. The panel concluded that due
to changes in land use, hydrology in the watershed had been altered. As a result, runoff volume during storm flows has
increased. This increases pollutant loading and increases erosion processes during storms. The panel noted that a
reduction in freshwater runoff would not have any significant effect on the diversity and density of species, but on their
distribution. This would have little effect on flora and fauna in the river, but might improve water quality. The panel also
noted that the most effective strategy for protecting water quality is to involve all of the communities impacting the system
and to implement overall land use planning in the watershed. It was suggested that the group needed to define their water
quality goals and how they want to manage the river and watershed to achieve those goals. Individual actions for
localized effects would require some additional modeling and research to determine the best options.
The panel concluded with a list of mitigation recommendations listed below:
. Pursue a study of the river to determine what, if any, actions should be taken to improve circulation up and
downstream of the highway.
. Examine options to manage stormwater in new and existing developments.
. Pursue the maintenance of buffers along creeks and streams.
. Pursue stricter enforcement of sediment and erosion control at construction sites.
. Endorse, encourage, and facilitate the use of BMPs in forests and farms.
. Work to develop a mechanism for watershed - based or coordinated land-use planning to address all of the
suggestions.
. Explore alternative waste management strategies for both single users and municipalities to reduce nutrients.
121
-------
The Board's Recommendations
Over the next several meetings, and as a consequence of these findings, the Board recommended the following
actions;
1. To reduce freshwater inputs to the estuary and possible negative impacts of highway runoff on water quality, the
Advisory Board recommended storm water runoff from bridge and highway expansion not be discharged into the
river and that the Department of Transportation (DOT) explore options to eliminate discharge into the waterways.
At a minimum, discharge from Highway 24 should be directed south (downstream) of the causeway to prevent
impacts to shellfish. In addition, it was recommended that amelioration of the velocity, volume, and quality of that
runoff be implemented, if feasible.
2. Historic maps showed that, prior to the 1930's, the mouth of the White Oak River was open and unrestricted,
allowing free tidal flow. In 1932 and 1933, Department of Transportation and US Army Corps of Engineers (ACOE)
projects closed approximately 80% of the mouth of the river and altered physical processes. The Advisory Board
recommended that to restore salinity regimes, increase tidal circulation, and reduce sedimentation, DOT take actions
to reopen the mouth of the river to the maximum extent possible. One option would be the creation of a north-south
channel connecting the estuary with the sound near the current location of the Flying Bridge Restaurant on the
Carteret County side of the river spanned by a bridge or connected by a culvert. Additionally, the Board
recommended that DOT and ACOE access ACOE ecological restoration funds and collaborate with each other to
mitigate the impacts of this expansion and past actions.
3. Since efforts to open the channel would not remain effective unless the State of North Carolina initiates an ongoing
maintenance program, the Advisory Board recommended that a long-term maintenance program supporting
improved circulation, reduced sedimentation, and restored salinity regimes be developed and implemented by
responsible agencies.
These recommendations were presented to and adopted by commissioners for Carteret and Jones Counties in May
and June of 1997. In addition, the White Oak River Watershed Advisory Commission of Onslow County (a group
appointed by the Onslow County Board of Commissioners to address water quality issues in Onslow County) endorsed
the recommendations of the Board at their May of 1997, meeting. This collaborative, consistent, watershed-based policy
statement became part of the public record for the NC-DOT hearings in May of 1997, and a preliminary draft was included
in the NC Division of Water Quality's Basinwide Water Quality Management Plan for the White Oak River Basin (North
Carolina Division of Environmental and Natural Resources, 1997).
Response to the Recommendations of the Board
At a joint meeting that included representatives of the NC-DOT, the White Oak River Advisory Board, the Extension
Project Team, and USCOE, the DOT agreed to support the Board's recommendations and revise their stormwater plans
to direct runoff away from the shellfish resource in the river.
Blueprints for construction were redrawn reflecting the following features. In the vicinity of the bridges carrying
Highway 24 over the White Oak River at Swansboro, NC-DOT agreed, to the extent possible, to direct the stormwater
runoff from the roadway to the Bogue Sound side of Highway 24 and away from the river. In Swansboro (west of the
island causeway), the existing stormwater collection system (which has outfalls on both the river and sound side of
Highway 24) will continue to be used for the runoff following roadway expansion, thus preventing the need for additional
outfalls in the river.
From the island causeway eastward for approximately 2.5 miles, the stormwater runoff from the highway will be
collected and piped to outfalls on the sound side of Highway 24. Also, NC-DOT has designed special channelization
islands for commercial driveways to accommodate some of the stormwater runoff from the highway and bridges. These
water quality islands are depressed inside the curb to allow the first inch of highway runoff to pond within these islands
and filter through the grassed areas located there. The filtered runoff from these islands is then collected and piped to
outfalls on the sound side of Highway 24.
122
-------
In addition to the stormwater design changes, it was agreed that DOT would cooperate with other state and federal
agencies in any efforts to improve circulation and tidal flushing. Currently, the Board is continuing to work on adding a
section to the Congressional Water Resources Development Act that would authorize the USCOE to conduct the study
necessary to determine what, if any, actions could be taken to improve flushing in the river.
Conclusions
Local stakeholder-based citizen groups can impact policies that affect their environment. Support for gathering,
summarizing, and delivering technical information to local citizens and governments is an important aspect of the success
of these processes. Knowing who to approach for answers to specific questions, and where to look for scientific
information is an important function of the group's technical support. In addition, translating the information gathered into
digestible and usable material is also critical.
References
Adams, David A.; Larry K. Benniger; Paul E. Hosier; Margery F. Over-ton; and James P. Reed. 1982. White Oak River
Management Project. Office of Water Resources, NC Dept. of Natural Resources and Community Development, Raleigh,
NC.
Benniger, Larry K. and Christopher S. Martens. 1983. Sources and fates of sedimentary organic matter in the White Oak
and Neuse River estuaries. WRRI Rept. # 194.
Danielson, L. E., 1998. Are We Really Prepared to be Honest Brokers and Conflict Resolvers in Controversial
Situations? Organized Symposia in Dealing with Controversy in Natural Resource Issues, at AAEA Annual Conference,
Aug 4, 1998. Salt Lake City, Utah.
Kelley, Cheryl A.; Christopher S. Martens; and Jeffery P. Chanton. 1990. Variations in the sedimentary carbons
remineralization rates in the White Oak River estuary, North Carolina.
Martens, Christopher S. and Martin B. Goldhaber. 1978. Early diagenesis in transitional sedimentary environments of
the White Oak River Estuary, North Carolina. Limnol. Oceanogr., 23(3), 428-441.
North Carolina Department of Environment and Natural Resources, 1997. White Oak River Basinwide Water Quality
Management Plan. North Carolina Division of Water Quality, Raleigh, North Carolina, 227 pps.
White, N. M., D. E. Line, J.D. Potts, William Kirby-Smith, Barbara Doll and W.F. Hunt. 1999. Jump Run Creek Shellfish
Restoration Project. In review, Journal of Shellfish Research.
123
-------
Community Responses to Stormwater Pollution:
Case Study Findings with Examples from the Midwest
George P. Aponte Clarke, Policy Analyst
Natural Resources Defense Council
New York, NY
Peter H. Lehner1, Chief
Environmental Protection Bureau
New York State Attorney General's Office
New York, NY
Diane M. Cameron', President
Cameron Associates
Kensington, MD
Andrew G. Frank3, Litigation Associate
Paul, Weiss, Rifkind, Wharton, and Garrison
New York, NY
Stormwater runoff threatens the nation's waterways and public health, and costs Americans hundreds of millions of
dollars each year. Concerns about urban runoff and interest in proposed new federal Stormwater regulations prompted
the Natural Resources Defense Council (NRDC) to document existing, effective Stormwater strategies. Our report aims
to encourage municipal action and empower communities to address this critical issue. More than 150 case studies from
across the nation were compiled and evaluated to highlight effective pollution prevention, administrative, and financing
strategies for addressing Stormwater runoff. The case studies show, on a practical level, that Stormwater management
can be environmentally effective, economically advantageous, and politicallyfeasible. The reportalsoforms the foundation
of a comprehensive outreach effort. Together, they help guide communities as they implement or improve Stormwater
management programs by providing detailed examples of proven tools and approaches used to prevent Stormwater
pollution. Collectively, the case studies offer an outline for further successful Stormwater management strategies.
Elements critical to the effectiveness of these programs include: a pollution prevention emphasis with structural treatment
measures when needed; a focus on preserving natural features and processes; programs that inform and involve the
public; a framework that creates and maintains accountability; a dedicated and equitable funding source to ensure long-
term viability; strong leadership; and effective administration. These broad themes translate into a set of nine local actions
for addressing the technical, social, and political issues associated with Stormwater runoff. The case studies show that
following these actions will help communities form a sound Stormwater policy.
Key Terms: urban Stormwater runoff, impervious surfaces, pollution prevention, best management practices, diffuse
pollution, accountability.
'formerly Senior Attorney and Clean Water Project Director, Natural Resources Defense Council
2 formerly Senior Scientist, Natural Resources Defense Council
3 formerly Policy Analyst, Natural Resources Defense Council
124
-------
Introduction
Currently, there is substantial concern about the impacts of urban and suburban runoff. Pollution from diffuse sources,
including urban stormwater, is the leading source of contamination in the nation's waters (U.S. Environmental Protection
Agency, 1997a). Stormwater runoff pollution is a particularly important issue since most of the population of the United
States lives in urban and coastal areas. Water resources in urban and coastal areas are highly vulnerable to and are often
severely degraded by stormwater runoff. Specifically, urban and suburban runoff is the second most prevalent source
of water quality impairment in the nation's estuaries after industrial discharges (U.S. Environmental Protection Agency,
1998b).
Economic impacts are an important aspect of this concern. Even a partial accounting shows that hundreds of millions
of dollars are lost each year through added government expenditures, illness, or loss in economic output due to urban
runoff pollution and damages (U.S. Environmental Protection Agency, 1998a). The ecological damage is also severe and
is at least as significant. In particular, uncontrolled urban runoff contributes to hydrologic and habitat modification, two
important sources of river impairment identified by the U.S. Environmental Protection Agency (EPA).
The polluted stormwater runoff problem has two main components: the increased volume and rate of runoff from
impervious surfaces and the concentration of pollutants in the runoff. Both components are closely related to development
in urban and urbanizing areas (Booth and Reinelt, 1993; Schueler, 1994; U.S. Environmental Protection Agency, 1997b).
When impervious cover (roads, highways, parking lots, and rooftops) reaches between 10 and 20 percent of the area
of a watershed, ecological stress becomes clearly apparent (Klein, 1979; Booth and Reinelt, 1993; Schueler, 1994).
Everyday activities can deposit on these surfaces a coating of various harmful materials. When it rains or when snows
melts, many of these pollutants are washed into receiving waters, often without any treatment.
The deposition of pollutants and the increased velocity and volume of runoff together cause dramatic changes in
hydrology and water quality (Klein, 1979; Jones and Clark, 1987; Booth, 1990; Galli, 1990; U.S. Environmental Protection
Agency, 1997b). These changes affect ecosystem functions, biological diversity, public health, recreation, economic
activity, and general community well-being (Bannerman ef a/., 1993; Novotny and Olem, 1994; Haile et a/., 1996;
Carpenter et a/., 1998). Urban stormwater is not alone in polluting the nation's waters. Industrial and agricultural runoff
are often equal or greater contributors. But the environmental, aesthetic, and public health impacts of diffuse pollution
will not be eliminated until urban stormwater pollution is controlled.
While urban and suburban runoff continues to be a critical issue, there is substantial evidence that the problems are
not intractable. Increasingly, communities are recognizing the causes and consequences of uncontrolled urban runoff and
taking action to control and prevent runoff pollution, often without any mandate. These innovative communities are
realizing the environmental, economic, and social benefits of preventing stormwater pollution. However, neither the extent
of these efforts nor the specific actions being taken have been well documented.
There is also a growing interest in proposed new federal stormwater regulations. Comprehensive stormwater regulation
is required under Section 402(p) of the Clean Water Act. Since 1992, cities with populations over 100,000, certain
industries, and construction sites over 5 acres have been required to develop and implement stormwater plans under
Phase I of the National Pollutant Discharge Elimination System (NPDES) stormwater regulations (U.S. Environmental
Protection Agency, 1990). In October 1999, EPA is expected to promulgate a new rule requiring municipalities with
populations fewer than 100,000 people located in "urbanized areas" (where population density is greater than 1,000
persons per square mile) to develop stormwater plans. Under what is known as the "Phase M" rule, the EPA and states
will develop "tool boxes" from which the smaller local governments can choose particular stormwater strategies to develop
their stormwater plans (U.S. Environmental Protection Agency, 1998a).
To address all of these issues and concerns, the authors developed a study to examine, document, and disseminate
information on environmentally effective and economically advantageous stormwater pollution prevention strategies. The
study resulted in a report, Stormwater Strategies: Community Responses fo Runoff Pollution, that highlights some of the
125
-------
most effective existing stormwater strategies from around the country (Lehner et a/, 1999). The report provides
substantial evidence that such programs exist and highlights a variety of innovative strategies actually being used. The
report also aims to provide guidance to communities addressing stormwater issues, encourage municipal action, and help
empower communities to be involved in this critical issue. This paper summarizes the study and presents its primary
findings and recommendations.
Study Design and Approach
The study was exploratory in nature, with the intent of presenting information on existing effective stormwater
management programs. To achieve this goal, we collected examples of environmentally beneficial and cost-effective
stormwater programs from across the country. We compiled this information into the case-study-based report described
above. This information and report have become the basis for a comprehensive outreach effort.
The first step was to gather information on programs and projects by examining existing programs (several begun
under Phase I as well as many that started earlier), reviewing literature, contacting regional and local stormwater
management experts and researchers, and interviewing representatives from stormwater management or other local
government agencies. We gathered information on over 250 programs. The information was then examined in detail and
narrowed down to a set of case studies that demonstrated elements of success. Three fundamental criteria for selection
were used: environmental gains, economicadvantages, and community benefits. Environmental gains included biological,
hydrological, or chemical improvements resulting from stormwater management. Economic advantages included cost
savings to the municipality or developers, or increases in property values related to the pollution prevention measure.
Community benefits included aesthetic or recreational enhancement, administrative or institutional successes, or
community relations improvements.
Seventy-seven programs and projects were selected as case studies for the final report. Another 88 programs were
annotated to provide additional programs/locations not fully evaluated for the report. The case studies represent
communities of all sizes, types, and regions throughout the United States. To help ensure accuracy, local experts or
people familiar with the program, called "groundtruthers,"were contacted to review the case studies and add information
from their own knowledge and experience.
The case studies were first organized geographically by dividing the United States into six regions based in part on
general rainfall patterns. Within each of the regions, case studies were then further subdivided into five categories of
stormwater management measures including, (1) addressing stormwater in new development and redevelopment, (2)
promoting public education and participation, (3) controlling construction site runoff, (4) detecting and eliminating improper
or illegal connections and discharges, (5) and implementing pollution prevention for municipal operations. These
categories roughly parallel those measures that large municipalities address under existing Federal regulations (40 CFR
parts 122.26 and 123.25) and small municipalities will address under pending Federal regulations (U.S. Environmental
Protection Agency, 1998a).
Case Study Findings
Through reporting over 150 examples of actual programs, the full report provides substantial evidence that stormwater
pollution can be reduced or prevented with proper planning and implementation in growing or re-developing areas. The
examples presented in the report also demonstrate that if some communities can measurably and cost-effectively reduce
stormwater pollution, so can other communities and states (Lehner, et al., 1999).
The Five Categories of Stormwater Management Measures
Individually, the case studies provide detailed examples of substantial water quality improvement, effective or
innovative stormwater control strategies to protect the natural environment, significant cost-savings, and important
ancillary benefits to the community. The programs and strategies highlighted come from communities of all sizes, types,
126
-------
and regions. They include efforts by municipal agencies, developers, and community groups. In many cases, several of
these groups worked together to create win-win outcomes. The case studies highlight a variety of strategies for
addressing the five categories of stormwater management measures previously enumerated, and are described in more
detail as follows.
Addressing Sformwaferin New Development and Redevelopment. By far the most important category of stormwater
strategies focuses on land use and development. It encompasses a wide range of measures including regional or
watershed planning, buffers and open space preservation, infill development, conservation design, and the use of site-
specific structural and nonstructural treatment measures. One of the best strategies a municipality or developer can
employ is to minimize the aggregate amount of new impervious surfaces. For example, developers of the Prairie Crossing
project in Grayslake, Illinois, prevented runoff pollution and saved money by using conservation design strategies. The
developers first reduced impervious cover by clustering 317 residences on only 132 acres of the site, which left 80 percent
as open space. They then designed the developed area around a natural drainage system consisting of vegetated swales,
restored prairie, and wetlands. Modeling indicates that this stormwater treatment drain system will remove approximately
85% of nutrients, metals, and suspended sediments and reduce peak flows by 68%. Eliminating curbs and gutters
resulted in savings of $1.6 to$2.7 million. The development is also very appealing to homebuyers, with sales comparable
to or better than conventional developments in the area (see Lehner et al., 1999, p. 224).
Promoting Public Education and Participation. Individuals play a key role in reducing stormwater impacts both in their
own day-to-day activities and in showing support for municipal programs and ordinances. The most successful highlighted
programs accomplished three goals: they educated the public about the nature of the problem, they informed the people
about what they can do to solve the problem, and they involved citizens in hands-on activities to achieve pollutant
reduction or restoration targets. One example of this success is in Minneapolis, Minnesota, where a decline in water
quality motivated the Lake Harriet Watershed Awareness Project. Monitoring revealed that lawn-care chemicals were a
significant contributor to the problem, which suggested focused education efforts. In turn, the project developed two
approaches: a volunteer master gardener program and the distribution of educational materials. Evaluation showed that
67% of watershed residents reported using the information presented and 30% reported a change in behavior. As a result,
concentrations of lawn-care pesticides have dropped by 50% or more since the program began (see Lehner et al., 1999,
p. 231).
Controlling Construction Site Runoff. The case studies demonstrate that effective construction site pollution prevention
is politically and economically feasible and can dramatically reduce pollution. The most effective programs rest on four
cornerstones laid in pairs: enforcement and education; erosion prevention and sediment control. However, the first and
over-arching necessity is a clear set of requirements. For example, Herzog et a/. (1998) found that in Geauga County,
Ohio, and St. Joseph County, Indiana, aggressive, widespread seeding and mulching reduced construction site erosion
by up to 86% and reduced phosphorus loadings by 80%. These measures can also benefit developers financially. They
found that homebuyers perceive these "green" lots to be worth $750 more than comparable "brown" lots (see Lehner et
a/., 1999, p. 236). While existing programs employ a wide variety of erosion and sediment control practices, virtually all
successful strategies require proper planning and phasing of construction activities to minimize land disturbance.
Defecting and Eliminating Improper or Illegal Connections and Discharges. Local governments have found that
identifying and eliminating illicit connections and discharges is a remarkably simple and cost-effective way to address
some of the worst stormwater pollution. The case studies show that two factors are critical to success of this element of
stormwater programs: finding illicit connections and discharges, and enforcement. In Washtenaw County, Michigan, the
Huron River Pollution Abatement project resulted in a 75% reduction in the river's fecal coliform levels in just 4 years. The
project focused on eliminating existing illicit connections and preventing future incidents through chemical storage surveys,
industrial inspections, water-quality monitoring, public education, and complaint and spill response. Over a six-year period,
the program dye-tested more than 3,800 facilities, after which 328 of the 450 illicit connections found were removed (see
Lehner et al., 1999, p. 239).
127
-------
Implementing Pollution Prevention for Municipal Operations. A wide range of municipal operations can affect
stormwater quantity and quality. The case studies reveal that some local governments have been able to manage their
municipal operations to reduce stormwater pollution. The municipalities highlighted have done so in a variety of ways
including reducing the use of harmful chemicals in the maintenance of municipal properties and vehicles, improving the
maintenance and cleaning of roads and stormwater infrastructure, and training staff in pollution prevention practices.
Several municipalities have taken these steps at their golf courses. For example, the Village Links Golf Course in Glen
Ellyn, Illinois, is preventing runoff pollution by incorporating integrated pest management, water conservation, stormwater
detention, native planting, recycling, and public outreach into its day-to-day management. The golf course relies on both
mechanical and biological pest controls and has significantly increased natural areas. The course collects runoff from
nearby streets and neighborhoods in its system of ponds and spillways. These ponds provide approximately 60% of the
course's irrigation water, and the course itself passively treats and filters all excess runoff from irrigation (see Lehner et
a/., 1999, p. 243).
Themes Common to Success Stories
Collectively, over 150 case studies present a clear model for success. Evaluation of the case studies revealed several
common elements among the highlighted programs. We distilled those elements into the seven broad themes listed below
to help guide communities as they develop or improve stormwater programs. Since they are based on actual programs,
these themes form a solid foundation for successful programs.
Preventing pollution is high/y effective and saves money. Pollution prevention measures dramatically and cost-
effectively reduce the quantity and concentration of pollutants "winding up" in stormwater. Common pollution prevention
measures include reducing or eliminating the use of harmful products, preventing erosion, reducing the amount of
pavement in new developments, and changing maintenance practices. In highly urbanized areas, however, such
measures may be difficult. In such cases, several communities have found treatment of runoff with structural measures
or retrofitting existing structures to be effective alternatives.
Preserving and utilizing natural features andprocesses have many benefits. Many communities and developers have
found strategies that rely on natural processes to be highly effective and economically advantageous. Undeveloped
landscapes absorb large quantities of rainfall and snowmelt and vegetation helps to filter out pollutants from stormwater.
Buffer zones, conservation-designed development, sensitive area protection, or encouragement of infill development all
enhance natural processes.
Educafingandinforming the general public and municipal staff improves program effectiveness. Providing information
and training to the general public and local businesses is a key component to many of the highlighted programs. Since
many sources of stormwater pollution are derived from individual activities such as driving and maintaining homes,
educating the public goes a long way to reducing stormwater pollution. Several communities involve the public in civic
activities, such as monitoring water quality or stenciling storm drains, which not only provide educational opportunities
but also save the municipality money.
Strong incentives, routine moniforing, and consisfenf enforcement establish accountability. Enforcement, or more
broadly accountability, is a key element to improving water quality. All actors need a clear statement of performance
goals, and they need to be held accountable by others for accomplishing these goals. We found that programs with high
accountability were the most effective, often achieving pollutant reductions of 50% or greater.
Financial stability helps ensure effective programs. Effective stormwater programs are financially viable and affordable.
Dedicated funding sources, such as stormwater utilities or environmental fees are equitable ways to build stability into
stormwater programs. Stability and equity are also important in gaining public support. Nearly 200 communities across
the nation are already realizing the benefits of implementing stormwater utilities as dedicated and equitable funding
sources.
128
-------
Strong leadership is often a catalyst for success. Success, at least initially, often requires an individual to champion
the project and make it happen.
Effective administration is critical. Regardless of which strategies a community chooses, those programs with clear
goals and objectives are the most successful. Such clarity enhances accountability, responsibility, and trust. Furthermore,
an established and understood institutional framework often improves administration by fostering collaboration among
different parts and levels of government, neighboring communities, and local citizens. Effective administration allows
implementation of broad-based, multi-faceted programs, which are often the most effective at controlling the diffuse
problem of stormwater pollution.
Authors' Recommendations for Local Action
To further guide communities addressing stormwater runoff issues, we translated the broad themes presented above
into an action plan based on nine key recommendations. These actions roughly parallel the broad themes presented
above. The case studies demonstrated that following the nine local actions outlined below will help build a strong
framework for effective, efficient, and successful stormwater management over the long term.
1) P/an in advance and set clear goals. Carefully plan programs, as opposed to simply reacting to provided
opportunities, crises, or transient pressures. Planning allows development of more effective and cost-effective
actions. An essential outcome of planning is addressing the issues and concerns of all stakeholders involved.
2) Encourage and facilitate broadparficipation. Program planning, development, and implementation should involve
multiple levels of government, key members of the community, and professionals from a variety of related
disciplines. A key to success is the public's understanding of the issue, how it relates to them, and what they can
do about it.
3) Promote public education opportunities. Implement broad-based programs that reach a range of audiences and
solicit different levels of public involvement. Remain committed to the education program and take advantage of
existing community organizations to enhance participation.
4) Work to prevent pollution first; rely on structural treatment on/y when necessary. Focus on prevention-based
approaches, through regional and watershed planning, local zoning ordinances, preservation of natural areas,
stormwater-sensitive site design, and erosion prevention as these are significantly more effective than treatment
of polluted runoff.
5) Establish and maintain accountability. Essential components of this process are setting clear standards, creating
strong incentives and disincentives, conducting routine monitoring and inspections, keeping the public informed,
promoting public availability of stormwater plans and permits, and consistently enforcing laws and regulations.
Strong enforcement is often key to significant water quality improvements.
6) Secure financial resources. Consider establishing a dedicated funding source such as a stormwater utility.
Combine with it budget-saving measures such as creative staffing, public-public and public-private collaboration,
and building off existing programs.
7) Tailor strategies to the region and setting. Recognizing that every case will be different, consider strategies that
are particularly tailored to the region, the specific audience, and the problem.
8) Evaluate and allow for evolution ofprograms. Set clear goals and priorities, and allow programs to develop over
time. Establish clear ways to check and see that goals and objectives are being met. This opens opportunities
for improvements and helps ensure long-term success.
129
-------
9) Recognize the importance of associated community benefits. Stormwater pollution prevention measures usually
offer ancillary quality-of-life benefits in addition to targeted improvements. For example, preserved areas offer
parks, ponds offer beauty and habitat, clean streets are more attractive, education helps empower people, and
sediment control improves fisheries and prevents flooding.
Conclusion
Many fine handbooks provide theoretical and technical guidance concerning the design and implementation of
effective stormwater pollution prevention and control measures. This study took a different approach and focused on
existing effective programs in a variety of settings. In doing so, it accomplished two key goals. First, the study
demonstrates that stormwater management is quite possible. The case studies show on a practical level that stormwater
management can be environmentally effective, economically advantageous, and politically feasible. Second, the case
studies enable communities developing or improving stormwater programs to learn from their peers. In doing so, the case
studies offer an outline for future successful stormwater management strategies.
Acknowledgments
The authors extend their appreciation to all the communities, organizations, agencies, and individuals who provided
information for this study; the technical consultants, peer reviewers, and groundtruthers for their helpful comments; and
the volunteers and Natural Resources Defense Council staff who assisted with this study. The study was supported by
a grant from the EPA under Section 104(b)(3) of the Clean Water Act.
References
Bannerman, R. T., D. W. Owens, R. B. Dodds, and N. J. Hornewer. 1993. Sources of Pollution in Wisconsin Stormwater. Water
Science and Technology, 28(3-5): 241-259.
Booth, D. B. 1990. Stream-Channel Incision Following Drainage-Basin Urbanization. Water Resources Bulletin, 26(3): 407-417.
Booth, D. B. and L. E. Reinelt. 1993. Consequences of Urbanization on Aquatic Systems-Measured Effects, Degradation
Thresholds, and Corrective Strategies. Proceedings, Watershed '93: A National Conference on Watershed Management,
Alexandria VA, pp. 545-550.
Carpenter, S. R, N. F. Caraco, D. L. Correll, R. W. Howarth, A. N. Sharpley, and V. H. Smith. 1998. Nonpoint Pollution of
Surface Waters with Phosphorus and Nitrogen. Ecological Applications, 8(3): 559-567.
Galli, J. 1990. Thermal Impacts Associated with Urbanization and Stormwater Management Best Management Practices: Final
Report. Metropolitan Washington Council of Governments, Washington, DC, 157 pp.
Halle, R. W., James Alamillo, Kevin Barrett, Ron Cressey, John Dermond, Carolyn Ervin, Alice Glasser, Nina Harawa, Patricia
Harmon, Janice Harper, Charles McGee, Robert C. Millikan, Mitchell Nides, John S. Witte, 1996. An Epidemiological Study of
Possible Adverse Health Effects of Swimming in Santa Monica Bay. Santa Monica Bay Restoration Project, Santa Monica, CA,
70 PP-
Herzog, M., J. Harbor, K. McClintock, J. Law, and K. Goranson. 1998. Are Green Lots Worth More than Brown Lots? An
Economic Incentive for Erosion Control on Residential Developments. Journal of Soil and Water Conservation, Accepted.
Jones, R. C. and C. C. Clark. 1987. Impacts of Watershed Urbanization on Stream Insect Communities. Water Resources
Bulletin. 23(6): 1047-1 055.
Klein, R. D. 1979. Urbanization and Stream Quality Impairment. Water Resources Bulletin, 15(4): 948-963.
130
-------
Lehner, P. H, G. P. Aponte Clarke, D. M. Cameron, and A. G. Frank. 1999. Stormwater Strategies: Community Responses
to Runoff Pollution. Natural Resources Defense Council, New York, N.Y., 269 pp.
Novotny, V. H. and Harvey Olem. 1994. Water Quality: Prevention, Identification, and Management of Diffuse Pollution. Van
Nostrand Reinhold, New York, 1054 pp.
Schueler, T. R. 1994. The Importance of Imperviousness. Watershed Protection Techniques l(3): 100-1 11.
U.S. Environmental Protection Agency. 1990.40 CFR Parts 122,123, and 124, National Pollutant Discharge Elimination System
Permit Application Regulations for Storm Water Discharges; Final Rule. Federal Register, U.S. Government Printing Office,
Washington, DC 55(222): 47992-47993.
U.S. Environmental Protection Agency, 1997a. Nonpoint Source Pollution: The Nation's Largest Water Quality Problem.
Accessed December 21, 1998, at URL http://www.epa.gov/OWOW/ NPS/facts/point1 .htm.
U.S. Environmental Protection Agency. 1997b. Urbanization and Streams: Studies of Hydrologic Impacts, Office of Water,
Washington, D.C., 841 R-97-009, 15 pp.
US. Environmental Protection Agency. 1998a. 40 CFR Parts 122 and 123 Part II, National Pollutant Discharge Elimination
System-Proposed Regulations for Revision of the Water Pollution Control Program Addressing Storm Water Discharged;
Proposed Rule, Federal Register, U.S. Government Printing Office, Washington, D.C., 63(6): 1536-1 643.
U.S. Environmental Protection Agency. 1998b. National Water Quality Inventory: 1996 Report to Congress, EPA841 -R-97-008,
521 pp.
131
-------
Integrated Urban Stormwater Master Planning
Eric Strecker, Vice President
URS Greiner Woodward-Clyde
Portland, OR 97201
and
Krista Reininga, Sr. Project Engineer
URS Greiner Woodward-Clyde
Portland, OR 97201
Abstract
Urban stormwater management agencies are increasingly being called upon to address water quality and natural
resources issues in addition to their traditional focus on flood conveyance. In response to this, stormwater drainage
master plans have been increasingly addressing stormwater quality and, in limited cases, natural resources and habitat.
This paper will describe some of the problems with traditional stormwater master planning approaches, including those
where water quality and natural resources have been included as "add-ons," and the urban stormwater problems we are
now trying to address which have resulted from these approaches. A framework for how communities can develop
integrated stormwater master plans that address multiple objectives, as increasingly mandated by public concern as well
as by regulations, will be presented. Given that the tools available for master planning are not equivalent in their
numerical evaluations, new procedures and project approaches are required. Especially important is how the
hydrology/hydraulic methods are performed, including both flood evaluations and evaluation of the smaller channel-
forming storms.
Communities are often not institutionally organized to address multiple objectives. Master planning has traditionally
been led and performed by engineers trained in hydrology and hydraulics, and they are usually in different departments
from those who are responsible for other environmental aspects of the drainage system. This paper will focus on the
technical, institutional, and process-oriented aspects of how master planning can be improved. Several case studies from
the Pacific Northwest of the United States will be discussed.
Introduction
The purpose of this paper is to discuss some of the attributes of urban stormwater master planning and how those
master plans can be improved to more fully address issues besides conveyance capacity and flood control. Stormwater
master plans go by a number of names, including storm drain master plans, stormwater infrastructure plans, and urban
catchment management plans. These plans are usually very focused on flood control and, until just recently, address
water quality minimally. This paper will discuss some of the attributes of traditional urban stormwater master planning
and its results, regulatory programs (which in the US and New Zealand are requiring adifferent approach), how integrated
master planning can be accomplished, and institutional barriers which often prevent integrated master planning from being
accomplished. In this paper, an Integrated Stormwater Master Plan is an infrastructure and management plan that not
only addresses flood control and property protection issues, but also considers stream stability and habitat, along with
water quality and aesthetics.
Urban Stormwater Drainage Problems
It has long been recognized that in urban areas, unplanned stormwater management systems result in damage to
property and sometimes people. As it will be well demonstrated by other papers in these proceedings, urbanization of
watersheds and the resulting impervious areas also cause changes to the hydrology and water quality of receiving waters
which ultimately result in other impacts to aquatic life and humans. Even some of our measures to control impacts can
have unplanned detrimental effects. Especially sensitive to these changes are stream systems and coastal embayments
132
-------
that are not well flushed. Almost always there are also direct impacts to stream riparian areas which also increase these
changes through canopy removal and channel modifications.
Urbanization usually includes impervious areas directly connected to efficient stormwater conveyance systems
(including roof drains and driveways connected to streets and curbs to inlets to pipes) which then are discharged to
streams directly or through engineered channels. This has resulted in stormwater being conveyed as fast as possible
to receiving waters (and away from properties). Increasingly, it is being recognized that because stormwater is drained
to streams in this manner, small storm hydrological changes that result in increased runoff flows can significantly increase
the frequency and duration of elevated flows. This energy change within the normal wetted channel often results in
channel cutting, widening, and/or sedimentation, which in turn can cause severe habitat and water quality degradation
(MacRae 1996; Severn and Washington, 1996). Often to "fix" these channel problems, streams are enclosed, hardened,
and/or straightened. Even without considering the water quality of stormwater, our stormwater systems are severely
impacted from a physical habitat standpoint, including habitat loss, higher velocities, and temperature changes. Figure
1 shows an example of how stream runoff can change with urbanization, including much higher and peaky flows as well
as increased volumes of runoff.
Rainfall and Runoff
0.9
0.8
0.7 •
£0.6
!rb.5 •
|o.4 •
CC
0.3 •
0.2
0.1
n n .
* MB Rainfall (in/hr)
I |\ — • Post Dev (c=.95)
j« \\ - - Pre Dev (c=.30)
n P ji Post Dev (with discharge control)
I |^ t h t i
i PN\
i \/\; \
I i^l u ^
' i / * ]i \
-. \ ' •• v i
;:\/-A- *. '- •' •.' l \
• '. ' ' V v . • \ , I ^v
•^ '. I \
0.09 -L \ \
JL.
^^^^^g^^^^^^^^^^^^^^^^^^ 1 1 1 t 1 1 1 1 1 1 1 1 1 1 Ill
0.08
0.07
0.06
0.05
0.04 -g
0.03 0
E
0.02
0.01
0
-0.01
00
CM
CM •* (O 00 O
CO CO CO CO "*t
Hours
Figure 1. Example Schematic of Changing Rainfall/Runoff Relationships with Development.
With urbanization also comes a dramatic change in water quality. Urban stormwater systems are the efficient
conveyance system of urban pollutants, both those discharged during storm events and those occurring during dry-
weather discharges. There are numerous ways that pollutants enter stormwater from those in the rainfall itself to
commonly thought of sources such as street dirt and car drippings. Stormwater often exceeds US EPA water quality
criteria. Figure 2 is a graph of the frequency that stormwater runoff from identified land uses in Oregon exceed US EPA
acute dissolved metals water quality criteria in runoff from identified land uses (Strecker et al., 1997). It should be noted
that most of this runoff was measured in pipes, while the criteria are meant to apply to receiving waters. A data set of
flow-weighted composite samples (representing average storm concentrations) from over 40 land use stations from
various areas of the Willamette Valley was utilized to develop the information displayed in the figure. The stations
included an open land use station in an urban area (Forest Park in Portland) for comparison. Note that dissolved copper
and zinc in developed land uses exceeded criteria for 30 to 65% of the storm events. Similar findings have been found
in other programs, including the San Francisco Bay area programs (Cooke and Lee, 1993).
133
-------
• Open
• Residential
• Commercial
D Transportation
U Industrial
Cadmium Copper Lead
Zinc
Based upon Oregon NPDES Stormwater Monitoring Data
Compiled by ACWA. Developed areas: 27 to 67 storm
events; Open space: 9 storm events
Figure 2. Frequency that Flow-weighted Composite Urban Stormwater Runoff Samples Metals Concentrations Exceeded US EPAs Acute Criteria for
Aquatic Life.
The water quality impacts together with the physical hydrology changes described above have caused our urban
stream systems to become severely degraded. Our traditional systems have not protected the resources nearly as well
as they have protected property. As many have now recognized, at about 10 to 25% imperviousness, the health of the
aquatic system is severely degraded (May et al., 1997; Schueler, 1994). In many cases because of the longer-term
channel stresses, property has been damaged as well, including under cutting of headwalls, etc. The plans typically only
identified solutions that solved large flooding problems, sometimes just temporally until what has been considered
"maintenance" problems such as head wall failures, occur.
Environmental Concerns and Regulatory Requirements
In the US, Congress has recognized that urban Stormwater plays a major role in affecting receiving waters when it
mandated in the revised Clean Water Act that urban Stormwater water quality be addressed through a permitting
(consent) program. New Zealand has similar requirements through its Resources Management Act of 1991. Both these
programs are still evolving. The Stormwater permit program in the US specifically requires that larger cities (over 100,000)
and soon smaller cities address Stormwater quality issues as they conduct flood control projects. New Zealand's program
also requires that municipalities obtain consents for Stormwater discharges.
Under the overall program, one area that has been slow to change is how urban Stormwater master plans are
developed and implemented. Although there are requirements to consider water quality in conducting flood control efforts,
for a number of reasons (including institutional inertia) agencies have been somewhat slow in actually giving water quality
and habitat protection equal weight with flood control in master planning. Some of this is due to the fact that Stormwater
master plans are typically the responsibility of engineers who are experienced in hydraulics, but that often lack experience
and knowledge in other aspects of environmental Stormwater management. To be fair, engineers have been told to plan
for managing Stormwater based upon land-use zoning that was selected without considering Stormwater issues. Another
major issue is the resources allocated to conduct integrated planning efforts which are more expensive; often agencies
do not recognize the value of better up-front planning compared to capital and maintenance costs.
Increasingly though, the public has started to demand that more environmentally sound and/or aesthetically pleasing
Stormwater management approaches be utilized. For example, with the endangered species act (ESA) listings and
proposed listings of salmon and trout species in the US Pacific Northwest, many neighborhood organizations are
134
-------
pressuring municipal agencies to change their stormwater management approaches. Some of these efforts are having
more success than the regulatory programs.
Stormwater Management Agency Functions
Understanding a stormwater management agency's function and history is important to understanding its approach
to stormwater management. Stormwater management agencies typically fulfill the following roles:
1. Stormwater System Maintenance
2. Development Standard
3. Stormwater Master Planning
4. CIP Design and Construction
5. Funding-Utilities/System Development Charges
6. Stormwater System Permitting and Environmental Impact Minimization
7. Education
The last two elements are the most recent. Many agencies began by responding to emergencies and problems, and
were then tasked to develop onsite design conveyance standards. Stormwater master plans for the most par-t were
developed in response to problems that arose after watersheds were developing with little or no stormwater planning.
They also were typically focused on just flood control and property protection. Most often they focused on the piped
systems and road culverts. Often creeks away from culverts were not evaluated unless there had been a particular
problem identified. In the US, the Federal Emergency Management Agency (FEMA) had separately developed flood plain
maps for larger systems, which communities relied upon for protecting structures from larger river and stream flooding.
This was done to meet requirements for participation in FEMA's flood insurance program. Therefore, flood plains and
the creeks themselves have not been a focus of master plan (e.g., creek sections were typically not evaluated to a great
extent).
Stormwater Drainage Master Plan Goals and Results
The traditional purposes of the Stormwater Drainage Master Plan were to:
. Guide a city's stormwater drainage system capital improvement project (CIP) program, (e.g., identify, select, cost,
and prioritize stormwater system construction projects.)
. Establish a maintenance program for the stormwater system (recommended stormwater system maintenance
practices and frequencies)
• Establish onsite conveyance requirements (design standards for level of peak flow conveyance by an engineered
stormwater system and, sometimes, requirements for street conveyance of stormwater beyond the onsite
requirements)
Master plans seldom included requirements for development with regard to stormwater system impacts (e.g.,
downstream flow and/or water quality impacts). Master plans were sometimes utilized to assess potential future problems
as well as to fix existing problems. Often systems were evaluated under current conditions and future planned zoning
to be able to assess costs to current rate/tax payers or new developments. Because master plans were not usually
completed prior to some significant level of development, attributing these costs was important to the development
community as well as to the residents.
135
-------
The traditional approach to stormwater master planning has been to focus on hydrology and hydraulics of the existing
stormwater systems, and proposed larger trunk systems to determine whether there is enough capacity. This is usually
is accomplished by the following steps:
. Route a designated large storm through system, assume worst case conditions (saturated, etc.) and determine
capacity deficiencies
• Develop an enlarged (or more efficient) system to handle larger flows or, when necessary, reduce peak flows by
detention (if the cost of detention is less than a conveyance upgrade)
. Sometimes consider water quality as a "add-on" (e.g., if detention is required, claim a water quality benefit)
This approach has certainly significantly reduced property damage (sometimes only for short-term), but has led to
more damage in streams. The damage has been a result of a significant increase and duration in small storm runoff
flows. The result of not planning for this increased energy, which is primarily contained within the stream channel, has
often been an increase in maintenance and property damage. For example, channel cutting that occurs upstream of
culverts often causes headwall and culvert failures. In other areas where channel cut sediments settle out (often in over-
designed or poorly designed culverts), areas are filled in with sediments. When this occurs (especially in a culvert), it can
lead to flooding. These problems (headwall failures, culverts filled in, etc.) are often called maintenance issues, when
they are in fact really failures of the master plan to adequately address stream impacts of development.
Typically smaller urban stormwater systems (e.g., 10 to 50 acre catchments) are dominated from a flooding
standpoint by shorter-duration, more-intense storms (thunderstorms), whereas, the larger urban watersheds are often
impacted by larger, but less-intense storms of longer duration. Master plans typically utilize a single large design storm
event based upon a rainfall depth (mm of rain over a watershed) for a specified duration and return period. This depth
is then assigned a conservative shape such as the SCS type IA shape shown in Figure 3). The storm shown is the 25-
year, 24-hour storm depth for Eugene, Oregon, with the SCS distribution applied to it. As an example of how overly
conservative the peak of
25-Year SCS Type lASynthetic Design Storm
Totd Rartai Volume = 5.28 in
Max Hourly Ranai Intensity = Q9 irVhr
Time (hi)
Figure 3. 25-Year, 24-Hour SCS Type 1A Synthetic Design Storm for Eugene, Oregon.
136
-------
the "design" hydrograph is, Figure 4 shows an actual 25-year storm hydrograph (based upon analysis of the Eugene
Airport rain gage). This storm was confirmed by long-term simulation modeling to have caused approximately the 25-year
return-period flows in the larger stormwater systems in the city. In reality, the 25-year return-period storm depth seldom
if ever arrives with the peaky "shape" given it in master plans.
National Weather Service Rainfall Data for February 5-8,1996
Total Rainfall Volume = 7.3 in
Max. Rainfall Intensity = 0.66 in/hr
Time (hours}
Figure 4. Rainfall Event that was Considered to Cause the Approximately 25-Year Return Period Peak Runoff Flows in Eugene, Oregon.
Many have justified this shape as being one that will also allow flood control effects of smaller thunderstorms on the
smaller stormwater systems to be adequately evaluated. When the peak is modeled in this fashion on a larger watershed
during an already large rainfall, the peak may greatly affect the larger system design. This conservative design approach
we believe has led many communities to determine that streams are undersized and must be widened, channeled, and/or
piped.
Of course in communities where there is the potential for combined phenomenon to cause severe flooding (e.g.,
snowmelt and frozen ground combined with a hard rain), there may be good reason to over size facilities. However, in
most cases, it may be more appropriate to utilize methods that account for this and to strive to preserve open channels
in more natural ways (e.g., larger stream buffers) to the extent possible.
Another assumption that is often made is that the watershed is saturated before the design storm arrives. This
assumption is made to be "conservative." However, it results in an uneven level of conservatism. This assumption would
tend to lead to the most over-designed conveyance systems in the least paved areas. That is, the saturation assumption
would tend to make systems most over-designed in low-density, single-family areas vs. less over-designed in the
downtown core area. The point here is that the levels of over-design are not consistent, nor targeted to the areas where
the greatest level of protection is desired (highest property value).
137
-------
Finally, water quality, and sometimes habitat, is only now being considered in master planning. This is most often
accomplished by adding water quality to a detention feature or specifically selecting and locating several demonstration
water quality projects. Some communities have chosen to not emphasize habitat by engineering their streams with the
purpose of providing flood conveyance as recreational amenities. The Denver area is a good example of this type of
design. In arid areas, where streams are seasonal or even just storm driven, this may be a good choice for communities.
However, some communities are considering the value of seasonal streams play for downstream resources from a
biological and water quality perspective. For example in Eugene, Oregon, the city has determined that seasonal streams
contain a rich fauna of aquatic invertebrates (WCC, 1995) which likely would benefit the health of downstream systems.
There are a number of reasons why the above approaches have continued to be employed. First, planning that
considers multiple objectives is much more difficult to accomplish, from the technical approaches, due to the need to
involve more parties in decision making. The traditional technical approach described above is straightforward, while
design of more natural systems is not (e.g., pipe flow equations are much easier to utilize then open channel flow in
natural streams). In addition, there are many more people to involve in making decisions than dealing only with
engineered physical structures within the stormwater system. Second, most municipalities are not organized well for the
purpose of urban watershed planning. The City of Portland, Oregon (which has been very progressive in many ways)
still has four separate departments (all in one bureau) that do: 1) facilities planning (stormwater system master planning),
2) site stormwater standards, 3) stormwater quality (permit compliance), and 4) watershed management. Each of these
groups has developed its own plans and programs that have understandably not been very well-coordinated or integrated.
Finally, and probably most important, is that integrated planning studies cost significantly more (on the order of 2 to 4
times as much).
Integrated Storm Drainage Master Plans - Approach
The new approach to stormwater master plans is the integration of flood control, water quality, natural resources, and
aesthetics of stormwater systems. This approach requires significantly more effort and should be thought of as one that
will entail adaptive management. That is, the master plan must include components that allow for changing conditions
as development occurs and the downstream systems react.
In completing a stormwater master plan, it is difficult to achieve "maximums" of flood control, water quality, natural
aquatic habitat, and aesthetics. It is somewhat analogous to the rule that it is hard to get a cheap price, good service,
and high quality. It is our belief that one of the problems with master plans has been a lack of recognition that streams
will change and that the plans should be developed to manage change in a positive fashion.
One of the keys to successful integrated master planning is that the planning approach places the proper emphasis
on the technical and decision-making processes employed. As mentioned above, master plans typically have been driven
by the hydrologic/hydraulic modeling of large storm(s) and usually begin with model data collection and analysis. Figure
5 presents a suggested flow diagram for an alternative way of sequencing the development of a master plan. It begins
by conducting an inventory of all aspects of the stormwater system, including all attributes related to the multiple
objectives mentioned above. The approach suggests utilizing multi-disciplinary teams to review conditions in the field to
look for opportunities for meeting objectives, as well as reviewing existing and suspected future problems. Next, before
any modeling is done, the project team and decision-makers should utilize the collected information to develop goals and
objectives for the plan. Then additional technical analyses, including where and what type of detailed hydrologic/hydraulic
modeling is appropriate, can be decided upon based upon these objectives. We have found this approach sharpens the
focus of modeling so that the model is not "driving" the master plan into solutions that focus primarily on conveyance
upgrades.
In developing an integrated master plan, it is generally understood that the right mix of multi-disciplinary technical
specialists should be involved. In addition, it is important to involve the "right" decision-makers and stakeholders early
in the process. It is also important to agree up-front upon the decision-making process that will be utilized. We have
found that utilizing an agreed upon set of factors to evaluate, select, and rank projects is very useful not only for guiding
the process more objectively, but also to serve as a history of why certain projects were recommended and why others
were not. This is very useful for future decision-makers for two reasons. First, when questioned by others, there will be
138
-------
, Flooding
Water Quality
Natural
Resources
NPDES
Natural Resources
Permitting
Problem/
Opportunity
Identification
Team
Evaluate
Flooding
Water Quality
Natural Resources
Alternatives
Evaluation and
Selection
Surface Water
Master Plan/CIP
Program
- updated
inventory of
problems and
opportunities
- problem
definition
modeling
- gather/assess
information
- goals, objectives, . fina| criteria
and criteria
options for
resolving problems
- options for
enhancements
- apply criteria
to alternatives
- selected Capitol
Improvement
Projects (CIPs)
- Multiple and
single objective
solutions
Figure 5. Suggested Integrated Stormwater Master Plan Project Approach.
some backing for why certain decisions were made. Second, as conditions that affect the factors change, selection of
projects can also change in a logical fashion.
The approach we recommend is to evaluate solutions that are primarily single objective rather than those that are
multiple objective. That is, a two-stage decision process is employed to make sure that good single-objective solutions
are not ignored because of the multi-objective nature of the factors. The factors employed include:
. Addresses flooding problems
• Addresses water quality pollutants of concern
. Meets community amenity objectives
• Habitat value
. Life-cycle costs
Meets regulatory requirements
Implementability
Reliability/sustainability
Other environmental impacts
Equability
Integrated Storm Drainage Master Plans - Hydrology
Integrated stormwater master planning includes evaluating and considering smaller storm hydrological impacts.
Figure 6 presents a storm-depth frequency curve for Portland, Oregon. The figure demonstrates that storms of a depth
of 1.5 inches and less dominate both the number of storms (more than 95%) and the volume of runoff (over 90%). It is
the smaller storms of about 0.3 to 0.8 inches in depth that change the most in their characteristics. In natural areas of
the Northwest, these often did not result in appreciable runoff or resulted only in slightly elevated flows for a long duration.
However, after urbanization, these storms are causing severe and rapid changes in flow levels with each storm. This kind
of analysis can be used to assist decision makers in deciding what level of water quantity and water quality control is
going to be the most cost-effective in reducing the impacts of urbanization.
The best hydrologic and hydraulic modeling approach for assessing and designing stormwater systems is likely the
use of continuous simulation models using long-term rainfall records to evaluate a system under a wide range of varying
hydrologic conditions. However, this is quite expensive. One of the approaches that we have been taking is to utilize
long-term simulations of stormwater systems to select design storms. We believe that this improves the consistency in
providing design storms that are closer to the level of protection that is being "advertised," without having to run long-term
simulations. This approach involves using real rainfall data with continuous simulation models (e.g., SWMM) to define
the resulting return frequency of runoff peaks in various parts of the stormwater system. Then, real storms are selected
139
-------
Portland Rainfall Statistics
o
.C1.5
Q)
3
0>
00.5
E
a
V)
95% of ev<
90*/o of
70%of,
Fits
0% 10% 20% 30% 40% 50% 60% 70%
% of annual runoff volume
80%
9 0 % 1C?
Figure 6. Cumulative Storm Event Rainfall Depth Analysis for Portland, OR Airport
that resulted in the return period of interest (using a partial duration frequency analysis). These "real" storms are then
utilized to design the system. Figure 7 presents an example of partial-duration frequency evaluation of peak flows in one
of the basins in the Eugene, Oregon area. From this frequency distribution, the storm that was closest to the intended
design level (25-year) was selected for design analysis of the system. Figure 8 shows a similar analysis for another basin
in Eugene, along with the design flows from an earlier master plan (which utilized the traditional SCS storm method with
saturated conditions).
What Figure 8 demonstrates is that in this basin, the more traditional approach would have resulted in what is likely
a significant over-design of the system. In most basins, this was found to be the case. However, there were several
basins that were close and a few where the real storm approach resulted in larger designs. Figure 9 and 10 compare
the resulting designs in the Flat Creek basin in Eugene. Note that the real storm approach resulted in fewer and smaller
projects in this basin. This means that the city can utilize more of its scarce resources to complete other types of multi-
objective projects. One of the advantages of the use of real storms is that the concept is very easy to communicate to
citizens. In addition, the city has found that some of its channels are over-designed compared to the stated level of
protection, and that they may be able to relax vegetation maintenance requirements to allow for more natural channels.
Overall, the city is finding that allocating sufficient resources to conduct an integrated plan will likely lead to a more cost-
effective program overall, in terms of multiple benefits.
Integrated Storm Drainage Master Plans - Water Quality
There are a number of stormwater quality models and approaches (Donigian and Huber, 1991). Some are quite
simple and straightforward, while others are much more complex. In general, water quality models currently cannot
accurately predict how pollutants get into stormwater. Although some researchers have made great strides in establishing
sources of pollutants in the urban environment (Pitt, 1993), there still are numerous pollutant sources that are not fully
understood. Most models rely on either some land-use-based concentrations to drive water quality predictions or they
use a build-up/wash-off function to describe pollutant concentrations (Donigian and Huber, 1991).
140
-------
Exanpte Frequency Distribution of Peak Flows
Bethel Danebo Basinette (Area=687 Acres)
1000
Upper 95% Orf tierce Lint
Special Fbak Row (cfs)
Lover 95% Cbrfidence Lint
10
FHlFNPBKr)(yr)
Figure 7. Example Frequency Distribution of Peak Flows in Eugene, OR.
1000-,
PEAK FLOW (cfs)
5 8
Peak Row Comparison
Santa Clara - River Road Basinette (Area = 301 Acres)
«
hr^r- •
-_ r~* j
)
^A
L J J .1
,
<-"^
-
1 •
~>->
— .^ . ^
(
-j^i^^^TTTT
. , ,
. — X
*- * ^" _
_
TT^
_ i. J J J
, . , .
'T^Tvr
• i i i
, , , ,
, , , ,
, , , ,
^«-ny e-j • •
Lpper 95% Confidence umt
— • — Bcpected Fbak Row (cfs)
, Lower 95% Confidence Unit
_-*-- 1990 Master Ran
1 10 100
RETURN PEROD(yr)
Figure 8. Peak Flow Comparison in Urban Runoff from Eugene, OR.
141
-------
Flat Creek Drainage System
Capital Improvements Proposed in 1990 Master Plan
Using lo-year SCS Synthetic Design Storm
City of Eugene
Open Channel
— Pipe or Cukert
• Transition Between Section
CIP Proposed in 1990 Master Plan
® 30" RCCP replaced by 4'x3' RCBC
® open channel replaced by 2-36" RCCP
® 30" RCCP replaced by 2-3'x3' RCBC
0 existing channel expansion
(S) existing channel expansion
® existing channel expansion
® 3-43"x27" CMP replaced by 2-6'x3' RCBC
® 2-36" RCCP replaced by 2-4'x3' RCBC
® 58"x3B" and 24" CMP replaced by 2-5'x3' RCBC
Total
tU r^
£y \
rJ14h,,f™-l/^Sl ' \
Enlarged Area
Cost
$27,300
$175,400
$31,800
$9,100
(3,900
121,000
$87,100
$81,100
$553,800 c; n
i iguit ;
Figure 9. Proposed Conveyance System Improvements Utilizing the SCS Type 1A Synthetic Design Storm and Assuming Saturated Conditions.
The build-up/wash-off of suspended solids (TSS) is modeled and then TSS concentrations are utilized to predict other
concentrations for such parameters as phosphorus and heavy metals. The first problem with this approach is that it
assumes that the build-up/wash-off of TSS is much greater than any other source pathway. This has not been found to
be the case (Pitt, 1993). When build-up based/wash-off models are calibrated to real data, the build-up/wash-off function
must be set to be much larger than it really is in order to match actual data. When a source control such as street
sweeping is applied, the model will then significantly overestimate its effectiveness no matter what the assumed street
sweepingefficiency is. This may explain why street sweeping has seldom if ever been found to be as effective as
predicted. The second problem with these models is the assumption that other constituent concentrations can be related
142
-------
Flat Creek Drainage System
Capital Improvements Proposed by Woodward-Clyde
Using lo-year Design Storm of November 23, 1960
City of Eugene
Enlarged Area
37cfs f s
Open Channel
Pipe or Culvert
Transition Between Section
-------
There have been a number of attempts to develop a better understanding of how stormwater quality BMPs work and
why (Strecker, 1992; Brown and Schueler, 1997). However, what we know about the effectiveness of stormwater best
management practices in improving water quality and ultimately aquatic health has been questioned with good reason
(Strecker 1994; Urbonas 1995; Maxted and Shaver, 1996). Some of the questions arise from the actual studies and how
they have not been completed as to be very useful in assessing effectiveness. In addition, there have been suggestions
that pollutant removal efficiencies may not be the best way to assess effectiveness (irreducible concentrations, etc.).
Finally, there have been some studies that have shown that downstream of some BMPs (e.g., detention systems),
aquatic invertebrate populations are no different from systems that do not have such in-stream ponds (Maxted and
Shaver, 1996).
What we know is the application of BMPs is an evolving science and that the exact cause and effect relationships
are not well known. However, we do know that BMPs have been effective at reducing concentrations. In cases where
there has been no downstream improvement in aquatic invertebrate health from BMPs, we should ascertain what the
limiting factors are and whether the BMP was able to mitigate some if not all of them before we dismiss a BMP. In
addition, we need to understand whether other attributes of the BMP may be contributing to downstream problems such
as demonstrated downstream temperature impacts (Galli, 1991) of on-line ponds, as well as the interruption of drift of
aquatic invertebrates downstream. It is becoming increasingly clear that within-stream detention systems need to be
very carefully evaluated before they are selected as BMPs. What we will need to do in master planning is to make good
subjective decisions regarding the appropriate application of BMPs for water quality. We do not have the data and models
to do otherwise.
Integrated Storm Drainage Master Plans - Stream Stability/Habitat
Unless a watershed has a great ability to infiltrate stormwater or evaporation is a viable technique, stream hydrology
will change (increased runoff) with development. While there are some great techniques to reduce the changes (e.g.,
Prince Georges Department of Environmental Resources. 1997), in many cases these techniques will not be able to
reduce the increased energy within a stream enough to stop channel cutting and downstream sedimentation from
occurring. A technique that has been employed in an attempt to prevent downstream damage is the requirement that
new development controls runoff from a one- or two-year event such that pre- and post- development peak flows for that
event are equaled. MacRae (1996) has demonstrated that this approach may actually cause more problems then it
solves. It usually leads to shifting over-bank flow energy to the wetted channel, further exacerbating channel down
cutting. Figure 1 demonstrates this. Suppose that the peak in hour 14 was the one-year pre-development flow for this
creek. Maintaining post-development flows to this level would significantly lengthen the time the creek is subject to this
channel-forming flow condition, while reducing over-bank flows. Even setting post-development peak runoff rates to one-
half pre-development, results in significant extended energy in the channel. One would likely have to set a requirement
that the flow rate be one-fourth or one-fifth to have a positive effect. This would require very large detention areas.
In many, if not most cases, we believe that the master plan must include within-stream structures to assist it in
changing with development (Severn, 1996). That is, the plan must move beyond just getting runoff to a stream and
making sure any culverts in the stream are "right-sized." Master plans should include a component to design in-stream
structures (habitat friendly ones, of course) and have an adaptive management program for them. This approach has
been successfully applied to the Pipers Creek and Thorton Creek watersheds in Seattle, both heavily urbanized
watersheds. What this can accomplish is much faster and more postive equilibrium for the stream system (e.g., the
increased energy can be utilized to create deeper pools and increased spawning gravels in the pool tailway).
Integrated Storm Drainage Master Plans - Capital Improve Projects (CIPs)
The above integrated stormwater master planning elements will result in changing the traditional definition of what
a CIP is. Traditionally it has been structural controls located within the municipally owned stormwater systems (e.g., the
streets and street drainage structures and at creek crossings, etc.) Now CIPs can include property or property rights
acquisition, buffer areas, protection and enhancement of natural resource sites and preservation of the open channel
drainage system.
144
-------
Integrated Storm Drainage Master Plans - Public vs. Private Solutions
Another element of master planning can include the evaluation of the trade-off of requiring private solutions (e.g.,
on-site design requirements) versus implementing public stormwater system measures. Figure 11 shows schematically
that on a watershed basis, one could employ a combination of both to achieve the overall most cost-effective system.
This can be addressed in modeling and cost-estimation for both approaches and then one or some combination
employed.
w
o
o
+*
v
o
O)
n
-------
Metropolitan Washington Council of Governments. Maryland Department of Environment. Washington D.C. 188pp.
MacRae, C. 1996. Experience from Morphological Research on Canadian Streams: Is Control of the Two-year
Frequency Event the Best Basis for Stream Channel Protection? In: Effects of Watershed Development and
Management on Aquatic Ecosystems. American Society of Civil Engineers. Edited by Larry Roesner. Snowbird, Utah.
Pp 144-1 62.
May, C.R., R. Horner, J. Karr, B. Mar, and E. Welsh. 1997. Effects of Urbanization on Small Streams In the Puget Sound
Lowland Ecoregion. In: Watershed Protection Techniques, 2(4): 483-494.
Maxted, J., and E. Shaver. 1996. The Use of Retention Basins to Mitigate Stormwater Impacts on Aquatic Life. In: Effects
of Watershed Development and Management on Aquatic Ecosystems. American Society of Civil Engineers. Edited by
Larry Roesner. Snowbird, Utah. pp. 494-512.
Prince Georges Department of Environmental Resources. 1997. Low Impact Development Design Manual. Department
of Environmental Resources, Prince Georges County, Maryland.
Pitt, R. 1993. Source loading and management model (SLAMM). National Conference on Urban Runoff Management,
March 30-April 2, 1993. Chicago, IL.
Schueler, T. 1994. The Importance of Imperviousness. In: Watershed Protection Techniques, l(3): 100-1 11.
Severn, D., and P. Washington. 1996. Effects of Urban Growth on Stream Habitat. IN: Effects of Watershed
Development and Management on Aquatic Ecosystems. American Society of Civil Engineers. Edited by Larry Roesner.
Snowbird, Utah. pp. 163-1 77
Strecker, E., 1994. Constituents and Methods for Assessing BMPs. Proceedings of the Engineering Foundation Conference
on Stormwater Related Monitoring Needs. Aug. 7-12, Crested Butte, Colorado. ASCE.
Strecker, E.W., Kersnar, J.M., Driscoll, E.D., and Horner, R.R. 1992. The Use of Wetlands for Controlling Storm Water
Pollution. The Terrene Institute. Washington, D.C.
Strecker, E., M. lannelli, and B. Wu. 1997. Analysis of Oregon Urban Runoff Water Quality Monitoring Data Collected from
1990 to 1996. Prepared by Woodward-Clyde for the Association of Clean Water Agencies.
Urbonas, B.R., 1995. "Recommended Parameters to Report with BMP Monitoring Data. J. Water Resources Planning and
Management, ASCE. 121(1), 23-34.
Woodward-Clyde, 1997. Willow Creek Physical, Biological, and Chemical Baseline Report. Prepared by Woodward-
Clyde, for the City of Eugene.
146
-------
Conservation Design: Managing Stormwater through Maximizing Preventive
Nonstructural Practices
Wesley R. Horner
Environmental Management Center
Brandywine Conservancy
Chadds Ford, Pennsylvania
Abstract
Unlike conventional methods of stormwater management that prioritize peak rate control to mitigate post-development
downstream flooding effects, Conservation Design first aims to prevent or minimize the creation of stormwater from the
outset. Preventive Conservation Design methods are defined in this paper as those that integrate stormwater
management into the initial stages of project design, instead of waiting to consider them in the final steps of the site
planning process. Mitigative Conservation Design techniques will be explored that use natural processes performed by
vegetation and soil to mitigate unavoidable stormwater runoff impacts once prevention has been maximized to the
greatest extent possible. Underlying these techniques-whether preventive or mitigative in nature-is a comprehensive
perspective of water resources that views stormwater as an asset to be managed, not a waste for disposal.
This paper summarizes a recent project which the Brandywine Conservancy undertook for the Delaware Department
of Natural Resources and Environmental Control, with support from USEPA Section 319 funding. For interested readers,
Conservation Design for Stormwater Management: A Design Approach to Reduce Stormwater Impacts from land
Development (Delaware Department of Natural Resources and Environmental Control with Brandywine Conservancy,
1997) further details all aspects of the Conservation Design program described here. This manual is referenced
throughout this paper and is available by contacting DNREC at 302-739-4411 in Dover DE.
Introduction
Most Stormwater management programs place a heavy reliance on implementation of structural stormwater
managementfacilities: detention basins, conveyance piping and inlet/outlet structures. These facilities-though created
to mitigate negative stormwater impacts by controlling flooding-cannot in and of themselves eliminate adverse impacts
of urban development throughout a watershed. In fact, because these systems fail to acknowledge and plan for critical
system-wide water cycle processes, stormwater management itself can become a problem, rather than a solution. This
is especially true when conventional stormwater management systems are combined with conventional large-lot
subdivision designs.
The negative effects of this type of development and conventional stormwater management have been described in
a variety of recent studies and reports, including the Pennsylvania Handbook of Best Management Practices for
Developing Areas (CH2MHJII, 1998) and a variety of other state stormwater manuals; Center for Watershed Protection
publications such as Better Site Design: A Handbook for Changing Development Rules in Your Community (Center for
Watershed Protection, 1998) and Planning for Urban Stream Protection (Schueler, 1995); the Northeastern Illinois
Planning Commission's Reducing the Impacts of Urban Runoff: The Advantages of Alternative Site Design Approaches
(Northeastern Illinois Planning Commission, 1997), and Urban Stormwater Best Management Practices for Northeastern
Illinois (Northeastern Illinois Planning Commission, 1993). These effects include:
. Altered site hydrology and reduced groundwater recharge
. Reduced stream base flows
. Altered stream geomorphology (resulting in damaged aquatic habitat)
147
-------
Loss of site area for other uses (e.g.; recreation)
• Single purpose: disregards site resource conservation benefits
. Lack of attention to water quality
. High construction costs
. Maintenance burdens and costs
. Negative visual appearance (e.g., basins often fenced off)
. Limited number of stormwater discharge points
• Less flexibility in design
Conservation Design reflects a totally different philosophy toward land development that integrates stormwater
management into the very core of site design, as opposed to considering it a problem to be resolved after the design has
been completed. This philosophy regards stormwater as a key component of the hydrologic cycle and critical to
maintaining the water balance-and groundwater reserves-for a particular watershed.
Recently we have come to realize that land development's impacts to water resources are not one-dimensional. They
include, in addition to flooding, the multiple concerns of water quality, groundwater quantity, stream and wetland
characteristics, in-stream habitat, and biodiversity. Therefore, stormwater management and site design must be
approached much more comprehensively. At the foundation of this comprehensive approach lies an understanding of
the relationship between land development and our water resources. In order to better comprehend this relationship, we
must understand the water cycle itself-the amount of rainfall, evapo-transpiration, groundwater infiltration, and
runoff-and how this cycle is affected by the characteristics of an individual site such as soil types, topography, and
vegetation.
The Water Cycle and Landscape Dynamics
Appreciation of the water cycle is especially important to achieve successful, comprehensive stormwater management
(Figure 1). In fact, only through understanding full water cycle dynamics, can we hope to achieve some sort of system
balance and minimize negative stormwater impacts. Figure 2 displays a generic flow chart of the water cycle that
highlights the various components of this cycle and how they are interconnected (Conservation Design for Stormwater
Management, 1997). It is important to appreciate that the system itself is a closed loop: what goes in, must come out.
If inputs to infiltration are decreased by 10 inches, then inputs to surface runoff and/or depression storage must be
increased by this same amount. Furthermore, infiltration outputs must also be decreased: following along on the flow
diagram, the groundwater reservoir, evapo-transpiration and soil moisture elements together will be reduced by this 10
inches, which will reduce stream baseflows.
The logical first step in any discussion of the water cycle is precipitation-h all its various forms. In southeast
Pennsylvania, and indeed throughout much of the Mid-Atlantic states, the climate is relatively humid (Conservation
Design for Stormwater Management 1997, based on Hydrosphere 1992 database). Substantial precipitation tends
to be distributed throughout the year in frequent events of modest size. This consistency in rainfall throughout the year
indicates that this region does not have a defined wet or dry season as do other areas of the country. This rainfall
potential throughout the year has significant implications for consideration of stormwater runoff. For example, having
rainfall throughout the year indicates that sediment laden runoff can occur at any time; therefore, it is important to
establish some sort of erosion-controlling groundcover during all seasons of the year.
Also important is the distribution of rainfall by size of event. Based on analysis of 35 years of data from a Wilmington,
Delaware rain gage (Conservation Design for Stormwater Management 1997), it is clear that the precipitation occurs
mostly in small "events" or storm intensities. Ninety-eight percent of the total number of events during this extended
period were classified in the "less than 2 inches" category. Even more important from a water cycle perspective, 96%
of the average annual rainfall volume occurred in storms of less than 3 inches (which is less than the 2-year, 24 hour
148
-------
Shallow Infiltration
Deep Infiltration
Groundwater discharge to lakes, streams and oceans
Figure 1. The Water Cycle.
storm). This understanding of storm size distribution is critical for a variety of reasons in stormwater management. For
example, if our concern is keeping the water cycle in relative balance, capturing and recharging the 1 - or 2-year storm
as the basis for design will encompass the vast bulk of precipitation and stormwater runoff volumes in the average year
and provide adequate water cycle balance. This leads to very different design criteria than if flooding (peak runoff rates)
is the only concern addressed.
Another key component of the water cycle is the linkage between stormwater infiltration, groundwater recharge and
stream baseflow. As land is developed and impervious coverage increased, less water is recharged to groundwater
aquifers (Thomas Dunne and Luna Leopold's Wafer in Environmental Planning [Dunne and Luna, 1978] is an
excellent background text in addition to the above referenced reports). As these subtractions continue acre-by-
acre, development-by-development, their cumulative effects grow larger. Also, as development occurs, more water is
often withdrawn from the underground reserves for drinking, irrigation, or commercial uses, As subtractions are made
from the groundwater reservoir flow, the impact will be seen in the form of a lowered water table and reduced stream
baseflow discharge. Headwater springs and first-order streams-the lifeblood of our stream systems-may even dry up.
The baseflow from headwater zones is critical to maintaining a diversity of aquatic plant and animal life, as well as
terrestrial animals dependent on certain aquatic species for survival. In some cases the groundwater reservoir does not
discharge to a stream, but ratherto a wetland. In these instances, reduced infiltration and a lowered water table ultimately
translate into a loss of wetlands themselves, and an elimination of their rich and vibrant ecological function.
A final component of the water cycle that must be addressed is overland runoff. This is the component most
frequently addressed in conventional stormwater management approaches, for it is the cause of increased downstream
flooding. Three major elements determine the volume and character of stormwater runoff for a given storm intensity: soil
type, land cover (including vegetation and debris), and slopes. Soils vary widely in their ability to infiltrate stormwater and
149
-------
Figure 2. Water Cycle System Flow Chart.
minimize runoff and are classified accordingly by the USDA Natural Resources Conservation Service (NRCS) into four
categories based on their permeability rates (Hydrologic Soil Groups A through D, with A having best permeability).
Land cover greatly affects the rate and volume of stormwater runoff and has significant water quality impacts as well.
Obviously, the landcover of greatest concern for stormwater management is impervious coverage created through the
development process. Interestingly, compacted lawns and cultivated fields can have significant runoff rates as well,
especially when no crop covers the bare soil. The landcover in this region best suited to retard stormwater runoff and
assist in its infiltration is the natural one: the piedmont forest. A mature forest can absorb much more water than an
equivalent area of turf grass due to the presence of an organic litter layer and herbaceous and woody plant material. The
organic litter layer on the forest floor provides a physical barrier to sediments, maintains surface soil porosity, and assists
in denitrification and other water quality functions. The vegetation, both herbaceous and woody, physically retards runoff
and erosion with its spreading root mats and also assists in maintaining soil permeability and water quality by taking up
nutrients through its root systems.
Finally, slopes are another critical component of the stormwater runoff equation. Steeper slopescan accelerate runoff
and increase the erosive force of the water. Therefore, removing vegetation on steeper slopes can have dramatic impacts
on downslope aquatic systems.
150
-------
As seen above, the water cycle and the implications for stormwater management are complex and comprehensive.
The process of urbanization dramatically impacts the functioning of this water cycle. Conservation Design has been
developed to address the issues of comprehensive stormwater management and to address the land use patterns that
impact it.
Land Use and Site Development Impacts
Throughout much of the United States, farmland and natural areas are converted to suburban development at an ever
accelerating pace. In fact there is hardly a city in America that does not occupy at least two to three times more land area
than in 1970, even if population has not increased proportionately. This history of land use change is certainly true of the
Mid-Atlantic states, where communities continue to grapple with the effects of unmitigated suburban sprawl.
The dynamic nature of wet-weather flow regimes and landscape ecology make it difficult to assess the impact of
urbanization on aspects of the water cycle such as groundwater reserves and aquatic habitat. However, studies have
indicated that the biological community in urban streams isfundamentallychanged to a lower ecological quality than what
was there before development occurred. In one study in Delaware, approximately 70% of the macroinvertebrate
community found in streams of undeveloped forested watersheds was comprised of pollution sensitive mayflies,
stoneflies, and caddisflies, compared to 20% for urbanized watersheds (Maxted and Shaver 1996). Other studies suggest
that the decay in stream quality is very rapid in the early stages of watershed urbanization; watersheds with less than 10%
impervious cover are the most susceptible to the adverse effects of urbanization. Therefore early intervention as a
watershed begins to develop is critical, and furthermore, this intervention should include measures to address stormwater
management and land use in a connected, comprehensive manner.
In addition to in-stream habitat impacts, the issue of land development and water resources also has great
implications for our human communities well beyond the issue of flooding. Reduced stream baseflows and groundwater
resources means decreased availability of drinking water supplies. Also, reduced baseflows result in less available water
for diluting the pollution output from industrial or municipal waste systems. As stormwater runoff increases, water quality
can be greatly impacted by stream bank erosion, resuspension of sediment, runoff of chemicals and fertilizers from lawns
and fields, and increased stream temperatures. Stormwater-linked pollutants vary with type of land use and intensity of
use and have been shown to include bacteria, suspended solids, nutrients, hydrocarbons, metals, herbicides and
pesticides, toxins and organic matter. Not only are these pollutants increased, but the landscape's natural capacity for
filtering and chemical uptake through vegetation is decreased as land is cleared and paved. All of these pollutants can
impact both drinking water supplies and natural aquatic systems.
Thus it becomes evident that if the negative effects of land development on our water resources are to be minimized,
we must find alternatives to the conventional structural approach to stormwater management. Moreover, these
alternatives must address the issue of land use and patterns of development in a comprehensive fashion, one that strives
to maintain a hydrologic balance on site and replicate the pre-development hydrologic regime to the greatest extent
possible. One approach-or collection of approaches-that can accomplish these goals is Conservation Design.
Conservation Design Principles
Stormwater management throughout the Commonwealth (and elsewhere) can be markedly improved by approaching
stormwater differently than has been the practice in the past, where "stormwater management" has been defined largely
as stormwater disposal. This different perspective challenges us to maximize prevention, even before stormwater
becomes a problem, and to avoid highly engineered structural solutions that are expensive to build and maintain. In their
place, Conservation Design focuses on utilization of natural systems and processes to achieve stormwater management
objectives where feasible. At the same time, this new approach is intended to work with site resources-woodlands, soils,
wetlands, etc.-to enhance their stormwater functions. The end result is a site design which minimizes stormwater
generation and then mitigates the remaining stormwater in a low-impact manner, with an emphasis on groundwater
recharge. Conservation Design is not so much a singular approach or solution as it is a collection of approaches and
practices that are flexible enough to effectively address any given site and development program. Common to all these
approaches and practices are several basic principles.
151
-------
Achieve multiple objectives. Stormwater management should be comprehensive in scope, with techniques
designed to achieve multiple stormwater objectives. These objectives include both peak rate and total volume control
(i.e., balance with the hydrologic cycle), as well as water quality control and temperature maintenance. These objectives
should include maintaining or improving the pre-development hydrologic regime.
Integrate stormwater management early into the site design process. Stormwater management tacked on at
the end of the site design process almost invariably is flawed. To optimize comprehensive stormwater management
objectives, stormwater management must be integrated into the first stages of the site planning. Stormwater impacts may
even be a factor in determining type of use, extent of use, and location of the development on a site.
Prevent first, mitigate second. Approaches to site design which can reduce stormwater generation from the outset
are the most effective approach to stormwater management. For example, effective clustering of units significantly
reduces length of roads when compared to conventional development. Reduction in street width and driveway length can
minimize impervious coverage. These type of approaches are rarely thought of as stormwater management practices,
yet they achieve powerful stormwater quality and quantity benefits.
Manage stormwater as close to the source of generation as possible. From both an environmental and
economic perspective, redirecting runoff back into the ground as close to the point of origin as possible, is preferable to
constructing elaborate conveyance systems that increase flows and suffer from failures over time. Avoid concentrating
stormwater. Disconnect, rather than connect, where feasible.
Engage natural processes in soil mantle and plant communities. The soil mantle offers critical groundwater
recharge conveyance and pollutant removal functions through physical filtration, biological action, and chemical
processing. Understanding how much of what type of soil is in place on any given site is essential when assessing
stormwater management/water quality impacts and opportunities. Vegetation similarly provides substantial pollutant
uptake/removal potential and can assist in infiltration by maintaining soil porosity and retarding runoff. In addition,
naturally vegetated areas improve their stormwater functions over time as leaf litter and debris builds a richer organic soil
layer. Areas of good soil permeabilities (A and B soils) and intact vegetative communities should be prioritized in
prevention strategies.
A Conservation Design Procedure
The Conservation Design principles outlined above, though greatly simplified, can offer valuable guidance when
approaching a particular land development project. In fact, these five principles form the basis for a Conservation Design
Procedure. This Design Procedure incorporates both Preventive Approaches and Mitigative Practices. Preventive
Approaches tend to be broader in geographic scope than other techniques and typically may influence some of the major
decisions regarding a particular development project. Approaches may even transcend the site itself, involving an entire
planning jurisdiction or area, or even an entire region. Also, Preventive Approaches attempt to reduce impervious
coverage or minimally disturb the existing vegetation and soils in prime recharge areas. For example, a reduction in road
width from 30 feet to 18 feet means an immediate 33.3% reduction in roadway imperviousness, which typically comprises
a large portion of site imperviousness.
Mitigative Practices include mitigative techniques which are often more structural in nature. These practices
encompass a rapidly growing array of biofiltration and bioretention methods that maximize the stormwater management
potential of soils and vegetation. Mitigative Practices include vegetated swales for stormwater conveyance, vegetated
filter strips and riparian buffers, grading, berming, terrraforming, and level spreading stormwater in natural areas. These
practices should mitigate as close to the source as possible and achieve multiple objectives. For example, a berm, which
is used to retain stormwater runoff on a forested slope, can double as a walking trail, thus decreasing the expense of two
separate individual systems.
Figure 3 graphically displays the Design Procedure as a flow diagram. The procedure itself can be thought of as a
series of questions which must be asked as Conservation Design is applied to each site. If site designers rigorously
address all of these questions, the "answers"-the Conservation DesignPreventive Approaches and Mitigative
152
-------
Conservation Design Procedure
Prevention
Building
Program
Lot
Configuration
Impervious
Coverage
Minimum
Disturbance
Conceptual
Stormwater
Management
Plan
Calculations
(peak and
volume)
Sizing
of
Controls
Figure 3. A Conservation Design Procedure.
153
-------
Practices-will successfully be identified for each site. The overriding objective ultimately is to achieve a new way of
thinking about site design. The procedure begins with an effective and complete Site Analysis, which can help identify
both areas of concern and resourcesforopportunity in regard to stormwater management. The procedure then flows from
macro, larger-scale preventive questions (i.e., how can the design be clustered to reduce site disturbance) to micro, small-
scale mitigative questions (i.e., can stormwater be infiltrated in bioretention areas?). Probably the most important aspect
of the procedure in Figure 3 is its positioning of the Conceptual Stormwater Management Plan as a concurrent task with
the entire site design process. This reinforces the notion that stormwater management should be an integral part of the
entire design process, including the site analysis.
In order to better understand the Conservation Design Procedure, each of its components (the Preventive Approaches
and Mitigative Practices) is discussed in more detail below.
Site Analysis
Three major aspects need to be addressed in the Site Analysis process:
Site Background and Context
What is the surrounding context?
What is its location in the watershed?
In which geologic/geographic region is it located?
What is the site size?
What are adjacent uses and landcover?
Critical Natural Features
Existing hydrology?
Wetlands? Floodplains? Riparian buffers?
Steep slopes? Special habitat areas?
Stormwater Opportunity Areas
Where are soils that are best suited for stormwater recharge? Worst?
Where is existing landcover optimal to prevent stormwater?
What opportunities exist to use vegetation and soils in mitigation?
On what soils and slopes is this vegetation?
What is depth to bedrock or water table?
Preventive Approaches
The Preventive Approaches include a range of hierarchical questioning:
Building Program
What is the current zoning and density for this tract?
154
-------
Is there currently an open space design option for the site?
Can the proposed building program be reduced in terms of density?
Can the type of unit or lot size be modified to promote open space?
What are the possibilities for water and sewer supply?
Lot Configuration
Have lots been reduced and open space been maximized?
Have lots been clustered to avoid critical areas of recharge?
Have lots been configured to take advantage of mitigative practices?
Impervious Coverage
Has development been clustered to reduce impervious surfaces?
Have road widths been minimized?
Have building setbacks been minimized to reduce driveway lengths?
Have parking ratios and needs been carefully examined?
Have needs and sizes of walkways been examined?
Minimum Disturbance
Has maximum total site area, including soils and vegetation, been protected from clearing and
disturbance?
Are zones of undisturbed open space maximized?
Have buildings been sited carefully to reduce vegetation removal?
Can no-disturbance buffers be installed to limit zones of soil compaction?
Mitigative Practices
The Mitigative Practices include a tool box of options that promote groundwater recharge and improve water
quality. These practices have been assigned to several groupings, although in many cases the practicesoverlap.
Virtually all of these techniques make maximum use of vegetation and soil functions, so although they are all
technically structures, they are of lower complexity and more rooted in natural process than conventional
approaches.
Vegetated Swales
Vegetated swales are effective means of stormwater conveyance. At low slopes, they can recharge
modest amounts of stormwater, filter it through vegetative processes, and slow it down.
155
-------
Terraforming
Terraforming comes in a variety of techniques. These include constructing subtle berms along contour
below undisturbed areas. The berms act as modest "dams" retaining the water for up-slope recharge.
Also, subtle grading of depression areas promote retention and recharge throughout a site.
Level Spreading/Natural Areas
With a level spreader, stormwater spills over the lip of a long trench or berm, creating sheet flow across
a broad area. The level spreaders slow down the intensity of runoff and discharge it over a large,
adjoining vegetated area with good soils, which in turn filter it and assist in groundwater recharge. Filter
strips are planted vegetated strips through which runoff passes that filter it and slow it down. Riparian
buffers are vegetated zones along stream corridors that filter the stormwater passing through it and help
minimize erosion. These techniques are most valuable when used in conjunction with preventive
strategies that leave larger natural areas undisturbed in order to handle these additional stormwater
inputs.
B ioretention/b iof i Itration
Bioretention is a popular name given to just about any type of device that utilizes vegetation and soil to
manage stormwater flows. They can be subtle depressions that exist naturally and receive stormwater
or depending on soil conditions, they may be physically constructed "pits" that are filled with permeable
soils and planted with native vegetation that adapt to both wet and dry conditions. These systems can
either be "on-line" (part of the stormwater conveyance flow) or "off-line" (separate from the rest of the
stormwater management/mitigation system). In either case, they have modest ponding storage that is
recharged over the course of time.
Other mitigative devices
Not all of the required volume storage to meet peak rate requirements for a given site may be attained
through the practices outlined above. At times, it may be necessary to put in "structural" systems such
as in-ground infiltration trenches, infiltration pipes, or stormwater wetlands. However, these systems
should be explored only after both Preventive Approaches and Mitigative Practices of Conservation
Design have been maximized to the greatest extent possible.
Conclusion
The Conservation Design Procedure is perhaps best characterized as a "check list" or protocol of questioning during
the site design process. The key to this approach is its range of innovative, yet effective options, not afforded in
conventional systems which tend to be standardized irrespective of the particular site. With Conservation Design, the
approaches and practices can be combined in a variety of ways to minimize the impacts of development on the water
cycle and still meet regulatory stormwater management criteria such as peak rate control. Often, because these
approaches and practices tend to favor multiple objectives and nonstructural techniques, Conservation Design can be
less expensive to install and maintain than conventional systems. Also, because they are largely based on soil and
vegetative processes, conservation design techniques tend to improve in function overtime, while conventional detention
basin systems tend to diminish in function over time. In terms of water quality, Conservation Design Approaches and
Practices can outperform conventional systems. For example, filter strips and biofiltration areas can remove over 90%
of the suspended solids, 40% of the phosphorous, and 20% of the nitrates (Dillaha et al. 1986 and 1989; Yu et al. 1993).
In addition, reduced yard areas and increased forested zones prevent chemical runoff from lawns-a great contributor
to non-point source pollution-at the outset.
Conservation Design is limited only by the creativity of the designer and the flexibility of the developer and regulatory
agencies. It must be emphasized that the Conservation Design approach will not eliminate a need for structural systems
in all cases; however, more often than not, Conservation Design can replace or reduce the need for structural practices
156
-------
while providing attractive site amenities. And in the process, the water cycle will be balanced, and forests and other
sensitive resources will be preserved. In short, Conservation Design can do more with less, and more for less, than
conventional approaches to stormwater management.
References
Center for Watershed Protection, 1998; Better Site Design: A Handbook for Changing Development Rules in Your
Community; Elliot City, MD.
CH2MHNI, 1998; for the Pennsylvania Association of Conservation Districts, Inc., and other agencies; Pennsylvania
Handbook of Best Management Practices for Developing Areas; Harrisburg, PA.
Delaware Dept. of Natural Resources and Environmental Control with the Brandywine Conservancy, 1997; Conservation
Design for Stormwater Management: A Design Approach to Reduce Stormwater Impacts from Land Development; Dover,
DE.
Dillaha, T.A., J.H. Sherrard, and D. Lee, 1986; Long-term Effectiveness and Maintenance of Vegetative Filter Strips;
Virginia Water Resources Research Center Bulletin No. 153; Blacksburg, VA.
Dunne, Thomas and Luna Leopold, 1978; Water in Environmental Planning; W.H. Freeman and Company; New York,
NY.
Northeastern Illinois Planning Commission, 1993; Urban Stormwater Best Management Practices for Northeastern Illinois;
Chicago, IL
Northeastern Illinois Planning Commission, 1997; Reducing the Impacts of Urban Runoff: The Advantages of Alternative
Site Design Approaches; Chicago, IL.
Schueler, Tom, 1995; Center for Watershed Protection; Site Planning for Urban Stream Protection; Silver Spring, MD
(relocated to Ellicott City).
Yu, S., S. Barnes, and V. Gerde; 1993; Testing of Best Management Practices for Controlling Highway Runoff; Virginia
Transportation Research Coundil; FHWA/VA 93-R16.
157
-------
Low-Impact Development Design: A New Paradigm for Stormwater
Management Mimicking and Restoring the Natural Hydrologic Regime
An Alternative Stormwater Management Technology
Larry S. Coffman, Associate Director
Department of Environmental Resources
Prince George's County, Maryland
Abstract
Whether complying with federal or state regulations or addressing local vital watershed protection/restoration
objectives, local jurisdictions are confronted with the daunting task of developing, administering and funding complex
effective multi-objective Stormwater management programs, Today's comprehensive Stormwater program not only has
to deal with runoff quantity and quality control but, may also have to address such complicated issues as ecosystem
restoration, combined sewer overflow reduction, fisheries protection, potable surface/ground water source protection, and
wetland, riparian buffer and stream protection. As our understanding of the technical and practical limitations of
conventional Stormwater management technology has increased over the past two decades, and as watershed protection
objectives have changed, many jurisdictions have begun to question the efficacy and cost-effectiveness of conventional
Stormwater approaches in meeting today's complex environmental/water resources objectives. Older communities with
existing extensive Stormwater management infrastructures are also struggling with the economic reality of funding the
high costs of maintenance, inspection, enforcement and public outreach necessary to support an expanding and aging
infrastructure. Still more challenging are the exceptionally high costs of retrofitting existing urban development using
conventional Stormwater management end-of-pipe practices to restore and protect receiving waters and living resources.
With growing concerns about the limitations of conventional technology and to address the changing objectives of
watershed protection, in 1990 Prince George's County's Department of Environmental Resources (PGDER) began
exploring alternative Stormwater management practices and strategies. The development of bioretention or "Rain
Gardens" (using the green space to manage runoff within small depressed landscaped areas) led to an understanding
of how to optimize and engineer the landscape to restore hydrologic functions by uniformly integrating micro-scale
management practices and impact-minimization measures into the development landscape. In 1997 PGDER released
the Low Impact Development (LID) Design Manual demonstrating the principles and practices of LID to create a
hydrologically functional landscape (PGDER, 1997).
LID Stormwater management technology can maintain or restore a watershed's hydrologic regime by fundamentally
changing conventional site design to create an environmentally and hydrologically functional landscape that mimics natural
hydrologic functions (volume, frequency, recharge and discharge). This is accomplished in four ways. First: minimizing
impacts to the extent practicable by reducing imperviousness, conserving natural resources and ecosystems, maintaining
natural drainage courses, and reducing the use of pipes and minimizing clearing/grading. Second: recreating detention
and retention storage dispersed and evenly distributed throughout a site with the use of open swales, flatter slopes,
depression storage, rain gardens (bioretention), water use (rain barrels), etc. Third: maintaining the predevelopment"time
of concentration" by strategically routing flows to maintain travel time. Fourth: providing effective public education and
socioeconomic incentives to ensure property owners use effective pollution prevention measures and maintain
management measures. With LID, every site feature is multifunctional (green space, landscaping, grading, streetscapes,
roads and parking lots) and helps to reduce Stormwater impacts or provide/maintain beneficial hydrologic functions. The
cumulative beneficial impact of using the wide array of distributed LID techniques allows the site designer to maintain or
restore watershed's natural relationship between rainfall, runoff, infiltration and evaporation.
The effective use of LID site design techniques can significantly reduce the cost of providing Stormwater management.
Savings are achieved by eliminating the use of Stormwater management ponds, using less pipe, inlet structures, curbs
158
-------
and gutters, less roadway paving, less grading and clearing. Where LID techniques are applicable, and depending on the
type of development and site constraints, stormwater and site development design construction and maintenance costs
can be reduced by 25 % to 30% compared to conventional approaches.
The creation of LID's wide array of micro-scale management principles and practices has led to the development of
new tools to retrofit existing urban development. Micro-scale management practices that filter, retain and detain runoff
can be easily integrated into the existing green space and streetscapes as part of the routine maintenance and repair of
urban infrastructure. LID retrofit techniques may lead to drastic reductions in the cost of retrofitting existing urban
development. Reducing urban retrofit costs will increase the ability of cities to implement effective retrofit programs to
reduce the frequency and improve the quality of CSOs and improve the quality of urban runoff to protect receiving waters.
LID represents a radically different approach to controlling stormwater runoff that provides effective tools to restore or
maintain a watershed's hydrologic functions for new or existing development.
In 1998 EPA provided grant funding to assist PGDER in their efforts to develop a general manual describing LID
principles and practices, and share this technology with other local governments throughout the nation. Efforts are
currently underway with EPA to further advance LID technology by improving the sensitivity of current hydrology and
hydraulic analytical models for application with small watersheds and sites and to develop new micro-scale control
approaches and practices for urban retrofit. Additional efforts are also underway to demonstrate how LID micro-scale
management and multifunctional infrastructure principles and practices can be used to control highway runoff within
existing rights-of-way. It is hoped that the LID national manual will help to stimulate debate on the state of current
stormwater, watershed protection and restoration technology and its future direction. The lessons learned about LID
planning, principles, practices and research are described in detail in the reference documents listed at the end of this
paper. Copies of these reference documents can be obtained by calling the Prince George's County's Department of
Environmental Resources at (301) 883-5832.
Background
Typically, adverse stormwater impacts are mitigated through conservation of natural resources (forests, streams,
floodplains and wetlands); zoning restrictions to direct densities and increase open space; and the use of structural or
non-structural control technologies (best management practices - BMP's) to treat and manage runoff quantity and quality.
Many conventional stormwater mitigation approaches, such as management ponds, exhibit a number of inherent practical,
environmental and economic limitations including inability to replicate predevelopment watershed hydrology, elevated
water temperatures, costly maintenance burdens, and accelerated stream erosion due to the increased duration and
frequency of runoff events. Furthermore, because current mitigation practices only lessen development impacts, there
is concern about the cumulative impacts of the widespread use of conventional mitigation practices that may
fundamentally alter a watershed's hydrologic regime and water quality, adversely affecting receiving waters and the
integrity of their ecosystems. Many highly urbanized jurisdictions are beginning to question the efficacy of current
technology and are finding it harder to ensure, enforce or fund stormwater programs and maintain the massive
infrastructure created by conventional approaches.
Currently every site is designed with one basic overriding goal - to achieve good drainage. As we develop a site
reshaping the landscape inch by inch, its hydrologicfunctionsare altered on a micro-scale level. The cumulative impacts
of micro-scale changes to the landscape drastically alter watershed hydrology. If sites can be designed to achieve good
drainage, destroying natural hydrologic functions, why not design sites with the opposite objective to maintain
predevelopment hydrologic functions? If inch by inch, sites are carefully and intelligently engineered to maintain hydrologic
functions, would the cumulative beneficial affects result in the preservation of a watershed's hydrology? Can a site be
designed in a way to remain as a functional part of a watershed's hydrological regime? To achieve a hydrologically
functional development there must be a radical change in our thinking. We must not think in terms of impact mitigation
as the stormwater management objective, but rather preservation of hydrologic and environmental functions. We should
design sites to maintain hydrologic functions not just to mitigate impacts. Can our current stormwater management
technology adequately meet our regulatory objectives and water resources/ecosystem protection needs? No one can
answer that question for sure. However, it has not been shown that conventional ponds replicate predevelopment
hydrology nor is there any evidence to suggest that conventional technology can ensure the ecological integrity of
159
-------
ecosystems. In fact, recent studies suggest that conventional approaches can not meet our water/natural resources and
ecological objectives.
Introduction
With growing concerns about the economicsand efficacy of conventional technology, in 1990 Prince George's County
Maryland's Department of Environmental Resources began exploring alternative stormwater management practices. The
success that was achieved through the development and use of bioretention (filtering or infiltration runoff in small
depressed landscaped areas) led us to understand that perhaps changing the form and function of the developed
landscape could be important in mitigating urban stormwater impacts. Later it was realized that through intelligent site
design and uniform distribution of LID micro-scale management controls it was possible to maintain or restore hydrologic
functions in a developed watershed. What is not known is how much of a watershed's hydrologic functions can be
maintained or restored within a given development type (residential, commercial or industrial)? The one limiting factor to
maintaining/restoring the hydrologic regime for highly urbanized development is the lack of available micro-management
tools. Much of the current research underway is to expand the number of practices applicable in highly urbanized areas.
LID's objective is to preserve the natural predevelopment hydrologic regime. If predevelopment hydrology and water
quality can be maintained, this would provide the best level of protection possible to receiving waters and aquatic living
resources. Experience over the last 20 years has demonstrated that maximizing the efficiency of conventional
conservation measures and the use of conventional end-of-pipe stormwater management practices can not reasonably
be used to restore watershed functions. What is needed is a new philosophical approach to site development, an
approach that will allow the designer to retain a site's hydrologic functions.
The approach used in LID designs is really an old one. LID borrows its basic principles from nature - uniform
distribution of micro-management controls. In a natural setting, stormwater is controlled by a variety of mechanisms
(interception by vegetation, small depression storage, channel storage, infiltration and evaporation) uniformly distributed
throughout the landscape. LID mimics these mechanisms by uniformly distributing small infiltration, storage, and retention
and detention measures throughout the developed landscape. What we soon began to see is that every development
feature (green space, landscaping, grading, streetscapes, roads, and parking lots) can be designed to provide some type
of beneficial hydrologic function.
Low - Impact Development General
LID controls stormwater at the source creating a hydrologically functional landscape that mimics natural watershed
hydrology. Low impact development (LID) achieves stormwater management controls by fundamentally changing
conventional site design to create an environmentally functional landscape that mimics natural watershed hydrologic
functions (volume, frequency, recharge and discharge). LID uses four basic management planning and design principles.
First: minimize impacts to the extent practicable by reducing imperviousness, conserving natural resources/ecosystems,
maintaining natural drainage courses, reducing use of pipes and minimizing clearing and grading. Second: provide runoff
storage measures dispersed uniformly throughout the landscape with the use of a variety of small decentralized detention,
retention and runoff practices such as bioretention, open swales and flatter grades. Third: maintain the predevelopment
time of concentration by strategically routing flows to maintain travel time and control discharge. Fourth: implement
effective public education and incentive programs to encourage property owners to use pollution prevention measures
and maintain on-lot landscape management practices. A developed site can be designed to become a hydrologically
functional part of the watershed with comprehensive and intelligent use of LID practices and principles.
LID Basic Site Planning Strategies
The goal of LID is to design the site in a way that mimics hydrologic functions. The first step is to minimize the
generation of runoff (reduce the change in the runoff curve number (CM)). In many respects, this step is very similar to
traditional techniques of maximizing natural resource conservation, limiting disturbance and reducing impervious areas.
The major difference with LID is you must carefully consider how best to make use of the hydrologic soil groups and site
topography to help reduce and control runoff. These considerations would include how to:
160
-------
1. maintain natural drainage patterns, topography and depressions,
2. preserve as much existing vegetation as possible in pervious soils; hydrologic soil groups A and B,
3. locate BMP's in pervious soils; hydrologic soil groups A and B,
4. where feasible construct impervious areas on less pervious soil groups C and D,
5. disconnect impervious surfaces,
6. direct and disburse runoff to soil groups A and B,
7. flatten slopes within cleared areas to facilitate on-lot storage and infiltration and
8. re-vegetate cleared and graded areas.
Where ground water recharge is particularly important (to protect well, spring, stream and wetland flows) it is
important to understand the source and mechanisms for ground water recharge. When using the LID design concepts
to mimic the hydrologic regime you must determine how and where ground water on the site is recharged and where
necessary, protect and utilize the recharge areas in the site.
LID Hydrologic Analysis/Response
The objective of LID site design is to minimize, detain and retain the post development runoff volumes uniformly
throughout the site close to the source to simulate predevelopment hydrologic functions. Widespread use and uniform
dispersion of on-lot small retention and/or detention practices to control both runoff discharge volume and rate is key to
better replicating predevelopment hydrology. Using LID practices also produces runoff frequencies that are much closer
to existing conditions than can be achieved by typical application of conventional BMP's. Management of both runoff
volume and peak runoff rate is included in the design. This is in contrast to conventional end-of-pipe treatment that
completely alters the watershed hydrology regime.
The LID site analysis and design approach focuses on four major hydrologically based planning elements. These
fundamental factors affect hydrologic and are introduced below.
1. Curve Number (CN) - A factor that accounts for the effects of soils and land cover on amount of runoff generated.
Minimizing the change in the post development CN by reducing impervious areas and preserving more trees and
meadows to reduce runoff storage requirements, all to maintain the predevelopment runoff volume.
2. Time of Concentration (Tc) - This is related to the time runoff travels through the watershed. Maintaining the
predevelopment Tc reduces peak runoff rates after development by lengthening flow paths and reducing the use
of pipe conveyance systems.
3. Permanent storage areas (Retention) - Retention storage is needed for volume and peak control, water quality
control and to maintain the same CN as the predevelopment condition.
4. Temporary storage areas (Detention) - Detention storage may be needed to maintain the peak runoff rate and/or
prevent flooding.
Minimizing the Change in CN
Reducing the change in CN will reduce both the post development peak discharge rate and volume. Calculation of
the LID CN is based on a detailed evaluation of the existing and proposed land cover so that an accurate representation
of the potential for runoff can be obtained. This calculation requires the engineer/planner to investigate the following key
parameters associated with LID including:
161
-------
1. land cover type,
2. percentage of and connectivity of impervious cover,
3. hydrologic soils group (HSG), and
4. hydrologic conditions (average moisture or runoff conditions).
The following are some of the LID site planning practices that can be utilized to achieve a substantial reduction in the
change of the calculated CN:
1. narrower driveways and roads (minimizing impervious areas),
2. maximized tree preservation and/or afforestation,
3. site finger-printing (carefully siting lots/roadways to avoid disturbance of streams, wetlands and other resources),
greater use of open drainage swales,
4. preservation of soils with high infiltration rates to reduce CN,
5. location of BMP's on high-infiltration soils and,
6. construction of impervious features on soils with low infiltration rates.
Maintaining the Predevelopment Time of Concentration Tc
The LID hydrologic evaluation requires that the post development Tc be close to the predevelopment Tc. Minimizing
the change in pre and post Tc will help maintain the same frequency of runoff discharges, assuming there is uniform
distributed micro-scale retention and detention of LID practices. The following are some of the site planning techniques
can be used to maintain the existing Tc:
1. maintain predevelopment flow path length by dispersing and redirecting flows using open swales and vegetated
drainage patterns,
2. increase surface roughness (e.g., preserving woodlands, vegetated swales),
3. detain flows (e.g., open swales, rain gardens, rain barrels etc.),
4. minimize disturbances (minimizing soil compaction and changes to existing vegetation /drainage patterns),
5. flatten grades in impacted areas,
6. disconnect impervious areas (e.g., eliminating curb/gutter and redirecting down spouts) and,
7. connect pervious areas to vegetated areas (e.g., reforestation, afforestation).
The combined use of all these techniques results in cumulative impacts that modify runoff characteristics to effectively
shift the post development peak runoff time and frequencies to that of the predevelopment condition, and lower the peak
runoff rate.
Maintaining the Redevelopment Curve Number and Runoff Volume
Once the post development Tc is maintained at the predevelopment conditions and the impact of CN is minimized,
any additional reductions in runoff volume must be accomplished through distributed micro-scale on-site stormwater
management techniques. The goal is to select the appropriate combination of management techniques that simulate the
hydrologic functions of the predevelopment condition to maintain the existing CN and corresponding runoff volume. The
target design volume is equal to the initial abstraction of rainfall that would have occurred in the predevelopment condition.
162
-------
LID site designs maximize the use of small retention practices distributed throughout the site at the source to provide the
required volume storage. The required storage volume will be reduced when the change in the pre and post CN is
minimized.
Retention storage allows for a reduction in the post development volume and the peak runoff rate. The increased
storage and infiltration capacity of retention LID BMP's allow the predevelopment volume to be maintained. The most
appropriate retention BMP's include:
1. bioretention cells (rain gardens),
2. infiltration trenches,
3. water use storage (rain barrels and gray water uses) and,
4. roof top storage.
Other possible retention BMP's include retention ponds, cisterns and irrigation ponds but it may be difficult to distribute
these types of controls throughout a development site.
As retention storage volume is increased there is a corresponding decrease in the peak runoff rate, in addition to
runoff volume reduction. If a sufficient amount of runoff is stored, the peak runoff rate may be reduced to a level at or
below the predevelopment runoff rate. This storage may be all that is necessary to control the peak runoff rate when there
is a small change in CN. However, when there is a large change in CN, it may be less practical to achieve flow control
using volume control only.
Potential Requirement for Additional Detention Storage
In cases where very large changes in CN cannot be avoided, retention storage practices alone may be either
insufficient to maintain the predevelopment runoff volume or peakdischarge rates or require too much space to represent
a viable solution. In these cases, additional detention storage will be needed to maintain the predevelopment peak runoff
rates. A number of traditional detention storage techniques are available that can be integrated into the site planning and
design process for a LID site. These techniques include:
1. swales with check dams, restricted drainage pipes, and inlet/entrance controls,
2. wide, low gradient swales,
3. rain barrels/cisterns,
4. rooftop storage and
5. shallow parking lot/road storage.
Determination of Design Storm Event
The hydrologic approach of LID is to retain the same amount of rainfall within the development site as was retained
prior to any development (e.g., woods or meadow in good condition) and then release runoff as the woods or meadow
would have. By doing so, it is possible to mimic, to the greatest extent practical, the predevelopment hydrologic regime
to maximize protection of receiving waters, aquatic ecosystems and ground water recharge. This approach allows the
determination of a design storm volume that is tailored to the unique soils, vegetation and topographic characteristics of
the watershed. This approach is particularly important in watersheds that are critical for ground water recharge to protect
stream/wetland base flow and ground or surface water supplies.
163
-------
LID BMP's
Site design techniques and BMP's can be organized into three major categories as follows; 1) runoff prevention
measures designed to minimize impacts and changes in predevelopment CN and Tc, 2) retention facilities that store runoff
for infiltration, exfiltration or evaporation and 3) detention facilities that temporarily store runoff and release through a
measured outlet. Table 1, below, lists some of a wide array of LID BMP's and their primary functions. Placing these BMP's
in series and uniformly dispersing them throughout the site provides the maximum benefits for hydrologic controls.
Table 1. Examples of LID BMP's and Primary Functions
BMP
Bioretention
Infiltration Trench
Dry Wells
Roof Top Storage
Vegetative Filter Strips
Rain Barrels
Swale and Small Culverts
Swales
Infiltration Swale
Reduce Imperviousness
Strategic Clearing / Grading
Engineered Landscape
Eliminate Curb and Gutter
Vegetative Buffers
Runoff
Prevention
X
X
X
X
X
Detention
X
X
X
X
X
X
Retention
X
X
X
X
Conveyance
X
X
X
Water
Quality
X
X
X
X
X
X
X
X
X
Water Quality
LID maximizes the use of the developed landscape to treat stormwater runoff. Not only can the landscape be used
to store, infiltrate and detain runoff, the unique physical, chemical and biological pollutant removal/
transformation/immobilization/detoxification capabilities of the soil, soil microbes and plants can be used to remove
pollutants from runoff. For example, bioretention basins or rain gardens are designed to use the upland soil/microbe/plant
complex to remove pollutants from runoff. Rain gardens which look and function like any other garden except they treat
runoff are designed with a layer of 2-3 inches of mulch, 2-3 feet of planting soil and vegetation (trees shrubs and flowers).
Figure 1 shows a parking lot landscape island rain garden (bioretention practice) that uses a high rate filter media with
plants to filter and treat 90% of the annual volume of runoff from the parking lot.
Figure 1. Parking Lot Rain Garden.
164
-------
Studies conducted by the University of Maryland have shown rain gardens to be very effective in removing pollutants.
The percent pollutant removal of various contaminants is shown below in Table 2. The results shown represent the
average removal rates under a wide variety of flow rates and pollutant concentrations.
Table 2. Percent Pollutant Removal by Rain Gardens
cu Pb Zn P TKN NH4+ NO3- TN*
93 99 99 81 68 79 23 43
. Removal varied as a function of depth in the soil. Percent removal shown is at a depth of approximately 3 feet.
Testing Conducted by the University of Maryland, Department of Engineering
The variety of physical, chemical and biological pollutant removal mechanisms available in the complex rain garden
system is staggering. A description or explanation in any detail of these mechanisms is beyond the scope of this paper.
A more detailed description can be found in the 1998 "Optimization of Bioretention Design" study conducted by the
University of Maryland. Mulch has been found to be very effective in removing heavy metals through organic complexing
with the hydroxyl and carboxyl sites on the organic molecules. Soil bacteria can metabolize (use as a carbon energy
source) oil, grease and gasoline into C02 and water in the presence of adequate nutrients and oxygen. Soil bacteria have
been used for years for the remediation of contaminated soils. Plants are known to uptake, transpire, accumulate and
detoxify heavy metals and many other toxic compounds. The physiologic and metabolic processes of plants are used to
clean contaminated soils through phytoremediation. A goal of LID is to maximize the use of upland landscape with its soil/
microbes/ plant complex to treat runoff. Using upland systems to trap and remove pollutants allows one to more easily
control the fate of contaminants and prevent them from entering the water column where they are almost impossible to
contain and remove.
Public Outreach and Pollution Prevention
Pollution prevention and maintenance of on-lot LID BMP's are two key elements in a comprehensive approach.
Effective pollution prevention measures can reduce the introduction of pollutants to the environment and extend the life
of LID treatment BMP's. Public education is essential to successful pollution prevention and BMP maintenance. Not only
will effective public education complement and enhance BMP effectiveness, it can also be used as a marketing tool to
attract environmentally conscious buyers, promote citizen stewardship, awareness and participation in environmental
protection programs and help to build a greater sense of community based on common environmental objectives and the
unique character of LID designs.
Costs
LID case studies and pilot programs show that at least a 25% reduction in both site development and maintenance
costs can be achieved by reducing grading and the use of pipes, ponds, curbs and paving. In one subdivision called
Somerset which used the rain garden LID technique for water quality controls, the developer saved $4,500 per lot or a
total of $900,000 by eliminating the need for curbs, ponds and drainage structures. Maintenance costs are also reduced
in scale and magnitude by using the small LID practices. LID site designs require only routine landscape care and
maintenance of the vegetation, This eliminates the high costs of pond maintenance associated with dam repairs and
dredging.
Road Blocks to LID
In the development and acceptance of the LID site planning approach, a number of roadblocks had to be overcome.
Regulating agencies, the development community and the public all had concerns about the use of new technology. The
165
-------
LID design manual represents the culmination of four years of work to address all of these concerns and issues. Some
of the major components of the LID approach, which addressed the many concerns, include:
1. develop an hydrologic analytical methodology to demonstrates the equivalence of LID to conventional approaches,
2. develop new road standards which allow for narrow roads, open drainage and cluster techniques,
3. streamline the review process for innovative LID designs which allow easy modification of site, subdivision, road
and stormwater requirements,
4. develop a public education process which informs property owners on how to prevent pollution and maintain on
lot BMP,
5. develop legal and educational mechanisms to ensure BMP's are maintained,
6. demonstrate the marketability of green development,
7. demonstrate the cost benefits of the LID approach,
8. provide training for regulators, consultants, public and political leaders and,
9. conduct research to demonstrate the effectiveness of bioretention BMP's.
Summary
LID is a viable economically sustainable alternative approach to stormwater management and the protection of natural
resources. LID provides tangible incentives to a developer to save natural areas and reduce stormwater and roadway
infrastructure costs. LID can achieve greater natural conservation by using conservation as a stormwater BMP to reduce
the change in CN. As more natural areas are saved, less runoff is generated and stormwater management costs are
reduced. This allows multiple use and benefits (environmental and economical) of the resource.
Additionally, developers have incentives to reduce infrastructure costs by reducing impervious areas, and eliminating
curbs/gutters and stormwater ponds to achieve LID stormwater controls. Reduction of the infrastructure also reduces
infrastructure maintenance burdens making LID designs more economically sustainable. Superior protection of aquatic
and riparian ecosystems can be achieved since a LID developed watershed functions in a hydrologically similar manner
as the predevelopment conditions. Recreating the predevelopment hydrological regime is a better way to protect the
receiving waters than the conventional end-of-pipe mitigation approaches.
LID promotes public awareness, education and participation in environmental protection. As every property owner's
landscape functions as part of the watershed, they must be educated on the benefits and the need for maintenance of
the landscape and pollution prevention. LID developments can be designed in a very environmentally sensitive manner
to protect streams, wetlands, forest habitat, save energy, etc. The unique character of a LID green development can
create a greater sense of community pride based on environmental stewardship.
References
Prince George's County, Maryland. Department of Environmental Resources. Low-Impact Development Design Manual.
(PGC, 1997)
Prince George's County, Maryland. Department of Environmental Resources. Low-Impact Development Guidance
Manual. (PGC, 1997).
Prince George's County, Maryland. Design Manual for use of Bioretention in stormwater Management. (PGC, 1993).
166
-------
Prince George's County, Maryland. Optimization of Bioretention Design for WaterQualityand HydrologicCharacteristics.
(Davis, et al. 1998).
167
-------
A National Menu of BMPs for the Phase II NPDES Storm Water Program
James H. Collins
Tetra Tech, Incorporated
Fairfax, VA
John A. Kosco, P.E.
USEPA/Off ice of Wastewater Management
Washington, DC
Introduction
Implementation of the Phase II NPDES Storm Water Program, which is presently "proposed," will be enhanced by
the development of several tools used to help reduce discharges of pollutants from regulated small construction activities
and from regulated small municipal separate storm sewer systems (MS4s). One of the tools being developed by EPA
is a national menu of best management practices (BMPs), from which regulated Phase II municipalities can select as they
develop their stormwater management programs.' The purpose of the national menu of BMPs is to provide a list of
options available to regulated Phase II municipalities as they develop a stormwater programs. The national menu of
BMPs may be adopted or modified by each NPDES permitting authority, or the permitting authority may develop its own
menu of BMPs for use by the Phase II municipalities under its jurisdiction. This paper describes the process of developing
the national menu of BMPs and measurable goals for each of the six minimum measures required to be in the stormwater
management programs.
Process of Development
The process used in developing the menu was to first list appropriate BMPs for each minimum control measure, with
subcategories under certain control measures.' Then, a basicformat for presenting the information about each BMP was
established. Information being provided for each BMP in the menu consists of BMP name, description, an illustration,
applicability and design considerations, limitations, operation and maintenance, effectiveness, cost, and references. The
menu is being prepared by EPA, with support from Tetra Tech, Incorporated, and the Center for Watershed Protection.
A peer involvement/peer review group has been selected and will provide review and input to the process of developing
the menu over the course of the next year. The menu is currently being reviewed and developed as a traditional hard
copy document. Following development, there are plans to make the menu available as an interactive Web-based tool.
The menu of BMPs is scheduled to be released by October, 2000.
Descriptions of Six Minimum Control Measures, with Lists of BMPs
1) Public Education and Outreach
This measure in the proposed rule calls for the creation of a public education program to inform citizens about the
impacts that stormwater runoff can have on water quality. It includes the preparation and distribution of educational
materials to the community, describing these impacts and steps that can be taken to reduce pollution from discharges
of stormwater. Examples of such steps include proper septic system maintenance, limitations on use and runoff of
household and garden chemicals, proper disposal of used motor oil or household hazardous wastes, and involvement
in local stream restoration activities. The following BMPs have been identified (in the Draft Rule) under four major
subcategory groups in the menu for the Public Education and Outreach minimum measure:
Public outreach/education for homeowners
. Lawn and garden activities, including proper pesticide use and disposal practices
168
-------
. Water conservation practices
. Proper disposal of used materials or household hazardous wastes
. Pet waste management
Targeting public outreach/education
• Education/outreach for commercial activities (parking lots, gas stations, etc.)
. Tailoring outreach programs to communities, including minority and disadvantaged, as well as children
. Classroom education on stormwater
. Distributing stormwater educational materials (how, to whom)
Public Outreach Programs for New Development
. Low-impact development (includes buyer awareness, legal documents, and settlement documents)
Pollution Prevention Programs for Existing Development
. Educational display, pamphlet, booklet, utility stuffer
. Using the media (includes newspaper, magazine, radio, television, public service announcements, and Internet
messages)
. Promotional giveaway
2) Public Participa tion/lnvolvemen t
This measure in the proposed rule includes compliance with state and local public notice requirements, but goes
beyond that to encourage municipalities to seek public involvement in the development and review of their stormwater
programs. Opportunities for members of the public to participate in the development of their municipality's stormwater
management program may include serving as citizen representatives on a local stormwater management panel, attending
public hearings, working as citizen volunteers to educate others about the program, assisting in program coordination with
other pre-existing programs, or participating in volunteer monitoring efforts. The following BMPs have been identified
under two major subcategory groups of practices in the menu for the Public Participation/Involvement minimum measure:
Activities/Public Participation
. Storm drain stenciling
. Stream cleanup
. Volunteer monitoring
. Reforestation program
. Wetland plantings
. Adopt-A-Stream program
Involvement/Public Opinion
. Watershed organization
169
-------
. Stakeholder meetings (includes local stormwater management ponds)
. Attitude surveys
. Community hotlines
3) Illicit Discharge Detection and Elimination
This measure envisions the creation of an illicit discharge detection and elimination program. Specific program
elements include developing a demonstrated knowledge base of the MS4, using maps or other documents to identify
major outfalls and pipe networks on a topographic basis; developing a plan to address illicit discharges into the MS4,
including appropriate enforcement procedures to the extent allowable by law; and developing a process for informing the
public about the hazards associated with illicit discharges and the improper disposal of waste. For example, recycling
programs and other public outreach activities could be developed to address sources of illicit discharges, including used
motor oil, antifreeze, pesticides, herbicides, and fertilizers. The following BMPs have been identified for the menu for the
Illicit Discharge Detection and Elimination minimum measure:
. Failing septic systems
. Industrial connections
• Recreational sewage
. Sanitary sewer overflows
. Identifying illicit connections
. Wastewater connections to the storm drain system
4) Construction Runoff Control
This measure provides for the enforcement of a program to reduce pollutants in storm water runoff from construction
activities resulting in the disturbance of one acre to five acres of land. The program would apply to the individuals
responsible for activities at construction sites and should include an ordinance to control sediment and erosion; a
mechanism to ensure control of other wastes at construction sites, such as discarded building materials, concrete truck
washout, and sanitary waste that could impact water quality; requirements for the implementation of appropriate BMPs,
such as silt fences, temporary detention ponds and hay bales; provisions for preconstruction review of site management
plans; procedures for receipt and consideration of comments and other information provided by the public; regular
inspections during construction; and penalties to ensure compliance. The following BMPs have been identified under 11
major subcategory groups of practices in the menu for the Construction Runoff Control minimum measure:
Minimize Clearing
. Land grading
. Permanent diversion
. Preservation of natural vegetation (includes tree preservation and protection)
Stabilize Exposed Soils
. Chemical stabilization
. Mulching
170
-------
. Permanent seeding
. Sodding
. Soil roughening
Protect Steep Slopes
. Geotextiles
. Gradient terraces
. Soil retention (includes slope stabilization, retaining wall, reinforcement)
. Temporary slope drain (a.k.a. - pipe slope drain)
. Temporary storm drain diversion
Stabilize Drainage Ways
. Check dam (a.k.a. grade stabilization structure)
. Filter berm
. Grass-lined channel
. Riprap
Protect Waterways
. Temporary diversion
. Temporary stream crossing (bridge, culvert)
. Vegetated buffer
Phase Construction
. Construction sequencing
. Dust control
Install Perimeter Controls
. Temporary diversion dikes, earth dikes, and interceptor dikes (includes temporary fill diversions)
. Sand fence and wind fences
. Silt fence
. Brush barriers
Install Sediment Trapping Devices
. Sediment basin/rock dam
. Sediment filters and sediment chambers
171
-------
. Sediment traps
Inlet Protection
. Stabilized construction entrances
. Storm drain inlet protection (includes block and gravel, excavated drop, fabric drop, and sod drop inlet protection)
Education and Awareness
• Contractor certification and inspector training
. BMP inspection and maintenance
5) Post-Construction Runoff Control
This measure uses post-construction controls as part of a program to address stormwater runoff from new
development and redevelopment projects using appropriate structural and non-structural BMPs. Non-structural BMPs
are preventive actions using management and source controls, such as policies and ordinances that result in protection
of natural resources and prevention of runoff. Non-structural BMPs might include requirements that encourage growth
in identified areas while protecting sensitive areas such as wetlands and riparian zones, minimizing impervious surfaces,
maintaining open space, and minimizing clearing, grading, or other disturbance of soils and vegetation. Some of the
typical structural BMPs include storage practices (wet ponds, extended-detention dry ponds, or other storage facilities
with outlets); infiltration practices (infiltration basins, infiltration trenches, and porous pavement); and filtration practices
(grassed swales, sand filters, and vegetated filter strips). This measure should also ensure effective and reliable
performance by providing for the long-term operation and maintenance of the selected BMPs. The following BMPs have
been identified under eight major subcategory groups of practices for the Post-Construction Runoff Control minimum
measure:
Ponds
. Extended detention dry basin or pond (with or without permanent pools or shallow marshes near the outlet),
includes tank storage
• Wet pond
Infiltration Practices
• Infiltration basin (a.k.a. - recharge basin)
. Infiltration trench (a.k.a. - infiltration galley)
. Porous pavement
Filtration Practices
. Bioretention
. Filters, including organic media filter (peat sand or compost-type), sand filter, multichamber treatment train (MCTT)
system, and inlet filtration systems
Vege ta five Practices
. Constructed wetland, shallow marsh
. Grassed swale
172
-------
. Vegetative filter strip
Runoff Pretreatment Practices
. Catch basin
. In-line storage, includes flow regulator information
. Manufactured systems for water quality inlets
Experimental Practices
• Alum injection system
On-lot Treatment
. On-lot treatment includes information on dry wells, roof downspout systems, rain barrels, exfiltration storage
systems, french drains, and dutch drains
Better Site Design
. Conservation easements
. Infrastructure planning
. Buffer zones/setbacks
. Open space development
. Narrow streets
. Curb elimination
. Green parking lot
. Alternative turn-around
. Urban forestry
. Alternative pavers
6) Pollution Prevention/Good Housekeeping
This measure in the proposed rule envisions the creation of an operation and maintenance/training program to prevent
or reduce pollutant runoff from municipal operations. The program should include training for municipal staff to address
prevention measures in government operations, such as park and open space maintenance, fleet maintenance, planning,
building oversight and stormwater collection system maintenance. Other possible pollution prevention activities that might
be relevant include controls for reducing or eliminating the discharge of pollutants from streets, roads, highways, municipal
parking lots, maintenance and storage yards, and waste transfer stations; programs to promote recycling and pesticide
use information; procedures for proper disposal of waste removed from municipal systems and public areas (such as
streets) including dredge spoil, accumulated sediments, floatables, and other debris; and new flood management projects
to assess the impacts on water quality and examine exiting projects to determine if they need additional water quality
protection devices or practices. The following BMPs have been identified under two major subcategory groups of
practices in the menu for the Pollution Prevention/Good Housekeeping minimum measure:
173
-------
Source con trols
. Animal waste collection
. Automobile maintenance
. Car washing
. Illegal dumping control
. Landscaping and lawn care
. Pest control
. Parking lot and street cleaning
. Roadway and bridge maintenance
. Septic system controls
Materials Management
. Alternative products
. Hazardous materials storage
. Household hazardous waste collection
. Road salt application and storage
. Spill response and prevention
. Used oil recycling
Conclusion
As part of the Stormwater Phase II Tool Box, the Menu of BMPs should help municipalities develop, implement, and
enforce the Phase II program. The menu will be available in time for regulated municipalities to use in complying with
stormwater management program requirements under Phase 2 permits and might also benefit other jurisdictions and
individuals.
References
USEPA, 1999. Storm Water Phase II Fact Sheet Series 1-4. April 1999. EPA 833-F-99-001 through 015.
USEPA, 1998. 40 CFR Parts 122 and 123; National Pollutant Discharge Elimination System-Proposed Regulations for
Revision of the Water Pollution Control Program Addressing Storm Water Discharges; Proposed Rule. Friday January
9, 1998.63 FR 1536.
174
-------
Determining Urban Stormwater BMP Effectiveness
Eric Strecker
URS Greiner Woodward Clyde
Portland, OR
Marcus M. Quigley
URS Greiner Woodward Clyde
Boston, MA
Ben R. Urbonas
Urban Drainage and Flood Control District
Denver, CO
Abstract
The overall purpose this US EPA funded cooperative research program with the American Society of Civil Engineers
(ASCE) is to develop a more useful set of data on the effectiveness of individual best management practices (BMPs) used
to reduce pollutant discharges from urban development. BMP performance data gathered at a particular site should not
only be useful for that site, but also be useful for comparing studies of similar and different types of BMPs in other
locations. Almost all BMP effectiveness studies in the past have provided very limited data that is useful for comparing
BMP design and selection among individual BMP types (e.g. sand filters). This paper overviews some of the problems
of past BMP effectiveness studies from the perspective of comparability between studies. It suggests some of the ways
that data should be collected to make it more useful for assessing factors (such as settling characteristics of inflow solids
and physical features of the BMP) that might have led to the performance levels achieved. It briefly presents the
database that has been developed by this project, which not only serves as a tool for storing data from existing studies,
but as a tool for entering and storing data collected from future studies. Discussed are considerations that affect data
transferability, such as effectiveness estimations, statistical testing, etc. It overviews the efforts to establish and analyze
the data base for existing studies and overviews proposed analyses for the future, when more studies that have followed
the protocols are available. The database has specifically pointed out the need for additional BMP performance studies,
as the current data is very sparse in terms of studies that have recorded enough information to be useful in assessing
BMP type performance.
Introduction
Many studies have assessed the ability of stormwater treatment BMPs (e.g., wet ponds, grass swales, stormwater
wetlands, sand filters, dry detention, etc.) to reduce pollutant concentrations and loadings in stormwater. However, in
reviewing and summarizing the information gathered from these individual BMP evaluations, it is apparent that
inconsistent study methods and reporting make wider-scale assessments difficult, if not impossible. For example,
individual studies often included the analysis of different constituents and utilized different methods for data collection and
analysis, as well as varying degrees of information on BMP design and inflow characteristics. Just the differences in
monitoring strategies and data evaluation alone contribute significantly to the range of BMP "effectiveness"that has been
reported. These differences make combining these individual studies almost impossible to assess what design factors
may have contributed to the variation in performance (Strecker et al., 1992). Urbonas (1994 and 1995) and Strecker
(1994) summarized information that should be recorded regarding the physical, climatic, and geological parameters that
likely affect the performance of a BMP and considerations regarding sampling and analysis methods.
Efficiency, Effectivness, and Performance
In order to better clarify the terminology used to describe the level of treatment achieved and how well a device,
system, or practice meets its goals, definitions of some terms often used loosely in the literature are provided here. These
terms help to better specify the scope of monitoring studies and related analyses:
175
-------
. Best Management Practice (BMP) -A device, practice, or method for removing, reducing, retarding, or preventing
targeted stormwater runoff constituents, pollutants, and contaminants from reaching receiving waters
. BMP System - A BMP system includes the BMP and any related bypass or overflow. For example, the efficiency
(see below) can be determined for an offline retention (Wet) Pond either by itself (as a BMP) or for the BMP system
(BMP including bypass)
. Performance - measure of how well a BMP meets its goals for stormwater that is treated by the BMP
. Effectiveness - measure of how well a BMP system meets its goals in relation to all stormwater flows
. Efficiency - measure of how well a BMP or BMP system removes pollutants
The ASCE project team is working with available data to determine efficiency of BMPs and BMP systems. In
addition, effectiveness and performance are being evaluated, acknowledging the limitations of existing information about
the goals of specific BMP projects. Quantification of efficiency only evaluates a portion of the overall performance or
effectiveness of a BMP or BMP system. Calculation of the efficiency helps to determine additional measures of
performance and effectiveness, for example the ability of a BMP to meet any regulatory goals. A list of typical goals and
the current ability of the ASCE/EPA project to help evaluate them is shown in Table 1.
Problem: BMP Performance Study Inconsistencies
Studies of BMP effectiveness have utilized significantly different:
. sample collection techniques (e.g., from sample collection types-grab, composite, etc., flow measurement
techniques, to how the sample was composited, etc.);
. water quality constituents, including: chemical species, methods (detection limits), form (e.g., dissolved vs. total,
vs. total recoverable, etc.), and treatment potential;
. data reporting on tributary watershed and BMP design characteristics ( e.g., tributary area or watershed attributes
such as percent impervious, land use categories, rainfall statistics, etc.);
. effectiveness estimation techniques (there are at least four common techniques that have been utilized to assess
effectiveness that can cause significant differences in pollutant removal reporting, with the same set of data), and
potential alternatives to reporting just concentration/loading reductions; and
. statistical validation of results (typical lack of statistical tests to determine if the reported removal efficiency can in
fact be shown to be statistically different than zero).
Monitoring strategies that could be employed to monitor BMP effectiveness include:
. New BMP installation with new development- input/output (e.g., monitor new detention pond of newly developed
watershed and evaluate inflow concentrations/loads vs. outflows) or conduct a "control" watershed comparison
. Retrofit of existing or new single BMP within existing watershed-input/output, and/or, before/after (e.g., retrofit of
an existing flood control basin for water quality)
. Watershed-wide new structural or non-structural-"control" watershed comparison (e.g., new BMP catch basins
in developing area)
. Watershed-wide structural retrofit or application of non-structural - before/after, and/or, "control" watershed
comparison (e.g., catch basin retrofit on watershed scale)
Input/output monitoring is the typical approach utilized. However, control watersheds and before/after approaches
have also been employed. All of the other potential factors that could be contributing to differences in performance must
identified and accounted. On the other hand, it is beneficial to be able to show that a watershed-wide difference is or is
176
-------
Table 1. Goals of BMP Projects and the Ability of the National Stormwater BMP Database to Provide Information Useful for Determining
Performance and Effectiveness
Goals of BMP Projects
Ability to Evaluate
Performance and Effective
Category
Hydraulics
Hydrology
Water Quality
(Efficiency)
Toxicity
Regulatory
Implementation Feasibility
Cost
"Imfftesttiappearance of site
Maintenance
• Uangg* functionality
Resources
Safety, Risk and Liability
Public Perception
'Improve flow characteristics upstream and/or downstream of BMP
'Flood mitigation, improve runoff characteristics (peak shaving)
'Reduce downstream pollutant loads and concentrations of pollutants
'Improve/minimize downstream temperature impact
•Removal of litter and debris
'Reduce acute toxicity of runoff
•Reduce chronic toxicity of runoff
'Compliance with NPDES permit
*Meet local, state, or federal water quality criteria
*For non-structural BMPs, ability to function within management and oversight
structure
'Capital, operation, and maintenance costs
'Operate within maintenance, and repair schedule and requirements
'Ability of system to be retrofit, modified or expanded
'Improve downstream aquatic environment/erosion control
'Improve wildlife habitat
'Multiple use functionality
'Function without significant risk or liability
•Ability to function with minimal environmental risk downstream
'Information is available to clarify public understanding of runoff quality, quantity
and impacts on receiving waters
/
/
/'
/1
/ can be evaluated using the ASCE/EPA Database as information source
/1 will be able to be evaluated using the database as primary source of information after enough studies have been submitted
/2 can be evaluated using the database the primary source of information combined with a secondary source of comparative data
not being detected with BMP implementation. These differences in monitoring approach certainly effect the ability to
compare studies.
Any of the above topics would require an in-depth discussion beyond the scope of this paper. Therefore, this paper
will present only a brief overview of each and some potential solutions for improving how data is collected. The ASCE
project team has developed a set of protocols and a database on BMP performance studies with the purpose of improving
the consistency of BMP monitoring information. The project includes:
. Developing protocols for BMP monitoring and reporting
. Developing a database on BMP performance studies
. Conducting an evaluation of existing information to assist EPA in providing guidance to the regulated community
177
-------
The database specifies a chosen set of reporting information, but does not guide users on how to develop such
information. For example, it does not specify in detail what a flow-weighted composite sample is and how it should be
collected. The next step beyond the EPA protocols and database effort should be a guidance document on monitoring
data collection strategies and techniques to improve their consistency and ultimate transferability. A few of the issues
related to proper guidance are discussed in the next two sections. It should be recognized that, with the development of
the database and the protocols, it will be a number of years (5 to 10) before a significant number of new studies on BMPs
are conducted utilizing the protocols. Therefore, a rigorous evaluation of BMP selection and design factors will need to
take place in the long-term future.
Recommended Parameters for Assessing BMP Performance
In developing a method for quantifying BMP performance, it is helpful to look at the objectives of previous studies
seeking such a goal. BMP performance studies usually are conducted to obtain information regarding one or more of the
following objectives:
. What degree of pollution control does the BMP provide under typical operating conditions?
. How does performance vary from pollutant to pollutant?
. How does performance vary with various input concentrations?
. How does performance vary with large or small storm events?
. How does performance vary with rainfall intensity?
. How do design variables affect performance?
. How does performance vary with different operational and/or maintenance approaches?
. Does performance improve, decay, or remain the stable over time?
. How does the BMP's performance compare relative to other BMPs?
. Does the BMP reduce toxicity to acceptable levels?
. Does the BMP cause an improvement in downstream biotic communities?
. Does the BMP have potential downstream negative impacts?
The monitoring efforts implemented most typically seek to answer a subset of the above questions. This often leaves
larger questions about the performance of the BMP, and the relationship between design and performance, unanswered.
Standardization of BMP data collection and evaluation methods (i.e., guidance and the ASCE/EPA database) allows this
broader set of questions to be examined.
There has been a very wide variety of pollutants analyzed in BMP and characterization studies. The protocols
established under the EPA-funded cooperative research program recommend a standard set of constituents for BMP
testing programs. Table 2 presents the recommended constituents developed from the review of previous studies with
an understanding of costs and likelihood of providing meaningful results. A discussion of how these constituents were
selected and a detailed description of each can be found in Strecker (1994).
There are some practical and technical considerations regarding data reporting which would facilitate data
usefulness, including consistent formatting of data, the clear indication of QA/QC results, standard comparisons to water
quality criteria, reporting of tributary watershed characteristics, and BMP design information. The last two items are
considered critical for evaluation of what contributed to BMP effectiveness in one location over another.
Data Reporting, It is recommended that all constituent concentration data be reported as event mean concentrations
(EMCs). These statistics should be based on use of the lognormal distribution. The NURP and FHWA studies (EPA,
178
-------
Table 2. Recommended Standard Analytical Tests for Urban Stormwater BMP Assessment
EPA Method Detection
Lab Analysis Number Limit (mg/l)
Conventional
TSS EPA 160.2 1
TDS EPA 160.1 1
TOO EPA 415.1 3
COD EPA 410.4 1
Total Hardness SM314-A 1
Nutrients
(NH3-N) SM417-AD 0.1
Total phosphorus (as P) SM 424-CE 0.005
Ortho-phosphate (as P) SM 424-E 0.05
Nitrate + nitrite (NO, + NO2 - N) EPA 353.1 or .2 0.05
Total Metals
Cd (cadmium) EPA 7131 0.0002
Pb (lead) EPA 7421 0.0003
Cu (copper) EPA 6010 0.001
Zn(zinc) EPA 6010 0.001
Dissolved Metals
Cd (cadmium) EPA 7131 0.0002
Pb (lead EPA 7421 0.0003
Cu (copper) EPA 6010 0.001
Zn (zinc) EPA 6010 0.001
1983a; Driscoll et al., 1990) identified that the lognormal distribution is suitable for characterizing EMC distributions. The
high degree of variability is why proper statistical techniques should be employed to evaluate whether a measured
difference between BMP before/after or input/output is significant. The recommended inclusion of outlet data as a part
of any paper or report will allow comparisons of typical outlet concentrations and may allow the determination of the lowest
or average expected concentration from a particular type of BMP. For example, it may be that wet ponds may only be
able to treat to some minimum concentration range at the outlet and the "effectiveness" is greatly impacted by the inlet
concentrations.
Quality Assurance/Quality Control (QA/QC). All monitoring studies should include a QA/QC program. The details and
results of the QA/QC program should be reported in monitoring study reports and summarized in papers. It is especially
important to discuss when data are characterized as estimates due to QA/QC results and when detection limits were
affected. Too often this information is not included.
Comparisons to Water Quality Criteria. A method to gage effectiveness could be to monitor how the BMP affects
the number of times (frequency) that EPA water quality criteria are exceeded in both the inflow and the outflow, to assess
how the BMP reduces (or does not reduce) the frequency of potentially toxic events. For heavy metals analyses, it is
recommended that hardness be collected for all storms monitored and that comparisons to criteria be made utilizing the
dissolved fraction with the computed aquatic criteria as modified by EPA (1993b).
Watershed and BMP Design Parameters. Table 3 presents a summary of these parameters. These parameters
(more detailed parameter lists are available on ASCE's Web page at http://www.asce.org/peta/tech/ nsbdOl .html) have
been selected with the purpose of being able to utilize this information to evaluate what BMP design attributes and
tributary watershed characteristics can be linked to BMP effectiveness information.
The primary goals of the ASCE/EPA database development process were to facilitate efficient data entry, provide
useful queries of stored data, and output information in a comprehensive and applicable manner through a user-friendly
interface. The database was written in Microsoft© Access incorporating Access relational database engine and features
and the Visual Basic®for Applications programming language for customization of the functional aspects of the front end.
Distribution will take place initially as an Access run-time version on CD-ROM, but will be available in the nearfuture over
the Internet.
179
-------
Table 3. Parameters to Report with Water Quality Data for Various BMPs
3arameter
Type
ibutary
atershed
eneral
ydrology
later
Parameter
Tributary watershed area, average slope,
average runoff coefficient, length, soil types,
vegetation types
Total tributary watershed impervious percentage
md percent hydraulically connected
Details about gutter, swer, swale, ditches,
larking, roads in watershed
•and use types (res., comm., ind., open) and
acreage
)ate and start/stop times for monitored storms
lunoff volumes for monitored storms
>eak 1 -hr intensity
)esign storm/flood recurrence intervals and
nagnitude
3eak flow rate, depth, and Manning's roughness
loefficient for the 2-year storm
)epth to seasonal high
jroundwater/impermeable layer
Saturated hydraulic conductivity, infiltration rate,
soil group
\verage annual values for number of storms,
>recipitation, snowfall, minVmax. Temp., from
appropriate weather station
Mkalinity, hardness and pH for each monitored
storm
A/ater temperature
Sediment settling velocity distribution, when
ivailable
:acility on- or off-line?
Bypassed flows during events
Retention (Wet)
Pond
t
.
Extended
Detention (Dry)
Basin
*
*
Wetland
Pond
Basin
<
<
Grass Swale/
Wetland Channel
<
<
Sand/Leaf
Compost Filter
<
<
Oil & Sand Trap/
Hydro-dynamic
Device
<
<
Infiltration and
Perc..
<
.
180
-------
eneral
rcility
ret Pool
etention
olume
retreatment
lant
Zetland
lant
Type and frequency of maintenance
Types and location of monitoring instruments
nlet and outlet dimensions, details, and number
Media or granular material depth, type, storage
volume, and porosity
/olume of permanent pool
.ength of permanent pool
Dermanent pool surface area
Jttoral zone surface area
Jolar radiation, days of sunshine, wind speed,
>an evaporation, from appropriate weather
station
detention (or surcharge) and flood control
volumes
Detention basin's surface area and length
Srimful and half-brimful emptying time
Sottom stage/infiltrating surface area and type
rorebay volume, surface area
Relationship to other BMP's upstream
/Vetland/swale type, surface area, and length,
side slope Ibottom width for swales and
;hannels)
Percent of wetland surface between 0-12", 12-
>4", and 24-48"
'lant species and age of facility
•
•
•
181
-------
Estimation of BMP Pollutant Removal Effectiveness and Effectiveness
BMP pollutant removal effectiveness estimations are not straightforward and a wide variety of methods have been
employed. Martin and Smoot (1986) discussed three types of methods to compute efficiencies, including an efficiency
ratio, sum of loads, and regression of loads. Many researchers have utilized an efficiency measure based upon storm
pollutant loads into and out of the BMP on a storm-by-storm basis. This weights the effectiveness considering that all
storms are "equal" in computing the average removal. However, it is readily apparent that all storm volumes and their
associated concentrations are not equal.
One factor that complicates the estimation of the effectiveness is that for wet ponds and wetlands, (and other BMPs
where there is a permanent pool), comparing effectiveness on a storm-by-storm basis neglects the fact that the outflow for
a particular event being measured may have little or no relationship to the inflow for that same event. Based upon a
national characterization of rainfall (Driscoll, et. al., 1989), if a basin were sized to have a permanent pool equal to the
average storm, about 60 to 70% of the storms would be less than this volume. Therefore, in many cases, flows leaving
may have little or no relationship to flows entering the pond. Storm-to-storm comparisons are probably not valid. It is
probably more appropriate to utilize statistical characterizations of the inflow and outflow concentrations to evaluate
effectiveness or, if enough samples are collected (i.e., almost all storms monitored), to utilize total loads into and out of
the BMP.
Table 4 compares three of the methods, including percent removal by storm with a statistical characterization of
inflow/outflow concentration and a simple comparison of total loads in and out for the sampled storms for an example site.
The removals estimated differ by up to 18 percentage points. In this record, there are several storm events where inflow
concentrations were relatively low and therefore the system was not "efficient." However, it was effective at maintaining
the effluent quality.
Table 4. Comparison of BMP Pollutant Removal Efficiency Techniques
Storm
1
2
3
4
5
Volume of
Flow (ft3)
lnflow= Outflow
445,300
649,800
456,100
348,111
730,261
Concentration
In Out
(mg/l)
352 24
30 25
99 83
433 141
115 63
Load
In Out
(Ibs)
9780 670
1220 1010
2820 2360
9410 3060
5240 2870
% Removal
by storm
93%
17%
16%
67%
45%
Med 139 65 A
Cov 1.48 .86 28,470 9,970 V 48%
Mean 249 85 G
Cone 66%
Loads 65%
note: 1 lbm = 2.2046 kg and 1 ft3 = 0.028317 m3
Based on these factors, it is recommended that a statistical characterization of inflows vs. outflows be utilized. Use
of the log-transform of EMCs is recommended. Tests of the applicability of a log transform should be made to support the
transform of data when sufficient data is available. Standard descriptive statistics, box-and-whisker plots, and normal
probability plots of the transformed data for both the inflow and outflow should be employed to clearly demonstrate not only
the differences in the mean EMCs, but also the effectiveness of the BMP throughout the range of influent and effluent
EMCs. This approach provides the ability to determine whether any apparent differences in inflow and outflow EMC
populations are statistically greater than zero. If enough data on storms is collected, (e.g., continuous samples over an
182
-------
extended period including base flow measurements where significant), the total loads in and out may also be an acceptable
method. A graphical look at the distribution of contributing storms will often provide insight into the applicability of the
method, (e.g., do a small number of large storms dominate the resulting effectiveness value).
The variability in runoff concentrations from event to event is large. In attempting to statistically characterize a BMP
influent concentration (and outflow), the more data the better. As mentioned above, there are a number of types of BMP
evaluations that can be conducted: (1) standard evaluation of a single BMP, testing input and output, (2) evaluation of
multiple BMPs within a basin (before/after or control basin), and (3) evaluation of a BMP with multiple inlets (where it might
be very difficult and expensive to evaluate the BMP utilizing input/output). All methods should require that a rigorous
statistical approach be applied in selecting the number of samples to be collected to help assure detection of a given level
of change.
As an example of the number of samples required to detect a "true" difference, Table 5 presents an analysis of two of
the Portland monitoring stations (WCC, 1993) where 10 flow-weighted composite samples were collected. The Fanno Creek
station is a large (about 1,200 acres) residential catchment that is in an open channel, while the M1 station is a smaller
(about 100 acres) mixed land use station that is in a pipe. An analysis of a variance-based test was utilized with the
existing data to determine how many samples are estimated to be needed to detect a 5%, 20%, and 50% change in the
mean concentration at the station. The test was performed considering an 80% probability that the difference will be found
to be significant, with a 5% level of significance (Sokal and Rohlf, 1969). This analysis does not consider potential seasonal
effects on the collection of data as a factor. Even so, quite a large number of samples would be required to detect a 5%
to 20% difference in concentrations. In many locations, given that there may be only 10 to 20 storm events per year that
are large enough to monitor, it would take a number of years of sampling all storm events to be able to detect small
differences.
Table 5. Analysis of Sample Sizes Needed to Statistically Detect Changes in Mean Pollutant Concentrations from 2 Stations in Portland, Oregon
Monitoring Site
R1 - Fanno Creek
Residential
M1 - NE 122nd
Columbia
Slough Mixed Use
Parameter
TSS
Copper
Phosphorus
TSS
Copper
Phosphorus
Number of Samples Required to Detect the Indicated %
Reduction in Site Mean Concentration*
5% 20% 50%
202 14 4
442 29 6
224 16 4
61 5 2
226 15 4
105 8 3
*80% certain of detecting the indicated % reduction in mean of the EMCs.
There are numerous examples in the literature where small differences (2 to 5%) have been reported based upon fewer
samples than indicated by this analysis. This highlights the need to be more rigorous with regard to statistical testing of
reported effectiveness estimates. To detect larger changes, the number of samples becomes reasonable. The mixed land
use catchment in Portland is currently being studied for the effectiveness of the implementation of a number of source
controls and other controls that do not lend themselves to input/output testing. Examples include maintenance changes
(catch basin cleaning, street sweeping); education (business and residences); tree planting, and others. Post-BMP
monitoring will be conducted along with qualitative evaluations.
Another approach that this study will be evaluating is the use of effluent data to compare to design criteria. It has been
suggested by some researchers that BMPs may be able to treat only to a given concentration and therefore, if relatively
clean water is entering a BMP, performance based upon efficiency may not fully characterize whether a BMP is well-
designed. An example of this is based upon Rushton et al. (1997). The pond was located at the Southwest Florida Water
Management District service office in Tampa. The drainage basin is 6.5 acres with about 30% of the watershed covered
by rooftops and asphalt parking lots, 6% by a crushed limestone storage compound and the remaining 64% as a grassed
storage area. The pond was modified twice after initial construction; therefore, there are three periods of performance data
for three different designs. The first pond had an average retention time of 2 days, the second 5 days and the third 15 days.
The second design added wetland features, while the third utilized deeper and larger pools.
183
-------
Figure 1 shows the input and output median concentrations in log based 10 scale as well as the 95% confidence
limits. The study reported that performance of the pond (defined as removal efficiency) decreased after the first
modification. What appears to be evident is that the average inflow concentrations were much lower during the second
period, while the outflow concentrations were about equal (less, but not statistically different from the first design). It
appears that with the original and first modified designs that the effluent level was not decreased. However, one could
not say that the BMP was any poorer in efficiency. The last design appears to have lowered the potential effluent
concentration, but the major difference in efficiency came from the significantly higher inflow concentrations during the
sampling period. This example points out the need to carefully think about whether pollutant removal efficiency is an
accurate representation of how well a BMP works or does not.
In many cases, there is a need to conduct dry weather analyses between storms on BMPs with dry weather flows.
It may be that pollutants captured during storms are slowly released during dry weather discharges.
Biological and downstream physical habitat assessments such as aquatic invertebrate sampling and habitat
classification should be explored as an alternative to just utilizing chemical measures of effectiveness (Maxted, 1999).
Long-term trends in receiving water quality, coupled with biological assessments, would likely be a much better gage of
the success of the implementation of BMPs, especially on an area-wide basis.
2.60
2.00
Median=
| 21mg/l
81mg/l
I191
. 1.50 i
O !
I I
1.00
0.50
i 1.32
llmg/l
1.03 T 1-05
llmg/l
lo.96
9mg/l
).86 i
5mg/lj
Inflow 1990 Outfbw Inflow 1993-* Outflow Inflow 1994- Outflow
1900 1004 1993-1OS4 1995 1994-1995
Figure 1. Inflow and Outflow Log Mean TSS Concentrations (mg/l) and 95% Confidence Limits for Different Designs of a Wet Pond Located at SWFWMD
Service Office in Tampa, Florida.
Summary and Recommendations
There is a great need to have consistency with the constituents and methods utilized for assessing BMP
effectiveness. This paper has presented only some of the consistency issues. It is recommended that researchers who
undertake BMP effectiveness studies consider the recommendations suggested here, by Urbonas (1995) and other
recommendations based upon further analysis of this subject. It is the authors' opinion that EPA should require studies
receiving federal funding to conduct BMP effectiveness studies that utilize standard methods as suggested here, together
with much still-needed detailed guidance on data collection and sampling methods to improve data transferability.
184
-------
Acknowledgments
The authors wish to thank Gene Driscoll, Jonathan Jones, Bob Pitt, Bill Snodgrass, and Larry Roesner for their helpful
discussions and comments on the subject. Jonathon Jones, Jane Clary, and John O'Brien from Wright Water Engineers
developed the Database Structure and Platform as members of the ASCE project team. The Urban Water Resources
Research Council has provided very beneficial support and feedback. Finally, the assistance by EPA in funding this
cooperative agreement is acknowledged as well as the helpful participation of Eric Strassler and Jesse Pritts in reviewing
our work.
References
Driscoll, E.D., P.E. Shelley, and E.W. Strecker. 1990. Pollutant Loadings and Impacts from Storm Water Runoff, Volume
III: Analytical Investigation and Research Report. FHWA-RD-88-008, Federal Highway Administration.
Driscoll, E., G. Palhegyi, E. Strecker, and P. Shelley, 1989. Analysis of Storm Event Characteristics for Selected Rainfall
Gages Throughout the United States. Draft Report. Prepared by Woodward-Clyde for the U.S. Environmental Protection
Agency. 43pp.
Martin, E.H. and J.L. Smoot. 1986. Constituent-Load Changes in Urban Stormwater Runoff Routed Through a Detention
Pond-Wetland System in Central Florida. U.S. Geological Survey Water Resources Investigation Report 85-4310.
Maxted, J. 1999. Proceedings of the First International South Pacific Conference on Urban Stormwater
Ruston, B.T., Miller, C.H., Hull, H.C., and J. Cunningham, 1997. Three Design Alternatives for Stormwater Detention
Ponds. Southwest Florida Water Management District. Brooksville, FL.
Sokal, R.R. and F. James Rohlf. 1969. Biometry: The Principles and Practice of Statistics in Biological Research. W.
H. Freeman and Company. San Francisco, CA.
Strecker, E., 1994. Constituents and Methods for Assessing BMPs. Proceedings of the Engineering Foundation
Conference on Stormwater Related Monitoring Needs. Aug. 7-12, Crested Butte, CO. ASCE.
Strecker, E.W., Kersnar, J.M., Driscoll, E.D., and Horner, R.R. 1992. The Use of Wetlands for Controlling Storm Water
Pollution. The Terrene Institute. Washington, D.C.
Strecker, E. and B. Urbonas, 1995. Monitoring of Best Management Practices. Proceedings of the 22nd Annual Water
Reources Planning and Management Division Conference, ASCE, New York. pp. 48-51.
U.S. Environmental Protection Agency. 1983a. Final Report on the National Urban Runoff Program. Water Planning
Division, U.S. EPA. Prepared by Woodward-Clyde Consultants.
U.S. Environmental Protection Agency. 1993b. Memorandum. Office of Water Policy and Technical Guidance on
Interpretation and Implementation of Aquatic Life Metals Criteria.
Urbonas, B.R. 1994. Parameters to Report with BMP Monitoring Data. Proceedings of the Engineering Foundation
Conference on Storm Water Monitoring Related Monitoring Needs. August 7-12, Crested Butte, CO. ASCE.
Urbonas, B.R., 1995. "Recommended Parameters to Report with BMP Monitoring Data. J. Water Resources Planning
and Management, ASCE. 121 (1), 23-34.
Woodward-Clyde Consultants. 1993. Final Data Report: Data from Storm Monitored between May 1991 and January
1993. Submitted to Bureau of Environmental Services, City of Portland, OR.
185
-------
Texas Nonpoint SourceBOOK Is Now On-Line!
http://www.txnpsbook.org
John Promise, P.E.
Director of Environmental Resources
North Central Texas Council of Governments, and
Chair, Water Resources Management Committee, American Public Works Association
Keith Kennedy
North Central Texas Council of Governments
Robert W. Brashear, Ph.D.,
Camp Dresser & McKee Inc.
Summary
Nonpoi
urce BOti
\
The Texas Nonpoint SourceBOOKis an internet-based resource that has been developed
to assist public works officials across Texas with storm water management. The
SourceBOOK'provides basic information about storm water quantity and quality impacts,
outlines how to develop and implement a local storm water management program, identifies
localized water quality issues, and provides an interactive database of more than 100 Best
Management Practices (BMP's) to use in a variety of situations.
The Texas Nonpoint SourceBOOK provides information for the novice as well as the
experienced storm water manager. The project was funded by the Environmental
Protection Agency and matching funds from 20 local governments across Texas. The North Central Texas Council of
Governments (NCTCOG) served as project administrator. The SourceBOOK was developed by a consulting team lead
by Camp Dresser & McKee Inc. (COM). A Project Management Committee of local governments provided project
oversight. After extensive review via the Internet, the SourceBOOK'was officially endorsed by the Executive Committee
of the Texas Chapter - American Public Works Association (APWA). Five training workshops were conducted across
the state. The SourceBOOK is intended to be a living resource, with additions and changes occurring continually in
response to input from users. A feedback page allows direct input from the Internet.
Why a Texas Nonpoint SourceBOOK?
Recognizing the need for improved communication, cooperation, and education statewide on stormwater issues, a
Statewide Storm Water Quality Task Force was established by the Executive Committee of the Texas Chapter -
American Public Works Association. At an organizational meeting in February 1994, a Steering Committee and
subcommittees were formed. The various subcommittees immediately tackled the task of identifying current issues and
needs regarding storm water quality and nonpoint source pollution, particularly with respect to the needs of public works
officials across Texas.
Already known was that nonpoint sources, including stormwater, contribute to water pollution problems. The Water
Quality Subcommittee began to review data from the Texas Clean Rivers Program, available nonpoint source monitoring
data, and the State's Nonpoint Source Water Pollution Assessment Report. They presented this assessment at
subsequent meetings of the Task Force. Water quality problems were known, but not how best to address them.
186
-------
What was not known was the applicability and cost-effectiveness of Best Management Practices (BMPs) for
addressing many of the typical water quality pollutants: bacteria, pesticides, nutrients, metals, toxic chemicals, and
others. The Best Management Practices (BMP) Subcommittee surveyed local governments across Texas on BMP
implementation but found little technical data. It was evident that until questions such as applicability and cost-
effectiveness could be answered, local governments would not invest limited public funds on storm water controls.
A project was formulated that would provide the assistance local governments needed by developing an internet-
based resource of storm water management information. At the time it was a striking idea, since the Internet was very
new and few local governments had any "on-line" experience. Using the emerging Internet would provide ready electronic
access and would allow for the use of new technologies in communication. This resource was to be called the Texas
Nonpoint SourceBOOK, and would be developed in both "hardcover" and electronic form. A grant application, submitted
to the Texas Natural Resource Conservation Commission under the Section 319(h) Nonpoint Source Program, was
awarded in the spring of 1996. Work on the project began in September, 1996.
How Was the Texas Nonpoint SourceBOOK Developed?
The North Central Texas Council of Governments provided staff support and general administrative oversight. To
guide the development of the SourceBOOK, a Project Management Committee was established from the Texas Chapter-
APWA membership. Among its first tasks was issuing a Request for Proposals for professional consultant assistance,
and selecting the consultant finalists. From the finalists the Committee selected a consultant team led by the firm Camp
Dresser & McKee Inc., in association with Espey Huston & Associates, Inc.; Center for Watershed Protection; Booth, Ahrens
& Werkenthin, P.C.; Carter Burgess; and Pavlik & Associates. Together, the committee and consultants used the State's
Nonpoint Source Water Assessment Report and supporting information to identify particular pollutants from priority
watersheds and related pollution prevention BMPs.
During FY97, the Project Management Committee worked with the consultant to establish the format of the Texas
Nonpoint SourceBOOK on the Internet. Presentations on local BMP experiences were made at the TX-APWA Short
Course at Texas A&M in February, 1997. Initial consultant materials were reviewed by the TX-APWA general
membership at its summer, 1997, Annual Meeting. A draft of the Texas Nonpoint SourceBOOK was presented to the
TX-APWA general membership at the February 1998 Short Course, and local government comments were solicited.
The TX-APWA Executive Committee endorsed the Texas Nonpoint SourceBOOKin February, 1999. It is available
through the Internet and on CD-ROM for use by local governments across Texas. The Committee and consultant
conducted technology transfer and training workshops on storm water management and the Texas Nonpoint SourceBOOK
at five regional one-day workshops across Texas during February and March of 1999.
How is the Texas Nonpoint SourceBOOK Organized?
The SourceBOOK is designed to make use of the capabilities of the Internet. This includes the ability to organize and
present textual and graphical information through common browser formats, as well as providing active links to related sites.
The design of the content of the SourceBOOK maximized the use of existing web sources wherever possible.
The content of the SourceBOOK consists of a set of modules:
Introduction and Overview
• About This Site
• Frequently Asked Questions (FAQs)
• Related Links
. Nonpoint Source News
187
-------
. Post Your Feedback
Module 1 - Nonpoint Source Management 101
. History of Nonpoint Source Management
. Urban Nonpoint Source Primer
. Controlling Urban Runoff-Guidance for Beginners
. Selecting the Right BMP - Guidance for Beginners
. Planning Your Stormwater Management Program - Guidance for Beginners
• Glossary
Module 2 - Urban Runoff Management Programs
. Introduction
. The Planning and Goal Setting Process
. Planning and Program Approaches
. Funding Mechanisms
. Measuring Effectiveness of Management Programs
. Implementation Strategies
. Case Studies
. Bibliography
. Additional Resources
Module 3 - Characterizing Urban Waterways
. Urban Runoff Flow and Water Quality
• Assessing Urban Waterways
. Water Quality and Other Watershed Physical Characteristics in Texas
Module 4 - Runoff Quality Best Management Practices
• Selecting Management Practices
. Housekeeping Practices
• Source Control Practices
. Treatment Control Practices
. Interactive BMP Selector
188
-------
What Does Each Module Provide in the Texas Nonpoint SourceBOOK?
Module 1, "Nonpoint Source Management 101," is a primer for beginners on urban stormwater management. It
quickly establishes that storm water quality £0° quantity management need to be addressed as one integrated program
within a local government. It provides guidance on regulatory issues, basic axioms of runoff control, and the use of
pollution prevention, source and treatment controls.
Module 2, "Urban Runoff Management Programs," describes the process to be used to manage urban runoff within
the overall framework of the city, county, or special district. Particular attention is placed on the key institutional and
financial components necessary for a successful ongoing program.
Module 3, "Characterizing Urban Waterways," begins with a generic discussion of urban runoff flow and water quality
relationships. Considerable attention is then given to proper techniques for monitoring urban waterways and stormwater
runoff. The majority of the module focuses on Texas-specific information. Descriptions of known water quality problems
can be accessed for the entire state. Each regional planning area and basin has specific information on water bodies,
B* &*_
Urban Nonpoint Source Management 1Q1
Ghaplrcr Contents Home
Examples of site controls: an extended detention basin with permanent pool (right) and an infiltration basin
Regional Controls serve a multi- acre drainage area, usually greater than 10 acres. Regional facilities can be
h^nor^r r»n,nn«l rnnfrnlyrg h*»»Pi
Urban Runoff Management Programs
Chapjar Cnntants Home
- monitor program effectiveness.
* figure M
THE URMP PLANNING PROCESS
Program Activities Technical Activities
INVOLVE Ml
,,^.,
_ __ _,
E
odule 1
$CW£-"
189
-------
tfc E* Y»» ao frnto H*
,*•••**OJ3<3
Characterizing Urban Waterways
5.0 Other Watershed Physical Characteristics by Region
Texas State River Basins, Council of Government Boundaries, and Urbanized Areas
(click on your «re« of Interest) ;
tie™
watershed characteristics, annual precipitation and runoff, major soil types, and the like. There are many "hot" links to
real-time gauging stations, local programs, and state/federal sites, such as EPA's Surf Your Watershed.
Module 4, "Runoff Quality Best Management Practices," provides guidance on the selection of Best Management
Practices for pollution prevention, source control, and treatment control. Considerable effort was placed on gathering
the most current information on more than 100 BMPs and review by the Project Management Committee of local
governments. Each BMP includes detailed information, such as performance data, photographs, and relevant reference
citations. An innovative BMP Interactive Selector was developed for the SourceBOOK. It enables the user to peruse
BMP's in each category, or to input several characteristics specific to their situation and request a set of the most
applicable BMPs.
190
-------
Lunoff Quality Best Management Practices
•Mptmr qof*ent* tUMD* S^P Cftflsnt? tl«CCb
Source Controls
UBS suras. Source controls appropriate for construction sites art designated as 'CS-' and source controls for perminaent us» are designated as
PS-'. All source controls are rated for their suitability on Residemial/CQrtvnarcial, Indus t rial/Comma rtcial, or Construction applications,
i Residential/Commercial applications include residential developments as i
commericial and/or industrial).
i Indus I rat,'Commercial applications are focused an individual sites whose activity
with storm water regulations or who have activities that could pollute ttormwattr runoff,
i Construction applications are those practices required during tne construction of residential,
larger developments that involve mixed land use (residential,
es are industrial or commercial in nature and who must comply
ercial, or industrial facilities.
Clicking once on the column headings in the table below will sort th« display in descending order Clicking again on that same column heading will
sort the table in reverse order. Click an the Number column to preview the description of the practice, or click on the BMP ID to view the BMP.
!>• Prrt HIM faum fim** Bat- R»"
Sort»4 by [R»WConB. Uujndti] Rx [I U>35] of 15 Gi
^1 CS-EC 15 Er°*°n Control - Channel StabfliMQ
1 , PS SW1 Swale
1 f PS-SW.O Hter Stop (cwpS-3)
i i' PS-IHO .Mto-ationBarinCcwpM)
i ,E' PS-EC.5 i&owio Control - Flow Controls
radices
in for the disturbed ci
J
Portion of Sediment Captured
(Sediment Rettlned» A- A x F
Undliturbed andJor
Offsrte Area
Sit.R.tinj-
Totil Stdimtnt R«t«in«d On SIM
Total S«dim«nt Loidlng Fr«m SIM
AI
ment conttDl plans
Runoff Quality Best Management Practices
Jar CoiUantt Home ilXii-l(KOlflI3£i Bq-mfl
Treatment Control Practice Overview
m
91 "on doaign alotm t
100 year »tor») . Dnt
•od vhan properly daeigned .
a? ii dDwnctrBB> ch«n
191
-------
A Comparison of the Long-Term Hydrological Impacts of
Urban Renewal versus Urban Sprawl
Jon Harbor, Suresh Muthukrishnan, and Shilpam Pandey
Departments of Earth and Atmospheric Sciences
Purdue University
West Lafayette, Indiana
Bernie Engel, Don Jones, and K.J. Lim
Departments of Agricultural and Biological Engineering
Purdue University
West Lafayette, Indiana
Abstract
Recent concern over environmental and economic impacts of urban sprawl has focused renewed attention on the
importance of making full use of existing urban areas. Revitalizing former industrial, commercial, and residential areas
often involves changes in land use type or intensity of use. It is important to have the ability to evaluate the long-term
hydrological impacts of such changes. These impacts can then be placed within the context of impacts that similar land
uses would have if a decision were made to place them in the urban fringe (urban sprawl) rather than in existing urban
areas (urban renewal).
In this study, we illustrate how the Long-Term Hydrological Impact Analysis (L-THIA) tool can be used to compare
the hydrological impacts of land use change in existing urban areas versus change in the urban fringe. L-THIA is a simple,
comparative tool that requires the user to provide information on land use and soil type for existing and future/planned
conditions. The tool combines this information with local rainfall data to calculate long-term average annual surface runoff
under existing and future/planned conditions. L-THIA analyses can be run directly at our web site for locations throughout
the U.S. where the curve number technique is already routinely used (http://danpatch.ecn.purdue.edu/-sprawl/L-THIA).
By performing analyses of renewal versus conversion of agricultural land at the urban fringe, it is possible to provide a
comparative assessment of impacts. This initial comparison can be helpful in educating the general publicand decision-
makers, thereby raising awareness of this element of the set of variables that are considered in land use decisions.
Introduction
Because almost every major North American city had been founded by 1900, the dominant form of urban
development during the 20th Century has been growth on the outer edges of existing cities, or just beyond city limits
(Orum, 1995). With improvements in transportation and communications, the need for people to be clustered in high-
density central areas has decreased (Chinitz, 1991), encouraging decentralization, suburbanization, and sprawl. In the
United States, 87% of the population now lives in metropolitan areas and their hinterlands (Angotti, 1995), and steady
infilling between urban areas has resulted in the development of megalopolises such as the Philadelphia - Boston -
Washington DC - New York urban corridor. Even metropolitan areas which are stagnating or declining in terms of total
population are still growing in terms of total built area because of low-density suburban growth (Johnston, 1982).
Decentralization and suburbanization have changed the relative importance of the core areas of cities (Richardson,
1982). Although these central areas were the sites of initial city growth and development, many cities are now faced with
the challenge of revitalizing these once vibrant central industrial, commercial, and residential areas that have been in
decline in recent decades:. The following quote reflects efforts to slow the tide of migration from urban centers.
192
-------
"To combat the number of people fleeing [Chicago] for the suburbs, developers have lured middle-class home
buyers back with promises of safe neighborhoods and affordable homes. Chicago also leads the nation in
converting office and warehouse property into residential space such as condominiums and rental units, often
targeted to low- and moderate-income buyers." (Heavens, 1999)
At the same time that city administrations have been coping with the challenges of urban core renewal, suburban and
rural communities have become increasingly concerned about the environmental, economic, social, and aesthetic impacts
of continued urban growth at the fringes of developed areas (these later concerns are often grouped under the term urban
sprawl). Preservation of prime farmland and protection of rural areas have become important concerns, alongside a
growing emphasis on combating the impacts of continued sprawl on flooding, groundwater recharge, air pollution, climate,
ecology, and habitat fragmentation (Schueler, 1994). Although there is considerable interests in revitalizing urban cores,
especially if this reduces urban sprawl, to accomplish this requires that the decision-making process for urban and
suburban planning include consideration of the environmental as well as the economic aspects of land use.
Land use decisions are highly complex, involving consideration of economics, infrastructure, politics, labor and
population dynamics, and the environment. The planning process requires collection and comparison of a wide array of
data, usually with the goal of providing a planned solution that meets goals based on sustainable growth in industry and
commerce. However, increasing public and political concern over the environmental aspects of urban development has
raised the profile of efforts to develop efficient and environmentally sustainable urban environments. The key components
of environmentally sound urban development include land use patterns that minimize environmental impacts (Arendt,
1996), efficient automobile and pedestrian traffic, and the use of energy saving and environmentally sound building
designs. When attempting to balance economic and environmental concerns, it is important to quantify the differential
environmental impacts of alternate land-use scenarios. Objective measures of differential impacts provide a rational basis
for decision-making. In addition, they can be used to educate the public and key decision-makers in government and the
private sector about the level of environmental benefit that can be gained from alternative land-use decisions.
The aim of the work presented here is to demonstrate the application of an impact assessment tool in evaluating the
long-term hydrologic impact of development consistent with urban renewal versus the impact of an identical development
located at the urban fringe. Although the general outcome of such a comparison is unlikely to surprise anyone, the
advantage of quantifying differential impacts is in providing an objective numeric measure that is much easier to include
in decision-making than vague subjective assessments of environmental benefits.
Long-Term Hydrologic Impact Assessment (L-THIA)
In response to concerns from local planners that they had no simple, objective way to assess the impacts of alternate
development plans on surface water runoff and groundwater recharge, a Long-Term Hydrologic Impact Assessment tool
(L-THIA) has been developed (Harbor, 1994; McClintock et al., 1995; Ogden, 1996; Grove, 1997; Bhaduri et al., 1997;
Bhaduri, 1998; Minner, 1998; Minner et al. 1998; Lim et al., 1999; Leitch and Harbor, in press). L-THIA uses readily
available data on soils, climate, and land use to estimate long-term surface water runoff. By running the model for current
conditions, and then with changed land uses, the user can simulate the potentialimpact of land use change. The method,
initially developed as a simple spreadsheet application (Harbor, 1994), is based on the U.S. Department of Agriculture's
curve number (CN) method for relating precipitation and runoff as a function of land use and soil type (USDA, 1983,
1986). The CN method was selected because it forms the basis of other commonly used hydrologic models, thus the
data required for its use is readily available in most planning settings. Because of the reliance on the CN method, L-THIA
applies directly to those areas where the CN method is routinely used. Subsequent development of the L-THIA method
has included provision of a Geographic Information System (CIS) version (Grove, 1997), addition of nonpoint source
pollution loadings to land uses (Bhaduri, 1998), and development of an Internet-accessible version of the method (Lim
et al., 1999).
In the curve number technique, the land use and hydrologic soil type of an area are used to derive a CN value (values
typically range from 30 to 98). For any given daily precipitation, surface runoff is then computed from empirically based
relationships between rainfall, CN, and runoff. Although most commonly used to estimate runoff for extreme storm
events, in L-THIA the CN technique is used to determine daily runoff for a 30-year time series of daily precipitation values.
193
-------
Average annual runoff is calculated for each CN to provide a measure of long-term average impact, rather than simply
impact on isolated extreme storm events. To compare different land use change options, pre-development and post-
development average annual runoff can be calculated for each scenario. The L-THIA method is freely available at
http://danpatch.ecn.purdue.edu/~sprawl/LTHIA. This site includes information on the technique and its application, as
well as access to US climate and soils data necessary to run analyses. Users can submit land use and soil information
through a spreadsheet-style interface (Figure 1). Analyses are performed on a server at Purdue University and results
are delivered back to the user in the form of tables and graphs.
fleets or Land Use Change on Hydrology and NFS pollution - Netscape
vvEPA
United States
Environmental Protection
Agency
LTHIA
(Long Term Hydrologic Impact Assessment) WWW_
Input
| Illimii
Kvn.son.cRon AM*
YEAKl Wail YEAJI3
nru
nm
Bte
* lament: DoneUi :T7ST * " "":, ' ~*** ""7"*";' !';/'°" t
Figure 1. L-THIA WWW Input Screen at http://danpatch.ecn.purdue.edu/~sprawl/LTHIA.
A Comparison of Core Renewal versus Fringe Development
Study Scenario
The L-THIA tool can be used to examine the relative impact of land use change in the form of an urban renewal
project; replacing underused or abandoned commercial, residential, and industrial buildings in an urban core region;
versus an urban sprawl project; replacing agricultural land at the edge of a city. For the sake of illustration, consider
planning a 70 Ha major commercial development with urban core and urban fringe location alternatives. Although the
location decision-making process will be driven by economic and infrastructure concerns, also assume that differential
environmental impact is important in decision-making, perhaps as a result of political or regulatory pressure. In the
context of improving urban environments then, an important question is the extent to which placing this development in
an urban core region would have different hydrologic impacts than placing it at the city fringe.
To simulate this situation, consider two possible sites in the Chicago area. The first is in the urban core, and currently
consists of a mix of residential, industrial, and commercial properties that are unused or underused (Figure 1). The
second possible site is on the urban fringe, and currently is used for agriculture. For simplicity we assume that both sites
are on the same type of soil (from a hydrologic perspective), although in a real world example this might not be the case.
In each case, we use the L-THIA web tool to analyze how average annual runoff will change if the site is converted to
194
-------
solely commercial use (Figure 1). In the L-THIA input and output, the urban core site is labeled "YEAR 1 ", the urban
fringe agricultural site is labeled "YEAR 2" and the commercial land use for both sites is labeled "YEAR 3." The L-THIA
web tool uses the "YEAR" designation for different scenarios because analyses are typically for land use changes over
time.
Results
For the example described here, placing a commercial development in an urban core region, replacing an existing
mix of urban land uses, increases average annual runoff by 58% compared to the initial situation (Table 1 and; Figure
2). Note that the levels of impact given in Table 1 do not depend on the size of the commercial development; the same
percent increase applies regardless of area. Runoff increases because land uses with less impervious cover, such as
residential, are replaced by commercial land use that has a higher percentage of impervious area, in contrast, for the
urban fringe location, replacing agriculture with commercial use increases runoff by 670% (Table 1 and Figure 2), a ten-
times greater impact. Runoff increases so dramatically because agricultural use on relatively permeable soil is replaced
by very extensive impervious surfaces.
Table 1. Averaae annual runoff deoths and chanae for commercial development (post-development) in the urban core versus the urban fringe. Results
are for the specific example described in the text."
Urban Core
Urban Fringe
Pre Development
Average Annual Runoff
(mm)
81.8
16.8
Post Development
Average Annual Runoff
(mm)
129.3
129.3
Increase in Runoff (%)
58
670
100000 -
g o^ 75000 -
1 "
> | 50000 -
1 1
§»
EC •§ 25000
^o,
A
Low Density
Residential
-*, High Density, .
- ' Ccwmeriaall^-
/
«HV*
Core
Fringe Commercial
Figure 2. Average annual runoff volumes for commercial development, the urban core mixed-use, and the urban fringe agricultural use. The much larger
difference between the fringe location runoff volume and the commercial case indicates that fringe development will have the largest hydologic impact.
Note that the runoff volume is simple the average annual runoff depth (Table 1) multiplied by the site area. Results are for the specific example described
in the text.
195
-------
Discussion and Conclusions
The straightforward example presented here indicates that developing a commercial site in an urban core versus and
urban fringe location can have a very significant impact on the level of disturbance of the hydrologic regime. For the
Chicago example presented here, the urban fringe location produces an approximately ten times larger impact than the
urban core location. Clearly, from a solely hydrological standpoint, the urban core location is a better choice than the
fringe location. Although this is a hypothetical example, it illustrates the relative ease of use of the L-THIA tool, and more
importantly demonstrates an accessible way to provide a quantitative estimate of the relative impacts of different land use
decisions. More complex land use mixes and soil types can be run on the L-THIA web tool, either in the spreadsheet
version or in a CIS version also available at the web site. Thus, more sophisticated comparative analyses can be
performed.
In most cases, an L-THIA analysis provides a result that shows that renewal of existing areas has less hydrologic impact
than development of an area with rural use. This is not a surprise, rather the value of the tool is that it provides a context
for understanding and considering the magnitude of this difference in the decision-making process. For areas where
problems such as groundwater supply and downstream flooding are important, the scale and magnitude of the hydrologic
impact can be of considerable importance and can be considered alongside other concerns, such as infrastructure and
economic viability. We suggest use of tools such as L-THIA as part of the planning process, to ensure that land use
decisions are made after consideration of a full range of concerns, including environmental parameters as well as
economic, infrastructure, and political issues.
Acknowledgements
Development of the L-THIA tool is being supported by grants from the Environmental Protection Agency, Region 5.
References
Angotti, T., 1995. The Metropolis Revisited. Futures, 27: 627-639.
Arendt, R., 1996. Conservation design for subdivisions, a practical guide to creating open space networks. Washington
D.C., Island Press.
Bhaduri, B. 1998. A geographic information system based model of the long-term impact of land use change on non-point
source pollution at a watershed scale. Unpublished Ph.D. dissertation, Purdue University, West Lafayette, Indiana 47907-
1397.
Bhaduri, B., M. Grove, C. Lowry, and J. Harbor. 1997. Assessing the long-term hydrologic impact of land use change.
Journal of the American Water Works Association. 89:94-1 06.
Chinitz, B. 1991. A Framework for Speculating about Future Urban Growth Patterns in the US. Urban Studies, 28:939-
959.
Grove, M. 1997. Development and application of a CIS-based model for assessing the long-term hydrologic impacts of
land-use change. Unpublished MS Thesis, Purdue University, West Lafayette, IN 47907-I 397.
Harbor, J. 1994. A practical method for estimating the impact of land use change on surface runoff, groundwater
recharge and wetland hydrology. Journal of American Planning Association, 60:91-104.
Heavens, A.J., 1999, Building Anew in the City. Philadelphia Inquirer (09/12/99) L1.
Johnston, R., 1982. The American Urban System: A Geographical Perspective. New York: St. Martin's Press.
Leitch, C. and Harbor, J., (In Press) Impacts of land use change on freshwater runoff into the near-coastal zone, Holetown
watershed, Barbados: Comparisons of long-term to single-storm effects. Journal of Soil and Water Conservation.
196
-------
Lim, K.J., Engel, B., Kim, Y., Bhaduri, B., and J. Harbor. 1999. Development of the long term hydrologic impact
assessment (L-THIA) WWW systems. Paper No 992009, ASAE, St. Joseph, Ml.
McClintock, K., J. Harbor, and T. Wilson. 1995. Assessing the hydrologic impact of land use change in wetland
watersheds, a case study from northern Ohio, USA. In: McGregor, D. and Thompson, (Eds.) Geomorphology and Land
Management in a Changing Environment, pp.1 07-I 19.
Minner, M., 1998. Sensitivity analysis and advanced applications of L-THIA. Unpublished MS Thesis. West Lafayette,
IN: Purdue University, Department of Earth and Atmospheric Sciences, West Lafayette, IN 47907-I 397.
Minner, M., Harbor J., Happold S., and P. Michael-Butler, 1998. Cost Apportionment for a Storm-Water Management
System, Applied Geographic Studies, 2:247-260.
Ogden, M., 1996. Development and Land Use Changes in Holetown, Barbados: Hydrologic Implications for Town
Planning and Coastal Zone Management. Unpublished M.U.P. Thesis. Montreal: McGill University, School of Urban
Planning.
Orum, A., 1995. City-building in America. Westview Press, Boulder, CO.
Richardson, J., 1982. The Evolving Dynamics of American Urban Development, in: Cities in the 21st Century. Gapped,
G. and Knight, R. (Eds.) Urban Affairs Annual Reviews, 23.
Schueler, T., 1994. The importance of imperviousness. Watershed Protection Techniques, 1:100-1 11.
U.S. Department of Agriculture, Soil Conservation Service, 1983. Computer programs for project formulation - hydrology.
Technical Release 20.
U.S. Department of Agriculture, Soil Conservation Service, 1986. Urban hydrology for small watersheds. Technical
Release 55.
197
-------
Comparative Nutrient Export And Economic Benefits of
Conventional And Better Site Design Techniques
Jennifer Zielinski, Deb Caraco, and Rich Claytor
Center for Watershed Protection
Ellicott City, Maryland
Better site design describes a fundamentally different approach to the design of residential and commercial
development projects. It seeks to accomplish three goals at every development site: to reduce the amount of impervious
cover, to increase the amount of natural land set aside for conservation, and to use pervious areas for more effective
stormwater treatment.
When designing new residential developments, planners have the opportunity to reduce stormwater runoff and
pollutant export through better site design techniques. The better site design techniques applied to these developments
are referred to here as "open space design," and present an alternative to conventional residential subdivisions. Also
known as cluster development, open space design concentrates density on one portion of a site in order to conserve open
space elsewhere by relaxing lot sizes, setbacks, frontages and road section and other lot geometry. Open space design
also consists of:
• installing narrower streets and shorter driveways
• spreading stormwater runoff over pervious areas
using open channels rather than curb and gutter
• clustering development to conserve forests and natural areas
reducing the area devoted to turf
. protecting stream buffers
enhancing the quality of septic system effluent in areas where sewage is disposed of on-site
When these techniques are applied together, the cumulative benefits of better site design can be impressive.
Documenting the precise benefits is difficult, however, since few developments incorporating better site design techniques
have been built, let alone monitored.
As most better site design techniques are non-structural in nature, the achievable benefits will vary depending on the
unique characteristics of each development site and the actual site planning practices applied. Also, since better site
design techniques are commonly applied together, it has been difficult to accurately quantify their individual nutrient
removal benefits. Many local governments, consultants, and developers have expressed a strong desire for clear
documentation of these presumed benefits.
To help meet this need, the Center for Watershed Protection (CWP) recently completed a study to document the
comparative nutrient export and economic benefits of conventional and better site design techniques. The simple
assessment methodology analyzed both the residential and commercial environment through four real-world development
case studies in the Chesapeake Bay watershed. This paper presents the results of the residential component of that
project, including the incorporation of open space design techniques into the redesign of two residential case studies; the
resultant hydrologic, nutrient export, and economic benefits; and finally, the implications of our findings for the watershed
manager.
198
-------
Methodology
The basic method used in the Nutrient Loading from Conventional and Innovative Site Developmentproject (Caraco,
et al., 1998) conducted by CWP is a redesign analysis that compares conventional and better site design at actual project
sites using a simplified model.
CWP first assembled plans of previously developed sites representative of typical development scenarios across the
Chesapeake Bay, including a medium-density residential development from Virginia's Piedmont, a large-lot single family
residential subdivision from Maryland's Eastern Shore, a retail strip mall from Frederick County, Maryland, and a
commercial office park located outside of the District of Columbia in suburban Maryland. Each site was then "redesigned"
using better site design techniques.
The Simplified Urban Nutrient Output Model (SUNOM) was then used to compare each conventionally designed site
to the redesign. SUNOM is a spreadsheet model that computes the hydrologic budget, infrastructure cost and nutrient
export from any site, using common site planning variables. The model provides watershed practitioners with a simple
tool to compare the costs and benefits of better site design. It is not meant to be used as a method for determining actual
stormwater runoff and nutrient loading from a development site. To obtain accurate numbers for this, a more detailed
model should be used or on-site monitoring should be conducted.
Model input includes basic site planning variables that can be directly obtained or measured from a typical
development submittal to a land use authority, including total drainage area, length of sidewalks, total impervious cover,
linear feet of roads, lawn cover, utilities (length and type), forest cover, size, type, and length of stormwater conveyance,
riparian forest cover, size and type of stormwater practices, soil type(s), and method of wastewater treatment. Default
data are provided for many parameters and many of these assumptions can be changed based on site specific
information.
SUNOM is governed by the principles of a simplified water balance. In addition to annual runoff and infiltration,
SUNOM computes the annual nutrient load from each development site in pounds. In brief, the surface nutrient export
from each site is estimated using the Simple Method (Schueler, 1987). This export is then adjusted to reflect the mean
removal capability of stormwater BMPs where present (Schueler, 1997). The subsurface component of the model utilizes
annual subsurface recharge rates (based on the site's prevailing hydrologicsoil group) and monitored baseflow nutrient
concentrations in the receiving water to estimate the annual subsurface nutrient export from urban areas. These values
are then adjusted for the area of the site that cannot recharge (i.e., impervious cover) or are hindered from infiltrating by
other conditions (e.g., compacted urban turf). The model also calculates the cost of development utilizing previously
published or user-specified unit costs and predictive equations for infrastructure, stormwater management, landscaping,
and septic systems.
For each case study, SUNOM was used to compare the annual hydrologic budget and annual nutrient export under
five development scenarios: pre-developed conditions, conventional design without stormwater practices (uncontrolled),
conventional design with stormwater practices (controlled), design incorporating better site design techniques without
stormwater practices (uncontrolled), and design incorporating better site design techniques with stormwater practices
(controlled). The cost of development associated with each design was also estimated.
Case Study #1: Duck Crossing, A Low Density Residential Subdivision
Duck Crossing, a large-lot residential development, is located in Wicomico County on Maryland's Eastern Shore.
Prior to development, the parcel was representative of the typical terrain on Maryland's coastal plain, with very little
gradient. The site contained tidal and non-tidal wetlands, natural forest, meadow, the 1 00-year floodplain, as well as three
existing dwellings with on-site sewage disposal.
The large-lot subdivision of single family homes, constructed in the 1990's, (Figure 1) contains eight new residential
lots, each of which are 3 to 5 acres in size with houses set far back from the street. The street is wide given the few
homes that are served, ends in a large cul-de-sac, and is lined with a sidewalk. Each lot has an on-site private septic
199
-------
Figure 1. The conventional design of Duck Crossing, a low density residential subdivision on Maryland's Eastern Shore.
system, with a septic reserve field of about 10,000 square feet. Individual home property lines extend to the protected
tidal marsh, which is the only common open space on the site. Stormwater management consists of street runoff
conveyed by curb and gutter to a storm drain system that discharges to a small wet pond.
The major better site design techniques applied when redesigning this site (Figure 2) included:
conservation of tidal and non-tidal wetlands and forested areas
• a 100-foot buffer along tidal and non-tidal wetlands
clustering development to provide additional open space
• identification of potential development and open space areas based on location of sensitive areas, 100-year
floodplain, and potential septic field areas
. distribution of stormwater treatment practices throughout the site
use of a narrower access road; shorter, shared driveways; and wood chip paths through community open space
instead of sidewalks along the road
• use of shared septic systems utilizing more advanced re-circulating sand filter technology
The open space design resulted in reduced impervious cover, reduced stormwater runoff, increased stormwater
infiltration, and reduced infrastructure cost over the conventional design.
Case Study #2: Stonehill Estates, A Medium-Density Residential Subdivision
Stonehill Estates is located in Stafford County just north of Fredericksburg, Virginia. The original site was almost
entirely forested in a mix of mature deciduous hardwoods, with perennial and intermittent streams, and non-tidal wetlands.
An existing network of public water and sewer lines serves the site and road access to the subdivision is by two existing
streets.
200
-------
Figure 2. The open space design of Duck Crossing.
The conventional design produced a total of 108 house lots, each of which are about 9000 square feet in size (Figure
3). The subdivision is quite typical of a medium-density residential subdivision developed in the last two decades in the
Mid
Figure 3. The conventional design of Stonehill Estates, a medium density residential subdivision in Stafford County, Virginia.
Atlantic with uniform lot sizes and shapes, and generous front setbacks. The streets were 34 and 26 feet wide, numerous
cul-de-sacs were used as turnarounds, and sidewalks were generally installed on both sides of the street. With the
exception of a small tot-lot, the majority of the open space is unbuildable land, such as floodplains, steep slopes,
wetlands, and stormwater management areas. Street runoff is conveyed by curb and gutter to a storm drain system that
discharges to the intermittent stream channel. It then travels to a dry extended detention pond, which is primarily used
to control flooding, but also provides limited removal of stormwater pollutants.
201
-------
The open space design also results in 108 lots, but these were slightly smaller with an average size of 6,300 square
feet. The design also incorporates many techniques of open space design as advocated by Arendt (1994). The design
techniques employed in the redesigned site (Figure 4) include:
Figure 4. The open space design of Stonehill Estates.
• identify sensitive natural features, including mature forest and wetland, to be protected
incorporate a minimum 1 00-foot buffer along all perennial and intermittent streams
maximize the amount of community open space and preservation of natural areas
maintain the same number of lots as the conventional design
. provide open space adjacent to as many lots as possible
incorporate stormwater management attenuation and treatment throughout the site
use narrower streets, loop roads, shorter driveways, and fewer sidewalks
allow for irregular shaped lots and shared driveways
manage stormwater in a "treatment train" with bioretention facilities that discharge to a small but more effective
wet pond
The open space design resulted in reduced impervious cover, reduced stormwater runoff, increased stormwater
infiltration, and reduced infrastructure cost over the medium density subdivision conventional design (Table 1).
The Benefits of Open Space Design
For both of these case studies, application of the open space design techniques resulted in reduced impervious cover,
which translates directly to reduced stormwater runoff. Other "redesign" studies recently conducted in Delaware,
Maryland, and Virginia have provided similar results. These combined results consistently demonstrate that better site
design can reduce impervious cover by 25 to nearly 60% and stormwater runoff by 4 to over 60% for a range of
subdivisions (Table 1).
202
-------
Table 1. Redesign Analyses Comparing Impervious Cover and Stormwater Runoff from Conventional and Open Space Subdivisions
Residential
Subdivision
Duck Crossing
Stonehill Estates
Remlik Hall 1
Tharpe Knoll 2
Chapel Run 2
Pleasant Hill 2
Prairie Crossing 3
Buckingham Greene
Belle-Hall "
Conventional Zoning for
Subdivision
3-5 acre lots
1/3 acre lots
5 acre lots
1 acre lots
1/2 acre lots
Va acre lots
Vz - 1/3 acre lots
1 /8 acre lots
High Density
Impervious Cover at the Site
Conventional
Design
8%
27%
5.4%
13%
29%
26%
20%
23%
35%
Open Space
Design
5%
21%
3.7%
7%
17%
11%
18%
21%
Net Change
-35%
-24%
% Reduction in
Stormwater
Runoff
23%
24%
-31 % I 20%
-46%
-41%
-58%
-20%
-7%
4%
31%
54%
66%
8%
20% - 43% 31 %
sources: 1 Maurer, 1996; * DE DNREC, 1997; 3Dreher, 1994; and "SCCCL, 1995.
For both Duck Crossing and Stonehill Estates, the conventional design results in the highest annual volume of runoff
and the lowest volume of infiltration, as was expected. Of particular interest is the fact that the controlled conventional
design results in a higher annual runoff volume and a lower infiltration rate than the uncontrolled open space design. This,
however, should not imply that better site design alone, without structural Stormwater management, is sufficient in
controlling Stormwater runoff from this site since the open space designs do not come close to replicating pre-developed
hydrology.
Less impervious cover and Stormwater runoff, in turn, translates directly to smaller pollutant loads. Reducing the
impervious cover, preserving natural areas, and providing multiple Stormwater practices in series reduced nutrient export
for both case studies. However, neither open space design meets pre-development nutrient loads.
One area of particular interest for Duck Crossing is the implication of on-site sewage disposal systems. The
conventional design included astandard septictankand field for each lot, which resulted in phosphorus and nitrogen loads
that far exceeded pre-development levels. Recirculating sand filters were used in the open space design, instead of
conventional septic systems, because they yield better nitrogen removal efficiencies and are actually less expensive to
construct. This resulted in a much lower nutrient output from the entire site. However, even in the open space design,
the septic systems are the predominant source of nutrients.
For both case studies, the total infrastructure costs include the sum of the estimated costs of Stormwater
management, storm drainage, paving, sidewalk, curb and gutter, landscaping and reforestation, water, sewer and septic
systems. In both cases, the open space design resulted in a cost savings. Costs associated with grading, erosion and
sediment control, building construction and other incidental costs associated with land development were not analyzed.
In general, these costs should be comparable between the two development options. If anything, the grading and erosion
and sediment control costs should be lower with the open space design since less land is disturbed.
Several other studies have also shown that open space development can be significantly less expensive to build than
conventional subdivision developments. Most of the cost savings are due to savings in road building and Stormwater
management conveyance costs. The use of open space design techniques at a residential development in Davis,
California provided an estimated infrastructure construction costs savings of $800 per home (Liptan and Brown, 1996).
Other examples demonstrate infrastructure costs savings ranging from 11 to 66%. Table 2 lists some of the projected
construction cost savings generated by the use of open space redesign at several residential sites.
203
-------
Table 2. Projected Construction Cost Savings for Open Space Designs from Redesign Analyses
Residential Subdivision
Duck Crossing
Stonehill Estates
Remlik Hall 1
% Construction
Savings
12%
20%
52%
Tharpe Knoll 2 56%
Chapel Run 2
Pleasant Hill 2
Buckingham Greene 2
64%
43%
63%
Notes
Includes roads, stormwater management, and
Includes roads, stormwater management, and
reforestation
reforestation
Includes costs for engineering, road construction, and obtaining water
and sewer permits
Includes roads and stormwater management
Includes roads, stormwater management, and
Includes roads, stormwater management, and
reforestation
reforestation
Includes roads and stormwater management
Sources:' Maurer, 1996; 2DE DNREC, 1997.
Implications for the Watershed Manager
Better site design reduces impervious cover, conserves larger contiguous natural areas, and incorporates more
advanced stormwater treatment, which results in reduced stormwater runoff, increased infiltration, and reduced nutrient
export. Hopefully, the results of this study, as well as other redesign analyses, will answer some of the questions of local
governments, consultants, and developers as to the benefits of better site design.
However, there may still be difficulties to overcome before better site design becomes a reality and common practice
in many communities. Once there is a willingness to incorporate better site design techniques into new developments,
many communities may find that their existing development codes and ordinances are in conflict with the goals of better
site design. For example, many local codes and ordinances require excessive impervious cover in the form of wide
streets, expansive parking lots, and large-lot subdivisions. In addition, there are generally few, if any, incentives or
requirements for developers to conserve natural areas. When obstacles to better site design are present, it is a sign that
a community may want to reevaluate and consider changing some of its local codes and ordinances.
In 1997, CWP convened a national site planning roundtable to address this very issue. During the 18-month
consensus-building process, a diverse cross section of national planning, environmental, home builder, fire and safety,
and public works organizations (as well as local planning officials) crafted 22 model development principles to help further
better site design at the local level. This national roundtable is serving as a model for local government implementation
of better site design principles.
Recently, Frederick County, Maryland, initiated a local roundtable to take a critical look at its own development rules.
Members of the development community in partnership with local planning and zoning and public works staff are meeting
to identify and overcome impediments to better site design that are embedded in the county's codes and ordinances.
The outcome of the consensus process should be development rules that encourage rather than discourage the
application of better site design techniques.
Changing local development rules is not easy. Progress toward better site development will require more and more
local governments to examine their current practices in the context of a broad range of concerns, such as how changes
will affect development costs, local liability, property values, public safety, and a host of other factors. Advocates of better
site design will have to answer some difficult questions from fire chiefs, lawyers, traffic engineers, developers, and many
others in the community. Will a proposed change make it more difficult to park? Lengthen response times for emergency
vehicles? Increase risks to the community's children? True change occurs only when the community addresses these
and other questions to the satisfaction of all interests.
204
-------
References
Arendt, Randall. 1994. Designing Open Space Subdivisions. Natural Lands Trust, Media, PA.
Caraco, D., R. Claytor, and J. Zielinski. 1998. Nutrient Loading from Conventional and Innovative Site Development.
Center for Watershed Protection. Ellicott City, MD.
Delaware Department of Natural Resources and Environmental Conservation. 1997. Conservation Design for
Stormwater Management. Delaware Department of Natural Resources and Environmental Conservation, Dover, DE.
Dreher, D.W. and T.H. Price. 1994. Reducing the Impacts of Urban Runoff: The Advantages of Alternative Site Design
Approaches. Northeastern Illinois Planning Commission, Chicago IL.
Maurer, George. 1996. A Better Way to Grow: For More Livable Communities and a Healthier Chesapeake Bay.
Chesapeake Bay Foundation, Annapolis MD.
Schueler, T. 1987. Controlling Urban Runoff: A Practical Manual for Planning and Designing Urban BMPs. Metropolitan
Washington Council of Governments. Washington, DC.
Schueler, T. 1997. Comparative Pollutant Removal Capability of Urban BMPs: A Reanalysis. Watershed Protection
Techniques, 2(4): 539-542. Center for Watershed Protection. Ellicott City, MD.
South Carolina Coastal Conservation League (SCCCL). Fall 1995. Getting a Rein on Runoff: How Sprawl and Traditional
Town Compare in SCCCL Land Development Bulletin (Number 7). South Carolina Coastal Conservation League,
Charleston, SC.
205
-------
Predicting Erosion Rates on Construction Sites Using the Universal Soil Loss
Equation in Dane County, Wisconsin
J.D. Balousek, A. Roa-Espinosa, G.D. Bubenzer
Erosion Control Engineer, Dane County Land Conservation Department (LCD), Urban Conservationist, Dane County
LCD, Professor, University of Wisconsin-Madison
Madison, Wisconsin
Tools for Urban Water Resources Management and Protection
Sponsored by Urban Water resource Conference
Northeastern Illinois Planning Commission
The Westin, Michigan Avenue
Chicago, Illinois
February 7 to 10, 2000
Abstract
The Universal Soil Loss Equation (USLE) was developed for estimating sheet and rill erosion from agricultural fields
under specific conditions. Parameters used to estimate erosion include rainfall energy, soil credibility, slope length,
steepness, surface cover, and management practices. Traditionally, urban conservation planners have not used the USLE
for estimating soil loss and evaluating conservation measures and have relied on intuition alone to locate erosion control
practices on constructions sites. The results of this process are often subjective and may vary with the skill of the planner.
A USLE-based equation would provide a valuable, objective method for all planners, regardless of skill, to tailor specific
construction site practices to existing conditions. A method to predict soil loss from construction sites was developed by
adapting existing data for USLE erosion calculations to construction site conditions. In addition, the construction site
procedure was used to create a user-friendly computer-based program to assist planners in developing erosion control
plans. The computer program was distributed to engineers responsible for erosion control planning in Dane County,
Wisconsin. Implementation of the USLE-based equation has proven to be a valuable tool for assessing alternatives for
site management and erosion control. Planners are able to uniformly implement the equation on construction sites
throughout the county, decrease the time necessary to complete a USLE calculation, and reduce human error.
Keywords:
Universal Soil Loss Equation (USLE), urban erosion control.
Background
Soil erosion, detachment of soil particles from the soil surface, results when soil is exposed to the power of rainfall
energy and flowing water. Soil erosion causes a loss of productivity in the land, delivers millions of tons of sediment into
waterways, and provides a substrate for toxic chemicals which are carried into receiving waters. Construction site erosion
has been identified as a significant source of suspended solids in runoff in many parts of the United States (Hagman, et
al., 1980; Yorke and Herb, 1976; Becker, et al., 1974). In the State of Wisconsin, sediment is the largest pollutant by
volume (Wisconsin Department of Natural Resources, 1994). When erosion is compared on a rate basis, construction
site erosion generates more erosion in ashortperiod of time than any other land disturbing activity (Johnson and Juengst,
1997). While it is not possible to urbanize a watershed without exposing soil to erosive forces, it is possible to plan
construction to control the production of sediment through the use of erosion prevention and reduction practices.
The Universal Soil Loss Equation (USLE) (Equation 1) was developed by the United States Department of Agriculture
(USDA) for estimating sheet and rill erosion from agricultural fields under specific conditions (Wischmeier and Smith,
1978). The USLE enables planners to predict the average annual rate of soil erosion for combinations of seeding and
management practices in association with a specified soil type, rainfall pattern, and topography. The equation groups
206
-------
interrelated physical and management parameters influencing erosion rate into six major factors whose site-specific
values can be expressed numerically. More than a half century of erosion research in many states has supplied
information from which the USLE factors were determined.
The Universal Soil Loss Equation.
A = RxKx(LS)xCxP (Equation 1)
Where:
A = average annual soil loss
R = rainfall and runoff factor
K = soil credibility factor
L = slope length
S = steepness factor
C = cover and management factor
P = support practice factor
A The computed soil loss in tons/acre/year.
R The rainfall and runoff factor is the number of erosion-index units in an average year's rain. The erosion index
is the storm energy in hundreds of foot tons times the 30 minute storm intensity.
K The soil credibility factor is the soil loss rate (tons per acre) of a specific soil type and horizon as measured on
a standard plot of land.
L The slope/length factor is the ratio of soil loss from the actual land slope length to that from a standard plot (726
feet in length) of land. Slope length is defined as the distance from the point of origin of overland flow to the
point where either the slope gradient decreases enough that deposition begins or runoff water enters a well
defined channel that may be part of a drainage network or a constructed structure.
S The slope/steepness factor is the ratio of soil loss from the actual land slope gradient to that from a standard
plot of land (9%).
C The cover and management factor is the ratio of soil loss from an area with specified cover and management
to the corresponding loss from a clean-tilled, continuously fallow condition.
P The ratio of soil loss with a support practice such as contouring, stripcropping, or implementing terraces
compared to up and down the slope cultivation. The support practice factor does not usually apply to soil loss
on construction sites.
Soil losses computed with the USLE are best available estimates, not absolutes. The USLE will generally be most
accurate for medium-textured soils, slope lengths of less than 400 feet, gradients of 3 to 18%, and consistent seeding
and management systems represented in the USDA erosion studies. The USDA research shows that in comparing actual
soil loss to computed soil loss, 84% of the differences in long-time average soil losses were less than 2 tons/acre/year
(Wischmeier and Smith, 1978). The accuracy of a predicted soil loss depends on how accurately physical and
management conditions on the particular site are described by the parameter values. Large-scale averaging of parameter
values on mixed drainage areas reduces accuracy.
Traditionally, urban conservation planners have not widely used an equation similar to the USLE for estimating soil
loss and evaluating conservation measures. They have relied on intuition alone to locate erosion control practices on
construction sites. A USLE-based equation provides a valuable, objective method for all planners, regardless of skill, to
tailor specific construction site practices to existing conditions. Erosion control is more efficient when it focuses erosion
control practices in areas on the site identified by the USLE as being the most susceptible to erosion.
207
-------
The objectives of this project were to: 1) develop a method to predict soil loss from construction sites by adapting
existing data for USLE erosion estimation to construction site conditions and 2) create a user-friendly computer-based
program to assist planners in developing construction site erosion control plans with the USLE.
Implementation Area
The project was conducted in Dane County, located in south-central Wisconsin. Dane County has extremely diverse
and vast water resources with 475 miles of rivers and streams and 37 lakes, but these resources are threatened by rapid
urban growth. Within the next twenty years, it is conservatively estimated that an additional 72,000 people will live in the
county. Residents recognize how impacts to water quality affect their standard of living, and are interested in protecting
water resources.
Due to the value that the citizens of Dane County place on water quality, a very restrictive erosion control ordinance
was adopted in 1995. Any land disturbance greater than 4000 square feet must comply with the Dane County Erosion
Control Ordinance (Dane County, 1999). As part of this ordinance, applicants must prove that the erosion rate on their
project will not exceed 15 tons per acre over the construction period for non-sensitive areas. In sensitive areas, including
sites adjacent to, or directly draining to, lakes, streams, and wetlands, the soil loss is limited to 7.5 tons per acre over the
construction period. In order to prove the soil loss rate is below the county standard, applicants need to calculate the
USLE for their site from the start of construction until the site is stabilized. The Dane County Land Conservation
Department reviews erosion control plans for accuracy of the plan and compliance with the ordinance.
Met hods
Adapting USLE to construction site conditions
Our first objective was to develop a method of predicting soil loss from erosion on construction sites based on the
guidelines given by the USDA for the USLE. In order to adapt the USLE to urban conditions, each variable in the equation
was examined (see Equation 1).
The rainfall factor, R, is the first factor modified. Published R values represent erosivity during an average year. Most
construction sites do not remain disturbed for exactly one year. In addition, the time of year that the site is open is critical
in determining the amount of rainfall energy that will occur. In the Midwest, over half of this rainfall energy occurs during
July, August, and September. Projects that take place in the summer will experience higher intensity storms than projects
constructed in the winter. For these reasons, the R factor needs to be adapted to the construction schedule of the project
(Table 1).
Table 1. Percent of R occurring after January T'for Dane County, Wisconsin.
January February March April May June
1st 0 0 2 4 9 20
15'" 0 1 3 6 14 28
July August September October November December
1" 39 63 80 91 97 99
15'h 59 72 87 94 98 100
Once the percent R is calculated for the interval of time that the land will be open, it is multiplied by the annual R
factor for Dane County (150).
R = (% of R to date) x (Annual R factor)
The soil credibility factor, K, represents a soil's ability to resist erosion. The factor is determined by documenting
erosion of a soil in a bare condition on a unit test plot. The higher the erosion rate, the higher the K factor. On construction
sites, the subsoil K factor is often used because the topsoil is usually stripped. Subsoil K factors can be found in USDA
208
-------
Soil Interpretation Records. The soil properties that affect credibility include: soil structure, soil particle size distribution,
permeability, organic matter content, and iron content.
The slope length/steepness factor, LS, relates the length and steepness of the slope (Equation 2). The rate of erosion
increases exponentially as the length of the slope becomes longer. Erosion rates rise even more drastically as the
steepness of the slope increases. The percent slope is a representative portion of the disturbed area, representing
overland flow, not channel flow. The slope length is measured along the flow path from the top to the bottom of the slope
of the disturbed area.
Formula used to calculate the LS factor.
LS = (L/76.6)M(65.41 Sin28+4.56Sin6+0.065)
(Equation 2)
Where: L = slope length in feet
6 = angle of slope (in degrees)
M = 0.2 for slopes < 1%
M = 0.3 for slopes 1 .0 to 3.0%
M = 0.4 for slopes 3.0 to 4.5%
M = 0.5 for slopes > 4.5%
The cover and management factor, C, is based on the type and condition of the cover on the soil surface. In
construction site erosion control, the cover is extremely important. The vegetative cover provides protection from rainfall
impact and runoff water. If the condition of the cover is poor, the C factor will be high. Conversely, when the vegetation
is well established, the erosion and C factor will be reduced. C factors for construction sites can be found in Predicting
Rainfall Erosion Losses (Wischmeier and Smith, 1978). The C factors for seeding, seeding and mulching, and sod
represent the average cover over the establishment period. Once the site is seeded or sod is installed, a period of sixty
days during the growing season is automatically assumed for cover establishment. If the end of the sixty-day cover
establishment period falls after the recommended seeding dates, the calculation must be carried out to the following
spring to allow for adequate growth.
Commonly Used C Factors: Bare ground 1 .00
Seeding 0.40
Seeding and Mulching 0.12
Sod 0.01
The support practice factor, P, is not used to calculate soil loss on construction sites.
The product of the R, K, LS, and C factors equals the computed soil loss per acre over the construction period. In
Dane County, if this number is greater than the required standard, the project must reduce erosion below the standard
by using erosion control practices or by changing the management schedule. This assumes that 100% of soil loss is
transported and deposited off-site for relatively small areas of less than 40 acres with no intervening obstructions or
flattening of the land slope.
Developing the Spreadsheet to Calculate the USLE
Implementation of the USLE in erosion control plans was required for all land-disturbing activities greater than 20,000
square feet in Dane County after January, 1995. The calculation of soil loss was difficult for the consulting engineers
responsible for submitting plans. In addition, the USLE calculations were often done incorrectly or the wrong data were
used as inputs. For these reasons, a user-friendly computer-based program was developed to assist erosion control
planners with the USLE calculation. The program uses Microsoft Excel 97*, a spreadsheet program that is commonly
used among the engineering community.
209
-------
The worksheet (Figure 1) uses the following variables and inputs (Table 2) which are either entered by the user or
automatically calculated in the non-shaded rows.
Universal Soil Loss Equation for Construction Sites
Dane County Land Conservation Department
Developer
Construction Project
Run Dale
6/7/99
L*nd DIMurbtag
Activity
•»
<*
•»•
T
•w
*•
D»t»
%Rto
p*t»
Period %R
Annual R
Factor
SoilM*p
UnK
$0(1 ErotHblHty
K Factor
Hop»*
S
Slop* Lttigth
(f"tl L
Land DMurblng Actlvlll«i Input definition
bare ground acbvily v*itch loaves the ground devoid of veget
mulch with seed application of straw at 1 5 tony acre mlh or with
seeding temporary or permanent seeding without the us
sod instaiation ol sod
end end ot 60 day cover establishment (required inp
LS Fatter
i >
Land C«v«f C
'< -
. **•
TOTALS
Soil lot*
A-%RxR*KxLS)iC
(ten*)»tr»)
*"H H
1 \- * * *
! ! : ! 1
' >• i'.
Porcont Reduction Roqulrod
t« ra**t Ordlnane*
7M tot>*(a«r«
^ 5
•H-S'.,.
if"
NONG
«0»n,,,,,.
? S f
i . -^ <
. H
'' \ ;
NONE
Won
out seeding
of mulch ng rnatenats
ul)
Figure 1. Screen-capture of spreadsheet.
Table 2. Variables used in the spreadsheet.
Column #
1
2
3
4
5
6
7
8
9
10
11
12
13
Variable
Land Disturbing Activity
Date
% R to Date
Period % R
Annual R Factor
Soil Map Unit
Soil Erodibility K Factor
Slope % S
Slope Length L
LS Factor
Land Cover C Factor
Soil Loss
Percent Reduction to Meet Ordinance
Type
entered by user
entered by user
automatically calculated
automatically calculated
automatically calculated
entered by user
automatically determined
entered by user
entered by user
automatically calculated
automatically determined
automatically calculated
automatically calculated
Variable/Input Descriptions:
Land- Disturbing Activity
The land-disturbing activity relates to the type of disturbance that is occurring on the ground and must be selected
by using a pull down menu. Activity Inputs:
*Use of the commercial product name is for the convenience of the reader and does not imply endorsement of the product by either the Dane County Land Conservation
Department or the University of Wisconsin.
210
-------
bare ground Usually the initial disturbance and occurs when the ground is left bare due to stripping
vegetation, grading, or other actions that leave the ground devoid of vegetation.
seeding The application of permanent ortemporaryseeding without the use of mulch. Seeding requires
that the user allows 60-days during the growing season for cover establishment.
mulch with seed The application of a minimum of 1.5 tons/acre straw or other comparable mulching. This input
is entered if the seeding and mulching are done at the same time. It is not necessary to also
enter
seeding if this input is used. This input also requires a 60 day cover establishment period
during the growing season.
sod The installation of sod for cover establishment.
end End is a required input at the end of the 60-day cover establishment period. If the site is
stabilized by a method other than vegetative cover, end should also be entered.
Date
The date the planned land disturbing activity begins, e.g. 5/1 5/99. The activity is assumed to continue until the next
activity is entered. When seeding dates are later than the dates recommended for permanent cover establishment, the
end date must be carried out to the next spring, rather than 60 days.
% R to Date
The percentage of the annual R factor from January 1 st to the entered date.
Period % R
The percentage of the annual R factor calculated for the period from one land disturbing activity to the next.
Annual R Factor
The rainfall factor, R, is the number of erosion-index units in a normal year's rain. The erosion index is a measure
of the erosive force of a specific rainfall. In Dane County, Wisconsin the rainfall factor is 150.
Soil Map Unit
The soil map unit for the predominant soil type in the area of the land disturbing activity.
Soil Erodibility K Factor
The erosiveness factor of the subsoil for the specified soil map unit.
Slope % S
The percentage slope for the representative portion of the disturbed area, representing overland flow and not
channel flow.
Slope Length L
Slope length (in feet) is measured along the overland flow path from the top to the bottom of the slope of the
representative disturbed area.
211
-------
LS Factor
The LS factor is calculated using the equation for LS described previously (see Equation 2).
Land Cover C factor
The cover and management factor is the ratio of soil loss from an area with a specified cover and management
practice to that of a unit plot of bare land. The input for the land disturbing activity corresponds to this factor.
Soil Loss
The predicted value of soil loss (tons/acre) which corresponds to the time period of each land disturbing activity.
This value is calculated using the equation:
A = %R x R x K x (LS) x C
Percent Reduction Required to Meet Ordinance
The percentage value in the total row corresponds to the reduction of soil loss necessary to comply with the Dane
County Erosion Control Ordinance. It is required that the cumulative soil loss rate not exceed 15 tons/acre for non-
sensitive areas and 7.5 tons/acre for sites that are located adjacent to or directly drain to sensitive areas.
Typical Spreadsheet Example for Dane County, Wisconsin
Figure 2 shows a sample USLE calculation using the spreadsheet. The assumptions are that construction will begin
on July 17, 1999, and the site will be seeded and mulched on October 31, 1999. The representative pre-existing slope
is 10% over 100 feet and the slope after grading will be 5% over 250 feet. The soil type is Dresden Silt loam (DsC2). The
estimated soil loss rate for this site is 15.9 tons/acre. If this site is located near a sensitive area, the soil loss must be
reduced by 53% to comply with the 7.5 tons/acre standard; on the other hand, if the site was not located near a sensitive
area, the soil loss only needs to be reduced by 6% (15 tons/acre standard).
Universal Soil Loss Equation for Construction Sites
Dane County Land Conservation Department
PRINT'(
HEU>'",
Developer Sxarrfse csiojiatcr Ph««t — euecT ftk^e \ "un Date
Construction Project SHEET I ,-?£¥* i 7/7/39
version 1 0
Lind Dltturblng
Activity
bareg^xnd •*•
™th w«h soeti *r
end -w
w
•v
••
D)i.
T 7's.S
To
3,-1'w-
TO
TV. ;
URto
Dat*
533%
79.5%
96.8%
262%
173%
32%
Annual R
150
150
150
SoHMap
DSC:
!>' 1
n=< i
Sell Erodlbilty
028
028
0 28
Slop* %
"i j%
r; '-%
') I,1*,
Slop. Lwiflth
;!o
/: )
138
085
,
OB5
' it
UndCov^rC
1 00
0.12
TOTALS
5«H fat*
A-ttRxRxKxL$xC
(WnTfierM
152
0 7
, n f t
15. B
Pm»nt Reduction R«qulr«d
to MMtOrdkiini*
T.B toniJacrt
'. *
••? * '* *
53%
15.D tantfecr*
< -1
'' ]e '-' l
* <\\
6%
Land Disturbing Actlvltl**
bare ground
mulch with seed
seeding
diflnttlor
activity which leaves the ground devoid of vegetsbon
application of slraw at 1 5 tons/acre with or without seeding
temporary or permanent seeding without the use of mulching materials
installation of sod
end of 60 day cover establishment (required input)
Figure 2. Sample USLE calculation.
212
-------
Results and Discussion
There are several advantages to using the adapted USLE for erosion control planning on construction sites. One
advantage is being able to locate areas with the highest erosion rates, which results in more effective erosion control. If
one portion of a construction site is predicted to have a higher erosion rate, more or larger erosion control practices may
be targeted in that area, while less intensive practices may be required elsewhere on the site. The adapted USLE also
facilitates the design of sediment ponds and other erosion control practices. The predicted amount of soil loss exceeding
the standard can be used to calculate the percent reduction necessary to comply with the ordinance.
Another advantage is that the adapted USLE brings in the important element of time. In Wisconsin, the majority
of the year's rainfall erosion occurs during the summer months. Summer is also the time of year that most construction
is occurring. The USLE accounts for the date and duration the development project occurs and predicts the soil's
vulnerability to erosion at that time. The USLE may show that staging the construction project will help to reduce the soil
loss on the site.
The spreadsheet program has proven to be a valuable tool for calculating the soil loss. The program has been
distributed for more than a year, free of charge, to the planners and consultants in Dane County. The County's review
of the calculation in the erosion control plans has become easier and quicker by having a printout that summarizes
the variables used. An advantage of having tables and formulas included in the spreadsheet, is the consistency that is
achieved by everyone using the same parameters. Not only have the calculations of soil loss been more precise and
time schedules more realistic, but planners and consultants have stated that it has saved them time and simplified the
calculation process.
References
Becker, B.C., Nawroki, M.A., and Sitek, G.M., (1974), An Executive Summary of Three EPA Demonstration Programs
in Erosion and Sediment Control, Hittman and Associates, Inc., Columbia, MD.
Dane County Board of Supervisors, Dane County Code of Ordinances (1999), Chapter 14, subchapter II, Erosion Control
System, Dane County Board of Supervisors, Madison, Wl.
Hagman, B.S., Konrad, J.G., and Madison, F.W. (1980), Methods for Controlling Erosion and Sedimentation From
Residential Construction Activities, Wisconsin Department of Natural Resources, Madison, Wl.
Johnson, C.D. and Juengst, D., (1997), Polluted Urban Runoff: A Source of Concern, University of Wisconsin-Extension,
Madison, Wl.
Wischmeier, W.H. and Smith, D.D., (1978), Predicting Rainfall Erosion Losses- A Guide to Conservation Planning,
United States Department of Agriculture, Washington, D.C.
Wisconsin Department of Natural Resources, (1994), The Wisconsin Water Quality Assessment Report to Congress,
PUBL-WR 254-94-REV, Wisconsin Department of Natural Resources, Madison, Wl.
Yorke, T.H. and Herb, W.J., (1976), Urban Area Sediment Yield Affects of Construction Site Conditions and Sediment
Control Methods, United States Geological Survey, Parkville, MD.
213
-------
Public Involvement Programs That Support
Water Quality Management
Josephine Powell
Wayne County Department of Environment
Detroit, Michigan
Zachare Ball
Karen Reaume
Environmental Consulting & Technology, Inc.
Detroit, Michigan
The Rouge River, a tributary to the Detroit River, in southeast Michigan, has been documented as a significant source
of pollution to the Great Lakes System. The Rouge River Watershed spans approximately 438 square miles in 48
communities and three counties and is home to over 1.5 million residents. The eastern portion of the watershed contains
much of the older, industrial areas of Detroit and Dearborn. The western and northern portions contains newer suburban
communities and areas under heavy development pressure.
This paper discusses the programs used by the Public Involvement Team of Wayne County's Rouge River National
Wet Weather Demonstration Project (Rouge Project) to (1) increase watershed awareness of Rouge River Watershed
residents and business owners, (2) educate them about pollution sources to the Rouge River and (3) involve them in
restoration of the Rouge River by showing them that small changes in their daily activities can help improve water quality
and restore the river.
The Rouge Friendly Neighborhood Program was piloted over a two year period in watershed neighborhoods in three
distinctly different areas of the watershed. All neighborhoods were surveyed to determine the initial level of knowledge
about water quality issues, lawn care maintenance, and pollution prevention practices. Survey results were used to
fashion a neighborhood program for each area. All three neighborhoods received Rouge Friendly brochures, newsletter
articles, and other materials.
The Rouge Friendly Business Program, a companion program to the neighborhood effort, sought to educate small-to-
mid-sized businesses about how they can positively impact the Rouge River by making small changes to daily business
practices. Since auto-related businesses are very common in the Rouge River Watershed, an automotive services
roundtable was convened. The partners included representatives of automotive service associations, the local chamber
of commerce, and businessmen who met periodically for a year to review draft materials, make suggestions about the
program's promotion, and to help mold the program before it was implemented. Once implemented, the industry
representatives promoted the program in their publications and recruited businesses to participate in the program.
This paper will describe both of these pollution prevention programs and discuss how the Brightmoor neighborhood
in the Rouge River Watershed was impacted by the Rouge Friendly Neighborhood and the Rouge Friendly Business
Programs.
The Rouge Friendly Neighborhood Program
The Rouge Friendly Neighborhood Program was designed to be carried out by responsible neighborhood
organizations. Preferred prerequisites were:
1. The group participating in the program must represent a defined area or neighborhood.
214
-------
2. The group would participate in the Friends of the Rouge River Watch Program. The river system need not pass
directly through the neighborhood for participation. An assigned segment could be identified for the group by
Friends of the Rouge.
3. The group would participate in the Friends of the Rouge Storm Drain Stenciling Program. The stenciling of storm
drains should include, but is not limited to, all the storm drains within their designated neighborhood or area.
4. The group should actively participate and/or encourage proper household hazardous waste management. This
could occur through:
Reduced purchasing of hazardous house chemicals
Proper use of household hazardous chemicals
Proper disposal of hazardous household chemicals
Use of less-toxic alternatives to household hazardous chemicals
The group can accomplish this requirement by distributing information concerning proper household hazardous
waste management to their designated neighborhood.
5. The group would facilitate education of residents regarding non-point source pollution. Information would be
provided by the Rouge Project Team for distribution to the designated neighborhoods.
6. Submittal of semi-annual reports discussing the activities that have been taking place could be a requirement to
maintain Rouge Friendly Neighborhood status.
Three Rouge Project area neighborhoods representing different demographics and development history were chosen
as pilots for the Rouge Friendly Neighborhood Program. They were the (1) Brightmoor area of Detroit, an older, developed
area of the watershed along the Main Branch of the Rouge River; (2) Golfview Manor subdivision in Dearborn Heights,
a newer subdivision along the Middle Branch of the Rouge River; and (3) West Bloomfield Place, a subdivision in West
Bloomfield Township, a developing area along the Upper Branch of the Rouge River. These three pilots represented
communities with diverse demographics and concerns.
The Brightmoor neighborhood was a deteriorating area with strong community activism regarding neighborhood
problems and concerns. The neighborhood also showed strong stewardship for the Rouge River, which serves as a
western boundary to the neighborhood and flows through a nearby park. Golfview Manor in Dearborn Heights was a more
upscale, manicured neighborhood that was very active through its subdivision association, but did not have a real
connection to the River. West Bloomfield Place in West Bloomfield Township was an upper income, less urban area
bounded by a wetland.
Meetings were held with a core group of representatives from each neighborhood to garner support from the
neighborhoods' leadership and to discuss what the program was and what the expected outcomes were.
The residents of all three pilot areas were sent a survey, distributed by mail or door-to-door, to document their
knowledge of Rouge River water quality, storm water issues, and household hazardous waste disposal. In addition,
respondents were asked for demographic information. The survey information was used to determine what the Rouge
Friendly Neighborhood Program should focus on in each particular neighborhood.
Educational materials that had been developed about storm water pollution, household hazardous waste (and its
disposal), and watershed awareness were reviewed by each group. Each core group helped develop the particular
program that would be implemented in their neighborhood, because Rouge Project staff knew that no program would be
a success without the core groups' support and endorsement. These core group members were relied on to explain the
program at neighborhood meetings.
215
-------
The Rouge Friendly Business Program
The Rouge Friendly Business Program was developed as a partnership between Wayne County, local government,
and the business community to restore and protect the Rouge River. To accomplish this goal, information and assistance
are provided to small business owners to teach them how they can positively affect the water quality of the Rouge River
by changing some of their everyday practices. The education process is not about major contaminants, but those little
things that slip the mind, such as keeping the dumpster lid closed and storing materials under cover. These simple actions
can affect water quality because they stop pollutants from entering the storm system. As an incentive to participate,
Wayne County embraced the concept that businesses in the watershed that demonstrate stewardship and a strong
environmental ethic should be recognized by the community for their voluntary participation. As such, these businesses
should enjoy greater name recognition through the efforts of the local and regional media as well as specific program
materials, such as decals and magnets that identify the business as Rouge Friendly.
The Rouge River Watershed has approximately 42,000 businesses in its 48 communities and three counties. To
design program materials that would have the greatest impact, three criteria were developed to target business types.
They are:
. The business has a high incidence of illicit connections to storm drains
. The business conducts a significant number of pollutant-generating activities outdoors
. The business is found in large numbers in the watershed
Using these criteria, six types of businesses were selected and specific activities identified. They are:
. Vehicle Service Industry
. Food producers, grocers, and eating establishments
. Metal Machining
. Earth Disturbing Construction
. Remodeling and Repair Contractors
. General Business
Pollution control criteria were established for each kind of business. These criteria were used to create a self-
assessment form to be used by business owners to evaluate how "Rouge Friendly" their businesses are. Best
Management Practices (BMPs) were also written that correspond to each activity and this information was put into the
booklet along with a self-assessment form for distribution.
Representatives of various trade organizations were invited to participate in a Vehicle Service Industry Roundtable.
The roundtable was asked to review and comment on the educational materials, the self-assessment form, the BMPs,
and the best way to conduct program outreach. Rouge Project staff sought to engage businesses in an ongoing dialogue
to determine what approach would work best, with a secondary goal of determining how to get businesses to participate.
Feedback from this group resulted in a name change from the "Clean Business Program" to the "Rouge Friendly Business
Program." This was not a quick process, but took approximately six months of meetings to (1) form a Vehicle Service
Industry roundtable, (2) explain the purpose of the Rouge Friendly Business Program, and (3) refine the program and
products. The Rouge Friendly Business Program elements were finalized as follows:
. Self-assessment form and action plan
. Best Management Practices
216
-------
. Site visits by technical staff
. Recognition materials for participating businesses (stickers and magnets)
. Business pledge and newspaper recognition
After these materials were finalized, Wayne County and Rouge Project staff promoted the Rouge Friendly Business
on a pilot basis and recruited businesses through the following mechanisms:
. Business Roundtable contacts
. Letter and telephone contacts
. Door to door contact with businesses
. Contact through homeowner/neighborhood associations
. Integration of Business and Residential Programs
By the end of the pilot period, the Wayne County Department of Environment had recognized 20 businesses as
Rouge Friendly.
The Brightmoor Community Pilot
The Brighmoor area of Detroit was developed in the 1920s as a neighborhood forworking-class families. Most houses
are frame, with the newer areas of the neighborhood (1940s and 1950s) of brick construction. Over the past 15 years,
the Brightmoor area has deteriorated. Its once vibrant business strip is dotted with boarded, vacant buildings, graffiti,
trash, and debris. Whole blocks of residential land are vacant and overgrown and illegal dumping is abundant.
Environmental abuses ranged from a myriad of abandoned vehicles to illegal car repair businesses on residential streets.
The Rouge Friendly Neighborhood Survey (Attachment A), distributed in Brightmoor in 1996, showed that the top two
environmental concerns in the area were illegal dumping and abandoned housing.
Despite these challenges, the Brightmoor neighborhood had two characteristics that made it a viable pilot for the
Rouge-Friendly Programs. One, Eliza Howell Park, located on its western edge, was traversed by two branches of the
Rouge River. Second, Brightmoor had a wealth of grassroots organizations who were working to make the neighborhood
better. Some annually removed log jams and other debris from the Rouge River in Eliza Howell Park during Rouge
Rescue, sponsored by Friends of the Rouge, a grassroots organization serving the whole watershed.
Initial contact was made with the Brightmoor Concerned Citizens and other neighborhood representatives in January,
1996. The group agreed that they would like to participate as a Rouge Friendly Neighborhood pilot. A month later, the
same group met again with Rouge Project staff. This time, city parks staff were present. They were told about the
possibility of grant funding for storm water projects by the Rouge Project. The group brainstormed the kinds of things
they would like to see happen at the park, which had suffered from spotty maintenance. They agreed that they would
like to see wildflowers and prairie grass planted, nature trails restored, and a community garden created. The parks
department later applied for and was awarded a $180,000 grant to plant wildflowers and prairie grasses and to install
nature trails in the lower end of the park, near the Rouge River.
The next step in the program was to survey residents about their knowledge of pollution entering the river and
household hazardous waste disposal, their neighborhood environmental concerns, and demographic information. The
survey was created with input by the core neighborhood group. The major data extracted from the survey were:
. 78% thought the Rouge River was polluted or very polluted, and 20% thought the river was getting worse.
. 38% did not know that the storm drains lead directly to the Rouge River. However, 56% understood that sanitary
sewers go to the wastewater treatment plant.
217
-------
. 18% correctly answered that industry pollutes the Rouge River the least and 66% thought stormwater pollutes the
least.
. 87% maintained their own lawn. There was an even distribution among those who never fertilize their lawns and
those who fertilize I-2 times per year.
. 75% did not change their own motor oil.
. 80% took their cars to a car wash instead of washing it themselves.
. 92% claimed indicated that they know what household hazardous waste is, and 73% correctly identified motor oil
as a household hazardous waste. However, 54% did not properly dispose of their wastes.
. 83% said they were committed/very committed to make small changes to prevent pollution.
Following are neighborhood issues, in order of importance:
1. Abandoned buildlings
2. Illegal dumping
3. Household hazardous material disposal
4. Infrequency of street sweeping and storm drain cleaning
5. Recycling
6. Do-it-yourself car repair/illegal car lots on residential streets
7. Overuse of garden/lawn pesticides
8. Overuse of fertilizer
9. Composting
Wayne County Rouge Project staff, Friends of the Rouge, and Brightmoor Concerned Citizens leadership made a
presentation, including survey results, to the general membership in May, 1996. The general membership was
enthusiastic about the program. The annual Rouge Rescue held in Eliza Howell Park on June 1,1996, was expanded
to include other activities, including storm drain stenciling, a tour of a newly constructed combined sewer overflow basin,
and children's games.
Subsequent meetings with the Brightmoor group were used to brainstorm what the specific program elements should
be and what outcomes were expected. The following elements were supported by the core group:
. Urban gardens on vacant lots
. Composting education
. Attempting to get rid of the massive log jam at the confluence of the Upper and Main Rouge River in Eliza Howell
Park
. A tour of the area for the Detroit Environmental Court judge
. Lawn signs that read "I support the Rouge Friendly Neighborhood Program"
. Early recognition of well-maintained lawns and gardens; Brightmoor's "Resident of the Month"
218
-------
. A Brightmoor Rouge-Friendly Business Program
. A renters' workshop to educate tenants about their rights and responsibilities and the responsibilities of landlords
. Educational materials in the various neighborhood newsletters
All of the activities were implemented except the lawn signs, the recognition of well-maintained gardens, and the
renters'workshop. By the fall of 1996, the focus had shifted to conducting a monthly combined resident/business owners'
meeting to include businesses, which were primarily vehicle service oriented, into the Rouge-Friendly initiative.
Results and Outcomes
Rouge-Friendly Neighborhood Program: Because of the enthusiasm and commitment of Brightmoor residents and
business owners, many activities were conducted. They were:
. Thousands of educational brochures and children's materials were distributed in community centers, businesses,
schools, and newsletters.
. A local business owner successfully sued a public utility that was pumping hundreds of gallons of polluted
stormwater into the Rouge River.
. The local community organization not only enthusiastically participated in the annual Rouge Rescue event, but
conducted another such event on its own.
. Through a partnership with the city parks department, the Greening of Detroit, and the Brightmoor Concerned
Citizens, 100 trees were planted in Eliza Howell park by 200 local elementary school students.
. City officials agreed to conduct an environmental ticket blitz in the neighborhood, which resulted in the following
tickets being written: 179 parking tickets, 71 abandoned cars tagged for removal, 8 stolen cars being towed, 15
public works tickets for bulk garbage being put at the curb too early, 2 environmental protection tickets, and 47
tickets for inoperable vehicles.
. A monthly meeting that included neighborhood residents, business owners, non-profit organizations, city officials
(including police commanders), and county officials focused on environmental issues.
Rouge-Friendly Business Program: While many outstanding initiatives were accomplished by meeting with the
Brightmoor stakeholders monthly, only one Brightmoor business was recognized as Rouge Friendly after ten months of
monthly meetings. The meetings were well-attended and business owners felt comfortable discussing their environmental
concerns. In June, of 1997, the approach for recruiting Brightmoor businesses as Rouge Friendly was changed. A
community leader was paired with a technical staff member and they proceeded to visit neighborhood businesses. They
visited 14 neighborhood businesses several times over a two-month period. Information about the program was left with
the business owners, as well as an offer from the technical staff member to help the business owner with the self-
assessment form. Through this effort, Wayne County recognized six additional Brightmoor businesses as Rouge-
Friendly. This was a successful (43% participation), but labor intensive, method of recruiting businesses.
219
-------
ATTACHMENT A
ME ROUGE RIVER PROJECT
A WORLD CLASS EFFORT
BRINGINGOlMIVEFBACKTaifE
Are you willing to prevent pollution
in your neighborhood?
Start today by filling out this questionnaire.
We are working with the Rouge River National Wet Weather Demonstration Project to make our
subdivision a more attractive place to jive. We have been chosen as one of three pilot neighborhoods in the
Rouge River Watersheds to participate in a pollution prevention program that may be used as a model for other
urban watersheds across the country. In order to design a program that best fits our needs, we need you to
answer a few questions. The following survey is voluntary and confidential. Use the enclosed pre-stamped
envelope to return the questionnaire by April 22, 1996.
2.
3.
Very polluted Somewhat polluted Not polluted
Do you think the Rouge River is polluted? 54321
Getting cleaner Staying the same Getting worse
Do you think the Rouge River Is getting 5 4 3 2 f
cleaner.staying the same, or getting worse?
Q To a storage tank under the ground
rj To the Rouge River
rj To the waste water treatment plant
n Don't know
Where does water go when it enters an
outside storm drain in your neighborhood?
Where does water go when it
is flushed down the toilet or sink?
5. What pollutes the Rouge River the LEAST?
6. How do you maintain your lawn?
7. How often is your lawn fertilized?
a. Where do you change the oil
in your car?
9. Do you usually wash your car
or take it to a car wash?
10. If you change your oil at home,
how do you dispose of it?
11. Do you know what household
hazardous materials are?
12. Which of the following is a
household hazardous material?
rj To a storage tank under the ground
BTo the Rouge River
To the waste water treatment plant
Don't know
a
a
a
a
B
Combined sewer overflows
(a mixture of sewage and stormwater that
flows into the river when it rains.)
Stormwater (water that runs off
the grvund and enters the river)
Industry
Paid professional company
Paid neighbor
Someone in the household maintains it
U 1 to 2 times per year
rj 3 to 4 times per year
rj More than 4 times per year
rj Never
B
Auto Repair Shop/Quick Oil Chnage
Yard rj Street Q Driveway
n Wash it myself
Q Carwash
d In the garbage
a In the sewer
Don't know
n Yes
n NO
Q Baking soda
rj Motor Oil
rj On the ground
fj Take to facility that
accepts used oil
rj Lemon oil
Q Vinegar
220
-------
13. How do you dispose of your 0 Put it in the trash Q Dump it down the sink
household's hazardous materials? rj Dump it on the ground Q Don't know
fj Take it to a Household Hazardous Drop-off Area
14. Indicate whether the following environmental issues are
very important, important, or not important to your subdivision.
Very important Important
Overuse of fertilizer 5432
Composting 5432
Abandoned buildings 5432
Frequency of street sweeping 5432
and storm drain cleaning
Overuse of garden/lawn pesticides 5432
Recycling 5432
Household hazardous waste 5432
Illegal dumping 5432
Do-it-yourself car repair / Illegal 5432
car lots on residential streets
Other (soecifv: ) 5432
15. You can make small changes to prevent pollution (i.e. the type of fertilizer you purchase,
pose of your motor oil, etc.). What is your level of commitment to make these changes?
Very committed Somewhat committed
5432
16. How many people, including yourself, Q ^ Q 5
live at this address? Q Q 6
Q 3 n More
D 4
17. How many of these are children? D 0 D 3
D 1 D 4
rj 2 a More
18. How many pets do you own? Ej Q 3
n o n 4
Q 2 Q More
19. What is your gender? rj Female • : Male
20. What is your age group? Q Under 18 years n 46-60
Not important
f
f
1
1
1
1
1
1
1
1
how you dis-
Not interested
I
than 6
than 4
than 4
years
rj 18-30 years Q Above 60 years
O 3 145 years
21. What was the last grade rj Some high school
you completed in school? rj Completed high school
Q Post-high school training
Q Some college
Q Completed college
l_In Graduate or professional school
Thank you for doing your part in cleaning up our subdivision! Remember, return
in the pre-paid envelope by April 22, 1996.
Any questions should go to John or Shelley Mlynarczyk at 533*3453
the questionnaire
221
-------
The Water-Wise Gardener Program:
Teaching Nutrient Management to Homeowners
Marc T. Aveni
Virginia Cooperative Extension
Manassas, Virginia
Introduction
The Water-wise Gardener program was developed by Virginia Cooperative Extension (VCE) seven years ago with
funding from the Cooperative State Research, Education, and Extension Service (CSREES) of the U.S. Department of
Agriculture (USDA). It is an educational program aimed at reducing non-point source pollution from suburban residential
areas. The educational focus is upon nutrients, especially nitrogen and phosphorus from lawn fertilizer over-application
or misuse. The program seeks to reduce such nutrient pollution to Virginia waterways, and eventually the Chesapeake
Bay, through the recruitment of homeowner participants from impaired watersheds. Participating homeowners attend
educational seminars on lawn best management practices, are partnered with a Master Gardener volunteer, and are
expected to keep accurate records and implement recommended practices. The program, which is currently being
implemented in 12 urban/suburban Virginia counties, is supported by a combination of local county funds, grants from
the Virginia Water Quality Improvement Act, and funds from USDA. Cooperative Extension Units in North and South
Carolina have replicated the program.
How the Program Works
The Water-wise Gardener program begins by recruiting homeowners from watersheds with impaired streams or
other identified problems to participate in a year-long lawn care educational program. The most successful recruitment
method to-date has been to conduct a "reverse search" on the Internet by street name. Once names and addresses are
identified, a recruitment letter is sent personalized for the watershed; e.g., "Dear Resident of the Bull Run Watershed."
The letter invites the homeowner to participate in the program and lists the benefits of participation, such as free
seasonal seminars with regional experts, visits from a Master Gardener (volunteers trained by VCE in various aspects
of horticulture), a free soil test, and Virginia Tech publications. In order to be enrolled in the program, the homeowners
must return a completed pre-survey and a signed agreement form that details their obligation to the program. The pre-
survey asks questions about their lawn care practices and attitudes before program involvement, as well as demographic
information such as race, gender, income, and education levels. A stamped, self-addressed envelope is included for
ease of return. For every 100 letters sent out, between 20-30 are typically returned. A simple database program keeps
track of participants and their lawn care data. A reporting system on the Virginia Tech Intranet is currently being
designed to record this information on a statewide basis by hydrologic unit.
Once enrolled, participants are assigned a personal Master Gardener. The Master Gardener schedules a visit with
the homeowner to discuss his or her lawn. All Master Gardeners are instructed to stay outside on the lawn and not to go
inside anyone's home. Some choose to bring along a spouse, friend, or another Master Gardener. At this visit, the
Master Gardener works with the homeowner to correctly measure the square footage of lawn area, determine the type
and variety of grass, collect a soil sample, and ascertain previous fertilization practices and amounts previously applied,
if known. This information forms the basis of a personalized lawn care plan for the homeowner. Master Gardeners also
answer other questions the participant may have; common questions include weed and pest identification, what plant
grows best where, and why certain plants are not thriving. The Master Gardener leaves a business card with a phone
number or e-mail where he or she can be reached for further questions throughout the program year. All Master
Gardeners receive 50 hours of classroom training as well as supplemental field training before being assigned to
homeowners. Typically, a Master Gardener will be assigned to between 5-I 0 homeowners.
222
-------
In addition to the one-on-one visits from Master Gardeners, homeowners attend seasonal seminars on timely topics
of interest to those with lawns. In Northern Virginia, where cool-season grasses like Fescue and Bluegrass predominate,
fall topics include soil testing, fertilization, core aeration, and over seeding. Spring topics include mowing and pruning,
integrated pest management, and proper watering and planting. Popular locations for seminars include parks with
covered pavilions, school auditoriums, county buildings, and libraries. Any easily accessible public location large enough
to hold 50 to 100 people comfortably, and accessible to wheel chairs, will work. If held inside, cold temperatures, rain,
or wind are not a problem; however, an outside area for demonstration purposes is essential. State and regional
Cooperative Extension experts are recruited for the seminars to answer questions. Master Gardeners are also present,
with various displays, to answer questions and to meet with their assigned participants.
A professional-quality newsletter is sent to all participants approximately six times per year. A grant-paid editor
solicits articles that reinforce or complement topics taught at the seminars. Articles on various aspects of watershed
management are also introduced. The newsletter is made available electronically to other Extension Agents for editing
and reproduction elsewhere.
After participants have attended fall and spring seminars, they are visited again by their Master Gardener to collect
final lawn data and conduct a post-survey of practices and attitudes. The most important piece of data collected is the
amount of fertilizer now being applied. Square footage of turf can be re-checked, if needed, and questions answered.
The homeowner may chose to participate again the following year, or to offer their lawn as a demonstration lawn,
and erect a sign in their yard to promote the program in the community. The post-surveys and data sheets are collected
from all participants annually. Results are compiled and analyzed by a grant-paid technician and a final report generated
for each Cooperative Extension unit as well as an overall report for statewide efforts.
Results
Data for the period March 1998 to June 1999 for the Virginia counties of Arlington, Loudoun, and Prince William
shows 326 individual homeowner participants. These 326 homeowners managed 57.1 acres of turf in 11 different
hydrologic units in the Northern Virginia area. Between 100-200 additional individuals attended seminars but did not
participate in the pre- and post-survey and data collection.
Accurate information on amounts of nitrogen and phosphorus applied by participants before program involvement
is difficult to get. Most did not remember how much fertilizer they had applied in the previous year. Many stated the
reason they joined the program was in order to understand how much fertilizer to apply. A total of 72 participants
reported pre-program fertilizer application of 1,062 pounds of nitrogen. The same 72 participants reported 762 pounds
of nitrogen applied after program involvement, or a reduction of 300 pounds. Information on pre- and post-phosphorus
was not collected.
Pre-surveys indicated that only 12% of all participants had soil tested for their lawns prior to applying fertilizer.
Homeowners not testing soil are more likely to apply excess fertilizer. For this reason, Virginia Tech recommends soil
testing as a nutrient management practice for home lawns. Post surveys show 95% of participants returning surveys
tested soils after program involvement. Another important nutrient management practice for homeowners with cool-
season turf is to fertilize in the fall, when uptake by roots occurs best. Pre-surveys indicated that only 32% were fertilizing
at this time of year, while post surveys indicated that 64% were fertilizing in the fall. Similar increases were also observed
for recommended practices such as aeration (from 34% to 83%), and over-seeding (from 35% to 76%). An increase in
the number of participants not watering the lawn at all in the summer also increased (from 18% to 44%) (Figure 1).
Demographics from the program indicate that 72% of participants were male and 28% female. Participants were
overwhelmingly white (89%); followed by black (7%), Asian (4%), and Hispanic (1%). The majority (42%) had a four-year
college degree and a gross family income of over $70,000 a year (54%). More than one-third of the lawns were between
5,000 and 10,000 square feet (35%).
223
-------
Pre-program D Post-program
w 100 p=r
c | 80
o o 60
CD "•= AC\
fT CO ^«
°-^ 20
° 0
Figure 1. Pre and Post Program Practice Adoption
Figure 1. Pre and post program practice adoption.
Conclusions and Lessons Learned
The Water-wise Gardener Program was successful at reaching the intended audience and achieving adoption of
nutrient management practices. Based on the success of the program, it appears that suburban homeowners can be
recruited to maintain their lawns according to recommended practices. Homeowners are willing, with the help of Master
Gardener volunteers in some cases, to keep records on their nutrient use as a part of program participation. Although
326 individuals and 57 acres of turf may seem low for an area like Northern Virginia, it is significant for a populace that
does not traditionally participate in water quality educational programs. Considering that most lawns in suburban
subdivisions have a turf area of around 5,000 square feet, clearly many individuals will need to be enrolled to reach
meaningful numbers.
The study showed that it is difficult to obtain information on pre-program nutrient use for most participants. Most
homeowners cannot provide accurate nutrient use data from the previous 6-12 months. They simply do not remember
how much nitrogen and phosphorous was in the fertilizer bag applied last spring or fall. However, after program
involvement, they do appear to understand how much nitrogen they applied and the square footage of their turf. From
a water quality public policy perspective, it may be preferable to record nutrient use after program involvement and
consider participants' turf square footage as the urban nutrient management measurement. In this way, the focus could
be upon recruiting more and more individuals to participate in nutrient management educational programs like the Water-
wise Gardener, thus increasing the number of acres addressed by urban nutrient management efforts. Such an
approach could easily be integrated into local Geographic Information Systems, providing localities a simple method of
accounting for and reporting on urban nutrient management. Localities interested in a program like the Water-wise
Gardener should contact their local Cooperative Extension office to see if a similar project is already occurring or could
be developed. As this program is being continued and expanded in the 1999-2001 time frame, the opportunity to better
define what is realistic as an urban nutrient management measurement for homeowners will hopefully occur.
224
-------
Chicago Wilderness; Toward an Urban Conservation Culture
John D. Rogner
U.S. Fish and Wildlife Service
Barrington, IL 60010
Chicago Wilderness - Origin and Purpose
We in the Chicago conservation community have been using the word "wilderness" in a highly unconventional context
since 1996. We have coined the term "Chicago Wilderness" to refer to the rich biodiversity which resides in and around
this huge, sprawling metropolitan area, extending from southeast Wisconsin, through the six-county metropolitan area
in Illinois, and around Lake Michigan to northwest Indiana. This is a region which most people think of as anything but
"untrammeled by man, where man is a visitor who does not remain," in the words of the Wilderness Act, which has
defined our modern concept of wilderness.
This is an area that is associated with-indeed, defined by-humans and their cultural footprint. Although the
"wilderness" is scattered throughout the region, mostly in parcels that would be considered slivers of land by conventional
wilderness standards, it totals over 200,000 acres of land protected within a complex of national tallgrass prairie, national
lakeshore, county forest preserves, city and township parks, and similar preserved public lands. Its protected lands and
waters range from half-acre remnants to the 15,000-acre Midewin National Tallgrass Prairie,
Within this system of preserves can be found some of the largest and best woodlands, wetlands, and prairies in the
Midwest. These lands are set in a much larger matrix of public and private, developed and undeveloped lands that
support nature and the region's 8 million people.
We have called these lands "wilderness,"in part to draw the attention of people who are focused on Chicago's cultural
attractions to the existence of these lands in their own metropolis, and in part to deliberately blur the distinction, or
conversely, emphasize the connections between formal wilderness in remote and inaccessible places and wild lands in
the places where people live and work. The biotic connections exist on the land, and they ought to exist in people's
minds, as well.
The boundaries of the Chicago Wilderness region do, in fact, capture a spectacular concentration of rare ecosystem
types. These ecosystems harbor a high diversity of species, including a large number of those listed as threatened or
endangered in the states of Illinois, Indiana, and Wisconsin. Outside of the metropolitan area, particularly in rural Indiana
and Illinois, diversity decreases sharply as agriculture dominates the landscape.
"Chicago Wilderness" is also the name we have given the collaboration of over 90 organizations in the Chicago region
that have banded together to better protect, restore, celebrate, promote, and publicize our rich biodiversity. An
unfortunate and perhaps somewhat inevitable consequence of urban life is a detachment from the land; thus, a principal
goal of the partnership is to reconnect a landless urban population, in Aldo Leopold's words, to the "raw material out of
which we have hammered that artifact called civilization."
Despite the richness of nature and opportunity for conservation in the region, evidence suggests the Chicago region
is experiencing a decline in native species and communities. Prior to protection, much of the region's current base of
protected land was subject to agriculture, drainage, and other human influences which reduced or eliminated native plant
and animal communities. These areas are often fragmented and isolated from healthy lands which could otherwise serve
as immigration sources for native species.
225
-------
Small fragments are also subject to influences beyond their boundaries, such as urban runoff. Keystone species like
wolves and mountain lions, predators which formerly kept prey species like whitetailed deer in check, have been
eliminated, and deer now threaten to destroy some of our finest lands through overbrowsing. Many of our protected sites
are too small to sustain populations of area-sensitive species, or to retain their full complement of species in the face of
random population processes like immigration and emigration. Exotic plant and animal species pose major threats to
nearly all of our native communities. Landscape level processes, like fire, that shaped the fundamental character of our
ecosystems do not occur with the frequency or to the extent they once did, resulting in shifts in community composition
that usually result in a decline in biodiversity.
Chicago area residents are the beneficiaries of farsighted leaders early in the 20th century who established a tradition
of setting aside natural land in the urban matrix for the public good, a tradition that our forest preserve districts continue
today. The early model was not based on sophisticated concepts of biodiversity conservation, or of ecological processes,
but on the museum approach of setting nature aside and not meddling. We now are the beneficiaries of the science of
ecology, which begins to tell us how the land mechanism is constructed and how it operates. It is dynamic, not static,
and changes occur when landscape processes are interrupted. The science of ecology also reinforces the connections
between humans and the rest of nature.
This allows us to reexamine the old model of setting nature aside and leaving it alone. That removes the most
immediate threat of development, but it does not address the aforementioned degenerative loss of biodiversity due to
fragmentation and alteration of landscape processes. These processes clearly must be reintroduced into our preserves
if biodiversity is to be preserved or restored. Prescribed fire must be intelligently applied, invasive species must be
controlled, plant and animal species must be reintroduced where they have been eliminated, hydrology must be restored
where altered, and science must be improved where our understanding of ecosystem processes is deficient. Perhaps
most fundamentally, the people who must support the greatly increased levels of land management and research
necessary to restore and maintain our public lands in a healthy condition must have a basic understanding of land health
and the value system to commit public resources toward attaining it. Chicago Wilderness, the coalition, is committed to
working on all of these fronts.
Quite understandably, the Chicago region's system of public lands was, and perhaps still is, the core of Chicago
Wilderness, the initiative. It is what members rallied around during the coalition's formation in 1996. But the vision quickly
expanded beyond public lands, for two reasons.
First, our public lands do not exist in isolation. They are part of a much larger land base, and the protected 200,000
acres are affected by what happens on the remaining 6 million acres of the Chicago metropolitan land area. The
preserves form the core, but they cannot preserve all the biological parts by themselves since much biodiversity resides
on private unprotected land and because they are subject to outside influences.
Biodiversity considerations need to infuse all of the region's land use decisions much more extensively that they do
now. Private lands work either in harmony or discordantly with our network of preserves. The link between the two is most
apparent in the case of wetland or aquatic habitats, which in many cases are sustained or impacted by runoff from distant
areas. Streams, rivers, lakes, and wetlands defy the "protect by fencing" approach. Overall watershed characteristics
determine aquatic and wetland habitat quality quite independently of whether the habitat is in a formal preserve or not.
Second, many high-quality, biologically rich pieces of nature persist outside of our preserve system and are
threatened by development, along with other stresses like lack of management. Identifying these biologically important
areas within proposed developments, redflagging them, designing development with their sustainability in mind, and doing
all this with equity for the landowner, is one of our greatest challenges. Nature in the places where we live contributes
so much to quality of life, yet maintaining it through the development process resists standard regulatory approaches.
There are questions now asked routinely in the subdivision design process: does the plan conform to drainage code, are
storm water basins sized properly, is it consistent with surrounding development, does it have proper standards of
landscaping? A standard question should be: does it leave the land biologically richer or poorer? We are not yet routinely
asking this question, although there are development approaches available that can allow us to answer this question
affirmatively.
226
-------
This question could properly be asked for aesthetic reasons alone, but there are practical reasons for doing so.
Native landscapes hold enormous potential for managing storm water and preventing flooding. They hold enormous
potential for cleaning up surface waters so that urban waters become fishable and swimmable, instead of the
neighborhood joke or eyesore. Finding and applying the template for development that preserves and restores biological
diversity, and which serves both aesthetic and utilitarian purposes, is one of the objectives of Chicago Wilderness.
Chicago Wilderness - Structure and Function
Chicago Wilderness formally began as an initiative with the signing of a memorandum of understanding (MOU) by
34 founding members. Members included landowners and land managers; local, state, and federal agencies; centers for
research and education; and conservation organizations, among others. These institutions pooled their resources and
strengths to form the Chicago Region Biodiversity Council, which has grown to include nearly 100 members.
By signing the MOU, the members of this innovative partnership have pledged a commitment to the protection,
restoration, and management of biodiversity in the Chicago region. Four teams focus on central lines of action: science,
land management, policy and strategy, and education and communication. The teams attract the participation of many
non-member institutions, which adds to the scope and strength of the coalition. Chairs of the teams and other member
organization staff form the nucleus of a coordinating group that develops central strategies and maintains momentum.
A steering committee of executives oversees the direction of the overall initiatives. Despite this organizational structure,
Chicago Wilderness has not become legally incorporated under state law, but remains a loose partnership bound by
common goals and objectives.
The potential for Chicago Wilderness to serve as a model for urban conservation attracted the early attention of
several federal agencies, including the U.S. Forest Service, U.S. Fish and Wildlife Service, and U.S. Environmental
Protection Agency, who have provided significant operating grants. State and private grants have supplemented federal
dollars. Direct grants have totaled over 4 million dollars since 1996. This total does not include members' matching funds
or funds attracted by members for projects catalyzed, but not directly supported by, Chicago Wilderness.
Chicago Wilderness Accomplishments
The Chicago Region Biodiversity Council funds projects on an annual cycle. The Council's four teams set priorities
for these projects; core staff ensure broad participation from team members. Reviewed and approved by a proposals
committee, funded projects result from collaboration between member institutions and address critical conservation needs
in the region. Since its launch in April 1996, Chicago Wilderness has funded over 130 collaborative projects. In addition
to projects funded directly by the Council, the work of our individual member organizations in their own initiatives is central
to the success of Chicago Wilderness. Projects completed or underway fall into six categories: characterization and
information management; ecological inventory and monitoring; ecological restoration; planning and policy; education,
outreach, and public participation; and communications and publications.
Individual projects have included a NASA-supported land cover mapping project; development of models of pre-
settlement savannas, woodlands, and forests to guide restoration; assessment of restoration effects on bird communities;
a vegetation monitoring workshop; assessment of garlic mustard impacts on native woodland ground flora; development
of model restoration interpretive programs; a biodiversity educators workshop; and creation of a Chicago Wilderness Atlas
of Biodiversity. An early pilot project supported by Chicago Wilderness was the launching of Chicago Wilderness
magazine, a glossy, popular publication on nature in the Chicago area which since has been incorporated as a 501 (c)(3)
and has over 7,000 paid subscribers.
Chicago Wilderness Biodiversity Recovery Plan
In 1909, the Commercial Club of Chicago released the "Burnham Plan," a landmark of urban planning that proposed,
among other things, a network of public parklands to be set aside for nature and passive recreation. This led to the
legislative establishment of a system of such publicly owned preserves for the Chicago region which has continually
expanded, and now forms the core of the protected lands that currently comprise Chicago Wilderness.
227
-------
The Biodiversity Recovery Plan, completed in late 1999, takes the open space component of the Burnham Plan to
the next step by creating a vision of sustainability, not only for the core of protected land, but for all of nature and its
human inhabitants in the urban area. The recovery plan is a comprehensive statement of what Chicago Wilderness is
about, and it is clearly the most ambitious and significant accomplishment of the coalition to date.
This plan is the result of three years of assessment and planning by representatives of the Chicago Region
Biodiversity Council. The plan identifies the ecological communities of the greater Chicago region, assesses their
condition, identifies major factors affecting them, and provides recommendations for actions needed to restore and protect
them into the future in a sustainable condition. In short, the recovery plan outlines the steps necessary to achieve the
overall goal of the Chicago Wilderness collaboration, which is to protect the natural communities of the Chicago Region
and to restore them to long-term viability, in order to enrich the quality of life of its citizens and to contribute to the
preservation of global biodiversity.
To achieve this goal, the recovery plan identifies the following objectives: 1) involve the citizens, organizations, and
agencies of the region in efforts to conserve biodiversity; 2) improve the scientific basis of ecological management; 3)
protect globally and regionally important natural communities; 4) restore natural communities to ecological health; 5)
manage natural communities to sustain native biodiversity; 6) develop citizen awareness and understanding of local
biodiversity to ensure support and participation; 7) foster a sustainable relationship between society and nature in the
region; and 8) enrich the quality of the lives of the region's citizens.
The plan has many recommendations, some specific and some general, and identifies roles and specific actions for
Chicago Wilderness members and the greater public that must be engaged to help implement the plan. The plan's
intended intended audiences include the many staff members and general members of Chicago Wilderness institutions,
publicagencydecision-makers, large landowners, and all concerned and active citizens who vote and otherwise influence
biodiversity conservation in the region.
The recovery plan is both a plan and a process guided by its many sponsors. It is intended as a living document that
will continue to evolve as new ideas and information arise. It is intended to complement the many other planning efforts
completed or underway in the Chicago metropolitan area that are guiding the region to a better and more productive
future. Its ultimate success probably rests on its successful integration into a broader, mainstream regional planning
framework that has economic, cultural, social, and environmental components.
Strategic Visioning
After the second year of operation, the Biodiversity Council saw the opportunity to step back and evaluate the
structure and function of the coalition during the first two years, consider expectations of members at the outset and
evaluate to what extent they were met, and reprioritize its work for the next two years. This process consisted of
development of a member questionnaire, convening of a focus group representing a cross-section of members, and a
weekend retreat by Chicago Wilderness Steering Committee members and other leaders. It culminated with the
development of six priority functions for the next two years, and associated budget requirements.
Some of these functions represent an intensification and refinement of activities the Council is already involved in;
in other cases, they represent new endeavors. They include 1)facilitate networking among Chicago Wilderness members,
including new orientation materials, workshops, symposia, and lectures; 2) establish an integrated information
clearinghouse, including the development of regionwide resource databases, enhancement of the existing web site, and
development of more communication resources; 3) increase publicity and outreach to broader audiences; 4) influence
key actors outside Chicago Wilderness, including the establishment of a Conservation Policy Committee to develop
position statements on regional issues; 5) develop and implement a funding strategy, focusing on large grants from
foundations; and 6) implement, promote, and monitor the Recovery Plan.
228
-------
The Urban Conservation Culture
Conservation efforts in urban areas are often frustrated by the complexity of land use issues, countless players,
tangled politics, ecologically wrecked land, and a public dispossessed of nature. Yet it is crucial that we focus on urban
areas because of the strong political forces concentrated in urban centers that need to be engaged in national
conservation decision-making, and because there is no other way to engage the great majority of people other than to
take the messages to them. Moreover, urban residents are still plain members of Leopold's land community, regardless
of how obscure the connections, and these connections are best illustrated in the places where they live. Fortunately,
the Chicago region has an added bonus of harboring world-class biodiversity, which creates a local, immediately
compelling reason for public involvement and action.
Some writers have argued that the American ideal of wilderness has tended to shape our dominant view of nature
itself as a place that can only be corrupted by human influence. In urban areas, this has created an assumption that "real"
nature cannot exist in these places and it tends to absolve urban residents from local responsibility. Thus, it seems that
Chicagoans are much more aware of the plight of Brazilian rain forests that they are of the plight of oak savannas, a
globally rare community, in local forest preserves. In remote areas, the standard approach has been to specifically
designate areas as wilderness, and then maintain as complete a separation between people and these areas as possible.
Chicago Wilderness proposes to redefine wilderness to include local plant and animal communities, which can only be
sustained through direct, creative human intervention. A premise of the recovery plan is that if we do not adequately enlist
people to directly or indirectly support management and restoration of our lands, they will not become or remain healthy.
It is appropriate to recognize that humans in the Midwest always have influenced landscapes, for better or worse, and
that people can be a positive force in maintaining ecosystem health. It may be that by calling a 200-acre patch of prairie
in a sea of development wilderness, and by involving people in its stewardship, we can promote a correct sense of unity
between the places that we live and remote places we may never see except as pictures on calendars. Restoration and
stewardship can be the antidote to dualistic thinking. Remote wilderness and Chicago Wilderness can perhaps then be
seen as simply examples of nature, as part of a single system that includes people.
From a relatively straightforward beginning that focused on public land management issues, this, I think, has become
the broader goal of Chicago Wilderness-to reconnect people with nature and to make a societal commitment to sustain
and nurture nature-for utility, for aesthetics, for spirituality, for all of the equally valid reasons for doing it, on all of our
urban lands and in all of our land-use decisions. It begins with a process of educating the public about the natural wealth
in the Chicago area, and hopefully ends sometime in the future with the development of an urban conservation culture
of concern and personal responsibility for the health of all of our lands, both public and private.
229
-------
A Survey of Resident Nutrient Behavior in the
Chesapeake Bay Watershed
Chris Swann
Center for Watershed Protection
Ellicott City, Maryland
In recent years a handful of communities have attempted to craft education programs to influence our watershed
behaviors. These initial efforts have gone by a confusing assortment of names, such as public outreach, source control,
watershed awareness, pollution prevention, citizen involvement, and stewardship, but they all have a common theme -
educating residents on how to live within their watershed.
Many communities will need to develop watershed education programs in the coming years to comply with pending
EPA municipal stormwater NPDES regulations. Indeed, half of the six minimum management measures prescribed under
these regulations directly deal with watershed education - pollution prevention, public outreach and public involvement.
Yet, many communities have no idea what kind of message to send, or what media to use.
In the following presentation, we review the prospects for changing our behaviors to better protect watersheds. We
begin by outlining some of the daunting challenges that face educators who seek to influence deeply rooted public
attitudes. Next, we profile research on the outreach techniques that appear most effective in influencing watershed
behavior. Special emphasis is placed on media campaignsand intensive training programs. Lastly, recommendations are
made to enhance the effectiveness of watershed education programs.
Challenges in Watershed Education
Watershed managers face several daunting challenges when they attempt to influence watershed behaviors. Some
of those challenges include:
A lot of minds to change
The most pressing challenge is that there are simply a lot of minds to change. Some notion of the selling job at hand
can be grasped from Table 1, which contains provisional, but conservative, estimates of potential residential "polluters"
in the United States by various categories. It is clear that we are attempting to change deeply rooted attitudes held by
millions of people. While most people profess to support the environment, only a fraction actually practice much of a
watershed ethic on the small parcels of the environment where they live.
Table 1. Provisional Estimates of Potential Residential Polluters in the United States
Watershed Behavior Prevalence
Population
in Overall Estimates of Potential Residential Polluters
Over-Fertilizers 35% 38 million
Bad Dog Walkers 1 5 % 16 million
Chronic Car washers 25% 27 million
Septic Slackers 15% 16 million
Bad Mechanics 1 to 5% 3 million
Pesticide Sprayers 40% 43 million
Hosers 15% 16 million
Notes: estimates are based on 1999 U.S. population of 270 million, 2.5 persons per household, and average behavior prevalence rates based on
numerous market surveys (See references).
230
-------
Most Residents are Only Dimly Aware of the Watershed Concept.
It stands to reason that if citizens are asked to practice a watershed ethic, they will need to know what a watershed
is. Surveys indicate, however, that the average citizen is unaware of the watershed concept in general, and does not fully
understand the hydrologic connection between their yard, the street, the storm sewer, and (finally) local streams. Resident
surveys also continue to show limited or incomplete understanding of terms such as "watershed", "stormwater quality"
or "runoff pollution". For example, a recent Roper survey found that only41 % of Americans had any idea of what the term
watershed meant (NEETF, 1999). The same survey found that just 22% of Americans know that stormwater runoff is the
most common source of pollution of streams, rivers, and oceans.
At the same time, most of us claim to be very environmentally aware. For example, a Chesapeake Bay survey
reported that 69% of respondents professed to be very active or at least somewhat active in helping to reduce pollution
in the environment (SRC, 1994).
Resources Devoted to Watershed Education are Inadequate.
In recent years, several communities have developed education programs to influence the watershed behaviors
practiced by their residents. Most of these efforts, however, are run on a shoestring. For example, CWP recently surveyed
50 local programs that have tried to influence lawn care, septic cleaning and pet waste behaviors (Swann, 1999). These
education programs are typically run by the cooperative extension services, local recycling or stormwater agencies, or
urban soil and water conservation districts. Most are poorly staffed (0.1 to 0.5 staff years), relatively new (within last five
years), and have tiny annual budgets ($2,000 to $25,000). Given these limited resources, most watershed education
programs have no choice but to practice retail, rather than wholesale, outreach techniques. Consequently, most
watershed educators rely heavily on low-cost techniques such as brochures, posters, workshops, and demonstration
projects to send their message out.
The Marketing Techniques We Can Afford Don't Reach Many People
Watershed managers need to send a clear and simple educational message that can attract the attention of the
average citizen who is simultaneously bombarded by dozens of competing messages every day. A number of surveys
have asked residents which outreach techniques are most influential in attracting their attention (Table 2). Messages sent
through television, radio and local newspapers are consistently more influential in reaching residents than any other
technique, with up to 30% recall rates by the watershed population for each technique. By contrast, messages transmitted
through meetings, brochures, local cable and videos tend to be recalled by only a very small segment of the watershed
population.
Table 2. Most Influential Methods of Getting Messages to Citizens, in 8 Citizen Surveys
TV
TV ad
Newspaper
Local paper
Video
Brochure
Local cable
Meeting
WA
(Elgin, 1996)
TV ad
TV
Newspaper
Radio Ad
Brochure
Radio news
Paper Ad
Billboard
OR
(AMR.1997)
Direct Mail
TV ad
Newspaper
Radio
TV
Bill Insert
Newsletter
Local paper
CA
(As-sing, 1994)
TV Ad
Stencils
Billboard
Local paper
Brochure
Radio Ad
Bus Sign
Direct Mail
CA
(Pellegrin,
1998)
TV
Paper
Radio
Magazine
Neighbors
School
Billboard
Brochure
Ml
(PSC, 1994)
TV
Paper
Cable TV
Local paper
News-letter
Video
Meetings
Brochure
Wl
(Simpson,
1994)
TV
Paper
Newsletter
Brochure
Site Visit
Video
Meeting
MN
(Morris et al.,
1996)
Newspaper
Direct Mail
TV
Neighbors
Ext Service
Radio
Meeting
Local cable
One clear implication is that watershed education efforts must utilize a mix of outreach techniques if they are going
to get the message across to enough residents to make a difference in a watershed. Most existing watershed education
programs, however, cannot afford to use the more sophisticated wholesale outreach techniques that are most effective
231
-------
I
1
+*
o
£
o
o
i
• Education Programs
D Residents
Outreach Method
Figure 1. Outreach Methods Preferred By Residents Compared to Those Used by Education Programs.
at reaching the public with their watershed message. This gap is evident in Figure 1, which compares the outreach
methods actually used by local watershed education programs with the outreach methods that residents prefer, based
on responses from the Chesapeake Bay survey (Swann, 1999).
Crafting Better Watershed Education Programs
The first step in crafting better watershed education programs is to compile some baseline information on local
awareness, behaviors and media preferences. Some of the key questions watershed managers should consider are:
Is the typical individual aware of water quality issues in the watershed they live in?
. Is the individual or household behavior directly linked to water quality problems ?
*• Is the behavior widely prevalent in the watershed population ?
> Do specific alternative(s) to the behavior exist that might reduce pollution?
> What is the most clear and direct message about these alternatives?
> What outreach methods are most effective in getting the message out ?
»• How much individual behavior change can be expected from these outreach techniques?
The best way to elicit this information is to conduct a market survey within the watershed. These market surveys are
useful for two purposes: to gauge the level of watershed awareness and interest within the general population, and to
determine if there is a segment of the population where education efforts should be focused to achieve the best returns
in behavioral changes for the money spent.
232
-------
Perhaps the most critical step in crafting an education program is to select the right outreach techniques to send the
watershed message. Several communities have recently undertaken before and after surveys to measure how well the
public responds to their watershed education programs. From this research, two outreach techniques have shown some
promise in actually changing behavior - media campaigns and intensive training. Media campaigns typically use a mix
of radio, TV, direct mail, and signs to broadcast a general watershed message to a large audience. Intensive training use
workshops, consultation and guidebooks to send a much more complex message about watershed behavior to a smaller
and more interested audience. Intensive training often requires a time commitment of several hours from residents.
Both media campaigns and intensive training can produce up to a20% improvement in selected watershed behaviors
among their respective target populations (Tables 3 and 4). Both outreach techniques are probably needed in most
watersheds, as each complements the other. For example, media campaigns cost just a few cents per watershed resident
reached, while intensive training can cost a few dollars for each resident that is actually influenced. Media campaigns are
generally better at increasing watershed awareness, and sending messages about negative watershed behaviors.
Intensive training, on the other hand, is superior at changing individual practices in the home, lawn and garden.
Table 3. Effectiveness of Media Campaigns in Influencing Watershed Behaviors: Four Surveys
Location and Nature of Targeted
Campaign
San Francisco Radio, TV and Buses
(BHI, 1997)
ILos Angeles Radio and Newspapers
(Pellegrin Research Group, 1998)
Oregon Radio, TV (Advanced
(Marketing Research, 1997)
(Oakland County, Ml
IDirect Mail (Public Sector Consultants,
1994)
Effectiveness of Campaign
Awareness increased 1 O-l 5%
Homeowners who reduced lawn chemicals shifted from 2 to 5%
Best recall: motor oil and litter (over 40%)
Worst recall: fertilizer and dog droppings (<10 %)
Drop in car washing, oil changing, radiator draining of about 5 to 7%
Greater self-reporting of polluting behaviors: dropping cigarette butts, littering, watering and letting water run
on street, hosinq off driveways into the street (1 0% or more)
1 9% reported a change in "behaviors"-changes included being more careful about what goes down drain,
increasing recycling and composting, using more nature-friendly products etc.
44% of mail respondents recalled lawn care campaign
50% desired more information on lawn care and water quality
1 0% change in some lawn care practices as a result of campaign
(grass recycling, fertilizer use, hand weeding). No change in other
lawn care practices as a result of campaign
Tat
Location and Nature of Training
Campaign
Maryland Direct Homeowner
(Smith, 1996)
Florida Master Gardener
(Knox, 1995)
Virginia Master Gardener
(Aveni, 1998)
le 4. Effectiveness of Intensive Training in Changing Watershed Behaviors
Effectiveness of Intensive Training
10% shift from self to commercial car washing.
No change in fertilizer timing or rates.
Better claims of product disposal.
No significant change in fertilization frequency after program.
Some changes in lower rates, labels, slow release (8 to 1 5%).
Maior changes in reduced pesticide use (1 0 to 40%).
30 to 50% increase in soil testing, fertilizer timing and aeration.
10% increase in grass clippings and 10% decrease in fertilizer rate.
233
-------
Both techniques work best when they present a simple and direct watershed message, are repeated frequently, utilize
multiple media and are directly connected to local water resources that are most important in the community.
Other important suggested considerations for effectively marketing a watershed message are to:
Develop stronger connections among the yard, the street, the storm sewer, and the stream. Outreach techniques
should continually stress the link between a particular watershed behavior and the undesirable water quality it helps to
create (i.e., fish kills, beach closure, algae blooms). Several excellent visual ads that effectively portray this link are
profiled in our watershed outreach award winners.
Form regional media campaigns. Since most communities operate on small budgets, they should consider pooling
their resources to develop regional media campaigns that can use the outreach techniques that are proven to reach and
influence residents. In particular, regional campaigns allow communities to hire the professionals needed to create and
deliver a strong message through the media. Also, the campaign approach allows a community to employ a combination
of media, such as radio, television, and print, to reach a wider segment of the population. It is important to keep in mind
that since no single outreach technique will be recalled by more than 30% of the population at large, several different
outreach techniques will be needed in an effective media campaign.
Use television wise/y. Television is the most influential medium for influencing the public, but careful choices need
to be made on the form of television that is used. Our surveys found that community cable access channels are much
less effective than commercial or public television channels. Program managers should consider using cable network
channels targeted forspecificaudiences, and develop thematicshows that capture interest of the home, garden and lawn
crowd (i.e., shows along the lines of "This Old Watershed"). Well-produced public service announcements on commercial
television are also a sensible investment.
Understand the demographics of your watershed. The middle-aged male should usually be the prime target for
watershed education, as he is prone to engage in more potentially polluting watershed behaviors than other sectors of
the population. Indeed, the most important audience for the watershed message includes men in the 35 to 55 year age
group with higher incomes and education levels. Specialized outreach techniques can appeal to this group, such as radio
ads on weekend sports events.
Another target group worth reaching includes what Pellegrin (1998) terms the "rubbish rebels"- 18 to 25 year olds
who tend to have low watershed awareness, engage in potentially polluting behaviors, and are often employed in lawn
care and other service industries. This age group is hard to reach using conventional techniques, but may respond to ads
on alternative radio, concerts, and other events that celebrate the watershed.
As communities become more diverse, watershed managers should carefully track the unique demographics of their
watersheds. For example, if many residents speak English as a second language, outreach materials should be produced
in other languages. Similarly, watershed managers should consider more direct channels to send watershed messages
to reach particular groups, such as church leaders, African American newspapers, and Spanish-speaking television
channels.
Watershed educators should also be careful about using the traditional environmental education model that uses
schools to educate children who in turn educate their parents. While this model was instrumental in achieving greater
rates of recycling, it may not be as effective in changing watershed behaviors. While it is important to educate the next
generation of fertilizers, dog walkers, septic cleaners, and car washers, we need to directly influence the "boomer'
generation now.
Keep the watershedmessagesimpleandfunny. Watershed education should not be preachy complex, or depressing.
Indeed, the most effective outreach techniques combine a simple and direct message with a dash of humor.
Make information packets small, slick, and durable. Watershed educators should avoid the ponderous and boring
watershed handbook that looks great to a bureaucrat but ends up lining the bottom of a bird cage. One solution is to
create small, colorful and durable packets that contain the key essentials about watershed behaviors, with contact
234
-------
information to get better advice. These packets can be stuck on the refrigerator, the kitchen drawer or the workbench for
handy reference when the impulse for better watershed behavior strikes.
Educate private sector allies. A wide number of private sector companies may potentially stand to benefit from
changes in watershed behavior. Better watershed behavior can drum up more sales for some companies, such as septic
tank cleaners, commercial car washes, and quick oil change franchises - although they may need some help in crafting
their watershed marketing pitch.
Clearly, the potential exists for lawn care companies and landscaping services to shift their customers toward more
watershed-friendly practices. Nationally, lawn care companies are used by up to 50% of consumers, depending on
household income and lot size. Lawn care companies can exercise considerable authority over which practices are
applied to the lawns they tend, as long as they still produce a sharp looking lawn. For example, 94% of lawn care
companies reported that they had authority to change practices, and that about 60% of their customers were "somewhat
receptive to new ideas" according to a Florida study (Israel et al, 1995). De Young (1997) also found that suburban
Michigan residents expressed a high level of trust in their lawn care company.
Indeed, a small, but growing proportion of lawn care companies feel that environmental advertising makes good
business sense and can increase sales (Israel et al, 1995). Clearly, intensive training and certification will be needed to
ensure that watershed-friendly ads reflect good practice and not just slick salesmanship. It needs to be acknowledged
that lawn care companies strongly committed to practices that reduce fertilizer and pesticide inputs need to be strongly
endorsed by local government. Right now, it is not likely that such companies would be selected by the average
consumer, as consumers primarily rely on direct mail, word of mouth, and cost when choosing a lawn care company
(Swann, 1999 and AMR, 1997). For example, in the Chesapeake Bay survey, only two percent of residents indicated that
they had chosen a lawn care company primarily on the basis that it was "environmentally friendly" (Swann, 1999).
Lawn and garden centers are another natural target for watershed education. Study after study indicates that product
labels and store attendants are the primary and almost exclusive source of lawn care information for the average
consumer. At first glance, national retail chains should be strongly opposed to better watershed behavior, since it would
sharply cut into lawn and garden product sales and the lucrative profits they produce (even at the expense of the
community and environmentally friendly image they often market). The key strategy is to substitute watershed-friendly
products for ones that are not, and to offer training for the store attendants at the point of sale on how to use such
products.
Summary
For the watershed manager faced with new regulatory requirements under Phase II of the NPDES program, the
creation of an effective watershed education program should be a high priority. Not only is public education a mandated
component of an NPDES permit, but in urbanized areas it may the most cost-effective tool available to achieve water
quality goals. For smaller communities with scant budget and staff resources, it is imperative that these education
programs be productive in terms of changing behaviors and raising awareness of individual actions on local water quality.
Perhaps the most important factor in creating an effective watershed education program is selecting the right outreach
methods. Market surveys will often answer questions regarding the level of environmental awareness of watershed
residents, what forms of informational outreach attract their attention, and resident willingness to change pollutant
producing behaviors. This information allows the watershed manager to tailor outreach methods to specific target groups
where behavior change is most likely. These surveys will also establish the demographics of the residents and determine
whether multilingual outreach is required.
Watershed managers should also consider innovative approaches to sending out their pollution reduction messages.
Pooling resources with other communities to create regional media campaigns and the use of outreach opportunities
through private sector education are just two ways that program managers can reach broader audiences without spending
large amounts of money.
235
-------
Continued development of productive outreach methods and innovative techniques is necessary to relay the basic
premise of watershed education - that we live in a watershed and how to properly live within it - in the most economical
and effective manner.
References
Advanced Marketing Research (AMR). 1997. Stormwatertracking study. City of Eugene, Oregon. Unpublished marketing
survey.
Assing, J. 1994. Survey of public attitudes. February and July, 1994. Russian Hill Associates. Alameda County Urban
Runoff Clean Water Program. San Francisco CA. 84 pp.
Aveni, M. 1998. Water-wise gardener program: summary report. Unpublished Data. Virginia Cooperative Extension.
Prince William County, VA.
Big Honking Ideas, Inc (BHI). 1997. Final report: Spring regional advertising campaign. Prepared for Bay Area Stormwater
Management Agencies Association. Oakland, CA.
De Young, R. 1997. Healthy lawn and garden survey: data analysis report. Rouge River National Wet Weather
Demonstration Project. Oakland County, Ml. 40 pp.
Elgin DDB. 1996. Public awareness study: summary report. The Water Quality Consortium. Seattle, WA. 24 pp.
Israel, G., S. Pinheiro and G. Knox. 1995. Environmental landscape management - assessing practices among
commercial groups. University of Florida. Cooperative Extension Service. Bulletin 307. Monticello, FL. 18 pp.
Knox, G., A. Fugate and G. Israel. 1995. Environmental landscape management-use of practices by Floridaconsumers.
University of Florida Cooperative Extension Service. Bulletin 307. Monticello, FL. 26 pp.
Morris, W. and D. Traxler. 1996. Dakota County subwatersheds: residential survey on lawn care and water quality. Dakota
County, Minnesota, Decision Resources, Ltd.
National Environmental Education Training Foundation (NEETF). 1999. National Report Card on Environmental
Knowledge, Attitudes and Behaviors: Seventh Annual Roper Survey of Adult Americans. National Environmental
Education Training Foundation, Washington, D.C.
National Service Research (NSR). 1998. Pesticide usage and impact awareness study: executive summary. City of Forth
Worth Water Department. Fort Worth Texas. 44 pp.
Pellegrin Research Group. 1998. Stormwater/urban runoff public education program: interim evaluation, resident
population. Los Angeles County Department of Public Works. 28 pp.
Public Sector Consultants, Inc (PSC). 1994. A strategy for public involvement. Rouge River National Wet Weather
Demonstration Project. 56 pp.
Simpson, J. 1994. Milwaukee survey used to design pollution prevention program. Technical Note 37. Watershed
Protection Techniques. 1(3): 133-1 34.
Smith, J. 1996. Public survey used to estimate pollutant loads in Maryland. Technical Note 73. Watershed Protection
Techniques. 2(2): 361-363.
Survey Research Center (SRC). 1994. The Chesapeake Bay attitudes survey. Communications subcommittee.
Chesapeake Bay Program. U.S. Environmental Protection Agency. 42 pp.
236
-------
Swann, C. 1999. A survey of residential nutrient behaviors in the Chesapeake Bay. Widener-Burrows, Inc. Chesapeake
Research Consortium. Center for Watershed Protection. Ellicott City, MD. 112 pp.
237
-------
Lawn Care and Water Quality: Finding the Balance
Jerry Spetzman
Minnesota Department of Agriculture
St. Paul, MN 55107-2094
email: Jerome.Spetzman©State.MN.US
Abstract
When land is converted from natural areas to developed urban areas, pavement and rooftops replace grass and trees.
Water flows over driveways, streets and parking lots taking with it particles and debris in its path and depositing them,
via storm sewers, into nearby lakes, creeks and rivers. This non-point source pollution can contain sediment, debris,
fertilizers, pesticides, leaves, grass clippings, motor oil, or pet wastes. Small amounts of these materials entering a lake
or river are not generally considered harmful. But when small amounts are multiplied by thousands or tens of thousands
they can cause serious water quality problems.
Since 1993, the Minnesota Department of Agriculture, in cooperation with several other organizations, has been
gathering information on homeowner use of pesticides and fertilizers in the Twin Cities metropolitan area. This
information includes the amounts of lawn care products used by homeowners, where they are purchased, how they are
applied, and whether or not they have an effect on nearby lakes, creeks or rivers.
Residents living in two watersheds were selected to participate in a focused study. Lake Harriet represents urban
watersheds and Lake Alimagnet represents suburban watersheds. Based on survey results and water quality monitoring,
education materials were developed to promote public awareness of lawn practices and their potential to affect water
quality. These educational materials incorporate the concept that "everyone lives in a watershed" and that everyone has
the potential to affect water quality, whether or not they actually live on a lake shore.
What are the Lake Harriet and Lake Alimagnet Watershed Awareness Projects?
The projects have two purposes: (1) to inform urban and suburban homeowners about living in a watershed, and (2)
to help them learn how their lawn care habits can affect the quality of Twin Cities water. The project's goal is to improve
water quality by reducing the quantity of pesticides and nutrients through responsible use of those materials.
How has this goal been achieved?
Project members have:
. Surveyed the current lawn care habits of homeowners and measured the effects of those habits by monitoring
pesticide and nutrient runoff into Lake Harriet
. Informed homeowners about how their lawn care habits affect Twin Cities waters
. Asked homeowners how the projects have affected their lawn care practices
. Monitored runoff into Lake Harriet to quantify changes brought about by homeowner actions
. Drawn on detailed Lake Harriet experience to design urban watershed education materials for use throughout
Minneapolis and Minnesota-these materials were tested in the Lake Alimagnet watershed
238
-------
What is a watershed?
It is an area of land over which rain and melted snow flows to lakes, rivers, and wetlands. There are 46 major
watersheds and 255 sub-watersheds in the Minneapolis-St. Paul metropolitan area.
The Lake Harriet watershed is a 1,139-acre area in a well-established residential neighborhood with almost 6,000
homeowners. The Lake Harriet study area is a 148-acre portion of the watershed. About 40% of the study area is
covered with hard surfaces, such as pavement and rooftops. About 700 homeowners live in the study area, most in
detached, single family houses built in the early 1900s. The water quality in Lake Harriet is very good for a Twin Cities
lake.
A survey of 105 Lake Harriet watershed residents most familiar with lawn care done on their own properties, showed
that:
. They are highly educated (college or post-graduate degrees predominate)
. Their average age is 47
. They have middle- to upper-level incomes
. They care for their lawns by mowing regularly and using fertilizers and herbicides
. A few use professional lawn care services
The Lake Alimagnet watershed has approximately 3600 suburban households. Of this number, a large proportion
is made up of townhouse residents. The Lake Alimagnet watershed was developed in the 1960s. Homeowners are well-
educated and have middle- to upper-middle incomes. The water quality in Lake Alimagnet is considered poor, algae
blooms are common, and the predominant fish species is bullhead.
Why are we studying the Lake Harriet and Lake Alimagnet watersheds?
Because they provide good examples (respectively) of how urban and suburban development affect our water
resources and because they are sources of year-round recreation for many Twin Cities residents. These residents are
vitally interested in keeping the water clean.
What are we finding out?
Though Lake Harriet had some of the highest quality water in the Twin Cities, that quality has also declined over time.
Lake Harriet has poor quality water and area residents want to reverse the trends.
These projects and others have monitored storm water, rainfall, and lake water to determine the levels of non-point
source pollutants in Lake Harriet. Specifically, this project monitored two types of pollutants: pesticides, which can affect
water quality, and phosphorus, which can increase lake algae growth and reduce water quality.
Pesticide monitoring
The Lake Harriet project monitored storm runoff, rainfall, and lake water. The water quality monitoring consisted of
a permanent automatic sampling station installed in a storm drain outlet which carried watershed runoff into Lake Harriet;
samples were taken during storm events, from which mass loading of pesticides and nutrients were calculated. In addition,
a rainfall monitoring station sampled rain events. Finally lake monitoring samples were collected during the growing
season. Several hundred samples were analyzed for more than 30 pesticides.
The education program included several methods of education, including homeowner meetings, direct mail, flyers,
billboards, utility bill inserts, local newspaper articles and visits by Master Gardeners. Because monitoring preceded the
239
-------
education, and also followed ^reductions in pesticides can be measured. There was a decrease in average pesticide
loads between the earlier and later monitoring periods. Therefore, the annual storm sewer runoff load of pesticides to Lake
Harriet was reduced during the Lake Harriet project. The largest decreases came from the four compounds listed on this
table:
Lawn Herbicides
MCPA
Dicamba
2,4-D
MCPP
Percent Decrease
(1992-1 995)
86%
59%
58%
56%
The most prevalent pesticides found during monitoring were herbicides (weedkillers). The eight herbicides listed
on the following table accounted for 95% of all pesticide detections.
Lawn Herbicides
MCPA
Dicamba
2,4-D
MCPP
Agricultural Herbicides
Alachlor
Atrazine
Cyanazine
Metolachlor
. Storm water runoff monitoring summary:
. Lawn herbicides were found in 80% of the storm runoff events sampled between April and October.
. Agricultural herbicides were detected in 35% of the storm events sampled
. Rainfall monitoring summary:
. The agricultural herbicides listed above (the only herbicides found in rainfall samples) were atmospherically
deposited by wind and rainfall onto the watershed and the accompanying water bodies.
• Lawn herbicides were not detected in rainfall samples.
. Lake monitoring summary:
. The three most commonly detected compounds in lake water were MCPP, Atrazine, and 2,4-D. They were also
the most frequently detected compounds in stormwater entering Lake Harriet.
. Phosphorus monitoring:
Analyses revealed the phosphorus in runoff peaks twice a year, in the spring and in the fall.
. In the spring, melting snow carries phosphorus attached to tiny particles of grit, sand, and organic matter as it
enters the storm sewers.
240
-------
. In the fall, phosphorus in leaves, grass clippings, and other organic debris enters the storm sewers. Studies
conducted by other researchers established that only a small percentage of phosphorus in runoff results from
appropriate use of fertilizer on lawns.
. Water Quality Education
Lake Harriet and Lake Alimagnet project participants have concluded that educating homeowners living in the
watershed is one of the best ways of reducing pollution in the lake. Billboards, brochures, and water bill inserts have
carried messages based on the following two concepts:
. A healthy lawn and landscape promotes healthy waters. Home landscaping with regionally adapted, healthy
plants can help absorb and filter rainfall, irrigation, and runoff from melted snow.
. Keep your lawn and landscape healthy as follows:
. Apply pesticides and nutrients according to recommendations
. Aim roof downspouts onto lawns and gardens to filter and absorb runoff
. Keep grass clippings and leaves off streets, sidewalks, and driveways
. Leave grass clippings on the lawn or compost them
. Use fallen leaves as winter or summer mulch, compost them, or shred them and leave them on the lawn
. Keep lawn care products on the lawn and always follow label instructions
. Clean up and reuse granular lawn care products that fall on streets, sidewalks, and driveways
Project Evaluation
. Based on feedback from homeowners living in the Lake Harriet and Lake Alimagnet watersheds, we have
concluded the following:
. Most homeowners in the Lake Harriet watershed apply significantly less lawn fertilizer than the University of
Minnesota's recommended guidelines.
. Most homeowners compost grass clippings or leave them on their lawns.
. Homeowners would rather spot-treat weeds than apply herbicide to their entire yard or use non-chemical weed
control methods.
. Top soil in the Lake Harriet watershed is significantly deeper than that found in the Lake Alimagnet watershed.
In many cases when suburban areas are developed, top soil is removed, but not replaced. This results in
decreased plant vigor, and an increase in the need to fertilize, water, and maintain turf.
. Most homeowners feel the educational initiative has increased their understanding of how lawn care habits affect
water quality. Neighborhood newspapers and direct mail, the most common source of lawn care and water
quality information, have the greatest impact.
. Messages that are quick to read and easy to understand are the most effective in changing lawn care habits.
. These messages are best delivered over an extended period. Homeowners have been reached through fliers,
newspaper articles, brochures, direct mail, handouts, and personal contacts.
241
-------
Feedback also shows that homeowners still need to hear the following messages:
. You can preserve water quality and have a healthy lawn by applying lawn care products in appropriate amounts,
at the right times, and during suitable (or appropriate) weather conditions.
. By keeping leaves out of storm sewers, you can help reduce the amount of phosphorus carried to the lake in runoff
water.
. Fall is the best season to apply turf fertilizers and lawn care products that control broadleaf weeds. The primary
growth of turf grasses is early fall until late spring.
. Erosion, leaves, grass clippings, yard waste, pet waste, and rainfall all contain pollutants that can end up in lake
water.
Lake Alimagnet Project Results
During the project period, the water quality in Lake Alimagnet improved significantly.
. Project cooperators achieved improvements in total phosphorus and chlorophyll-a.
. The best secci disk reading ever on Lake Alimagnet was recorded.
• The Citizens Assisted Monitoring Program improved the ranking of Lake Alimagnet from a "D" to a "C."
These results may be a result of the project, a curly leaf pond weed cutting, or may be credited to El Nino.
For more information on the Lake Harriet or Lake Alimagnet Watershed Awareness Projects and samples of
homeowner education materials, please contact the Minnesota Department of Agriculture at (651) 297-7269.
242
-------
San Francisco Bay Area's Pesticide
Toxicity Reduction Strategy
Geoff Brosseau
Bay Area Stormwater Management Agencies Association (BASMAA)
Oakland, California
Introduction
Water quality research conducted by San Francisco Bay Area stormwater programs and wastewatertreatment plants
over the last several years has identified widespread toxicity in local creeks and wastewater treatment plant effluent
(California Regional Water Quality Control Board, 1997; San Francisco Bay Area Pollution Prevention Group, 1998). The
toxicity problem was ultimately traced to diazinon and chlorpyrifos-commonly used organophosphate pesticides available
in hundreds of consumer products (Alameda Countywide Clean Water Program, 1997). Study results indicated that
pesticide use according to label instructions could not be ruled out as a cause of wastewater and stormwater toxicity
(Regional Water Quality Control Plant-Palo Alto, 1996). In May 1999, San Francisco Bay and 35 Bay Area urban creeks
were listed by the U.S. Environmental Protection Agency (USEPA) as impaired by diazinon (USEPA, 1999).
Impact of 303(d) Listing on Local Governments
In its action, USEPA listed 53 waterbodies in California as impaired due to diazinon in urban runoff and 7 waterbodies
as impaired due to chlorpyrifos in urban runoff. By definition under the Clean Water Act, this action means that there is
a water quality problem, regardless of the problem definitions under the Federal Insecticide, Fungicide and Rodenticide
Act (FIFRA) (i.e., "unreasonable adverse effect") or the Food Quality Protection Act (FQPA). The listing action put over
100 municipalities in the San Francisco Bay Area and Central Valley at immediate regulatory, legal, and financial risk.
. Regulatory risk -The State Water Resources Control Board and USEPA can take enforcement action against, and
fine, these municipalities for violating their NPDES (National Pollutant Discharge Elimination System) stormwater
permits.
. Legal risk - Citizen and environmental groups can sue municipalities for the same reasons.
. Financial risk - These municipalities must now spend local public tax dollars proactively addressing this problem,
and potentially reacting to fines and lawsuits.
Municipalities' Response
To comply with their NPDES stormwater permits, municipalities must meet two broad goals:
1. Effectively prohibit non-stormwater discharges into storm sewers.
2. Reduce the discharge of pollutants to the maximum extent practicable (MEP).
To meet these goals and to address the 303(d) listing, there are a number of actions Bay Area stormwater programs
have taken or plan to take that may reduce pesticide-related toxicity in surface waters. These actions are packaged in
a Pesticide Toxicity Reduction Strategy (BASMAA, 2000). The Strategy is a multi-faceted effort including:
. Education/outreach including:
243
-------
. limiting or prohibiting pesticide use by municipal staff and contractors and/or requiring use of best management
practices (BMPs) such as Integrated Pest Management (IPM)
. providing adequate and convenient options for disposal of unused pesticides and pesticide containers through
household hazardous waste collection programs
. educating residents about pesticide-related toxicity and proper use and disposal through distribution of
educational materials, and development and implementation of media and advertising campaigns
. educating residents about alternative methods and products through such programs as demonstration gardens
and point-of-purchase campaigns in hardware stores and nurseries
. educating businesses about proper use and disposal, as well as alternative methods and products for use around
their own properties and facilities
. educating pest control operators and working with them to develop BMPs protective of surface waters
. Regulatory- Identifying opportunities to reduce toxicity and advocating state and federal agencies to seize these
opportunities through regulation and re-registration
. Monitoring - Investigating the extent and causes of toxicity, and assessing impacts on beneficial uses
The IPM Store Partnership
One exemplary part of the Pesticide Toxicity Reduction Strategy worth describing in more detail is the IPM (Integrated
Pest Management) Store Partnership. In 1997, the Central Contra Costa Sanitary District (CCCSD), a wastewater
treatment plant located in Martinez, California, jointly developed and successfully piloted the IPM Store Partnership with
the Regional Water Quality Control Plant (Palo Alto, California) in four locally owned garden centers and hardware stores.
In 1998, the Bay Area Stormwater Management Agencies Association (BASMAA) and the San Francisco Bay Area
Pollution Prevention Group (BAPPG) joined together to fund the expansion of the IPM Store Partnership to more stores
in the San Francisco Bay Area. By spring 1999, 116 stores in eight Bay Area counties were participating in the
Partnership.
The Partners
BASMAA is a consortium of seven San Francisco Bay Area municipal stormwater programs. These programs
represent more than 90 agencies, including 79 cities and 6 counties, and the bulk of the watershed immediately
surrounding San Francisco Bay. BASMAA agencies agree to a memorandum of understanding and each year collect
dues, prorated by population, from their members for a "baseline" program that provides for staff and finances
projects-like the IPM Store Partnership-that are endorsed by all member agencies.
The BAPPG is a voluntary association of 39 wastewater treatment plants working together to prevent water pollution
in the San Francisco Bay. These agencies represent all of the publicly owned municipal wastewater agencies that
discharge into San Francisco Bay in the nine Bay Area counties, and almost all of the watershed immediately surrounding
San Francisco Bay. BAPPG's decision-making is done by consensus. Each year, a work plan-with an associated
budget-is developed. The budget is allocated among the 39 plants based on the average amount of treated wastewater
discharged each day. Contributions are voluntary, although all agencies do contribute. These contributions are used to
fund projects like the IPM Store Partnership.
Integrated Pest Management
There are many definitions of Integrated Pest Management. The definition used to guide the IPM Store Partnership
was the following:
244
-------
Integrated Pest Management is an approach that uses regular monitoring and cultural, biological, and physical
methods to keep pests at acceptable levels. Only less toxic chemicals are used and only as needed.
IPM was used as the basis for the program because it: (1) focuses on effective alternatives to traditional chemical
pesticides; (2) does not substitute another pesticide that may become tomorrow's problem, and (3) does not preclude the
use of chemicals in all situations.
Although promotion of IPM was the basis for the program, the term itself is somewhat problematic. The terms "IPM"
and "Integrated Pest Management" were used in speaking with experts and agencies familiar with the jargon, but these
terms were avoided in communications with the non-initiated (e.g., general public). In addition, the term IPM was not very
representative of the situation in the store between the customer and the store employee. Most customers go to a store
for help when pests have already reached unacceptable levels, so store employees must start with controlling a pest
problem, rather than preventing it, which is the first step in IPM. Despite this challenge, customers were exposed to the
full range of IPM methods through the fact sheets and display materials, as well as less-toxic products.
Goals
The goals of the IPM Store Partnership are to:
. Educate the public about the value of IPM approaches to pest control and safe use and disposal of pesticides, when
used
. Deliver IPM-related messages without negative messages about any products
. Develop partnerships with retailers so that they can help spread the word about water quality problems related to
residential pesticide use
. Provide consistent messages
• Capitalize on economies of scale
. Prepare the stage for regional program expansion into chain stores
Program Elemen ts
The IPM Store Partnership is an education program for employees and customers of locally owned garden centers
and hardware stores. The project elements include:
. development and production of eight fact sheets on less-toxic pest management strategies for the public (Naturally
Managing Pests, Controlling Ants, Controlling Aphids in Your Garden, Keeping Cockroaches Out of Your House,
Keeping Fleas Off Your Pets and Out of Your Yard, Living with Spiders, Tips for a Healthy Beautiful Lawn, and Safe
Use and Disposal of Pesticides)
. development of an extensive list of less-toxic methods and products preferable to diazinon and chlorpyrifos for
various applications
. training sessions for store employees focusing on principles of Integrated Pest Management and successful
application strategies for products on the less-toxic list
. design and production of a program logo and in-store promotion materials including "end cap" displays, posters,
shelf-talkers, shelf signs, and vinyl banners
. program evaluation by a San Francisco State University-affiliated survey research and data analysis firm
245
-------
Results and Discussion
Although the final evaluation will not be complete until the end of the 1999 in-store promotion season (late 1999), the
sponsoring agencies expect it to be successful based on preliminary findings and on the success of the pilot project.
Feedback from store owners and employees that participated in the pilot was uniformly positive. The sales data in the
pilot IPM Store Partnership showed variations from store to store. One store found that sales of all but one diazinon and
chlorpyrifos product dropped. At the same time, less-toxic product sales experienced an overall 17% increase and profits
were not affected.
It is the hope of the participating agencies that the final evaluation and report on the 1999 Partnership will be useful
as a model and primer for other agencies and jurisdictions concerned about pesticide-related toxicity in surface waters
and interested in building educational partnerships with local businesses. While Bay Area water pollution prevention
agencies have been coordinating their public education efforts since the early 1990s, the IPM Store Partnership is the
first point-of-purchase program implemented regionally. All of the general benefits of inter-agency coordination (support
for smaller agencies, cost savings, options for pooled advertising and media relations) are magnified in such a large
undertaking.
Based on the partnership's success, all of the agencies that participated in the 1999 Partnership allocated funds for
continuation of the program in 1999-2000. BASMAA and the BAPPG again contributed funds to regional coordination.
Brainstorming sessions were held in late summer 1999 to determine how to improve the program, and minor modifications
were made for 2000.
The IPM Store Partnership is one example of BASMAA and the BAPPG's commitment to use public resources
efficiently. Given that philosophy, materials developed by the IPM Store Partnership are available to agencies interested
in implementing a similar program.
Other Aspects of the Strategy
Despite the success of the IPM Store Partnership and many of the other educational aspects of the Pesticide Toxicity
Reduction Strategy, it is clear to Bay Area water pollution prevention agencies that their efforts alone will not be enough
to solve the problem. Study results indicate that less than 1% of applied diazinon runs off, yet it takes less than a fluid
ounce of active ingredient flushed into stormwater runoff to cause toxicity in urban creeks (Regional Water Quality Control
Plant-Palo Alto, 1996). Educational programs run by Bay Area water pollution prevention agencies are some of the most
developed in the country and they have won numerous awards for their quality and effectiveness. Nevertheless, even
the best education programs are not 100% effective. It is clear that education alone will not solve this problem.
San Francisco Bay Area stormwater programs are and will continue to address the problem of pesticide-related
toxicity in surface waters by way of meeting the MEP requirement in their NPDES permits. These agencies have gone
so far as to develop the Pesticide Toxicity Reduction Strategy described above that includes three elements-education,
regulatory, and monitoring. The authority and ability of local governments to implement the strategy varies with each
element. The most cost-effective and appropriate aspect for local governments to implement is education. For the
regulatory and monitoring elements, local governments can, and have, identified the issues and opportunities to reduce
pesticide-related toxicity, but they have limited ability or authority to actually implement corrective actions.
Regulatory
Regulation of pesticides including their registration for use in the Unites States is the responsibility of USEPA.
California's Department of Pesticide Regulation (DPR) has responsibility for regulating the sale and use of pesticides in
California. California DPR, with few exceptions, registers pesticides only after they have been registered by USEPA.
California DPR can not register pesticides which have been denied registration by USEPA. At the local government level,
the California Food and Agriculture Code grants some authority to county agricultural commissioners for local enforcement
of pesticide regulations, record keeping, and outreach to applicators. However, with the exception of county agricultural
246
-------
commissioners, local governmentsare prohibited from regulating the registration, sale, transportation, or use of "economic
poisons." This regulatory structure means that the ability and authority of local governments is limited to:
. identifying opportunities to reduce toxicity, such as eliminating potentially problematic uses, and advocating that
state and federal agencies seize these opportunities through regulation and re-registration
. in the case of wastewater and stormwater agencies, regulating the discharge of pesticides to the sewer or storm
drain to ensure local agencies' compliance with state and federal laws (e.g., Clean Water Act)
Monitoring
Local governments have some ability, authority, and responsibility to use monitoring to address the problem of
pesticide-related toxicity of surface waters. To-date, San Francisco Bay Area municipalities have used monitoring to:
. identify and define the problem (Alameda County Urban Runoff Clean Water Program, 1995; Regional Water
Quality Control Plant-Palo Alto, 1996; California Regional Water Quality Control Board, 1997; San Francisco Bay
Area Pollution Prevention Group, 1998)
. characterize sources (Alameda Countywide Clean Water Program, 1997)
. recommend corrective actions (Alameda County Flood Control and Water Conservation District, 1997)
A review of monitoring data from around the country shows that municipalities in the San Francisco Bay Area and
California Central Valley are not alone in their identification of this environmental problem.
. Orange County, California (Lee, et al., 1999) - Multi-year studies of stormwater runoff in San Diego Creek as it
enters Upper Newport Bay have shown that the problem is not restricted to Northern California. Runoff from each
stormwater event has been shown to be toxic, and about half of the observed toxicity is due to diazinon and
chlorpyrifos used in urban areas for structural termite and ant control, and lawn and garden pest control.
. NAWQA (USGS, 1998) - Results from the United States Geological Survey's (USGS) National Water Quality
Assessment Program from 1992 through 1996 show that the problem is in fact a national one. Over 300 samples
have been taken from eleven urban streams scattered across the country, from Florida to Connecticut to Oregon,
as part of the Pesticides National Synthesis Project. In a recent report on the first cycle of the program, USGS
concluded that "urban and suburban areas are substantial sources of pesticides to streams" and that "most urban
areas have similar pesticides in streams...and many urban areas may benefit from similar strategies for reduction."
. Publicly-Owned Treatment Works survey (USEPA, 1989) - Results from a survey done 10 years ago by USEPA
show that pesticide-related toxicity is a wastewater problem as well as a stormwater problem. USEPA's
Environmental Research Laboratory in Duluth, working through the National Effluent Toxicity Assessment Center
(NETAC), reported on the occurrence of diazinon in 28 POTW effluents. Diazinon was found in sixteen (62%) of
the effluents, and levels were greater than or equal to 250 ng/L for nine (32%) of the effluents. NETAC concluded
in part "The frequency with which we have observed diazinon in the past, in this survey, and continue to find it in
effluents is indicative of a widespread problem."
Clearly this is a national problem caused by products that are registered at the national level and sold across the
country.
The pesticide registration process provides a built-in mechanism to use monitoring and science to address this
national problem. During the registration process, USEPA must review and summarize the findings of studies conducted
on each pesticide. During this step, USEPA may request that "registrants" (e.g., pesticide manufacturers) submit specific
studies for review. Based on its review, USEPA can confirm, deny, or change the pesticide's registration including
approved uses, sites of application, formulations, and label directions.
247
-------
Local governments are willing to use monitoring and science to further investigate local impacts and sources, and
to host case studies, if USEPA will provide financial and other support, with the goal of conducting representative case
studies whose results can be extrapolated across the country. But given the established mechanism in the pesticide
registration process, it would be inappropriate and ineffective for local governments to do more. USEPA must exercise
its federal authorities and use monitoring and science information to make more informed, up-to-date registration
decisions.
Conclusion
Rather than being a tool in and of itself, the Pesticide Toxicity Reduction Strategy is really a toolbox. It includes a
number of effective tools for reducing pesticide-related toxicity of surface waters-an increasingly important part of urban
water resource management and protection. Every job has its tool and in the right pair of hands, the job can be easy and
cost-effective to complete. The wrong tool or the wrong hands can make the job difficult, if not impossible to finish. It
is the responsibility of government agencies to be clear and disciplined about which tool and which pair of hands go with
which job when fixing environmental problems. The extent to which they implement that concept will determine how
successful the work of environmental protection will be.
References
Alameda County Flood Control and Water Conservation District, 1997. Strategy to Reduce Diazinon Levels in Creeks
in the San Francisco Bay Area. Prepared by J. Scanlin and S. Gosselin.
Alameda County Urban Runoff Clean Water Program, 1995. Identification and Control of Toxicity in Stormwater
Discharges to Urban Creeks. Prepared by S.R. Hansen and Associates.
Alameda Countywide Clean Water Program, 1997. Characterization of the Presence and Sources of Diazinon in the
Castro Valley Creek Watershed. Prepared by J. Scanlin and A. Feng.
Bay Area Stormwater Mgt. Association, 2000. Strategy for Reducing Organophosphate Pesticide-Related Toxicity in San
Francisco Bay Area Urban Creeks.
California Regional Water Quality Control Board, 1997. Diazinon in Surface Waters in the San Francisco Bay Area:
Occurrence and Potential Impact. Prepared by T. Mumley (RWQCB) and R. Katznelson (Woodward Clyde Consultants).
Lee, G. Fred, et al., 1999. Evaluation of the Water Quality Significance of OP Pesticide Toxicity in Tributaries of Upper
Newport Bay, Orange County, CA. Environmental Toxicology and Risk Assessment: Recent Achievements in
Environmental Fate and Transport: Ninth Volume, ASTM STP 1381, American Society for Testing and Materials, West
Conshohocken, PA.
Regional Water Quality Control Plant-Palo Alto, 1996. Diazinon in Urban Areas. Prepared by A. Cooper.
San Francisco Bay Area Pollution Prevention Group, 1998. Diazinon & Chlorpyrifos Quantitative Identification for San
Francisco Bay Area Wastewater Treatment Plants. Prepared by T. Chew, K. Easton, and A. Laponis (Central Contra
Costa Sanitary District).
USEPA, 1989. Results of Diazinon Levels in POTW Effluents in the United States. National Effluent Toxicity Assessment
Center (NETAC) Technical Report 14-89. Environmental Research Laboratory-Duluth. Prepared by T. Norberg-King,
M. Lukasewcyz, and J. Jensen.
USEPA, 1999. 1998 California 303(d) List and TMDL Priority Schedule.
USGS, 1998. Pesticides in Surface and Ground Water of the United States: Summary of Results of the National Water
Quality Assessment Program (NAWQA) (http://water.wr.usgs.gov/pnsp/allsumm).
248
-------
Administering the NPDES Industrial Storm Water
Program at the Municipal Level
Michael J. Pronold
Bureau of Environmental Services
City of Portland
Portland, Oregon
Abstract
As part of the EPA Phase 1 stormwater requirements, certain classes of industries are required to obtain Industrial
Storm Water permits. The EPA, or a state agency that has been delegated by EPA, administers these permits. The
Phase 1 regulations also require that municipalities develop a program to monitor and control pollutants in storm water
runoff from industrial facilities. These are potentially non-coordinated requirements and can result in redundant efforts
and a less than efficient program. In addition, EPA and/or state agencies may not have the resources to adequately
administrate and enforce the permitting program while leaving the municipality liable for the discharges from the municipal
separated storm sewer system (MS4).
The City of Portland, Oregon (City), met the requirement in its municipal storm water permit to control industrial
stormwater sources of pollution by developing a Memorandum of Agreement (MOA) with the Oregon Department of
Environmental Quality (DEQ), (which is the delegated authority) to administer the permit program. The MOA provided
the City with the mechanism to administer the industrial stormwater permits for those facilities that discharge to the City's
MS4. The City pursued this approach since it was responsible for the discharge from the MS4 and wanted to ensure that
it had adequate oversight of these discharges. By coordinating this effort with other ongoing industrial water quality
programs, the City could provide a more cost-effective program, considering the regulatory costs as well as cost to the
industry. City Code was developed to support this approach.
When the City took over the administration of the permits in 1994, over 50% of the facilities with a permit had not met
the requirements for the development of a storm water pollution control plan, the main requirement of the permit. In
addition, nearly 60% of the permitted facilities had not performed the required stormwater sampling. Of the samples
taken, approximately 30% violated standards in the permit. It was also evident that not all facilities required to obtain a
permit had done so. Efforts since 1994, have shown that only 25-30% of the facilities required to obtain a permit had
applied. A benefit of the local administration of the program is the detection of illicit discharges to the MS4.
Approximately 15% of all industrial inspections have identified illicit discharges.
The City has also identified certain classes of industries and activities that can be significant sources of pollutants
to the MS4. This has helped streamline the program efforts and redirect resources to where the greatest cost benefit will
be realized.
Introduction
Stormwater discharges have been increasingly identified as a significant source of water pollution in numerous
nationwide studies on water quality. To address this problem, the Clean Water Act Amendments of 1987 required EPA
to publish regulations to control storm water discharges under NPDES. EPA published storm water regulations (55 FR
47990) on November 16, 1990 which require certain dischargers of storm water to waters of the United States to apply
for NPDES permits. These regulations established NPDES permit application requirements for storm water discharges
249
-------
associated with large- and medium-size MS4s. The regulations also established NPDES permit application requirements
for storm water discharges associated with industrial activity. EPA has defined this phrase in terms of 11 categories of
industrial activity.
A requirement of the City's application process was "A description of a program to monitor and control pollutants in
storm water discharges to municipal systems from municipal landfills, hazardous waste treatment, disposal, and recovery
facilities, industrial facilities that are subject to Section 313 of Title III of the Super-fund Amendments and Reauthorization
Act of 1986 (SARA), and industrial facilities that the municipal permit applicant determines are contributing a substantial
pollutant loading to the municipal storm sewer system." (40 CFR 122.26(d)(2)(iv)(C). This creates the potential for
redundant efforts and a less than efficient program.
The stormwater regulations envision that NPDES permitting authorities and municipal operators will cooperate to
develop programs to monitor and control pollutants in storm water discharges to MS4 from certain industrial facilities.
The NPDES permits for industrial facilities establish requirements such as controls, practices, and monitoring of
stormwater discharges, as well as provide a basis for enforcement actions. An integral part of the requirement is the
adequacy of legal authority. This will allow the municipality to implement its program, which should include inspections,
review of stormwater pollution control plans, monitoring, and implementation of control measures.
The municipality is ultimately responsible for discharges from its MS4. To meet the requirement in its municipal
stormwater permit, and to provide the oversight necessary to protect itself from liability, the City developed new legal
authority and entered into an MOA with the authorized NPDES state authority (DEQ), to administer the permits for those
discharges to the MS4.
Program Elements
Legal Authority
The City did not have adequate legal authority to oversee discharges to the MS4. In response to this, the City
developed code in February 1994. Some of the major provisions of the code are:
. Authority of the Director of Environmental Services to Adopt Rules
. General Discharge Prohibitions
. Discharge Limitations
. Reporting Requirements
. Storm Water Pollution Control Plan (SWPCP)
. Storm Water Discharge Permits
. Inspection and Sampling
. Enforcement
Key elements of the code include the requirement for permit holders to submit their SWPCP and monitoring results
to the City, the authority for the Director to adopt administrative rules, make inspections, and undertake enforcements.
Memorandum of Agreement
The City entered into a MOA with the DEQ in March 1994. The MOA delineates the responsibilities for the
implementation of the program between the two agencies. The MOA also prioritizes the implementation of the program
to address those facilities that are of most concern first. Key elements of the City's responsibilities include:
250
-------
. Development of an inspection and monitoring program
. Informing DEQ of any new or existing facilities that require a permit
. Enforcement of City Code
Key elements of DEQ's responsibilities include:
. Issuance of NPDES Industrial Storm Water permits upon referral or approval by the City
. Denial of permit applications for process wastewater discharge into the MS4
. Enforcement where the City lacks authority
Inspections and Monitoring
The section responsible for the implementation of the program is an autonomous work group. It is housed in the
Source Control Division, which includes the pretreatment program. In November 1994, two inspectors were hired to
implement the program under an existing supervisor. DEQ, which had been issuing permits since September 1992,
provided a list of facilities with stormwater permits. A letter was sent to the permit holders requesting that they submit
their SWPCP and all monitoring results. The letter referenced the MOA and included code citing the City's authority.
Inspections were prioritized based on problematic outfalls as determined from information gathered in the Part 1 and 2
application process.
Inspections are usually scheduled in advance with the facility operator but can be performed without notice.
Inspection forms are filled out during the inspection and any readily noticeable issues addressed during a post inspection
meeting. Technical assistance is provided and information on Best Management Practices given in the form of verbal
suggestions and reprints. Facilities are also evaluated for the presence of illicit discharges. All inspections are followed
up with correspondence outlining the findings of the inspection and expectations of the industry. Any item where the
industry is not in compliance with the permit is highlighted with a deadline to meet compliance before escalating
enforcement is pursued. It is the goal of the program to perform annual inspections, at a minimum, of all permitted
facilities.
Industries are also inspected if they are identified as potentially needing a permit. These facilities are identified
through a systematic search using storm water outfall basins prioritized based on problematic outfalls. The basins are
delineated for drainage, the industrial facilities identified within the basin using our database, and facilities selected by
SIC Code. Inspections are also performed in response to referrals, drive-bys, complaints, and responses to an industrial
survey performed in support of the pretreatment program. Prior to an inspection, building records, existing files from the
pretreatment program, and plumbing records are reviewed.
Stormwater sampling of permitted facilities is performed by collecting grab samples at the sample point(s) identified
in the facility's SWPCP. Analyses are performed by the City lab. This sampling does not relieve the facility of its
stormwater sampling responsibilities. Files are developed on the facilities and maintained separately from the
pretreatment files. An electronic database has been developed and is used by both pretreatment and storm water staff.
Enforcement
Enforcement capability was developed in City Code. Where the City does not have enforcement authority, it seeks
voluntary compliance and refers enforcement to the permitting authority when necessary.
Funding
The program is entirely funded through a surcharge on the storm water fee for industrial and commercial accounts.
The storm water fee is currently based on impervious area. This surcharge also funds portions of other programs that
251
-------
have work related activities because of industrial and commercial storm water discharges. The current staffing level is
one supervisor and three technicians. Program costs amount to approximately $280,000 per year. This is primarily
composed of salaries and benefits, but also includes approximately $25,000 for sample analyses The budget also
contains funds for the development of BMPs and educational materials.
Findings
Legal Authority
It is essential that legal authority be developed in order to be able to implement and support a municipal program.
However, what is contained in the enabling legal authority (code) can vary drastically. It is important that the municipal
permit holder review the NPDES Industrial Storm Water permit to determine its adequacy in meeting the municipal permit
requirements. Most industrial storm water permits are general permits and they may not adequately address issues for
which the municipal permit holder is responsible.
For instance, if the municipality is responsible for meeting TMDLs for a particular water body, the industrial permit
may not even require that the facility monitor for these pollutants in its discharge. Provisions should be placed in code
that allow the municipality to require the facility to conduct this monitoring. Another example would be the requirement
to submit SWPCP and monitoring results to the municipality if this is not included in the permit. Nothing in federal
regulations prohibits the municipality from requiring additional controls beyond the permit requirements. A review of the
industrial stormwater permits can help identify elements that should be included.
Another provision that should be considered is the ability of the municipality to require a facility to obtain a permit.
Currently, federal regulations base the requirement for a permit on SIC Code and exposure. There is a caveat that allows
the permitting authority to require a facility to obtain a permit regardless of its SIC Code if that facility is impacting water
quality. However, this could require that the municipality undertake sampling and additional work to prove an impact.
This reduces the efficiency of the program in terms of resources and uniformity. It may even be necessary to include
provisions in the code that allow the municipality to develop its own permit. Such a tactic is time consuming, however,
and could create confusion for the regulated community.
Memorandum of Agreement (MOA)
The MOA should be developed to clearly outline the responsibilities between the permitting authority and the
municipality. Language should be broad enough not to constrict how the municipality implements the inspection and
monitoring program. This allows the municipality to alter the program as information is obtained from inspections without
having to alter the MOA. Probably the most important element of the MOA is the delineation of enforcement. Since the
municipality does not have authority to enforce permit conditions, language should specify that the municipality will enforce
applicable requirements of the Code and seekvoluntary compliance where it has no independent enforcement authority.
If compliance is not obtained using these methods, enforcement would be referred to the permitting authority.
Inspections
The City has placed the responsibilities for implementation of the program within the Industrial Source Control
Division. The section also houses the pretreatment program. It was felt that the responsibilities needed to be separate
because of the large number of facilities that are to be addressed. The City has over 24,000 commercial and industrial
facilities. Of these, nearly 3,000 have the SIC Code that potentially places them in the permitting program. In addition,
a Stormwater Work Group is responsible for addressing other discharges to the MS4, such as pumped groundwater,
boiler blow-down, water supply line flushing, washwater, and others.
For the City's situation, this arrangement has worked very well. The Work Group is able to develop expertise in the
area while having access to existing information from the pretreatment program. Other municipalities have adopted this
approach while others have incorporated the responsibility into the pretreatment program or other existing programs
252
-------
including fire and safety inspections. The municipality needs to consider several items when determining who will be
responsible for implementing the program.
. The number and type of industries
. Existing oversight of the industries (pretreatment, hazardous materials,. . .)
. Existing programs within the municipality
If the municipality decides to place the responsibility in a Work Group that is not dedicated for this purpose, it needs
to ensure that adequate resources exist to implement the program and meet the conditions of the MOA. The stormwater
program may not be the priority of the assigned work section and if resources become inadequate, this work may be
viewed as low priority and may not be addressed at the level that makes it effective.
The City has developed several "partnerships" to expand the inspection program. Informational flyers and a poster
were developed for county sanitarians to use when they inspect restaurants. A simple storm water checklist was
developed for City commercial recycling staff to use when inspecting retail establishments. In both of these cases, it is
important to note that the facilities targeted would not ordinarily be inspected for storm water issues (unless a complaint
was received) and that any issues of consequence would be addressed by storm water staff.
Permits
The DEQ has been issuing permits since September 1991. When the City took over administration of the permits
in the fall of 1994, 63 facilities that discharged to the MS4 had permits. Since that time, an additional 65 facilities have
been identified through inspections of non-permitted facilities. Non-permitted facilities are inspected based on SIC Code
and prioritized by outfall basins that have been identified as problematic. This approach was necessary due to the large
number of industries within the City that have the SIC Code included in the federal regulations. To perform a general
survey of all facilities would have generated much more work than resources allowed. Each site would have to be
evaluated prior to the issuance of a permit as the City is a mixture of combined sewers, sumps, and separated storm
sewers. Staff members spend a considerable amount of time determining where stormwater drainage discharges. A
municipality may be able to utilize this approach if the industrial base is smaller. Federal guidance states that a system-
wide approach to establishing priorities for inspections should be developed.
Based on inspections of non-permitted industries to date (approximately 15% require a permit), and the remaining
facilities that require inspections, it is estimated that an additional 50-100 facilities will be permitted. Based on these
numbers, only 25-30% of facilities requiring a permit had applied when the City took over administration of the program.
However, a large percentage of the facilities not requiring permits still had issues that needed to be addressed or were
given BMPs that they were requested to implement.
SWPCP
The original general permit developed by DEQ did not require that the permit holder submit the SWPCP. When the
City took over administration of the permits, the plans were to be submitted using provisions of the City Code. Over 50%
of the facilities (33 of 63) had not developed a plan within the 180 days allowed in the permit, and of these, 14 (22%) had
not even developed a plan. It is imperative that the municipality includes provisions in the code to obtain these plans if
the provision does not exist in the permit. The requirement to submit the plan allows the City to track its development
and review the plan prior to an inspection. Currently, only 5% of facilities have not met the requirement to develop the
plan in the required time period.
Unfortunately, there is no requirement in the permit that the plans need to be approved. As long as they contain the
necessary elements required in the permit they would be in compliance. This has proven problematic in the quality of
some plans. It also restricts the City's ability to require that the facility implement certain pollution control activities. This
emphasizes the need to include these provisions in the legal authority. The City has taken the approach of strongly
253
-------
suggesting that certain activities be implemented and incorporated into the plan. Once it is in the plan, it becomes part
of the permit and provides a mechanism to require the facility to implement these measures. The City is then able to take
the position of providing assistance in evaluating compliance with the permit. By noting these deficiencies and seeking
voluntary compliance, the City believes it is assisting the industry in meeting the conditions of their permit and benefiting
the environment through the implementation of the SWPCP.
Monitoring
When the City took over administration of the program in 1994, monitoring results were requested from permitted
facilities. Nearly 60% (36) of the facilities had not performed the required monitoring for the previous year. Of these, 22
had not taken the required two samples, while the remaining 14 did not perform the complete analyses. Of the samples
taken, 30% violated standards in the permit. Within the first year, the City was able to raise compliance on sample
collection to over 80% and reduce violations of standards from 30% to 23%. Currently, over 90% of the facilities are in
compliance with sample collection. It is more difficult to compare compliance with standards because a new permit was
issued in 1998, that includes benchmarks for metals that were not in the original permit.
Monitoring results have limitations because they are grab samples taken from a discharge that is short-term in nature
and highly variable. However, they can be used as a tool to measure effectiveness of BMPs and to identify sources of
pollutants. Based on sample results, the City has identified several classes of industries that pose significant pollution
concerns. These are, in order:
. Automotive recyclers (SIC Code 5015) - metals, oil and grease;
. Recycling industry (5093) - metals, oil and grease;
. Transportation facilities (various 4000) - metals, oil and grease, TSS;
. Heavy manufacturing (33--, 34-) - metals;
. Food industry (20--) - TSS, BOD, oil and grease.
Other SIC Code groups either represent a lower threat as a whole or are not present in the MS4. The City is now
using this information to reprioritize their efforts in identifying industries that require a permit. While the City is still
pursuing efforts based on outfalls, they are also developing a parallel effort to inspect all the facilities in these classes.
In addition, investigation efforts by the City identified the Wholesale Distribution of Construction Equipment (5082) and
Heavy Construction Equipment Rental (7353) as significant sources of pollutants. These classes are not included in the
federal regulations, but any municipal program should evaluate these facilities.
Enforcement
Enforcement capabilities have been developed in code for those discharges to the MS4. However, the City does not
have enforcement capability on permit provisions. The City must seek voluntary compliance and refer those matters to
the permitting authority for which they don't have enforcement capability. This has worked to date, but requires
coordination between the City and DEQ. When seeking voluntary compliance, the City uses the threat of referral to the
permitting authority or third party lawsuits to obtain compliance. To make this effective, the permitting authority must be
ready to follow up with enforcement upon the municipality referral.
Funding
As with a number of environmental programs, especially regarding storm water, it is very difficult to measure the
cost/benefit until the program has been in place for a period of time. Costs have been identified, and certain benefits have
been realized. Compliance with permit conditions, for both industry and the City, have been, for the most part, met.
However, has this resulted in a benefit to the environment? City data have shown that industrial land use areas have
254
-------
significantly higher concentrations of pollutants than other land uses. Whereas the industrial land area in the MS4 is only
4%, it accounts for 11% of TSS, 15% of oil and grease, and 24% of metals. It would reason that a program aimed at the
highest concentration of pollutants would produce a good return on the investment, Another benefit of the program has
been the identification and removal of non-stormwater discharges. Approximately 15% of the inspections have identified
non-stormwater discharges, primarily washwater, that were of concern.
Conclusions
The development of a program to monitor and control pollutants from industrial facilities is not one of the six BMPs
that Phase II permit holders will be required to be developed. This may be due, in part, to the assumption that all
industrial permits would be in place because of Phase I requirements. However, our efforts have shown that only 25-30%
of the industries requiring permits had applied prior to the administration of the program by the City.
If a municipality decides to develop and implement a program, it is recommended that it utilizes the accomplishments
of Phase 1 applicants. Phase 1 applicants can provide inspection forms, BMPs, MOAs, code language, and other
necessary components to develop the program. They can also share results of their work to help prioritize the efforts of
the municipality and help decide how to incorporate the work into existing programs. A municipality may also become
a co-applicant with Phase 1 permit holders. If this occurs, the applicant will become subject to an industrial control
program but may be able to utilize the existing program of the permit holder.
If a municipality does not develop a program, it is recommended that it at least work with the permitting authority to
identify who has a permit and the status of their compliance. The municipality should also evaluate the industrial base
in the MS4 and provide this information to the permitting authority if it identifies a facility that may be subject to the
program. It may be prudent to incorporate these activities into the illicit discharge elimination program, which is a
requirement of the permit. Whatever the municipality chooses, it needs to understand that it is ultimately responsible for
discharges from its MS4.
255
-------
Lessons Learned from Three Watershed-Sensitive Development
Demonstration Projects in the Great Lakes Basin
Sarah Bennett Nerenberg
The Conservation Fund
Great Lakes Office
Chicago, Illinois
Introduction to The Conservation Fund
The Conservation Fund (TCF) is a national, non-profit conservation organization that purchases and protects land
- more than 1.6 million acres since 1985. TCF also assists local communities, private landowners and government
agencies with programs that balance conservation with economicdevelopment. TCF works with communities to improve
water quality, build sustainable economic opportunities, and develop leadership skills, activities that put it at the forefront
of conservation across America.
TCF has been active in the Great Lakes Basin since it opened a regional office in 1995. The initial focus of its work
was the Great Lakes Watershed Initiative. This basin-wide effort was designed to raise the local visibility of the nonpoint
source water pollution issue. The Initiative adapted many of the innovative solutions showcased in the National Forum
on Nonpoint Source Pollution. TCF worked with many local partners to launch a network of community-based projects
addressing nonpoint source water pollution in urban and rapidly urbanizing areas in eight states and Canada. The
Initiative was conducted in partnership with the Council of Great Lakes Governors with major funding from the Great
Lakes Protection Fund and Kraft Foods.
TCF expanded several projects as an outgrowth of the Initiative including the watershed-sensitive development work
outlined in this paper and a sustainable development effort in Michigan. In Michigan, TCF facilitates a broad, community-
based sustainable development effort in the Saginaw Bay watershed. The goal of the initiative, which engages local
businesses, community groups, and government agencies, is to better link the environmental and economic well being
of Saginaw Bay communities in order to sustain and improve the region's overall quality of life. This year, the project
received the National Award for Sustainability from the President's Council for Sustainable Development and Renew
America.
Introduction to Conservation Development Project
Currently, TCF is targeting one of the remaining threats to natural resource quality, enhancement, and preservation
in urbanizing areas - conventionally designed subdivisions. In partnership with local developers, community groups,
and government agencies, TCF is working in the Great Lakes Basin on the Conservation Development project. This
project is designed to demonstrate the environmental and economic benefits of watershed-sensitive design through a
series of model developments. In particular, we are working to demonstrate the benefits of watershed-sensitive site-
planning and best management practices that reduce impervious cover and conserve open space. The current model
projects are being developed in Huron, OH, Germantown, Wl, and Niles, Ml. The George Gund Foundation and the
Great Lakes Protection Fund have provided major funding for this project.
We define watershed-sensitive development to include: open space design, significant reduction in impervious
coverage, natural stormwater conveyance and storage to the greatest extent possible, and appropriate construction
mitigation measures. Watershed-sensitive design can be used to build the same number of houses and still preserve
a significant portion of the subdivision's original landscape. These open spaces serve important community and
environmental functions. Agricultural land can be farmed, residents can enjoy recreational and aesthetic benefits, and
important natural areas and systems can be preserved. Alternative designs also reduce the amount of impervious cover.
256
-------
Techniques including narrower streets, porous surface parking areas, stream buffers, and open channelsforstormwater
conveyance minimize runoff from new development and its negative impacts on water resources.
When evaluating potential conservation development projects, The Fund considered the following criteria:
. Local community must be interested and open to new techniques, including flexibility on zoning and subdivision
code issues;
. Property already slated for development and conventional development would have significant negative impacts
on the site itself or adjacent natural resources;
. Project partners represent one of the dominant development paradigms in the Great Lakes (i.e., professional
developer building homes in farm fields, lay developer seeking to hold and protect family or other special lands,
government agency seeking to encourage sound practices); and
. Project site is suitable for demonstrating broad array of site design techniques and best management practices
(BMPs).
Through TCP's work in the Great Lakes, we gathered many lessons-learned that may be applied to other regional
and national efforts, This paper will review many of these lessons with the hope that other communities and
organizations will be able to benefit from our experiences. The paper is organized into the four sections listed below:
I. Overall Lessons Learned
1. Not "One Size Fits All"
2. Measurable Criteria for Watershed-Sensitive Development
3. Adequate Oversight and Inspection
4. Incentive System Needed
5. Relationship to Other Smart Growth Movements
II. Lessons Learned about the Development Process
1. Pace of Development Often Incompatible with Innovative Site Design
2. A Greater Initial Investment in the Baseline Information is Necessary
3. Initial Cost of Watershed-Sensitive Developments
4. Deed Restrictions
5. Need Additional Lay Developer Education
6. Aesthetics Do Not Equal Ecology
III. Lessons Learned about Engineering/Site Design
1. Educate the Engineers
2. Lot Size Often Dictated by Septic Issues
3. Need Hard Science
257
-------
IV. Lessons Learned about Working with Communities
1. Community Initiative
2. Local Official Knowledge Varied
3. Strong Local Partner is Key
4. Final Lessons Learned
Overall Lessons Learned
Not "One Size Fits AH"
The approach used to create a watershed sensitive development must be tailored to the individual, organization, or
developer creating it. The assistance needed by a private landowner that is seeking to preserve portions of family lands,
for example, is quite different from that needed by a professional developer. We found that the models we developed
need to take the different skills and goals of the project's initiator into account very early in the process. For example:
Lay developers (i.e., the individual landowners), not surprisingly, need help with the business aspects of the project,
and are more inclined to make frequent changes to the preliminary site plan and architectural style of the development.
These changes often reflect something the developer has "just learned" or "just considered." These new ideas can add
value to the project, but they also require the technical assistants (e.g., landscape architects and engineers) to be more
patient, more flexible, and firmer than they might be with professional developers.
Professional developers demand immediate turnaround on requests for assistance, and are looking for "the facts"
on what is required to make a development watershed-sensitive. They can be somewhat impatient with the notion that
there are not a fixed and specified set of best management practices and site design practices that, if employed, will
"always" result in an "environmentally friendly" development.
Measurable Criteria for Watershed-Sensitive Development
As we began to design the model projects, it became apparent that there were no specific criteria available to
measure the benefits of the watershed-sensitive design. A tool was needed to encourage developers to fashion
environmentally friendly site designs, to help communities add flexibility to their local ordinances, and to provide a
standard that can be understood by both homebuyers and existing community residents.
In response to this, TCF developed the Conservation Development Evaluation System (CeDES) as a rating system
to evaluate a conservation development over the development's lifetime with emphasis on water quality and landscape
impacts. The purpose of CeDES is to encourage developers to think about environmental concerns earlier in the
planning process and to provide consumers and communities with a means of assessing the impact of better site design
practices. It was developed with input from over thirty national professionals skilled in planning and evaluating
conservation developments. It may be viewed at http://www.conservationfund.org/conservation/sustain/gloindex.html.
Adequate Oversight and Inspection
One of the biggest challenges is ensuring that the contractors are building in an environmentally responsible manner.
Even if the developer is committed to minimizing the impact of the development on the environment, if contractors are
not educated and committed it may not happen. This is a challenge for many local agencies and municipalities who have
limited staff for constant inspections. Even if the communities have ordinances that require construction erosion controls
etc., without constant inspection many contractors do not follow the requirements. The nonpoint source pollution from
construction, especially the sediment loadings, can negate any benefits from the alternative site design. One
recommendation is for the community to require that an environmental inspector be hired specifically for the site. The
inspector may be from a consulting firm or from a local Soil and Water Conservation District.
258
-------
Incentive System Needed
We expect that this process will proceed much more quickly in communities that have recognized the threats
conventional developments pose and have begun developing strategies to address them. In order to expand
conservation development practices to a broader constituency, state or county agencies may need to develop incentive
programs that prompt local developers to undertake these projects. With each community, we encouraged the creation
of incentives for watershed-sensitive development. These included density bonuses for the developer through credits
for land preservation and minimization of impervious coverage. We also investigated the use of the Clean Water Act
State Revolving Fund (SRF). Among other uses, these funds are used to reduce nonpoint source pollution and could
encourage watershed-sensitive development. The State of Ohio has successfully used the SRF for this purpose and
we hope to pilot the same use of the SRF loans in other states. Incentives such as Ohio's loan program, coupled with
the higher financial returns these developments are expected to generate, are making watershed sensitive developments
more the norm in the Great Lakes Basin.
Relationship to Other Smart Growth Movements
There are many "Smart Growth" movements currently being debating and promoted throughout the country.
Watershed-sensitive development is just one part of the equation. At times, we were challenged to show how this fits
into overall community sustainability efforts. The work that the National Site Planning Roundtable completed to develop
"Better Site Design" principles has been invaluable in demonstrating how these different movement can work together
(Center for Watershed Protection, 1998). We often say that watershed-sensitive development is one option for Smart
Growth but that a community needs to find the correct planning principles to work for their residents and issues. Those
in the Traditional Neighborhood Design (TND) movement challenge putting sidewalks only on one side of the street,
which we recommend for reducing impervious coverage. We also suggest that if sidewalks are on both sides of the
street at least one should be made of pervious materials. There also are environmental groups that challenge us for
encouraging greenfield development instead of infill development. Again, watershed-sensitive development is only one
option of many and if the market is going to demand suburban fringe growth, at least we can work with the communities
and developers to ensure that it is done with maximum possible protection and enhancement of the natural resources.
Lessons Learned about the Development Process
Pace of Development Often Incompatible with Innovative Site Design
The pace of development and the pace of government decision-making often are absolutely incompatible.
Developers with outstanding loans on land need to move quickly to ensure a development is economically viable.
Government agencies, on the other hand, are very concerned about the impacts of development, but move very
cautiously, especially when they are undertaking something new. The result is that it is easier for both government and
developer to create conventional, environmentally harmful developments than to do something better.
On the demonstration projects, TCF took special care at the outset of the process to communicate the timelines of
each participant to the other. In this way, we hoped to keep the parties from throwing up their hands and giving up. For
the region, however, we explored possibilities to get communities to adopt "fast track" approvals for watershed-sensitive
communities. The first need is to show municipal authorities that these developments deliver tangible benefits, then
we can help them develop mechanisms such as a streamlined review process and updated subdivision and
zoning ordinances that encourage their creation.
A Greater Initial Investment in the Baseline Information is Necessary
Before planning a watershed-sensitive development, fairly detailed baseline information including topography, soils,
and wetlands delineations is needed. Although developers hope that they get enough of the baseline site information
before beginning design work, inevitably the risk/benefit of doing extensive baseline work (e.g., soil borings) may
preclude the developer from getting all of the necessary baseline data. The common practice is to use "engineering
259
-------
judgment" based on existing data and extrapolation to the rest of the site. Unfortunately, especially when the drainage
plan is an integral part of the initial site design, relying on "engineering judgment" simply is not sufficient. For example,
on the Ohio site we relied on the existing soils information to design the swale system. After presenting a preliminary
site plan, the developer discovered through additional research and sampling that the available soil information did not
accurately represent the existing soil conditions and the drainage plan had to be reconstructed.
Initial Cost of Watershed-Sensitive Developments
Planning and developing a watershed-sensitive development takes time and costs money. Both lay and professional
developers often underestimate these initial costs. Professional developers often leave site planning to their engineers.
The engineers typically obtain a wetlands delineation and examine soil and topography maps, but do not evaluate the
site from a watershed or ecological perspective. Although lay developers may be more familiar with the special features
of their properties than professional developers, both need help to catalog all the features and to understand the site's
role in the surrounding landscape. Quite reasonably, professional developers often are unwilling to undertake these
expenses until they have a sense of the project's scale and niche in the market. We believe, and existing watershed-
sensitive developments indicate, that the costs of evaluating a property from an ecological and a watershed perspective
will be recovered when the development is sold out.
Deed Restrictions
The deed restrictions (i.e., covenants, conditions, and restrictions) necessary to ensure that the development will
continue as a watershed-sensitive development in perpetuity are a lot more extensive than typical deed restrictions.
Early in the process, sample restrictions for various developments should be presented to the developer and to the
community so that they understand the consequences of using some of the watershed sensitive techniques. The
developer will gain an appreciation for the long-term commitment necessary for a successful development and local
officials may be put at ease when they recognize that major additional responsibilities (e.g., swale maintenance) rest with
the homeowners association and not the local government.
Weed Additional Lay Developer Education
Private landowners need to be assisted and educated through the process. Although these initiators often have a
deeper environmental commitment than professional developers, they often do not understand what kinds of activities
on their properties will have negative watershed impacts. For example, on one of our projects, the lay developer
suggested that a pond be built each time an area of low-lying ground is found to be wet most of the year. Once informed
about the relationship of these areas to more prominent wetlands on the site, the developer agreed to treat these areas
more appropriately (i.e., preserving and enhancing the existing wetlands). The professional developers understand
stormwater and wetlands issues better because they operate in the regulatory arena. The lay developers may need to
be educated about the significance of these issues and other issues that are common knowledge to professional
developers.
Aesthetics Do Not Equal Ecology
Another aspect of landowner education is the principle that aesthetics do not equal ecology. Just because a
development preserves or creates attractive green spaces does not necessarily indicate that it is not harmful to the
surrounding watershed. Accordingly, the criteria we developed for watershed-sensitive development (see discussion
above under Measurable Criteria) incorporate appropriate baseline evaluation of the site to insure key resources are
protected, and a thorough analysis of the stormwater impacts after development.
260
-------
Lessons Learned about Engineering/Site Design
Educate the Engineers
If any of the county, township, or city engineers are not comfortable with the techniques being used, they can turn
down the project at any point in the review process. In all three of our projects, the "old-timer-" engineers were extremely
conservative and feared change more than any other local officials. We found that the developers' engineers need
constant oversight and education to design the sites using the watershed-sensitive techniques. Unfortunately, without
a broad effort to educate engineers, they will have to be educated one community or county at a time. Once these
techniques become more commonplace, we assume that such a great initial effort will not be necessary.
Lof Size Often Dictated by Septic Issues
Wastewaterissuesoften control the form, location, and economicfeasibilityof a new residential subdivision. In many
parts of the Great Lakes region, heavy clay soils strictly limit the functioning of conventional septic systems. For this
reason, lot size is frequently dictated by septic issues as much as by local zoning. Although there are some alternate
systems (e.g., constructed wetlands and community systems) being piloted and used in the region, local health officials
are very cautious about permitting them. This caution arises both from concerns about their technical functioning and
about long term maintenance issues. Communities already feel burdened by the need to monitor individual septic
systems. They are skeptical about a homeowner association's ability to reliably maintain a community treatment system.
Wastewater treatment issues should be considered up front in evaluating the feasibility of clustering homes on a
particular site. If a public sewer does not serve the site, clustering probably will not work as well (i.e., the individual lot
sizes will not be able to be reduced as much). There is the possibility of placing the leach fields in the common property
to increase the overall open space percentage.
Weed Hard Science
Although there is a great deal of national literature detailing watershed-sensitive development techniques, there is
not a lot of research documenting the extent of the water quality benefits they provide in the field. The Center for
Watershed Protection (CWP) recognizes that there is a lack of water quality monitoring data that evaluates the
techniques in varied site conditions and is working to develop and encourage more studies. Through consultation with
the CWP, the Northeastern Illinois Planning Commission, and the Wisconsin Department of Natural Resources (WDNR),
we found that funding for long-term monitoring of these techniques is scarce. Without this data, many of these
techniques may be challenged successfully by skeptical local officials. With the assistance of Old Woman Creek
Estuarine Research Reserve, one of our local partners, we are monitoring the water quality at the Ohio site. We hope
that they will be able to continue the monitoring after our grants are over. We also are working with the WDNR to secure
funding for long-term water quality monitoring at the Germantown site. It is our hope that this information will continue
to back up many of the claims of watershed-sensitive development and that funding will continue to support these efforts.
Lessons Learned about Working with Communities
Community Initiative
Without community buy-in and interest in these concepts, even the most enlightened developer is not going to be
able to get a project approved. When we first started this project, we thought that the developers were going to be the
"hard sell." In two out of three of the communities, it actually has been the communities that needed more education.
In the Huron project, the developer was sold on many of the alternative site design techniques until he kept getting
negative feedback from the township board. This site was chosen because of the commitment of the developer and the
obvious benefits to the surrounding water resources. What was not realized was how much resistance there would be
in the political arena. At this project, we had several informal meetings with local officials prior to presenting a conceptual
plan, but because the process was developer-initiated, they continued to be resistant throughout the process.
261
-------
Local Official Knowledge Varied
Municipal, county, and state officials with similar regulatory responsibilities often have very different views about the
appropriateness of new techniques. Although there are no hard and fast rules about who is likely to be more
progressive, disagreements are common and a primary cause of frustration among developers.
As development is now regulated, it is more expensive and time-consuming for a lay or professional developer to
create a watershed-sensitive development. The only way for a developer to address this situation is to inform local
regulators and planning officials about the project early on, and to involve them in the process Unfortunately, this
involvement will probably not speed up the process for the individual developer, but after a few such projects are
launched, we believe the barriers for these kinds of developments will be lowered.
Getting everyone with a regulatory or permitting role on a project involved at the very beginning is absolutely vital.
If a project that includes techniques that have not been implemented in the region before gets too far along before all
the regulators and municipal officials are brought in, the "stranger to the deal" can feel left out and derail the project.
Much of this problem will be allayed once a few watershed-sensitive developments are built, but until then, developers
and regulators pushing for these practices need to make special efforts to get everyone to the table early. Of course,
this process increases the costs of doing the development initially, but it can keep it from falling apart after significant
site planning and related costs are incurred.
Strong Local Partner is Key
Throughout this project, TCP acted as a facilitator between the communities and developers and as a representative
of the silent third party, the environment. We believe that as each community begins to look at this type of development,
this third party is key to the success of a project. Although there are many merits to approaching communities as a
national organization, without a primary local partner who is well-versed in the trials and tribulations of the development
process (or willing to learn them), it is difficult to proceed. A preferable arrangement would be a local organization, such
as a land trust, leading the effort with support from a regional or national organization or technical assistance center.
A local organization will have a greater vested interest in and knowledge of the local environment, will know the local
officials and political and personal histories, and will be able to track and monitor the day-to-day activities surrounding
the development. In the long term, local land trusts may become a key player in this area. They understand land
conservation and watershed issues, frequently have close ties with both local landowners and local government officials,
and have some comprehension of the development industry.
Final Lessons Learned
Several realities of the development process that have little to do with the challenges of watershed-sensitive
development are important to mention for groups and communities considering this type of project. One is that the
personalities and reputations of the developers can make or break a project. On our project in Ohio, the developer
apparently had a "history" with several of the plan commissioners. Our partners in the community think that the plan
commission and the engineers were being unduly unfair during the review process. Also, one of the developers in
Wisconsin has a reputation for "low-end" development. Because of this reputation, the Village is afraid that the
developers will do their typical development in their town.
Another reality of the development process is that the Village Planner of Germantown estimated that 60% of
submitted development plans are reviewed by the plan commission and less than 50% of zoning requests are approved.
All of our projects include a zoning request because the current local ordinances do not include a provision for
watershed-sensitive development.
Conclusion
At all three of our model sites there are already signs of new developments being proposed with many of the
watershed-sensitive techniques. In Wisconsin, the developer was approached by neighboring communities to design
262
-------
similar subdivisions. In Michigan, several local officials have stated interest in adding language in their new ordinances
that would encourage this type of development. In Ohio, our local agency partner, Old Woman Creek National Estuarine
Reserve, was approached by a developer who has been watching the process and is interested in using some of the
techniques at an adjacent site. While the review processes for all three projects have not been as easy as anticipated,
it is expected that the next round of developments will have an easier time because of the trailblazing work done before
them.
References
Arendt, Randall. 1997. Conservation Subdivision Design. Natural Lands Trust, Media, PA.
Center for Watershed Protection (CWP). 1998. Better Site Design: A Handbook for Changing Development Rules in
Your Community. Prepared for the Site Planning Roundtable, CWP, Ellicott City, Maryland.
Center for Watershed Protection (CWP). 1998. Consensus Agreementon Model Development Principles to Protect Our
Streams, Lakes, and Wetlands. Prepared by the Site Planning Roundtable, CWP, Ellicott City, MD.
263
-------
Continuous Deflection Separation (CDS) for Sediment
Control in Brevard County, Florida
Justin Strynchuk, John Royal, and Gordon England, P.E.
Brevard County Surface Water Improvement
Viera, FL 32940
Abstract
In July 1997, Brevard County's Stormwater Utility Program installed a new type of trash and sedimentation control
device called a Continuous Deflection Separation (CDS) unit. This was the first American installation to use the CDS
technology, which was developed in Australia. After installation, autosamplers were placed upstream and downstream
of the CDS unit and a year's duration of sampling data collected. Monitoring has shown that the CDS unit has provided
an average 52% removal efficiency for total suspended solids and 31% removal efficiency for phosphorus.
Introduction
Stormwater sedimentation is a primary source of pollution to Florida's Indian River Lagoon. Suspended solids and
turbidity reduce sunlight penetration in the lagoon which negatively impacts seagrass growth. Where land is available,
detention ponds effectively reduce most of the suspended solids from Stormwater flows. When land is not available,
alternative, less effective, treatment methods must be used.
The CDS technology was initially developed in Australia to provide an effective method for trash and solids removal
from Stormwater flows. The screening action within the unit provides for 100% removal of trash and particles down to
4700 microns. In addition, the unique circular design creates centrifugal action within the round concrete box which
propels suspended solids to the center of the box and down into the storage chamber.
Methods
The location chosen for the CDS unit installation was along a ditch at the north end of Brentwood Drive, south of Port
St. John. The drainage basin for this location was 24.87 hectares (61.45 acres) in area. This basin has Type A soils
along a sand ridge. The land uses are 24.87 hectares (6.7 acres) of roadway (US Highway 1), 8.04 hectares (19.87
acres) of industrial park, 9.47 hectares (23.39 acres) of vacant land, and 4.65 hectares (11.49 acres) of commercial
property. The industrial area has a permitted Stormwater system. A significant land feature is a 2.02 hectares (5 acre)
dirt parking lot, 152.4 meters (500 feet) upstream of the site around a local restaurant. This parking lot has a steep slope
and is composed of fine white base material. There is evidence of heavy silt buildup in the inlets and pipes downstream
of this parking lot, along US 1.
There is an earthen ditch running eastward 76.2 meters (250 feet) upstream from the project location. At the project
site, there is an existing 122 centimeter (48 inch) RCP driveway culvert in the ditch which discharges to a concrete
channel running 152.4 meters (500 feet) eastward to the Indian River. The time of concentration to the site is 63 minutes,
with a 1 0-year flow of 1,557.2 L/sec (55 cfs) and mean annual flow of 1,177.9 Us (38.2 cfs). In Brevard County, the 10-
year storm is 20.1 centimeters (7.9 inches) of rainfall and the mean annual storm is 13.97 centimeters (5.5 inches) of
rainfall. There is no base flow at this location.
264
-------
A diversion weir 68.58 centimeters (27 inches) tall is placed in front of the 122 centimeter (48 inch) culvert so as to
divert flows over 254.8 L/sec (9 cfs) around the unit. In 18 months of observations, the water level has risen over the
weir one time.
A 76.2 centimeter (30 inch) concrete pipe was constructed adjacent to the existing 122 centimeter (48 inch) pipe in
order to divert flows to the CDS unit. The 76.2 centimeter (30 inch) pipe enters the CDS unit tangentially to start the
circular flow within the unit.
The CDS unit consists of three circular, concrete chambers stacked on top of each other. The top chamber, where
the water enters the unit, has a 1.524 meter (5 feet) inner diameter and is 188 centimeters (74 inches) tall. The middle
chamber has a 2.44 meter (8 feet) inner diameter and is 127.54 centimeters (51 inches) tall. In the middle chamber is
a 1.524 meter (5 foot) diameter stainless steel screen matching the walls of the top chamber. The screen has 4700
micron holes to filter larger materials. The bottom chamber has a 1.22 meter (4 foot) inner diameter by a 1.22 meter (4
foot) tall sediment sump.
Water enters the unit in a clockwise rotation. When the water passes through the screen, it flows counter-clockwise
between the screen and outerwall until it reaches a 76.2 centimeter (30 inch) concrete pipe. This exit pipe is tangentially
placed for smooth exit flows. The elevation of the exit pipe rises 96.52 centimeters (38 inches) from the lower chamber
to the outflow channel downstream of the 122 centimeter (48 inch) culvert. This rise in elevation keeps the normal water
level in the unit near the top of the second chamber at all times. There is no base flow at this location.
The top of the unit is flush with the surrounding ground and has a 0.91 meter (3 foot) square, lockable, stainless steel
access cover. This feature allows for easy access with a vacuum truck for cleaning purposes.
The CDS unit was installed on July 17, 1997. Installation took two days with the precast structures. A large crane
was required to lift the chambers into place. A 4.57 meter (15 foot) deep hole was excavated for the structure.
In conjunction with the CDS unit installation, County forces cleaned the ditch upstream of the unit. Two days later,
a significant rainfall event occurred and 2,294 kilograms (6,600 pounds) of sediment from the upstream ditch wastrapped
in the unit. After that storm, the ditch was reworked and sod was laid. The sod greatly reduced the volume of sediment
washing into the unit.
Cleanouts were also performed on November 17,1997, removing 626.84 kilograms (1,382 pounds) of sediment and
2.88 meters (34 cubic feet) of trash and debris, and again on May 6, 1998, with 998 kilograms (2,200 pounds) of
sediment. The solids removed from the unit are taken to the Brevard County landfill for disposal. The volume of water
stored in the unit is greater than the vacuum truck capacity, so decanting is performed on nearby sandy soils to avoid
a second trip to the landfill for disposal.
Evaluation of the CDS Unit During Storm Events
The intent of the sampling was to evaluate the effectiveness of the CDS unit in removing pollutants from a storm
event prior to discharging stormwater into the Indian River Lagoon.
Five storm samples were collected at the CDS unit between April 1998 and March 1999. The storm events occurred
after dry periods ranging between 7 and 75 days. Protocol for this program dictated that if the sample collection devices
(autosamplers) were triggered at intervals of less than three days between storms, the samples were to be discarded.
This situation did not occur during the year, and near-drought conditions were observed in the sample area throughout
most of the year-long monitoring program.
Rainfall was measured at the sampling site by a tipping bucket rain gauge, and additional rainfall data obtained from
the Orlando Utilities Commission (OUC) power generating plant 5.6 km (3.5 miles) to the north of the CDS installation.
Review of the rainfall data collected indicates that the majority of the water passing through this BMP was from
precipitation falling on the upland, 18.72 hectare (46.25 acre) watershed. The variation noted in both coverage and
265
-------
amount of rainfall helps illustrate the localized nature of the storms occurring along the Lagoon coastline. During this
sampling period, water flowing off the drainage basin contributed much more flow through the CDS unit than would have
been expected based on the rainfall recorded at the sample site.
Samples were collected through the use of automated storm water samplers; one at the inlet and another at the
outlet pipe of the CDS unit. All samples, associated blanks, and duplicates were collected in accordance with our state-
certified Comprehensive Quality Assurance Plan.
The stainless steel intake strainers for the samples were mounted on the reinforced concrete pipe, slightly off center
bottom, and both angled away from the flow. This was to prevent the strainers from becoming silted over by sediments
and allow collection of representative water samples. Flow rates during the storm events were measured initially utilizing
water level meters (ISCO bubbler type) in conjunction with a 90-degree V-notch weir, but eventually replaced with a
Doppler area-velocity flow meter which provides a more accurate flow assessment. Initially, two bubbler meters were
installed with both bubbler tubes mounted on the upstream weir. However, this led to difficulties in estimating just when
to trigger (time delay) the downstream sampler in order to collect samples from the same "plug" of water.
During the first three sample events, water levels recorded were correlated to flow, and the samples were manually
composited to give a flow-weighted composite sample from each sampler. Both inlet and outlet sample sets were
composited identically, in accordance with the EPA NPDES Stormwater Sampling Guidance Document (July 1992).
Discreet samples were collected for the fourth and fifth events.
It was intended that the third sample event would include a mass balance calculation. The CDS unit sump was
thoroughly cleaned utilizing a VAC-truck to ensure that the material collected was a result of the one storm to be
evaluated. Inlet and outlet stormwater composite samples were again collected, with the addition of a sediment (Table
1) and water column sample from the sump. Sediment depths were measured at five locations; four from the corners
of the lid opening and one in the center. Based on a depth of 13.21 centimeters, a sump diameter of 1.22 meters (4 feet)
and an estimated 1,410.6 kg/m3 (88 Ib/ft3), (based on previous sediment weight evaluation), approximately 217.3
kilograms (479.2 pounds) of sediment was collected in the unit from storm three. Based on the concentrations
measured, 126.07 grams (4.443 ounces) BOD 5, 33.587 grams (1.184 ounces) of metals, and 122.81 grams (4.33
ounces) of Total Kjeldahl Nitrogen (TKN) were removed.
Table 1. Sediment Chemical Analysis For Storm #1
Parameter
Arsenic
Barium
Benzo(b)fluoranthene
BODS
Cadmium
Chromium
Copper
Iron
Lead
Nickel
Silver
Total Kjeldahl Nitrogen
Total Phosphorus
Zinc
Sediment
Grab
0.096
3.4
260
650
0.03
1.1
1.2
220
2
0.4
0.16
450
79
14
Grab Duplicate
0.11
2.9
ND
510
0.033
1.1
0.95
260
2.2
0.36
0.059
680
230
14
Average
Value
0.103
3.15
250
580
0.0315
1.1
1.075
240
2.1
0.38
0.1095
565
154.5
14
Detection
Limit
0.069
0.14
240
2.7
0.014
0.027
.0.027
0.55
0.041
0.069
0.014
37
9.2
0.27
Units
Mg/Kg
Mg/Kg
Ug/Kg
Mg/Kg
Mg/Kg
Mg/Kg
Mg/Kg
Mg/Kg
Mg/Kg
Mg/Kg
Mg/Kg
Mg/Kg
Mg/Kg
Mg/Kg
Notes:
Equipment Blank Water Yielded ND for all listed analytes.
The benzo(b)fluoranthene mean value was calculated with the
RDL as the lower value for the duplicate.
Only parameters with values above detection limit are listed. Many others were tested below detection limits.
266
-------
For this third sample event, the upstream, or intake flowmeter bubble tube was mounted on the 90-degree V-notch
inlet weir, as it was for previous sample events. The downstream bubbler, however, was moved and attached to the
downstream discharge pipe. This change was necessary to account for the lag time between when the first sampler
received flow at the beginning of the storm, the time required to fill the sump with 8,115 liters (2,144 gallons), and
discharge to flow past the second sampler several minutes later. The problem encountered with this revised setup was
that the upstream V-notch weir used to determine the flow was overtopped, allowing flow around and over it, preventing
an accurate flow measurement. This led to disparity in the estimation of actual flow through the unit. Due to the
questionable flow measurements, it was not possible to calculate the mass balance.
For the fourth sample event, an ISCO Doppler area-velocity flow meter was mounted in the bottom of the outfall pipe
of the CDS unit. Upon registering a water level rise of one inch, this unit triggered both upstream and downstream
autosamplers. The autosamplers were synchronized, collecting a bottle set in each ISCO at the same time. With this
methodology and placement, overtopping the weir, flow bypassing, and pressurization were no longer potential sources
of error. Since the samplers now triggered only when the sump was full, it was also somewhat easier to accept the
premise of "what went in, must have come out."
Appropriate trigger points were selected in orderto allow sufficient water depth forthe velocity meter probe to operate
properly. We found that the Doppler area-velocity flow meter probes appear to function erratically when covered by less
than one inch of water, and believe that measurements taken when the water was at this depth are suspect. Two-bottle
sample sets were collected at sampler initiation, and at 1 0-minute intervals during the storm. During previous sample
excursions, samples were manually composited. Due to a high suspended solids content, (heavy particles including
sand) that rapidly settled in the sample container, it was questioned whether the composite samples were truly
representative of the solids collected. Therefore, discrete two-bottle sets collected every 10 minutes were sent to the
laboratory without being composited.
For the fifth sample event, two-bottle sample sets were again collected at sampler initiation, and at lo-minute
intervals during the storm. As with the previous sample event, sample sets were not composited but sent for analysis
as six individual, two-bottle sets. The sample bottles for bottle sets six were not collected due to insufficient water to
cover intake strainers, as the storm was not of adequate duration to produce the last 1 0-minute sample. Because of
numerous problems encountered in the previous storm event samplings, along with refinements in sampler setup and
flow measurement, the fifth storm sample event is considered the most accurate to determine what pollutant reduction
is provided by the CDS unit for that storm. The individual two-bottle sets showed the variation in pollutant loadings
throughout the storm event and the corresponding removal under the varying loads. Unfortunately, this was the lowest
flow storm encountered, which may account for higher than normal removal efficiency. Maximum flow was estimated
to be only 136 liters/sec (2.16 gpm). The average pollutant reduction between inlet and outlet samples for this event was:
BODS 53%, COD 52.6%, TP 36%, TSS 56%, and Turbidity 74.8%.
Sample results are presented in Tables 2 through 4 for the five sample events. Storm event 2 showed a 23%
reduction in turbidity, but no reduction in the other parameters. Storm 4 showed an increase in most parameter
concentrations between inlet and outlet that could not be attributed to resuspension due to a full sump, since the sump
had been cleaned prior to the third event. Data for these two storms are therefore suspect. For events 1, 3, and 5, the
average removal efficiencies for those parameters that showed a reduction were: TSS 52%, Turbidity 46.9%, BOD
34.2%, COD 35%, and TP 30.6%
After each sample event, field observations were made of the appearance of the sample jars, each containing a
water sample that had been collected at progressive ten-minute intervals throughout the storm flow. Outlet samples
typically appeared to be less turbid than the corresponding inlet samples, and also had less sediment on their bottoms.
An observation was also made of the water surface inside the CDS unit proper. There was typically a thick layer of
floating grass and other vegetation, an oil sheen, glass and plastic bottles, plastic sheets and bits, seeds and nuts, sticks,
and a surprising amount of Styrofoam cups and particles.
267
-------
Table 2. Storm #1-#3 Test Results - Composite Samples
STORM 1
CDS Inlet
CDS
Outlet
Change
Percent
Reduction
pH Total
Turbidity
s u Suspended NTU
Solids
mg/l
7.6 220
7.4 110
0.2 100
3% 50%
Maximum flow rate = 5.488 literslsec (87 GPM,
Storm Duration
Rainfall @ OUC
STORM 2
CDS Inlet
CDS Outlet
Change
Percent
Reduction
= 67 minutes
0.254 cm (0. 1 inch), @ SITE
pH Total
s u Suspended
Solids
mg/l
8.4 350
8.2 350
0.2 0
2% 0%
Maximum flow rate = 8.39 liters/sec (133 GPM,
Storm Duration
Rainfall @ OUC
STORM 3
CDS Inlet
CDS Outlet
Change
Percent
Reduction
= 68 minutes
1.778cm (0.7 inch),.® SITED.
pH Total
s u Suspended
Solids(mg/1 )
7.6 300
7.6 150
0 150
0% 50%
180
100
80
44%
0. 19 cfs)
not recorded
Turbidity
NTU
440
340
100
23%
O.Scfs)
0762 cm (0.03 inch)
Turbidity
NTU
110
86
24
21 .8%
BOD5-Day COD
mg/l mg/l
28 150
23 110
5 40
18% 27%
BOD5-Day COD
mg/l mg/l
8.2 20
8.2 20
0 0
0% 0%
BOD5-Day COD
mg/l mg/l
12 71
8.2 53
3.8 18
31 .7 % 25.4
Total
Phosphorous
mg/l
1.4
1
0.4
29%
Total
Phosphorous
mg/l
0.86
0.86
0
0%
Total
Phosphorous
mg/l
1.3
0.95
.35
27%
Maximum flow rate = 149.75 liters/sec (2374 GPM, 5.29cfs)
Storm Duration
Rainfall @ OUC
= 113 minutes
4.064cm (1.6 inch), ©SITE 1
.27 cm (0.5 inch)
268
-------
Table 3. Storm #4 Test Results - Discrete Samples
Set 1
@ initiation
Inlet 1
Outlet 1
Change
Percent
Reduction/Gain
Inlet 2
Outlet 2
Change
Percent +/-
Inlet 3
Outlet 3
Change
Percent
Reduction/Gain
Inlet 4
Outlet 4
Change
Percent
Reduction/Gain
Inlet 5
Outlet 5
Change
Percent
Reduction/Gain
Inlet 6
Outlet 6
Change
Percent
Reduction/Gain
Maximum flow rate =
Storm Duration = 55
BOD5-Day
(mg/1)
2.1
5.4
+3.3
+61%
6.6
7
+0.4
+6%
6.7
6.7
0
0%
6.3
NT
Na
Na
5.6
6.4
+0.8
+13%
6
6.3
+0.3
+5%
60.30 liters/sec
minutes
Rainfall @ OUC 2.794 cm (1 .1 inch),
COD
(mg/1)
2
2
0
0*%
15
18
-3
+17%
25
24
-1
-4%
45
NT
Na
Na
t
33
30
-3
-9%
39
33
-6
-15%
(956 GPM, 2.13
PH
(SU)
8
7.8
-0.2
-3%
8.3
8.4
+0.1
+1%
8.2
8.3
+0.1
+1%
8.1
NT
Na
Na
8
8.2
+0.2
+2%
7.9
8.2
+0.3
+4%
cfs)
@ SITE 0.006 cm (0.002
Total
Phosphorous
(mg/1)
0.32
0.19
-0.13
41%
1.2
0.94
-0.26
-22%
1.2
1.5
+0.3
+20%
1.6
NT
Na
Na
1.6
1.6
0
0%
1.6
1.5
-0.1
-6%
inch)
Total Suspended
Solids
(mg/1)
690
320
-370
-54%
1400
1600
+200
+13%
830
550
-280
-34%
330
NT
Na
Na
290
170
-120
41%
220
270
+50
+19%
Turbidity
(NTU)
99
120
+21
+18%
1800
1000
-800
44%
530
430
-100
-19%
200
NT
Na
Na
300
260
40
-13%
120
230
+110
+48%
269
-------
Table 4. CDS Storm #5 Test Results - Discrete Samples
Inlet 1
Outlet 1
Change
Percent
Reduction/Gain
Inlet 2
Outlet 2
Change
Percent
Inlet 3
Outlet 3
Change
Percent
Reduction/Gain
Inlet 4
Outlet 4
Change
Percent
Reduction/Gain
Inlet 5
Outlet 5
Change
Percent
Reduction/Gain
Average Percent
Change
BOD5-
Day
(mg/1)
4.6
4.0
-0.6
13%
10
3.8
-6.2
62%
13
4.7
-8.3
64%
9.9
3.9
-6
61%
9.6
3.4
-6.2
65%
53%
COD
(mg/1)
68
18
-50
74%
51
23
-28
55%
55
33
-22
40%
53
29
-24
45%
53
27
-26
49%
52.6%
PH
(SU)
7.8
7.91
+.1
1 %
7.81
7.9
+.1
1%
8.2
7.6
-0.6
7%
9.2
7.7
-1.5
16%
9.4
7.6
-1.8
19%
-%
Total
Phosphorous
(mg/1)
0.23
0.18
-0.05
22%
0.25
0.18
-0.07
28%
0.3
0.18
-0.12
40%
0.35
0.18
-0.17
49%
0.29
0.17
-0.12
41%
36%
Total Suspended
Solids
(mg/1)
49
11
-38
78%
59
19
-40
68%
23
21
-2
9%
39
15
-24
62%
35
13
-22
63%
56%
Turbidity
(NTU)
16
4.3
-11.7
73%
38
6.9
-31.1
82%
23
12
-11
48%
61
7.2
-53.8
88%
56
9.4
-46.6
83%
74.8%
Maximum flow rate 0.136 liters/sec (2.16 GPM, 0.005 cfs)
Storm Duration =50 minutes
Rainfall @ OUC 1.016 cm (0.4 inch), @ SITE, 0.5842 cm (0.23 inch)
Conclusions
While none of the sample events were a perfect combination of a good flow and everything working right, the data
collected, and our observations, certainly indicate that the CDS unit is operating as intended and removing significant
quantities of debris and suspended materials prior to discharge to surface waters. It was quite impressive to prevent this
trash and sediment from washing out into the lagoon during a normal rain.
The phosphorus removals observed for the CDS Unit, as with any BMP of this type, will not have a high degree of
accuracy, due to leaching of nutrients from grass, leaves, and other organic debris. If there are no base flows, these
leached nutrients will be washed out with runoff and skew sample readings. A much more comprehensive analysis is
available in the library of the web site www.stormwater-resources.com.
Future Evaluations
More data are necessary to further evaluate this BMP. Due to the inherent inaccuracies in water quality sampling,
additional determination of the efficiency of this type of BMP could be made by conducting a mass loading and sediment
evaluation. Much of the sediment collected in this type of BMP is invisible to current testing techniques since it is
comprised of large particles that roll along the bottom of the pipe. Yet, known quantities of sediment are being collected.
270
-------
A previous study of baffle boxes resulted in the same conclusion. Future sediment analysis from the CDS unit could
be compared to the baffle box data previously collected. Brevard County will be conducting a sediment evaluation at
three baffle box sites over the next 12 months that will provide additional comparison. As time permits, Brevard County
will also collect additional sediment data from he CDS unit.
271
-------
Use of Automated Technologies in Watershed Management Planning
Lake County Stormwater Management Commission (SMC)
Libertyville, IL
Introduction
The Lake County Stormwater Management Commission's (SMC) is working with many agencies to develop
comprehensive watershed plans. These watershed plans involve data collection and collation, problem analysis,
alternative solutions identification and action plan development. The watershed assessment includes hydrologic and
hydraulic modeling; floodplain and floodway mapping; and water resource assessment. As part of a watershed
management plan, one of the end results is to update floodplain maps and to map depressional storage areas.
Other end products of this effort include location maps of water resources, including wetlands and regional detention
sites, with identification of those needing preservation, enhancement or restoration. With this information, projects
can be prioritized and cost estimates determined in order to assist local governments in implementing the action
plans.
Lake County, Illinois is located in the northeastern corner of Illinois and is one of the fastest growing counties in the
country. The county has 61,000 acres of wetlands (12) and 400 miles of streams
and rivers throughout its 480 square miles. The combination of growth and the
need to protect natural resources is driving the Lake County Stormwater
Management Commission's (SMC) comprehensive watershed planning efforts.
Plans are currently being developed for urbanizing watersheds between 2 and
50 square miles in area.
With limited personnel and funding, SMC is utilizing in-house computer
capabilities and staff technical expertise to save time and money as we increase
our ability to model and display watersheds. The Squaw Creek Plan is an
example of how SMC is currently utilizing automated technology for watershed
planning purposes. The Squaw Creek watershed is 25.5 square miles and is
75% undeveloped (includesagriculture, vacant and open space). The watershed
is 17.3 percent wetlands. TheNorthern Illinois Planning Commission forecasts
a 155% population change between 1990 and 2020. The Squaw Creek
Watershed is located in the western portion of the county and drains into Fox
Lake, on the Fox River.
Map 1: Lake County, Illinois, Sub-Watershed
SMC is integrating Geographic Information System (CIS) (2) technology with Computer Aided Design (CAD) and
the Army Corps of Engineer's HEC-1 (10) and HEC-RAS (11) models to create an "automated" watershed closely
resembling the existing Squaw Creek watershed characteristics. SMC used a variety of vendor software packages that
include Environmental Systems Research Institute, Inc.'s (ESRI) ArcView (1) and its Spatial Analyst and Hydrologic
Extensions, and Bentley's MicroStation.
Data Collection
It is very important to determine the methodologies for collecting, calculating, and analyzing data early in the
automation process. Methodologies were determined for mapping floodplains, inventorying and analyzing water
resources (8), and estimating runoff water quality. The floodplain mapping variables included time of concentration,
precipitation runoff, stream storage, stream routing, sub-basin boundaries, and water surface elevations. Thesevariables
272
-------
had to be determined before final data could be formatted and collated. We also had to determine how data could be
documented in the report early in the study process,
Since considerable map data was available digitally, it was economical to perform many tasks on the computer rather
than on hard copy. The Northern Illinois Planning Commission (NIPC) provided the digital land use map. Lake County
Map Services provided digital copies of the Soil Conservation Service (SCS) hydrologic soil groups (HSGs) map, hydric
soil map, United States Geological Society (USGS) orthophotos, Lake County Wetland Inventory boundary map, and
Lake County parcel boundaries. In addition to this digital data, SMC contracted to obtain 2-foot topographic contours,
detailed orthophotos, stream cross-sections, and field-surveyed hydraulically significant structures. Bridge and culvert
information and stream cross-sections were also delivered digitally from Illinois Department of Natural Resource's (IDNR)
land survey crew using Global Positioning System (GPS) and conventional surveying. Photogrammetry and cross-
section control points were collected in the field utilizing a GPS with accuracy of 1:50,000 horizontal and +/- 0.03 feet
vertical (5). Each USGS digital orthophoto map covers one quarter of a quadrangle and used 45 MB of computer
storage. The topographic maps weredelivered in CIS and CAD formats. Contracted data were delivered by square mile.
This created a reasonable size data file, including:
Two foot topographic contours and breaklines 1.2 to 3 MB per square mile.
Orthophotos 35 MB per square mile
Digital Elevation Model 1 MB per square mile
The cost for the two foot contours overlaid on an orthophoto varied between $2200 and $3300 per square mile.
Additional record drawings of hydraulically significant structures, such as road crossings and detention basin outlets,
were collected from county and state highway departments and local communities. The townships and communities
seldom had detailed information, so field investigations were undertaken, where necessary, using topographic mapping
to establish a reference elevation.
The water resources inventory included a stream assessment, wetland inventory, and a wetland restoration
assessment. The stream assessment data were collected in the field along with short community interviews. The stream
inventory used an existing methodology created by NIPC (8). SMC created a methodology to identify potential wetland
restoration locations.
Surface runoff water quality was estimated using typical measured pollutant loading data for several general land
uses. NIPC had an existing procedure that assigned non-point runoff pollutant loads to general NIPC land uses. The
typical pollutant loadings were entered into a worksheet so this procedure could be automated.
Creating Hydrologic and Hydraulic Data
Several hydrologic and hydraulic parameters and other data were used to analyze the surface water runoff and
generate floodplain boundaries. These included delineating sub-basin boundaries; determining a runoff curve number,
time of concentration and Clark's coefficient of runoff for each sub-basin; calculating reservoir data; formatting HEC-1
model; and creating HEC-RAS model geometry.
Sub-basin Delineations
The sub-basin boundaries were produced automatically using the following steps. First, a Digital Elevation Model
(DEM) was produced from photogrammetry by a consultant. A DEM is a list of equally spaced data points with a defined
easting, northing, and elevation. A spacing of the DEM points of 10 and 30 feet was evaluated. The lo-foot spacing
would slightly increase the accuracy of the automated sub-basin boundary's but it used ten times the disk storage as the
30-foot spacing. Therefore, a 30-foot grid DEM was used to determine the sub-basin boundaries due to storage space
limitations. Second, the DEM was loaded into ArcView and converted to a DEM grid using ArcView's Spatial Analyst.
Third, the flow paths and the preliminary sub-basin boundaries were created using ArcView's Spatial Analyst and
Hydrologic Extension along with the DEM grid, which delineated 180 preliminary sub-basin boundaries in 2.5 hours.
273
-------
Fourth, these preliminary boundaries were edited with the digital contour map in the background to better model storage
areas and road crossings. This editing entailed splitting basin boundaries and joining basins together to produce more
accurate boundary lines. Editing was performed on portions of approximately 50% of the preliminary sub-basins that
were automatically created and ultimately reduced the number of sub-basins. Edited boundaries were checked against
hard copy maps and a field investigation of storm sewers and field tiles. A check of maps and field investigation identified
three boundaries that needed additional modifications including the addition of one sub-basin. Finally, ArcView was used
to automatically calculate each sub-basin's area and a sub-basin identification was assigned to each of the 140 sub-
basin areas.
Runoff Curve Number
SMC created a methodology to estimate precipitation runoff. This required converting SMC defined land use
categories to Soil Conservation Service (SCS) runoff curve numbers (RCNs) (9) using ArcView and Excel (6). RCNs
were calculated using the following sequence. First, the 1990 NIPC land use polygons were converted to SMC land use
categories based on land cover using 1996 orthophotos as a backdrop. Land cover was divided into six categories: 1)
impervious, 2) graded grass, 3) natural grass, 4) graded forest, 5) natural forest, and 6) agriculture. Typically graded
grass and graded forest land cover categories have increased runoff compared with their natural conditions as soils are
compacted and depressions are removed during grading. A SMC land use was created for the calibration year of 1996
and for the model year of 2000. Second, concurrently with the land use conversion, the digitally mapped soil numbers
were converted to HSGs using CIS queries. Third, the HSG map was intersected with the SMC land use categories to
automatically create a land cover map. Fourth, the land use categories table and a land cover conversion table were
joined so there was one RCN for each of the four HSGs.
Runoff Data
HEC-1 requires specific input data to generate runoff volumes for each sub-basin. The minimum input parameters
for each sub-basin were identification, area, the time of concentration (Tc), Clarks Coefficient of Runoff(R), and weighted
RCN. Sub-basin area was delineated as previously described.
The weighted RCN was determined in two steps. First, intersecting the finalized sub-basin boundaries with the
RCNs boundaries using ArcView. This splits the RCN polygons with the sub-basin boundaries. This calculation took
just twenty minutes. Then this table of RCN attributes for each sub-basin was exported from ArcView into Excel where
the weighted curve number for each sub-basin was calculated in one day.
In addition, each sub-basin requires a length and slope of travel to generate the Tc and R. To determine the length
and slope, a line with two points were needed, one upstream and one downstream. The line represented the direction
of runoff from the farthest ridge to the outlet of the sub-basin. GeoAnalytics, Inc., a consultant, created a program to
automatically generate a distance point 10% and 85% from the sub-basin outlet along this digitized line in 30 seconds.
The point locations along the line were determined by the methodology used to estimate Tc and R. These points were
queried individually with the DEM grid and checked against the topographic map to determine their elevation, which was
entered into a table, ArcView calculated all sub-basin line lengths in less than a minute. The stream line and its two
elevation points were associated with the sub-basin identification throughout this process. Next, the sub-basin
boundaries, the associated line, and two points were joined into one table and exported as a database file. This
database file was imported to an Excel worksheet where the slope, Tc, and R were calculated for each sub-basin.
Reservoirs
To model reservoir routing, the reservoir volumes were determined using ArcView and the 2-foot digital contours.
The reservoirs consisted of a series of polylines in ArcView after conversion of the CAD contour map. The polylines were
modified so they were completely connected and then converted to a polygon in ArcView. This documented the location
of every reservoir that was modeled explicitly, as not all reservoirs could be modeled within the scope of our project.
Second, the elevation for each contour was entered into a table. The topographic contractor now performs steps one
and two. After all the elevation polygons were created, ArcView calculated the area of the polygons with one command.
274
-------
Finally, the elevation and area tables were opened in an Excel worksheet to calculate the elevation versus storage
relationship. This worksheet was referenced by the HEC-1 formatted worksheet described in the next section. Stage
versus discharge relationship was determined for each reservoir when data was available using HEC-RAS or HY-8.
Hydrology Model Development
All of this data was combined into one Excel workbook to generate the input needed for HEC-1. The sub-basin data
entry included: identification, area, weighted RCN, Tc, and R. Most sub-basins also needed reservoir or stream routing
data. An Excel worksheet was edited with HEC-1 formatted column widths so the data could be saved into a file that
the HEC-1 FORTRAN program can accept. Sub-basin data were entered automatically by referencing other worksheets
in the same workbook. Once the first sub-basin referenced the other worksheets properly, the first formatted sub-basin
data were copied to create another set of HEC-1 data for the next sub-basin. After the sub-basin identification was
entered for this new HEC-1 input data set, the remaining data were automatically retrieved in the worksheet and correctly
formatted, to avoid data translation errors.
Hydraulic Data
The stream cross-section data were initially generated in MicroStation. Each section was digitized as a series of
connected line segments that were exported to a comma-delimited file of easting, northing, and elevation which was then
imported into HEC-RAS's "Import/Export Files for Geospatial Data." The culvert and bridge data had to be coded in
separately. The channel stationing was determined automatically using Intergraph In-Roads. This procedure not only
provided data formatted to be exported directly in HEC-RAS, but also created a 3D map of the channel cross-sections
and stream centerline to document the model spatially using MicroStation and ArcView. The cross-section segments
had to be manually identified for use in the automated floodplain mapping.
Floodplain Development
Stream cross-sections and hydraulic structures were modeled using HEC-RAS to determine the water surface
elevation along the stream. ArcView's Spatial Analyst Extension or Arclnfo could be used to delineate the floodplain from
the HEC-RAS output. Final maps were generated in ArcView.
The HEC-RAS generated water surface profiles were exported by HEC-RAS's "Import/Export Files for Geospatial
Data." GeoAnalytics Incorporated, Madison, Wisconsin, imported this data into an Arclnfo project that uses a 1 0-foot
DEM grid. Arclnfo needs a line and an elevation for each cross-section to map the floodplain. The cross-section line
and its identification were created in MicroStation, exported as comma delimited points, and then referenced into ArcView
to create the cross-section line. The line with its cross-section identification was associated with the water surface
elevation. The grid was then "flooded" between the two cross-sections with a linear slope between the appropriate water
surface elevations. This creates a grid of the flooded area. For each flood profile that was to be mapped, a separate grid
of the flooded area must be completed.
Reservoirs, such as lakes, ponds, and depressions, that have their Base Flood Elevations determined using HEC-1,
were mapped automatically. The storage areas had the water surface elevation defined using HEC-1 then the grid was
"flooded" for all points at or below that elevation.
The flooded grids were then converted to polygons and "smoothed" in Arclnfo for use in ArcView. Last the polygons
were reviewed against the digital two-foot contours and adjusted as needed before final map production. Every reservoir
outlet had to be manually mapped between its outlet and the downstream reservoir or stream floodplain.
Water Resource Assessments
A water resources inventory was completed that included a stream assessment, a wetland inventory, and potential
wetland restoration site identification. All of the stream assessment data were collected in the field along with short
community interviews and entered into a database. Several key stream characteristics were mapped using CIS. The
275
-------
stream inventory data were queried for specific stream conditions and key characteristics were mapped such as degree
of bank erosion or sediment accumulation.
A county wetland Advanced Identification (ADID) study was completed in 1992 prior to the assessment. One of the
criteria reviewed for each wetland was its storage potential, which was related to the area of the wetland. Querying the
spatial data for specified wetland areas and creating a new set of data easily identified these wetlands. The wetland
restoration and mitigation bank site identification methodology was developed by SMC in 1999. Several data sets were
queried to identify the former wetland sites that have the greatest number of characteristics necessary to make them
restorable and usable as a wetland bank. A less stringent set of criteria was used to define all former wetland sites with
restoration potential. The potential wetland restoration sites included all Advanced Identified (ADID) wetlands. Potential
wetland banking sites excluded ADID wetlands and restorable sites less than 20 acres.
Surface water quality 'hot spots" were estimated using non-point pollutant loading rates for several general land uses
via NIPC methodology. Twelve pollutants were evaluated. The pollutant "hotspots" analysis employed land use,
impervious surface area, annual runoff coefficients, and storm sewer conditions as surrogates to determine the annual
pollutant loading by sub-watershed. The pollutant loading database was attached to the land use map database. It was
then mapped in ArcView resulting in 12 maps, one for each pollutant.
The watershed advisory committee and NIPC identified which level of pollutant loadings should be labeled as
detrimental. The pollutant load data were then grouped, using natural breaks in the data set, as low, medium or high.
These were mapped and queried to determine where water quality enhancement projects would be most beneficial and
highest priority.
Summary
The Lake County Stormwater Management Commission has invested a significant amount of time and funding in
developing the hardware, software, and database necessary to perform floodplain analysis. By making this commitment
and establishing the methodologies for manipulating data and analyzing watershed parameters, we have created a
powerful analysis tool. Mapping accuracy, display flexibility and a wide range of CIS analysis ability has been created
through this process. This technology coupled with other resource assessment efforts has created a strong foundation
for future watershed planning in this watershed. The technology is transferable and will be used throughout Lake County
as our agency resources allows.
References
Environmental Systems Research Institute, Incorporated (ESRI) ArcView, 1995, "Using ArcView CIS 3.0a" and "Spatial
Analyst"
Environmental Systems Research Institute, Incorporated (ESRI) ArcView, 1996, "Using the ArcView CIS Spatial Analyst"
Lake County Stormwater Management Commission, 1994, Watershed Development Ordinance
Lake County Stormwater Management Commission, 1998, General Conditions of Topographic Mapping
Lake County Stormwater Management Commission, 1997, Attachment A, Field Survey and Data Collection
Lake County Stormwater Management Commission and GeoAnalytics, Inc., Madison, Wl 53704, 1998, Lake County,
Illinois Stormwater Management Commission, CIS Data Handling Procedures for Input into HEC-1
Lake County Stormwater Management Commission, Land Use to Land Cover Conversion Methodology
Northern Illinois Planning Commission, 1994, "Methods and Procedures for the Stream Condition Form," from Appendix
A of Sequoit Creek Watershed Management Project.
276
-------
Soil Conservation Service, Washington, DC, 1985, National Engineering Handbook Section 4 - Hydrology, PB86780494
U. S. Army Corps of Engineers, 1990 "HEC-1 Flood Hydrograph Package User's Manual Version: 4.0"
U. S. Corps of Engineers, 1997 "HEC-RAS 2.0 River Analysis System"
U. S. Environmental Protection Agency, Region 5, Lake County Stormwater Management Commission, and Northeastern
Illinois Planning Commission, November, 1992, "Wetland Advanced Identification (ADID) Study, Lake County, Illinois"
277
-------
Sediment and Runoff Control on Construction Sites Using
Four Application Methods of Polyacrylamide Mix
A. Roa-Espinosa
Urban Conservationist
Dane County Land Conservation Department
Assistant Visiting Professor
Biological Systems Engineering
University of Wisconsin-Madison
G.D. Bubenzer, Professor
Biological Systems Engineering
University of Wisconsin-Madison
E. S. Miyashita, Former Project Assistant
Biological Systems Engineering
University of Wisconsin-Madison4
Abstract
Fifteen small bare plots (1 meter x 1 meter) on a 10% slope were analyzed for runoff and sediment yield on a
construction site. A rainfall simulator applied 6.32 centimeters of rainfall per hour to each plot after a polyacrylamide mix
(PAM-mix CFM 2000*) treatment was applied. The following treatments: No PAM-mix applied to dry soil (control), PAM-
mix in solution applied to dry soil, dry PAM-mix application to dry soil, PAM-mix in solution with mulch/seeding applied
to dry soil, and PAM-mix in solution applied to moist soil. Each treatment was repeated on three plots. When a solution
of PAM-mix with mulch/seeding was applied to dry soil and compared with the control (no PAM-mix application to dry
soil), we found an average reduction of 93% in sediment yield. An average reduction of 77% in sediment yield was the
worst performing PAM treatment, and occurred when PAM-mix in solution was applied to moist soil. The application of
dry PAM-mix to dry soil reduced sediment by 83% and decreased runoff by 16% when compared to the control. Our
results show that regardless of the application method, PAM-mix was effective in reducing sediment yield in the test plots.
The ease of application, low maintenance, and relatively low cost associated with PAM make it a practical solution to
the costly methods being implemented today
Keywords:
Soil binders, soil erosion, polyacrylamides, flocculation, infiltration, water retention.
*Use of a product name is for the convenience of the reader and does not imply endorsement by the authors, Dane
County Land Conservation, or the University of Wisconsin.
Introduction
One effect of rainfall is the initiation of the erosion process where individual raindrops fall and impact the soil surface.
Soil detachment and particle transport by raindrop splash can lead to serious soil deterioration. Once soil is eroded and
transported by surface runoff to lakes, rivers, and streams, a degradation of the aquatic habitat occurs. Sediment is the
largest pollutant, by volume, in the State of Wisconsin (WDNR, 1994). In order to maintain a healthy watershed, it is
critical to control erosion and sediment yield.
Maintaining soil structure and aggregate stability helps control erosion by increasing infiltration and maintaining less
erodible-sized aggregates. Stable soil structures also help maintain a healthy environment. The use of polyacrylamides
278
-------
and polyacrilamide mixes (PAMs and RAM-mix) is a new tool to help maintain soil aggregate stability and reduce erosion
caused by surface runoff. Such materials can be applied at a rate of 20 to 30 Ibs per acre on construction sites, to
stabilize such sites against erosion until they can be permanently protected through vegetation.
Water-soluble polymers and water polymer mixes do not create aggregates when applied to soil. However, they can
stabilize existing aggregates when the aggregates are saturated with a solution of water soluble polymer mix. Increasing
the aggregate stability with polymers reduces the effect of raindrop impact on the soil, thereby reducing erosion. Polymer
application to the soil may also retard surface sealing, reduce particle soil detachment, reduce sediment in suspension,
and compensate for low residue.
Objectives
The objectives of this study are to determine the optimum application methods and the effectiveness of the PAM-mix
under moist and dry soil conditions. The different application methods were applied to a construction site in Middleton,
Wl. Data were collected to determine the most effective method of application and the effectiveness of the PAM-mix on
construction sites.
Literature Review
The use of polymers as soil conditioners has been studied for decades. The most conclusive studies, done by Lentz
et al. (1992), determined that negatively charged PAM is an excellent soil erosion deterrent for furrow irrigated fields.
It was found to be a cost-effective and safe technology. Sojka and Lentz (1994) found that PAM, when applied in
irrigation waters at rates greater than 0.7 kg/ha, reduced furrow erosion by an average of 80 to 90% and increased
infiltration on Portneuf silt loam by an average of 15%. Trout et al. (1995) reported a 30 to 110% increase in cumulative
infiltration. Roa et al. (1996) found that soils treated with PAM had infiltration volumes more than double that of untreated
soils over a two-hour period. The infiltration volumes for the untreated soils averaged 231 ml/38.5 square centimeters
while those for treated soils averaged 490 ml/38.5 square centimeters, or 98% of the volume of water to be infiltrated.
Roa et al. (1996) also found that the high infiltration rate of the treated sample was associated with low concentration
of sediment in the effluent or infiltrated water.
Nadler et al. (1994) found that PAM mobility in sandy loam, as well as clay loam soils, was limited to the top 25 cm
10 months after application. Clays were attached to anionic polymers more easily when salts were present in solution.
With anionic polymers, flocculation was easier and more complete. When polysaccharides are present with anionic
polymers in solution, fixation was also easier and more complete. Khamraev et al. (1983) reported that clay fixation is
best achieved for PAMs with 30% anionic charges. The cementation provided by the clay flocculation stabilizes the
aggregate at the surface. Roa et al. (1997) found that using polysaccharides, a calcium source with anionic polymers
orpolysaccharides with calcium nitrate and anionic PAM, increased the infiltration rate in saturated cores 5 times greater
than with no soil treatment.
PAM use for erosion control provides a potent environmental benefit by halting furrow erosion by about half a ton
of soil per ounce of PAM used. PAMs remove most sediment, phosphorus, and pesticides from return flows, and greatly
reduce return flow BOD (Sojka and Lentz, 1996). The consequences of reducing sediment and nutrient loading of
construction areas can ultimately be expected to reduce the frequency and intensity of algae blooms and reduce turbidity
and sedimentation of stream channels.
Lentz et al (1992) in Kimberly, Idaho, reported that when applied at 10 ppm, PAM provided a 94% reduction in runoff-
sediment in three years of testing. When used properly, PAM has no measurable toxicity to humans, plants, or aquatic
organisms. Molash et al. (1997) state that the Polyacrylamide Allocation Standard for Reduction of Soil Loss is
necessary because other best management practices (BMPs) are available and have varying degrees of effectiveness.
Sojka and Lentz (1996) summarized several advantages of PAMs over other erosion control BMPs: (1) PAM can
be applied using irrigation equipment and can be effective for controlling erosion over large areas, as demonstrated in
eastern Washington and Idaho; (2) PAM is very effective on fine silt/clay soils; (3) preliminary research conducted in
279
-------
Kansas and California has indicated that RAM is effective at abating wind erosion; (4) RAM enhances precipitation of
fine silts and clay particles, providing water quantity benefits; (5) RAM increases soil infiltration capacity that reduces
runoff volumes; and (6) high benefit to cost ratio.
The types of RAM used for erosion control should have an approximate molecular weight of 12-1 5 Mg/mole, with
an 8-35% negative charge density, and contain no greater than 0.05% Acrylamide monomer (Sojka and Lentz 1996).
A recent study done by King et al. (1996) focused on comparing the uses of polyacrylamides and straw mulch on
dry bean yields. It was shown that the sediment loss was reduced for both straw mulch and the RAM treatment.
In three years of studies in construction sites using RAM for controlling soil loss, RAM has provided a 60-97%
reduction in runoff-sediment (Roa et al. 1997).
Method
Five treatments were applied to soil test plots: (1) No RAM-mix application to dry soil [control], (2) RAM-mix in
solution applied to dry soil, (3) Dry RAM-mix application to dry soil, (4) RAM-mix in solution with mulch/seeding applied
to dry soil, and (5) RAM-mix in solution applied to moist soil. Three replications of each treatment were performed using
a randomized block design on 1 m x 1 m non-vegetated plots in the Middleton Hills Development, Middleton, Wl. The soil
was a Dodge silt loam. The average slope of the test site was 10%.
Plot preparations included large boulder, cobble, and excess debris removal. The surface was raked priorto testing.
Soil moisture prior to testing was about 9%.
The RAM-mix is a high molecular weight anionic granular polymer. The RAM-mix (2.25 g of RAM-mix added to 5
liters of water) was applied at a rate of 22.5 kg/ha, to the appropriate plots using a garden sprinkler. For the dry RAM-mix
application, 2.25 grams of the RAM-mix was applied using a sifter. For the RAM-mix applied to moist soil treatment, the
soil was pre-moistened by a 6.4 cm rainfall six hours before testing.
The sprinkler infiltrometer (Bubenzer and Patterson, 1982 ) was used to collect data for this study. A rainfall
simulator was used that produces 6.4 cm per hour. Actual rainfall depths were recorded using eight rain gauges for each
replication. Runoff from each plot was collected into a tank where the depth of the water was recorded at approximately
2-minute intervals during each test. The average trial time was 40-50 minutes or until the runoff collection tankwasfilled.
Runoff samples were extracted at approximately 1 0-minute intervals by diverting runoff into a collection container
during each replication to determine sediment yield. A representative sample was also taken at the end of each
replication from the tank. The samples were dried at 11 0°C for 24 hours and weighed to determine an average sediment
load for each trial.
Results and Discussion
Mean sediment yield, infiltration, and runoff depth for the three replications and the controls are presented in Tables
1, 2, 3 and 4. For Replication 1, the RAM-mix solution was prepared the evening before field testing. It was noted that
the viscosity of the solution decreased throughout the day. This change may have been due to UV light, reaction with
the mix, and/or oxidative and photolytic interaction. Thereafter, the solution was prepared immediately before the rainfall
simulation. After analyzing the results, a lower viscosity of the RAM-mix solution was determined to be less effective in
controlling sediment yield. This difference is presented in Table 4. Future recommendationsforcommercial applications
may need to take into account the time of preparation of the solution and handling before application.
During the first replication of testing, the largest sediment reduction occurred when RAM-mix in solution was applied
to moist soil. The control yielded 184.4 grams per square meter and the RAM-mix in solution applied to moist soil yielded
36.4 grams per square meter resulting in a reduction of 80% in sediment yield (Table 1). The sediment yield reduction
for the dry RAM-mix application to dry soil and RAM-mix in solution with mulch/seedling applied to dry soil were
280
-------
Table 1. Summary of rainfall, infiltration runoff, and sediment yield for Replication #1 .
Treatment
Control
Dry PAM-mix/Dry Soil
Solution PAM-mix/Dry
Soil
Solution PAM-mix/Moist
Soil
Solution PAM-
mix/Mulch/Dry Soil
Rainfall
(cm)
5.64
5.79
5.64
5.72
5.72
Infiltration
(cm)
1.70
1.91
1.60
0.05
1.57
Runoff
(cm)
4.01
3.89
4.11
5.66
4.14
Sediment
184.4
68.3
103.7
36.4
67.3
Soil Loss Runoff Rainfall
% of Control % of rainfall
100% 71%
37% 67%
56% 73%
20%
36%
99%
72%
approximately 64%. The sediment yield for RAM-mix in solution applied to dry soil was reduced by 44% when compared
to the control.
In Replications #2 and #3, the lowest sediment yield occurred for the treatment of RAM-mix in solution with
mulch/seeding applied to dry soil. A sediment reduction of 97% and 89% occurred, respectively. A sediment reduction
for the treatment of RAM-mix in solution applied to dry soil was 87% and 57% respectively (Table 2 and 3).
Table 2. Summary of rainfall, infiltration runoff, and sediment yield for Replication # 2.
Treatment
Control
Dry PAM-mix/Dry Soil
Solution PAM-mix/Dry
Soil
Solution PAM-mix/Moist
Soil
Solution PAM-
mix/Mulch/Dry Soil
Rainfall
(cm)
4.57
5.72
4.72
4.14
4.55
Infiltration
(cm)
0.51
1.57
0.61
0.13
0.38
Runoff
(cm)
4.06
4.14
4.11
4.01
4.17
Sediment
377.67
178.36
48.77
242.4
12.04
Soil Loss
% of Control
100%
47%
13%
64%
3%
Runoff Rainfall
% of rainfall
88%
73%
87%
97%
92%
Table 3. Summary of rainfall, infiltration runoff, and sediment yield for Replication # 3.
Treatment
Control
Dry PAM-mix/Dry Soil
Solution PAM-mix/Dry
Soil
Solution PAM-mix/Moist
Soil
Solution PAM-
mix/Mulch/Dry Soil
Rainfall
(cm)
5.05
5.38
4.50
4.42
4.39
Infiltration
(cm)
1.12
1.96
0.61
0.28
0.38
Runoff
(cm)
3.94
3.43
3.89
4.14
4.01
Sediment
(gm)
231.34
43.29
98.59
47.65
26.58
Soil Loss Runoff Rainfall
% of Control % of rainfall
100% 78%
19% 64%
43% 86%
21%
11%
94%
92%
When Replication #1 is excluded from the results, the average sediment reduction for RAM-mix in solution with
mulch/seeding applied to dry soil increased from 87% to 94% (Table 4). The sediment reduction for RAM-mix in solution
applied to dry soil was 76%. For dry RAM-mix applied to dry soil, the sediment reduction was 17%, and the sediment
reduction of RAM-mix in solution applied to moist soil was 77%.
Conclusion
Our results show that, regardless of the application method, RAM-mix was effective in reducing sediment yield in
the test plots. The most effective method of soil treatment throughout this study in reducing sediment yield is RAM-mix
in solution with mulch/seeding applied to dry soil. The ease of application, low maintenance, and relatively low cost
281
-------
4.01
3.81
4.04
4.60
4.11
264.51
96.65
83.71
108.82
35.32
100%
37%
32%
41%
13%
304.51
110.83
73.68
145.03
19.31
100%
36%
24%
48%
6%
79%
66%
81%
97%
84%
associated with RAM-mix makes it a practical solution to costly existing methods being implemented. The evidence from
the field application in this study reflects that RAM-mix is a tool to reduce soil loss on bare soil until vegetation cover is
established.
Table 4. Average summary of rainfall, infiltration runoff, and sediment yield for Replications 1, 2, and 3 and Replications 2 and 3, excluding
Replication 1.
Treatment Runoff Sediment Soil Loss % Sediment (Gm) Soil Loss % % of Rainfall
(cm) (gm) Replication Excluding Replication 2 And 3
1, 2, and 3 Replication 1
Control
Dry PAM-mix/Dry Soil
Solution PAM-mix/Dry Soil
Solution PAM-mix/Moist Soil
Solution PAM-
mix/Mulch/Dry Soil
The primary factor that must be considered in future studies is the time of polymer solution preparation and
application. It was noted that the optimal application procedure is to prepare the solution immediately prior to application.
This procedure is necessary in order to limit the amount of degradation and maximize the performance of the RAM-mix.
References
Bubenzer, G.D., and Patterson, A.E., Intake Rate: Sprinkler Infiltrometer, Method of Soil Analysis, Part 1, Physical and
Mineralogical Method, Second Edition, Chapter 33, pp. 845870. (Agronomy Monograph Series #9, 1982)
Khamraev, S.S., Artykbaeva, K., Nizamova, F. and Akhmedova, V. (1983): Leaching of salts from takyr soils conditioned
by amide-containing polyelectrolytes. Uzb. Khim. Zh. 4,16-20 (Russian). Chem. Abst. 99, 138730 (1983)
King, B.A., B. Izadi, MS. Ashraf, R.H. Brooks, and W.H. Neibling. 1996. On-farm comparison of polyacrylamide and
straw mulch on dry bean yields, irrigation performance and erosion. University of Idaho Publication No. 101-96, pp.5559
Lentz, RD., I. Shainberg, R.E. Sojka, D.L. Carter 1992. Preventing irrigation furrow erosion with small applications of
polymers. Soil Sci. Soc. Am. J. 56:1926-1932.
Lentz, R.D and. Sojka, R.E. 1996. Five year Research Summary Using PAM in Furrow Irrigation. University of Idaho
Publication No. 101-96, pp. 1 I-20.
Nadler, A.,M. Magaritz, and L. Leib. 1994. PAM application techniquesand mobility in soil. Soil Science 158(4): 249-254.
Roa, A., 1996. Screening of Polymers to Determine Their Potential Use in Erosion Control on Construction Sites,
University of Idaho Publication No. 101-96, pp.77-83.
Roa, A., Bubenzer, G.D, and Miyashita, E., 1997. Determination of PAM Use Potential in Erosion Control. Proceedings
of the first European Conference of Water and the Environment. Innovation Issues in Irrigation and Drainage. Edited by
Pereira et al. Institute of Agronomy, Technical University of Lisbon, Portugal.
Molash, E. 1997. WSDOT Stormwater Management Plan Page 148 V 5.3 3.25148
Trout. T.J., Sojka, R E. and R.D. Lentz. 1995. Polyacrilamide effect on furrow erosion and infiltration. Trans. ASAE. 38(3):
761-765.
Sojka, R. E. and Lentz, R.D., 1994. Polyacrylamide (PAM): A New Weapon in the Fight against Irrigation Induced
Erosion, USDA-ARS Soil and Water Management Research Unit Station Note #01-94.
282
-------
Sojka, R.E. and Lentz, R.D., 1996. A RAM Primer: A Brief History of RAM and RAM-Related Issues, University of Idaho
Publication No. 101-96, pp. 1 I-20.
Stephens, S.H., Final Report on the Safety Assessment of polyacrylamides, J. Am. Coll. Toxicol., 10:193-202.
Wisconsin Department of Natural Resources (WDNR), 1994. The Wisconsin Quality Assessment Report to Congress,
PUBL-WR 254-94-REV, Wisconsin Department of Natural Resources, Madison, Wl.
283
-------
Construction Site Planning and Management Tools
for Water Quality Protection
Thomas Mumley
California Regional Water Quality Control Board
San Francisco Bay Region, Oakland, California
California is seemingly a developer's paradise. Population is on the rise, the economy is good, and there is little
or no rain to interfere with construction for nearly eight months of the year. To top off these benefits, the California
Regional Water Quality Control Board, San Francisco Bay Region (Regional Board) has a comprehensive Construction
Site Planning and Management Program (Program). It is based on the integration of a strong regulatory and
enforcement posture, an outreach and education strategy, and technical assistance. The keys to the success of the
program are the balance of actions among these elements and implementation tools for actions within them.
Background
The Regional Board is the state agency in California responsible for protection of water quality and enforcement
of water pollution control regulations, including National Pollutant Discharge Elimination System (NPDES) permits. The
California Water Code provides the Regional Board with strong enforcement authority. This authority ranges from a
notice to comply, to a notice of violation, to enforcement orders, to monetary penalties. Penalties can be as high as
$10,000 per day of violation or $10 for each gallon of waste discharged. The Regional Board may also suspend part
of a penalty in exchange for an environmentally beneficial project.
In the San Francisco Bay Region, the Regional Board is responsible for enforcement of a general NPDES permit
for stormwater discharges from construction sites of five acres or greater. The general permit requires implementation
of an effective Stormwater Pollution Prevention Plan (SWPPP) that includes best management practices (BMPs) for
erosion and sediment prevention and control and management of equipment, materials, and wastes. The Regional
Board is also responsible for enforcement of NPDES permits for municipal stormwater discharges that have been issued
to all municipalities (regardless of population) in the urban areas of the region. These permits include requirements to
control discharges from construction sites (regardless of size). There is an inherent overlap of Regional Board and
municipality authority over construction of five acres or greater. The Regional Board's Program recognizes and takes
advantage of this overlap of authority.
Inspections
The Regional Board initiated an aggressive construction site inspection and enforcement effort in 1997. This
resulted in discovery of significant water quality problems associated with sediment discharges caused by minimal or
token erosion and sedimentation control actions. Some of the most common observations were:
. No permit.
. SWPPP not developed, not implemented, or deficient, especially in terms of timing.
. Mass grading allowed to continue throughout winter months until rain and muddy conditions make further work
impossible.
. Mass grading continues past the time when grasses will grow and germinate; first rains simply carry seed/mulch
away with eroded soil.
• No erosion control measures; reliance solely on sediment basins.
. Sediment basins are frequently undersized, improperly designed, and not maintained.
284
-------
. Site not monitored to assess BMP effectiveness.
. SWPPP not updated to reflect changes in site conditions.
. Hillsides stabilized with hydroseed, but no mulch (resulting in rains carrying seed material away with eroded
soils).
. Control measures driven by "tokenism" with control measures intended to demonstrate good intentions rather
than real effectiveness.
. Willingness to risk fines in order to maximize work during winter (rainy season) months.
. Local agencies, specifically planners and engineers with plan-approval authority, often unaware of "best"
management practices.
. Sites approved by local authorities for mass grading during rainy season.
. Local authorities review and approve erosion control plans but do not inspect sites.
Enforcement Actions
Several types of enforcement actions evolved from these findings. The first consisted of the development and
issuance of a "Notice to Comply" (Figure 1). Often (25 - 35 % of the time) operators at a site are unaware of their
requirements or appropriate BMPs. The Notice to Comply is essentially a "fix-it" ticket that results in no further
enforcement action if corrective action is implemented. Regional Board inspectors are authorized to issue Notices to
Comply in the field, and use of this simple enforcement tool has proven to be an effective mechanism to gain timely
corrective action at construction sites.
Other enforcement tools are used in circumstances where the severity of the problem warrants more intensive
enforcement action. These include, in terms of progressive severity: a Notice of Violation, a Cleanup and Abatement
Order, and a Cease and Desist Order. Violations of any of these actions typically lead to more aggressive enforcement
action. The most aggressive enforcement action is imposition of administrative civil liability (monetary penalties).
During the 1997/98 rainy season the Regional Board imposed over $1 million in penalties, ranging from $10,000
to $230,000. A major consideration in determining penalty amounts is ensuring that it does not pay to pollute. Due to
the economic and time pressures associated with many development projects, minor penalties may simply constitute
a cost of doing business. The Regional Board has clearly stated its intolerance to this circumstance and intends to
severely penalize repeat offenders. Clearly, such penalties not only get the attention of the violator, but the building
industry as a whole. Substantive penalties have also provided opportunities to fund environmental education projects
in lieu of direct payment of penalties. The Regional Board has favorably accepted development of technical assistance
tools as appropriate mitigation projects.
Education and Outreach
The Regional Board recognizes that regulation without education is ineffective. Consequently, its program includes
extensive outreach efforts. These include:
. Mass mailing to construction projects of more than five acres and projects permitted for winter grading
summarizing requirements and findings on inadequate performance
. Meeting with development community and local agencies prior to the rainy season (August through September)
to better communicate concerns and requirements and to establish a dialogue
. Providing detailed guidance and training for both developers and municipalities on their responsibilities and on
effective control approaches
285
-------
San Francisco Bay Regional Water Quality Control Board
1515 Clay Street, Suite 1400, Oakland, CA 94612 /Phone (510) 622-2300 - FAX (510) 622-2460
You are hereby notified thaL
Order No.
a NPDES Permit No. (if applicable).
a California Water Code Section _
• I Other
NOTICE TO COMPLY
(hereafter Discharger) has violated provisions of:
Federal, State, and Local Agency Contacts:
I. FACILITY INFORMATION
Inspection Date:
Discharger Contact:
Site Name & Location:
Headquarters/Owner Name & Address :
Time:
Title:-
Prior Notification: a Yes a &a a Unknown
Phone: ( )
County:.
II. NON-COMPLIANCE INFORMATION
Nature of Violation :
Recommendation to Correct:
Time to Comply (Not to exceed
30 days)
III. SIGNATURE SECTION
I acknowledge receipt of this Notice (must be owner, operator, or duly designated representative of facility):
RECIPIENT NAME (print): TITLE:
SIGNATURE: DATE:
STAFF NAME:
SIGNATURE:
IV. CERTIFICATION OF COMPLIANCE
Sign and return by mail or fax within 5 working days of achieving compliance
I certify under penalty of perjury that the above violation(s) have been
corrected.
I am aware that there are significant penalties for submitting false information.
PHONE:
DATE:
Recipient Signature:
Print Name:
Date:
FOR REG. BD. USE ONLY
Receipt Date: Acceptable:
/ / a sn
Reviewed by: D Yes
Recommendation:
Date:
Title:
Figure 1. Notice to Comply
The objectives of these outreach efforts are:
286
-------
. Commitment from the construction industry to include erosion control in their planning, scheduling, and (most
importantly) project implementation
. Commitment from municipalities to play a greater role in SWPPP review and implementation, including training
inspectors so that builders, municipal staff, and Regional Board staff are all on the same page - thereby allowing
for consistent regulation of construction activities by applying a uniform standard.
In response, the building industry and municipalities have collaborated with the Regional Board on the production
of training workshops on construction site planning and management for both building industry and municipal staffs.
The workshops provide a review of regulations and responsibilities including:
. State responsibilities
. Permits for work in or near streams
. Local agency responsibilities
. Plan approval authority and requirements
. On-Site responsibilities (plans, permits,inspections)
. Inspector responsibilities
. Enforcement
. Field inspection coordination (i.e., state agency/municipality)
The workshops also include training on BMPs for erosion and sediment control (principles, tools, corrective
measures, inspections, monitoring, reporting), non-stormwater discharge prevention and management, and a field trip
to an active construction site where vendors demonstrate both proper and improper installation practices.
Production of the workshops has been funded in part through mitigations associated with administrative civil liability
fines. Similarly, penalty mitigation funds have been used to develop education tools including:
. An 18 minute training video entitled "Hold on to Your Dirt: Preventing Erosion from Construction Projects" which
provides information on BMPs for grading projects and for stabilizing disturbed land
. An 18 minute training video entitled "Keep it Clean: Preventing Pollution from Construction Projects," which
provides information on BMPs to prevent water pollution from non-stormwater discharges from activities such as
painting, stucco, concrete washout facilities and saw cutting
. A booklet of "Guidelines for Preparing a Storm Water Pollution Prevention Plan"
Erosion and Sediment Control Field Manual
The centerpiece of the Regional Board's Program is an Erosion and Sediment Control Field Manual (also
developed with penalty mitigation funds). The Field Manual was produced in response to a common complaint by "field"
personnel that there is a need for information on cost-effective and proven BMPs, and that existing references were too
technical and difficult to read. The Field Manual contains concise descriptions of BMPs for erosion and sediment control
and site management (Table 1). Overviews of regulatory requirements and inspection and monitoring responsibilities
are also provided.
287
-------
Table 1. Erosion and Sediment Control Field Manual BMPs
Erosion and Sediment Control Field Manual BMPs
Erosion and Sediment Control Practices
. Scheduling
. Preservation of Existing Vegetation
. Slope Grading
. Temporary Seeding and Mulching
. Permanent Seeding and Mulching
. Hydromulching - Hydroseeding
. Dust Control
. Erosion Control Blankets and Geotextiles
. Fiber Rolls
. Temporary Stream Crossing
. Stabilized Construction Entrance
. Entrance/Exit Tire Wash
. Outlet Protection - Energy Dissipation
• Check Dams
. Silt Fencing
. Temporary Straw Bale Dike
. Sand/Gravel Bag Barrier or Rock Filter
. Storm Drain Inlet Protection
. Catch Basin Inlet Filters
. Sediment Basin
. Sediment Traps
. Dewatering: sediments/toxic pollutants
. Secondary Filtration
General Site and Materials Management Practices
. Water Conservation Practices
. Solid and Demolition Waste Management
. Hazardous Waste Management
. Spill Prevention and Control
. Vehicle and Equipment Service
. Material Delivery, Handling, and Storage
. Paints and Liquid Materials
. Handling and Disposal of Concrete and Cement
. Pavement Construction Management
. Contaminated Soil and Water Management
. Sanitary/Septic Waste Management
. Landscaping Management
BMPs are described in a user-friendly format that features full-color graphics, including do and don't illustrations
(Figure 2). Each BMP description includes its purpose, application, limitations, practices, inspection, and maintenance.
There is a section on Corrective Measures that discusses what can go wrong and common installation problems. This
latter section is essentially a troubleshooting guide that contains a table of common problems and corresponding
corrective measures. Overviews of regulatory requirements and inspection and monitoring responsibilities are also
provided. The Field Manuals waterproof 9" x 9" binder and coated pages make it ideal for use in the field. As such, it
provides the essential connection between the enforcement, outreach, and technical assistance components of the
Regional Board's Program.
Overlap of State and Municipal Authorities
The Regional Board's Program provides a clear demonstration of how the Storm Water Phase II Program's
construction requirements may be implemented. The Phase II rule allows states to recognize compliance with municipal
program construction requirements as equivalent to compliance with a state-issued NPDES permit for construction,
if it can be demonstrated that the municipal program requirements are equivalent. In such situations, a construction site
deemed in compliance with a municipality's requirements would be deemed in compliance with the state-issued NPDES
permit. The key is demonstration that the municipal program qualifies as equivalent.
In the San Francisco Bay area, as previously noted, the Regional Board has issued NPDES permits for municipal
stormwater discharges that include requirements to control discharges from construction sites. In essence, there is an
overlap of Regional Board and municipal authority where municipalities are in compliance with their permit requirements.
288
-------
SHORT CIRCUIT -^|^
DO ?
BAFFLES WILL INCREASE DETENTION TIME
Figure 2. Sediment Basin Design.
Unfortunately, what may seem equivalent on paper may not be equivalent in practice. The case in point is that the
Regional Board's inspection program noted above identified many construction sites out of compliance with their
construction NPDES permits. Consequently, these same construction sites would be deemed in non-compliance with
municipal requirements, In addition, the same inspection findings can be applied to the municipality. Since the
municipality's NPDES permit requires it to control discharges from construction sites, construction site non-compliance
means the municipality is not in compliance with its NPDES permit. In these circumstances, the Regional Board may
(and has) taken enforcement action against both the construction site and the municipality.
To date, the primary enforcement tool used for the municipalities has been Notice to Comply. The Notice to Comply
requires the municipality to report on the failure of its construction control program and to implement timely corrective
actions. Most municipalities have been very responsive to this "wake-up-call," and have made improvements to
demonstrate the desired "equivalency." The net result is a negative turning into a positive. The Regional Board's
Program, with its balance between enforcement and education, has provided a cfe facto mechanism for recognizing
municipal program equivalency allowed by the Phase II rule. By its design and implementation, the program essentially
289
-------
requires municipalities to demonstrate such equivalency. Consequently, the Regional Board will significantly reduce or
eliminate its inspections in municipalities where Regional Board inspections find construction sites in compliance, thus
providing incentive and reward to both the building industry and municipalities.
Lessons Learned
Lessons learned in the development and implementation of the Regional Board's Program are summarized in the
following points:
. The only effective means of controlling erosion is erosion prevention, which requires careful planning and
adherence to seasonal time-lines. Sediment capture should be used only as a secondary or back-up plan.
. Regulation without education is ineffective. Often, noncompliance is due to lack of awareness of the regulatory
requirements and cost-effective, proven BMPs.
. Education without enforcement is impotent. Despite good intentions, the building industry is constantly trying to
maximize its investment dollars, and environmentally sound BMPs are often superseded by time pressures to
complete a project.
. Enforcement actions must be severe enough that they cannot be accepted as a cost of doing business.
. The balance between regulation and education is dependent on readily available technical assistance and
implementation tools.
. Outreach and technical assistance needs to be directed to the right audiences. Workshop agendas and
attendance were initially misdirected toward planners and local decision makers. Key attendees are municipal staff
who actually review SWPPP plans and perform on-site inspections and building industry staff who are onsite.
Evaluations revealed attendees wanted more technical information on installation and less time spent on municipal
general plan/environmental plug. Audiences are especially responsive to builders discussing their experiences
in implementing BMPs.
. Both the building industry and municipalities have historically short shrifted training. Workshop attendees
expressed relief that practicable training is finally available - especially information on vendors, cost comparisons,
and practical BMPs. The building industry and municipalities now realize costs of training are minimal relative to
the benefit.
Conclusions
The bottom line is that environmental regulators, municipalities, and the building industry have different priorities that
must be reconciled. Regulators seek no adverse impacts to waters. Municipalities seek economic growth. Builders want
unfettered development. In the case of construction-related erosion, the means to each end is the same...effective
erosion and sediment control. A little more work on the part of each party involved has proven that their different
priorities are attainable and even harmonious.
Since the Regional Board made enforcement a top priority and began a collaborative effort with the building industry
and municipalities to provide cost-effective outreach and training, construction site compliance with NPDES permit
requirements has risen from 20% three years ago to greater than 90% today. Municipal compliance has risen similarly.
290
-------
Regulating Sedimentation and Erosion Control into Streams:
What Really Works and Why
Seth R. Reice
Department of Biology
University of North Carolina at Chapel Hill
Chapel Hill, NC 27599-3280 (address all correspondence to this author)
JoAnn Carmin
Department of Public Policy
Virginia Polytechnic Institute and State University
Blacksburg, VA
Abstract
The overall objective of this project was to determine the effectiveness of different environmental policies,
regulations, and incentives in reducing the ecological risks and consequences of sedimentation to streams. We were
trying to learn which sets of regulations, enforcement strategies, and landscapes result in effective protection of stream
communities from degradation, resulting from erosion and sedimentation from construction sites. By connecting erosion
control efforts to environmental impacts, our aim was to create more effective management strategies that ultimately
provide environmentally sustainable social and economic development in our watersheds.
We chose four replicate construction sites in each of three regulatory jurisdictions that varied in stringency of
regulations and enforcement activities. At each site, we conducted instream assessments of water quality and
biomonitoring of macroinvettebrates and fishes to determine the success of the regulators in protecting stream
ecosystem health. We combined these results with evaluations of the regulatory environment to link the policies and
management styles of the regulators to the effectiveness of protection of the streams. While all construction sites did
some damage to the steams, we found that enforcement style and frequency of inspections were far more important than
the nature of the regulations in preventing sediment pollution of streams.
Keywords: Development, enforcement, rivers, sedimentation, streams, regulations, regulatory effectiveness.
Introduction
A critical problem in American rivers and streams is sedimentation. Sedimentation degrades water quality, alters habitat
for fish and macroinvettebrates, limits ecosystem functions and services, and reduces the aesthetic and economic value
of rivers and streams. Many regulations and policy incentives have been devised to control sediment pollution of our
rivers and streams. Yet there has rarely been an attempt to reconnect the policies with the ecology of the rivers. That
was the goal of this research. This work integrates the regulatory environment, sediment ordinances, and policies with
resultant ecological impacts of sedimentation on rivers and streams. The question the ressearch sought to answer was
"What combinations of policies, regulations and on-site interactions between regulators and developers really work to
enhance stream biota and stream ecosystem health?"
Research goals were accomplished by comparing similar streams in different regulatory jurisdictions (a comparative
watershed approach). We tested the effectiveness of different intensities of sediment control regulations and
enforcement. We used the streams to tell us what matters ecologically. The selected political jurisdictions differed in
the stringency of their erosion and sediment control requirements and the nature and intensity of enforcement of the
regulations. We chose 17 construction sites along streams in three different jurisdictions. We interviewed the regulators
291
-------
and developers at each site and we studied the regulations and the attitudes of the regulators and developers. At each
site, we sampled the streams being impacted. Some projects are still unfinished. Therefore, we will present only the
results of the 'before construction' and 'during construction' samplings.
We asked "Which erosion and sediment control regulations really work and why?" We have analyzed the erosion
and sedimentation control regulations and compared them among the respective jurisdictions. Then we surveyed the
attitudes and enforcement activities at all levels within each jurisdiction. This paper will briefly outline our findings and
focus on what can be done to minimize sedimentation into streams from construction sites.
Methods
S/te Selection
We selected three regulatory jurisdictions so they would vary across a range of two critical variables: (1) stringency
of regulations (how strict and how rigorous the rules are) and (2) stringency of enforcement (i.e. frequency of inspections,
severity of punishment of violations). A summary of some of the salient characteristics of the three regulatory jurisdictions
is given in Table 1. Construction sites were selected from the array of applications for grading permits filed with the
erosion and sediment control offices in each jurisdiction. The biggest constraint in locating study sites was the availability
of construction sites on streams with riffle zones, One jurisdiction (District IV) extends eastward into the coastal plain
as does Eastern Wake County. Therefore, many otherwise promising sites, which had sandy bottomed, slow flowing
streams, were eliminated from our study. To be selected for this study, the construction sites had to have certain critical
characteristics. For example, streams had to be within 100m of the site. There also had be a significant slope from the
construction site down to the stream, so that if erosion occurred it would impact the stream. These factors made site
selection extremely difficult. In this paper, we will discuss only the impact of large construction sites (>100 acres
disturbed). We have located and sampled ten large sites.
Table 1. Selected Characteristics of Erosion and Sediment Control Jurisdictions Used for this Project
Minimum Disturbed
Area Requiring
Erosion Plan Staff
Orange County
Wake County
District 4'
(16 Counties)
# Field Total Area
(Miles') Projects Ratio
0.5 Acres
1 .0 Acres
1 .0 Acres
# Active Site/Staff
3
4
4
400
858
8,116
-100
-400
-1000
33.3
100
250
* District 4, of the NC Division of Land Quality oversees all construction projects in all 16 counties without a Local Erosion and Sediment Control
Program. It covers all governmental construction in the District 4 area, including Orange and Wake Counties. So, a single stream can have adjacent
construction sites along the banks, one supervised by District 4 and the other by the Local Program.
Stream Sampling Procedures and Variables Sampled
We monitored at least three replicate sites per jurisdiction for the large construction sites. We sampled before,
during, and after construction. We cannot control the timing of the construction projects, and since sampling must follow
a rain of >1/2" in 24 hours (i.e., a rain with the potential to produce erosion and sedimentation), our sampling was
dependent on the weather and the contractors. This means that the time between the before, during, and after sampling
is highly variable. Since upstream and downstream controls were sampled on the same day as the "at the site" samples,
this did not cause a significant analytical problem.
We sampled three sites on each stream, including >1 00m upstream, at the site, and >1 00m downstream. We took
two replicate Surber samples for macrobenthos, identified to species whenever keys permitted, including chironomids.
Chironomids are essential because they often constitute >90% of the individuals sampled, especially in the impacted
reaches. The number of samples is small since our objective was not to analyze any one stream in detail, but to treat
292
-------
streams as replicates. In the overall analyses, multiple samples per site are pseudoreplicates. The mean of the replicate
Surber samples was used in the analyses. We electroshocked for fish along one 50m reach of riffles and pools. We
collected basic water chemistry data. Water quality parameters included D.O., turbidity, conductivity, Total N, NH,+, NO3"
1, Total PO,-', SRP, pH, and temperature. We also studied leaf litter decomposition rates. Five g leafpacks of Comus
florida (dogwood) leaves were incubated for two weeks in situ at all three sites in the "during construction" period to
assess the critical ecosystem process of litter decomposition.
Environmental Policy Analysis
Surveys and semi-structured interviews were used to investigate both the regulatory agencies and developers. The
surveys focused on the capacityof the agency, the external commitmentthat the agency receives, as well as the internal
commitment toward the environment, and the control measures that are used. The surveys and interviews achieved
a 100% response and participation rate. Although it has been harder to get their cooperation, we have nearly completed
data collection from developers. The survey data is being augmented with documentary data from the sediment and
erosion control offices in each of the counties.
The evaluation of implementation focused on (1) the extent to which developers comply with sediment and erosion
control regulations and (2) the way that regulatory and organizational factors interact to shape compliance behaviors.
The examination of outcomes combines social science and biological data to examine associations among regulatory
styles, agency activities, and stringency of policy enforcement. We further analyzed how variations in sediment and
erosion control enforcement are related to the ecological outcomes (including biological, chemical, and physical factors)
in the impacted streams.
Hypotheses
Hypothesis 1. Greatest degradation will be evidenced at the construction sites, compared to upstream controls, with
moderate to complete recovery downstream.
Hypothesis 2. Tighter enforcement of erosion and sediment control laws will result in less damage to streams.
Hypothesis 3. Stronger erosion and sediment control regulations will result in less damage to streams.
Results
Nearly all biotic and environmental variables measured tell the same story. Figure 1 shows the changes in the EPT
Index for the during construction sampling. That is the species richness of the Ephemeroptera (the mayflies), Plecoptera
(the stoneflies), and Trichoptera (the caddisflies). The tally of EPT taxa (i.e., EPT Richness or the EPT Index) is a well-
established and universally accepted measure of stream health. These groups of aquatic insects are particularly
sensitive to (and highly intolerant of) high temperature, low oxygen, toxic substances, a wide range of pollutants, and
burial by sedimentation. An abundance of EPT species and individuals and high EPT diversity are clear indicators of
good stream health. Reductions in EPT values demonstrate degradation of stream conditions.
EPT richness follows a pattern. The differences between jurisdictions are clear. The greatest decline in EPT values
from upstream to at-the-site occurs in District IV. The EPT Index in Orange County changes little at any site. Wake
County actually shows some enhancement of the EPT richness as you go from upstream to at-the-site. We sampled
many other variables but the results parallel the EPT richness.
A short summary of the enforcement activities and attitudes of the regulators in the various jurisdictions is found in
Table 2. These data show that these agencies differ in these aspects. Orange County had the strictest enforcement,
penalizing nearly 25% of all construction projects, while Wake penalized -22% and District IV penalized only ~ 4.5% of
the projects they inspected. Orange County is most likely to use stop-work orders to halt construction due to
sedimentation violations, while District IV relies on fines. District IV is perceived as being so understaffed that it is unable
to make sufficient inspections. Consequently, some contractors do not feel obliged to follow their approved plans. Some
293
-------
contractors agree to a plan and then cut costs by not following the sediment controls. This laxity is detectable from the
stream data (see Figure 1).
Total EPT Richness
a
x
(8
z
0
Error bars = 1 std. error
District IV Orange Wake
Upstream
District IV Orange Wake
At The Site
District IV Orange Wake
Downstream
Figure 1. Total EPT Richness.
Table 2. Regulatory Environment
Agency/Variables
Enforcement Action
Penalties enacted past year
Stringency of Penalties
Orange County
Very Strict
24
High
NC District IV
Average
44
Medium
Wake County
Strict
88
Medium
Attitudes of Regulators
Perception of official commitment
Percent of developers that regulators
Supportive
8
Very Supportive
40
Indifferent
10
Orange County and Wake County regulators generally think that developers will try to avoid complying with erosion
and sediment control regulations. As the regulators' workload increases, their task becomes more difficult. This may
result in regulators adopting a more forgiving attitude toward developers and less vigorous enforcement of the
regulations. District IV regulators think that fully 40% of developers are trying to comply with the regulations.
Discussion
There is a clear link between the attitudes and enforcement activities of the regulators of erosion and sediment
control ordinances and environmental outcomes in the streams near construction sites. If the regulations are completely
effective, all sites should be similar to the upstream controls when the construction is completed and the site has been
294
-------
stabilized (i.e., revegetated). In our analysis, the degradation is clearly detectable in the benthic community data (see
Figure 1). Benthic communities at the site are dramatically negatively impacted in District IV, unchanged in Orange
County, and actually enhanced in Wake County. The effect is sometimes reduced downstream but the degradation
persists downstream in District IV.
Wake County and District IV have identical regulations, while Orange County's regulations are more stringent.
Comparison between the two jurisdictions with the same rules but different inspection and enforcement intensities will
help us tease apart these factors. The stream data suggest that the laws, as written, are not particularly important. Wake
County has the best environmental results while District IV has the worst stream degradation. Our analysis suggests that
differences in laws and regulations have limited impact on the degree of degradation of stream biota.
The key factors seem to be the attitudes and enforcement behavior of the regulatory agencies. The frequency of
on-site inspections is particularly important. In Orange County, every construction site is inspected every week. If it is
a problem site, the inspectors may visit daily. In Wake County, the inspections are closer to every other week. In District
IV, the goal is to visit every site once in the entire duration of the project. They also seek to respond to any citizen
complaints within one week. In Orange County a complaint generates an inspection within one day. Another critical
factor is topography. A very steep, erodible slope can undermine the best attempts at enforcement of erosion and
sediment control regulations.
Our analysis suggests that differences in the nature and frequency of enforcement and inspections does matter.
Developers tell us that a rigid, command and control approach to enforcement is less palatable to them than a flexible
problem-solving cooperative approach. If the developers perceive that the regulators are really tying to help them keep
sediment on site and out of the streams, they do a better job. Flexibility enters in as follows. If the sedimentation
inspectors have enough time to analyze a sedimentation problem in detail, their suggestions will be better. Very often,
the inspectors need the authority to implement solutions which are not exactly "by the book." When inspectors propose
innovative solutions, which can really solve the problem, this encourages the developers to be more cooperative. More
frequent inspections and a cooperative, flexible approach by regulators does ameliorate the stream damage among
similar streams in different jurisdictions.
On the other hand, if the developers know that the regulators will in fact shut them down (with a stop-work order or
a court injunction), it is easier for the regulators to get developers' attention. Fines are notoriously ineffective penalties
in North Carolina. Presently the maximum fine is $500 per day. When developers are pouring millions of dollars into
a project, this amount of fine is trivial. As one said, "It's just a cost of doing business." In essence, the effectiveness of
erosion and sediment control depends more on enforcement than on how the regulations are written. Even with weak
laws, the success of Wake County's Erosion and Sedimentation Control Program plainly depends on their on-site
enforcement actions.
Recommendations
. Provide sufficient inspectors to visit each construction site at least weekly.
. Give inspectors the authority and knowledge to implement innovative solutions to erosion problems on a site-
specific basis.
. Empower the inspectors to issue severe penalties (stop-work orders) in the case of sedimentation violations.
. Raise the maximum level of fines to a meaningful amount (we suggest $10,000 per day).
. Educate the development community to the damage that sedimentation does to stream communities.
295
-------
Effectiveness in Erosion and Sediment Control:
New Initiatives in Indianapolis
James Hayes
Marcia Mathieu
Marion County Soil and Water Conservation District
Indianapolis, Indiana 46237
marcia-matthieu @ iaswcd.org
Greg Lindsey
Center for Urban Policy and the Environment
School of Public and Environmental Affairs, Indiana University
Indianapolis, Indiana 46204
glindsey@iupui.edu
Since the late 1960s, when the severity of pollution from sediment from construction sites was first documented,
many states and municipalities have worked to develop effective programs for erosion and sediment control. These state
and local programs were augmented in 1987, when Congress required in the Clean Water Act that operators of all
construction sites over five acres prepare erosion and sediment control plans and obtain National Pollution Discharge
Elimination Permits (NPDES). At that time, some states, such as Maryland and North Carolina, already had well-
supported, comprehensive approaches that were developed largely in response to state law. Other states, including
Indiana, have relatively new programs that were adopted only after the federal mandate. In general, these newer
programs are not as comprehensive, and managers are still working to develop systematic and effective methods for
implementation.
This paper describes a new initiative in Indianapolis, Indiana, to increase the effectiveness of erosion and sediment
control programs. The paper describes a general framework for evaluating erosion and sediment control programs.
Next, it describes an intergovernmental, "S.W.A.T." team approach to inspection that was used in Indianapolis in the
summer of 1998. The paper summarizes the results of the inspections and concludes with a discussion of the
implications for managers of erosion and sediment control programs.
Effectiveness in Erosion and Sediment Control Programs
Managers and analysts in Maryland and North Carolina have used a general framework for evaluating erosion and
sediment control programs (Clevenger, n.d.; Departments of Civil Engineering and City and Regional Planning, 1990).
The framework comprises five criteria, each of which must be satisfied for sediment pollution to be controlled effectively:
. Complete coverage
. Competent plans
• Careful installation
. Continual maintenance
. Consistent enforcement
Overall effectiveness requires that the coverage rate (the proportion of construction sites with controls) approach 100%.
Operators of development sites must know of regulatory requirements and make efforts to comply. Second, erosion and
sediment control plans must be competent. Best management practices (BMPs) incorporated into plans by engineers or
technicians must be able, if constructed properly, to control erosion and sedimentation. Third, BMPs must be installed
completely and correctly. Improper installation may result in failure and off-site sedimentation. Fourth, BMPs must be
296
-------
maintained for the duration of the construction process. Finally, consistently effective approaches to enforcement must be
developed to ensure compliance with substantive criteria. If any one of these criteria is not met, the objectives of erosion and
sediment control may not be achieved.
When new programs are developed, these criteria can be considered sequentially. That is, when building a new program,
managers must first make sure that developers and builders are aware of regulatory requirements. Next, they must work to
ensure that developer's engineers are preparing good plans. If developers are aware of requirements and are submitting good
plans, attention can turn to installation and maintenance. Use of enforcement tools always is a last resort.
An Initiative in Indianapolis
Although the City of Indianapolis has a sediment control ordinance that predates federal requirements, erosion and
sediment control programs in Indiana have largely been developed in response to a state regulation [Title 327-IAC 15-5
(Rule 5)], that was adopted in 1992 to comply with EPA regulations. Since the adoption of Rule 5, managers generally
have seen improvements in efforts to comply. Most developers and builders are now aware of requirements, and
coverage is approaching 100%. With respect to plan review, the Division of Permits in the Department of Capital Asset
Management (DCAM) is responsible for plan review pursuant to the city's ordinance, while, under a memorandum of
understanding with the Indiana Department of Environmental Management (IDEM) and the Department of Natural
Resources (IDNR), the Marion County Soil and Water Conservation District (District) is responsible for plan review
pursuant to Rule 5. MCSWCD reviewers estimate that the quality of plans is improving, but that as many as 60 to 70%
of all plans still must be returned and revised before approval. Most plans are approved on the second iteration.
Although the review process now assures that competent plans are being prepared, installation often remains
inadequate, BMPs often are not maintained, and resources for inspection and enforcement are limited. IDNR has only
seven inspectors in the Division of Soil Conservation for all 92 counties and 550 municipalities. IDNR inspectors
generally work individually within regions, inspecting sites sequentially and in response to complaints. District personnel
lack enforcement authority and mainly visit sites in response to complaints. In Indianapolis, sediment control has been
a low priority with DCAM, which has no inspectors trained in or assigned exclusively to enforcement of sediment control
requirements.
Managers have struggled to find ways to overcome resource limitations and to increase the effectiveness of
implementation. In 1998, IDNR and District staff conceived of a "S.WAT." team approach to inspection. In this
approach, all IDNR inspectors and District staff together focused their efforts on all open construction sites in the county.
The objective was to visit all sites in a brief time period, thereby increasing the visibility of the program. Managers
believed that intensive scrutiny of the county, if only for a brief time, would result in greater efforts at compliance. One
of the assumptions on which this approach was based was that there are both formal and informal networks among
developers and builders and that this approach would stimulate discussion about compliance issues.
In Indianapolis, IDNR and District personnel completed a county-wide survey of construction sites on June 23 and
24,1998 (Hayes and Matthieu 1998). DCAM staff was invited to participate. IDNR, District, and DCAM staff visited more
than 300 construction sites. Of these sites, 177 were active and were evaluated for compliance with Rule 5.
Construction had not yet begun at 23 of the sites, construction had been completed at 61 sites, and the remainder were
not evaluated because they were inaccessible or because construction was just beginning. This summary is restricted
to the sites under active construction. The results provide a good picture of the current status of implementation and the
general level of effectiveness of erosion and sediment control requirements in Indianapolis.
Inspectors evaluated sites for compliance in nine categories using a standardized checklist developed by IDNR.
Sites also were checked for obvious evidence of off-site sedimentation. The nine categories were: (1) proper installation
of erosion and sediment control measures; (2) perimeter erosion control measures; (3) erosion and sediment control
measures on individual building sites; (4) protection of storm-sewer inlets; (5) stabilization of disturbed areas, (6) proper
stabilization of drainage channels; (7) stabilization of drainage outlets; (8) maintenance of existing erosion and sediment
control measures; and (9) tracking or accumulation of sediment on roadways. These criteria generally can be grouped
within the installation and maintenance stages of the evaluation framework outlined above, although most involve
297
-------
aspects of both installation and maintenance. The first seven criteria primarily concern installation of BMPs; only two,
maintenance and tracking, primarily concern maintenance.
Inspectors rated each applicable criterion at each site on a scale of Satisfactory, Marginal, Unsatisfactory or not
applicable (NA). Items in compliance with Rule 5 were rated S, items that were in danger of becoming out of compliance
were rated M and items in violation of Rule 5 were rated U. Because all criteria were not applicable at all of the sites,
the number of sites evaluated for with respect to each criterion varies.
Disturbing Results from Disturbed Sites
The results of the inspections are summarized in Figure 1 (Hayes and Matthieu 1998). Overall the results show that
installation is inadequate and that maintenance is worse. Improvements in implementation clearly are needed.
Discussion of each of the nine items reviewed follows.
Percent Satisfactory D Percent Marginal D Percent Unsatisfactory
Figure 1. Rule 5 Compliance Summary-Percent of Applicable Sites.
Installation of Erosion Control Measures
Erosion control depends upon installation of appropriate control practices in given situations. Examples
of these practices include silt-fence perimeter controls, sewer inlet and outlet protection devices, and the use
of stone or mulch to stabilize slopes. Proper installation of these devices and practices helps reduce the risk
of failure that may result in erosion and off-site sedimentation. Erosion and sediment control measures were
installed correctly at only 32% of the active sites. Installation was marginal at 24% of the sites and had been
done incorrectly at 44% of the sites. Proper installation was marked not applicable in cases where no erosion
control practices were in use.
Perimeter Erosion Control Measures
Perimeter erosion control measures are designed to keep sediment from leaving a site directly at its perimeter
through sheet or gully erosion. Perimeter erosion control devices/practices such as silt fence or buffer strips should be
installed before land disturbance begins. The most effective and cost-efficient perimeter control practice is to leave
existing vegetation in place, especially along waterways. Perimeter measures were installed and in compliance at 48%
of the sites. Marginal conditions were found at 21% of the sites, and 31% of the sites were found to be out of
compliance. Perimeter erosion control measures were not applicable at level or inward-sloping sites.
298
-------
Individual Building Sites
Under Rule 5, erosion control is the responsibility of the site operator, either the developer or builder, throughout
construction. At some sites, after the infrastructure has been installed and the lots have been sold to individual builders,
the developer no longer has direct control over erosion and sediment control practices on those lots. Builders and
contractors may or may not install and maintain erosion control practices. Erosion control on individual building sites
is a serious problem in Marion County. Erosion and sediment control measures on individual building sites at
developments were found to be adequate at only 9% of the active construction sites. Measures were in marginal
condition at 21% of the sites, and 70% of the sites were found to be out of compliance. Most of these sites lacked proper
construction entrances, storm-sewer protection, and perimeter protection. This category was not applicable for sites that
had not yet begun construction of homes.
Storm-sewer Inlet Protection
Sediment entering storm-sewer inlets significantly reduces the capacity of retention/detention basins and drainage
channels to store and convey stormwater away from flood prone areas effectively. If sediment is not removed prior to
site closure, the specified volume and dimensions of retention/detention basins that were approved by the City can
change. Inlet protection measures are especially important when sediment is tracked into or allowed to accumulate in
roadways where it is conveyed directly to sewer inlets. Using measures such as seeding and silt fence adjacent to inlets
will prevent sediment from clogging inlet protection devices and accumulating in the streets. Storm-sewer inlets were
adequately protected from sediment at just 14% of the construction sites. Sewer inlets were marginally protected at 17%
of the sites, and inlet protection measures were inadequate and not in compliance at 69% of the sites. Inlet protection
was not applicable to sites that had not completed sewer installation.
Stabilization of Disturbed Areas
Stabilization of disturbed areas on construction sites may be the single most important practice for reducing erosion
and off-site sedimentation. The best practice for achieving stabilization is to leave vegetation in place wherever possible.
If soil must be disturbed, stabilization is relatively easily accomplished through temporary seeding or application of
erosion control blanket. Rule 5 requires that disturbed areas that will be inactive be temporarily seeded. Stabilization
by seeding results in higher perceived value by potential buyers, offering developers a financial incentive to vegetate
land as soon as possible once the infrastructure is in place. Of active sites that were visited, 30% were in compliance
with Rule 5 with respect to stabilization of disturbed areas, while 32% of the sites were marginal and 38% were not in
compliance. This category of compliance was not applicable to sites that were being actively cleared or nearly completed
at the time of the visit.
Drainage Channel Stabilization
Ditches and swales designed to convey storm water away from development to natural drainage ways or storm-
sewers are subject to severe erosion and deterioration if not adequately protected. Erosion and damage to conveyance
channels results in off-site sedimentation of waterways. This can be avoided by stabilizing the soil in conveyance
channels immediately with permanent seeding of grasses, or with stone, mulch, or straw cover. Conveyance channel
stabilization was satisfactory at 44% of the sites, Approximately 23% of the sites had marginally protected channels and
33% of the sites had channels in unsatisfactory conditions. Conveyance channel stabilization was not applicable at sites
that did not have or require channels or at those that did not yet have them constructed.
Outlet Stabilization
Storm-sewer and drainage channel outlets from a site need to be properly stabilized to prevent erosion and
sedimentation of the banks and waters they empty into. Outlet stabilization is best accomplished by protecting the soil
around the outlet with stone riprap, geotextile fabric, or with well-established vegetation. Outlet stabilization was
satisfactory at 73% of active sites. Outlets were in marginal conditions at 18% of the sites and unsatisfactory at 9%. This
299
-------
category was not applicable at sites that did not have outlets on the site or where infrastructure development was not
yet completed.
Maintenance of Erosion Control Practices
Erosion control practices that have been installed properly must be maintained to be effective. In most cases, lack
of maintenance results in the same effects as not employing erosion control practices at all. Examples of maintenance
of erosion control practices include removing accumulated sediment from behind silt fence and reinforcing inlet protection
after storms. Failure of erosion control practices allows sediment to leave construction sites via storm-sewers, drainage
channels, roadways and sheet and gully erosion, An often-overlooked aspect of maintenance is removal of devices after
work is completed. Maintenance of erosion control practices was satisfactory at only 18% of the sites. Maintenance was
marginal at 27% of them, and there was little or no evidence of maintenance at 55% of the sites. Maintenance of erosion
control measures was not applicable at sites that did not employ erosion or sediment control practices.
Sediment Tracking and Accumulation in Roadways
Soil and sediment in streets and roads are readily washed into sewers and drainage channels and can be a
significant source of pollution. In addition, the sediment can be a traffic hazard with the potential for costly litigation
against the local governments or developers. Sediment accumulated in roads is also unsightly and may discourage
potential home buyers. Tracking and accumulation of soil in roads was kept to an acceptable level at 24% of the sites.
Approximately 28% of the sites exhibited marginal compliance with the rule for keeping roads clear of sediment. Sites
that were out of compliance with the rule made up 48% of this category. Large industrial sites where equipment was
usually kept on site and residential sites that did not yet require extensive coming and going of vehicles were rated not
applicable for sediment tracking.
Off-site Sedimentation
Sediment is the most abundant pollutant, by volume, in Indiana waters. Residential and commercial development
sites are potential sources of high volume, sudden discharges of sediment that can cause problems for land owners
down-stream of development. Besides the drainage and flooding problems caused by off-site sedimentation, sediment
can obstruct and widen streams and erode stream banks. Sedimentation of the state's streams and rivers also causes
habitat damage for many aquatic species. There were obvious signs of off-site sedimentation at 21% of the active sites.
This figure is believed to be low, however, due to the large number of sites surveyed in a very short time. Only the most
obvious cases were checked as displaying off-site sedimentation.
Observations and Implications: Priority-problem Solving
A number of observations that have important implications for managers of erosion and sediment control programs
can be drawn from this inspection initiative. First, it is useful to consider the initiative in the more general framework for
effectiveness in erosion control. Indiana regulations for erosion and sediment control first were adopted in 1992. Faced
with implementation of a new regulation with few resources, IDEM, IDNR, and District staff first devoted efforts to
education and ensuring complete coverage and competent planning. In late 1997 and early 1998, program managers
determined that the plan review process was fairly well established and that additional effort needed to be devoted to
installation and maintenance of BMPs. Because resource shortages preclude regular, periodic inspection, IDNR officials
developed a S.W.A.T. team approach. Teams of state, district, and available municipal officials focused inspection
efforts, visiting and inspecting as many sites as possible in a short time.
In Indianapolis, the results show that implementation generally is poor. Installation of erosion and sediment controls
was unsatisfactory on 44% of all sites, and satisfactory on less than one-third. With the exception of outlet stabilization
practices, which had been installed properly at nearly three-fourths of the sites, no practice was installed properly on
more than half of the sites. Perimeter controls, a basic practice, were installed satisfactorily on fewer than half of the sites
and they were unsatisfactory at almost one-third. Stabilization was satisfactory at less than one-third of the sites, inlet
300
-------
protection had been installed properly at less than 15% of the sites, and controls on individuals lots had been installed
properly at just 9% of the sites where they were needed. It is clear that installation is deficient and that additional effort
is needed to ensure that practices identified on plans are installed properly.
The inspections show that maintenance of erosion and sediment controls is even worse than installation. Inspectors
determined that maintenance of controls was unsatisfactory on 55% of the sites and satisfactory on only 18%. Mud is
being tracked on streets and washed into sewers and drainage channels on almost half the construction sites. Additional
field work to ensure proper maintenance of BMPs is a critical need.
Although these results were disturbing, they were not unexpected. Program officials knew that implementation was
inadequate and devised the S.W.A.T. team approach to provide a quick, comprehensive assessment of the status of
implementation. Since the inspections, program managers have used the results as part of overall efforts to increase
understanding of requirements for erosion and sediment control and to build commitment to the programs. City staff
agreed to mail copies of inspection reports to all developers, and the district provided a summary of results to all city-
county councilors.
The results provide information that program managers can use to establish priorities for problem solving and
education. For example, installation of perimeter controls appears better than efforts to stabilize disturbed areas on site.
Future inspections and educational efforts therefore can focus on the importance of stabilization. Similarly, since it
appears that site operators are doing a fairly good job at stabilization of outlets, this requirement can be de-emphasized,
and additional effort can be devoted to solving problems like installation of controls on individual lots that are not
controlled by practices on the larger development site. More generally, as more people understand the different steps
in the process of erosion and sediment control, implementation should become more effective.
The survey did not focus on discovering reasons behind compliance or non-compliance, but several inferences can
be drawn from these data. First, the data and experience indicate that some developers are unaware of their obligation
to control erosion and sedimentation and leave the permitting and erosion control planning to engineers and contractors.
This can result in a lack of commitment to implementation. Second, some developers, engineers and contractors clearly
do not yet understand the purpose and importance of implementing erosion and sediment control practices. Education
is needed to increase their understanding and commitment. Third, some operators know the requirements of Rule 5,
but do not take them seriously, ignoring the Erosion and Sediment Control Plan. For these individuals, enforcement
action may be required. In addition, a general problem that was observed has to do with sequence of construction. All
too often, land disturbance is beginning before erosion and sediment control measures are installed. More emphasis
must be placed on installation of practices prior to earth disturbance, and site operators must learn to follow the sequence
described on plans.
Given that resource shortages are likely to continue, problems in implementation are likely to continue and regulatory
programs are likely to remain less effective than they could be. Steps that may be taken to increase effectiveness
include making sure that the regulated community participates in on-site, pre-construction meetings that underscore the
scope and importance of controls: increasing the visibility of IDNR and District staff and the frequency of their site visits;
educating developers, engineers and contractors about erosion and sediment control practices and how to install and
maintain them; and emphasizing the need for erosion and sediment control throughout the entire development process.
The S.W.A.T. team approach clearly does not solve the problems of a relatively new, understaffed erosion and
sediment control program. But the approach is an effective way to obtain a significant amount of information in a short
time, raise the visibility of erosion and sediment control programs, and help establish priorities for problem solving.
References
Clevenger, B. Miscellaneous unpublished materials. Maryland Department of the Environment, Sediment and Stormwater
Administration, Baltimore, MD n.d.
301
-------
Department of Civil Engineering and the Department of City and Regional Planning. Evaluation of the North Carolina erosion
and sedimentation control program. North Carolina State University, Raleigh, NC, 1990.
Hayes, Jim and Marcia Matthieu. "S.W.A.T. Summary." Marion County Soil and Water Conservation District, IndianapolisJN,
1998,
302
-------
Using Constructed Wetlands to Reduce
Nonpoint Source Pollution in Urban Areas
Jon Harbor
Department of Earth and Atmospheric Sciences
Purdue University
West Lafayette, IN 47907
Susan Tatalovich
FMSM Engineers, Inc.
Columbus, Ohio 43229
Ron Turco and Zac Reicher
Department of Agronomy
Purdue University
West Lafayette, IN 47907
Anne Spacie and Vickie Poole
Department of Forestry and Natural Resources
Purdue University
West Lafayette, IN 47907
Abstract
Potential pollutants carried in stormwater runoff from urban surfaces are a major component of Nonpoint Source (NPS)
pollution. NPS pollution is a leading cause of reduced water quality in US rivers and lakes, and there are major efforts
underway to find innovative approaches to reducing NPS pollution from a wide range of sources, In urban areas, where
much of the land has existing structures, a major challenge is to find ways to retrofit built sites to reduce NPS pollution
associated with stormwater runoff. One component of this may be more widespread use of constructed wetlands that have
value not only in terms of water quality improvement, but also in terms of urban ecology, aesthetics and education.
We have begun a long-term monitoring program of the performance of constructed wetlands in two settings: 1) On
a commercial site where surface runoff is dominated by stormwater flow from parking lots and store roofs, and, 2) On a
golf course that receives considerable surface flow from adjacent commercial, residential and highway areas. Monitoring
includes both continuous measurements of flow, temperature, conductivity, pH and dissolved oxygen, and automated
sample collection during storm events for more complete chemical analyses. Initial results suggest that the commercial
site constructed wetland acts as an efficient trap during the spring and summer for suspended sediment and some
dissolved matter. During the fall and winter dormant season trap efficiencies are much lower, and in some cases negative.
The golf course site constructed wetlands also function as efficient traps during the summer, and plant growth in these
wetlands has been helped considerably by the regular water supply provided by golf course irrigation. Both wetland
systems also provide value in terms of improved aesthetics, their use by local educators, their diverse ecological
assemblages, and the public relations benefits associated with visible efforts at improved environmental management.
Replacing portions of existing parking lots with carefully designed constructed wetlands, and adding constructed
wetlands to urban recreational sites (such as golf courses and parks) should be viewed as one of several elements of an
integrated approach for effective retrofitting urban areas to reduce NPS pollution from stormwater runoff.
303
-------
Introduction
One of the major challenges facing urbanized areas is to find ways to improve environmental management in ways
that do not involve major, costly impacts on existing infrastructure. Increasing recognition of the environmental impacts
of built areas on parameters such as runoff amount and quality has increased regulatory and public pressure to develop
and implement effective management practices. However, many of the best management approaches that can be
integrated into the design of new developments cannot be implemented in existing built areas without prohibitive costs.
Thus there is considerable interest in best management practices that can be used to retrofit urban areas for improved
environmental performance.
Wetlands have the ability to store large amounts of water, reducing flooding of surrounding areas and in some cases
recharging groundwater (Mitsch and Gosselink, 1993). In addition, wetlands are capable of improving runoff quality in
many situations (Perry and Vanderklein, 1996) because they trap both solid and dissolved pollutants. Wetlands also can
have considerable aesthetic benefits, and provide habitat for a wide range of plants and animals. Constructed wetlands
are wetlands specifically designed and built for hydrologic and water quality management, as opposed to either natural
wetlands or created wetlands. Created wetlands are designed and built to replace lost wetlands or to compensate for
destruction of natural wetlands. Using constructed wetlands for water treatment attempts to take advantage of the benefits
of wetlands without compromising natural wetland areas.
In urban areas there are unique challenges to be faced in proposing and designing constructed wetlands, Existing
built areas rarely include extensive undeveloped space that can be converted to constructed wetlands. However, there
are several opportunities that arise in many areas, including: 1) Making space by reducing the size of an existing parking
lot; 2) Adding a constructed wetland to a redevelopment or urban renewal project; 3) Adding a constructed wetland to a
park or green space; 4) Adding constructed wetlands to existing recreational facilities such as golf courses. A second major
challenge in proposing constructed wetlands in built urban areas is to maintain adequate hydrology for long-term wetland
survival. The extensive impervious surfaces of built areas generate large amounts of runoff during storm events, but this
water is usually routed quickly away from the built area to prevent flooding. Because there is little opportunity for rainfall
to infiltrate into the soil in urban areas (because most soil is covered by impervious surfaces), shallow groundwater flow
is reduced. This means that wetlands in urban areas will receive far less between-storm water recharge from shallow
groundwater than would be expected for a similar non-urban setting. In essence, wetlands in urban areas will experience
a "flood and drought" hydrologic regime, which is poorly suited to an ecosystem that is based on extensive periods of wet
conditions. One way around this problem is to look for locations where water is applied regularly to adjacent areas, in
particular where extensive irrigation is used. Golf courses and lawns and gardens of major corporate complexes are
potential sites where between storm irrigation might provide excess runoff and soil water drainage to adjacent constructed
wetlands.
Given the potential use of constructed wetlands to improve water quality in built areas, it is important to evaluate how
well wetlands function as pollutant traps in such settings. Such studies can be used to drive design improvements, and
to evaluate the cost-effectiveness of using constructed wetland for NPS pollution control. Although there has been less
work done in the area of stormwater constructed wetlands, in comparison to wetlands used as part of a wastewater
treatment system (e.g. Hicks and Stober, 1989), limited results so far suggest that wetlands can be effective in treating
stormwaterfor nonpoint source (NPS) pollution (Mitsch and Gosselink, 1993; Witthar, 1993; Livingston, 1989). Few data
sets are available because of poorfollow-up of constructed wetland performancethrough appropriate monitoring programs
(Perry and Vanderklein, 1996). However, available studies to date and theoretical reasoning suggest that NPS pollution
control is enhanced by maximizing the distance between the wetland's inlet and outlet, including deep and shallow sections
in the wetland, selecting vegetation on the basis of climate and water quality and supply conditions, maximizing the ratio
of treatment area to base flow, and minimizing the slope along which the water travels (Horner, etal., 1994; Witthar, 1993).
The idea in such a design is to model the constructed wetland after a natural wetland, which not only has the ability to slow
down the flow of water (as does a detention or retention pond), but also can remove pollutants from the runoff water. The
most important factor in the design and maintenance of constructed wetlands is hydrology (Mitsch and Gosselink, 1993).
Without the proper water inflow and outflow, the newly created wetland will fail and be unable to accomplish its task of
stormwater treatment.
304
-------
Aims and Objectives
We are monitoring the performance of urban constructed wetlands in two settings, a constructed wetland incorporated
into site development for a commercial facility and a series of constructed wetlands built into a recently renovated golf
course that receives runoff from an adjacent urban area. Both sites are in West Lafayette, Indiana. The goal of long-term
monitoring is to provide insight into seasonal and longer-term variations in trap efficiency, both as the basis for improved
scientific understanding of constructed wetland processes and controls, and to form the basis for future improvements in
design.
Study Areas
The commercial constructed wetland site occupies approximately 0.51 ha, with a water surface area of 0.26 ha and
volume of 1300 m3. This wetland is intended to treat the "first flush" of runoff, and so was designed to accommodate the
volume of water corresponding to first half-inch of precipitation on the store's impervious surfaces (the parking lot and the
rooftop). The mean depth of the constructed wetland is 0.5 m but this includes two deeper pools with a maximum depth
of 1.8 m (Tatalovich, 1998). Conventional wisdom (which may not be correct) states that 90% of the annual pollutant load
is transported in the runoff produced by the first 1.3 cm of precipitation (known as the first flush), and this has been shown
to be true for the transport of most pollutants over impervious surfaces (Chang, 1994). At this commercial site, runoff that
exceeds the first-flush equivalent is routed to a separate basin.
One motivating force behind use of a constructed wetland on this site was concern over potential impacts on a natural
wetland (Celery Marsh) adjacent to the property. In addition to the constructed wetland, this site includes: elimination of
a proposed auto care center, abstinence from chemical ice-clearing methods, and construction of additional ponds to treat
stormwater runoff that could potentially include harmful pollutants. The constructed wetland receives runoff primarily from
the 4.1 ha commercial parking lot, as well as minor additional input from an adjacent store, local access roads, and US
Highway 52.
The golf course created wetlands are part of Purdue's new Kampen Golf Course and are positioned to intercept both
runoff from much of the golf course and the adjacent urban area. The developed area includes two residential highways,
a section of state highway, the parking lot of a motel, a gas station, and 200 residences. The water flowing through the
Kampen Course eventually enters Celery Marsh, but prior to reconstruction this water flowed directly through drainage
tiles and overland transport to the marsh, with no treatment. The golf course constructed wetlands serve several purposes:
providing a water hazard and aesthetic component of the course, and enhancing environmental quality that can also be
used in environmental education. Runoff from the urban area travels through three constructed wetlands prior to leaving
the course. One particularly notable aspect of these constructed wetlands is that they have flourished even during long
dry summer periods. Frequent watering of the greens and fairways, common on most courses, has the added advantage
that it provides runoff and tile drainage to the wetlands throughout the summer.
Methodology
To determine the effectiveness of each constructed wetland in trapping potential pollutants, water samplers were
installed at the inlet and outlet of the commercial constructed wetland (Figure 1), and at six locations in the golf course
constructed wetland complex to track the progress of water as it enters the course, moves through the wetland system,
and exits to the Celery Marsh. The samplers are equipped with ISCO® Submerged Probes that measure water levels, used
in conjunction either with a weir or pipe of known geometry. The sampler uses these levels and the corresponding
geometry of the sampling sites to calculate the flow into and out of the wetland. Each sampler also has a YSI® 600 Multi-
Parameter Water Quality Monitor that measures dissolved oxygen, conductivity, temperature, and pH. The samplers record
flow and water quality parameters every five minutes and are programmed to take water samples during storm events.
Storm sampling is triggered in most cases by a change in water level, and at two locations, by rainfall intensity as
measured with an automatic tipping bucket rain gauge. The trigger points were determined empirically, so that inlet and
outlet samplers begin to sample at approximately the same time. The sampling programs for each sampler are split into
two sections. The interval of time between samples in part A of each routine is closer together than those in the
corresponding part B routines, so that sampling occurs more often during the "first flush." After that, the second stage
305
-------
Figure 1. Sampling equipment at a constructed wetland. The laptop computer is downloading monitoring data from the sampler, and in the foreground
is a set of 24 sample bottles for storm sampling.
of each routine samples at larger intervals to guarantee samples at times coinciding with the downward slope of the
hydrograph.
Overall, the design of the experiment is to track flow and water quality into and out of the constructed wetlands
continuously, both during storms and between storms, for a multi-year period. This allows for determinations of storm,
seasonal and multi-year trends in constructed wetland trap efficiency. Trap efficiency can be defined in a number of ways,
depending on the likely application of the results. In this work we are interested in concentration trap efficiency (percentage
change in potential pollutant concentration between the inlet and outlet, both maximum and average values) and load trap
efficiency (percentage change in potential pollutant load between the inlet and outlet for given points in a storm, for storm
totals, and seasonally and annually). Selected samples from each precipitation event are analyzed by a Purdue University
laboratory for total suspended solids (TSS), hardness, total Kjeldahl nitrogen (TKN), and total phosphorus (TP). These
parameters are the same as those measured for seven other local sites as part of a larger analysis of water quality in rural
and urban settings. In addition to the analyses performed at the Purdue laboratory, more complete chemical scans are
performed once per season on selected samples by Heritage Environmental Services in Indianapolis, Indiana. The
selection of tests is based on the pollutants that might reasonably occur at each site. The reason for this more complete
scan is to determine whether any potential pollutants not routinely measured at the Purdue laboratory show up at unusually
high levels. Any parameters that were not detected in the Heritage samples could potentially be excluded from future
testing, but those parameters considered to be problems would need to be monitored on a consistent basis in the future.
Results and Discussion
To illustrate possible types of analyses and some major trends in the performance data, without reviewing the entire
data sets available, this discussion includes three examples from the two sites. These include a complete storm record
at the commercial site, between-storm sampling at the commercial site, and first-flush storm sampling at the golf course
site.
Sample Storm at the Commercial Site
A 0.97 cm-storm occurred on 26 October 1997, with a double peak in intensity (Figure 2). As expected, the wetland
acts to damp peak flows, so discharge values at the outlet slightly lag those at the inlet and are lesser in magnitude. Water
temperature in the constructed wetland inlet is high and uniform (no diurnal variations) prior to the storm (Figure 3),
306
-------
0.35 r—
0.30-
0.25'
Total
Hourly
Rainfall
(cm)
Average
Hourly
Discharge
(103 cmr/s)
0.10"
0.05 ''
0.00
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
Time afttr Midnight, 25 October 1997 (hours)
Figure 2. Rainfall and inlet and outlet runoff records for a storm event at the commercial site constructed wetland.
14.0 T
12.0
I
5'
<
10.0
80 -
60
4.0
20
0.0
DO (Inlet)
DO (Outfel) \
Temperature (Inlet) •
Temperature (Outlet);
16.0
140
12.0 £
10.0\
8.0 ft
"Z
»
6.0 £
&
4.0 u
<
2.0
0.0
0 5 10 15 20 2i JO 35 40 4$ 50 55 60 65 70
Time after Midnight, 25 October 1997 (hours)
Figure 3. Dissolved oxygen and temperature records for the storm shown in Figure 1.
drops 6°C during the storm, and slowly climbs back2°C in the 20 hours after precipitation stops. The outlet temperature
shows a 4°C diurnal cycle prior to the storm, and a lower amplitude cycle after the storm. At the same time, the dissolved
oxygen (DO) values climb during the storm (Figure 3). Inlet DO values vary within the 0 to 4 mg/L range before the storm,
jump up to 9 to 13 mg/L during the storm, and fall during the 24 hours following the storm event. The outlet DO varies from
1 to 8 mg/L prior to the storm, is very stable between 8 and 9 mg/L during the storm, and has strong variations from 3 to
9 mg/L post-storm. High DO values during the storm are due to the increased mixing of the water, which causes oxygen
to be introduced to the wetland, as well as the addition of "new" water that is higher in oxygen to the stagnant water.
307
-------
50
45
40
35
25
20
15
10
5
0
Inlet
I Outlet
50 100 150 200 250 300 350 400 450
Time after Start of Sampling (minutes)
30 200 400
Time after Start of Sampling (minutes)
Figure 4. Total suspended sediment load for the storm shown in Figure 1 (upper), and calculated trap efficiency (TE) if only grab sampling had been
used (lower). The actual load TE for the storm was 84%.
Initially, hardness data from the 26 October 1997 storm showed much higher values for the outlet than the inlet. This
discrepancy relates to the movement of water through the wetland. Soon after the start of a storm, water begins to flow
over the weir at the inlet and inlet sampling begins. This new runoff from the site has a low hardness, reflecting the
naturally low hardness of rainwater. At the outlet, high hardness values show that the water initially being sampled is not
new inlet water being displaced from the wetland; rather it is water that was stored in the concrete outlet box prior to the
storm. Hardness values at the outlet fall throughout the storm, showing that hardness is lower in the wetland than in the
308
-------
concrete outlet box. Because hard water occurs when concentrations of Ca2+ and Mg2* are high (Zumdahl, 1989), the
concrete surroundings themselves can add to the hardness of the water. Ninety-five percent of all concrete used is made
from Portland cement, which is prepared using finely ground limestone (Mindess and Young, 1981). The cement is made
into concrete using admixtures such as calcium chloride. Also, the interaction of the surface water with groundwater, which
in this geographic location is very hard due to infiltrating rainwater that dissolves the calcite-rich till from limestone in the
region, can add to the hardness of the water in the wetland (Davis and Cornwell, 1991).
The hardness data suggest that the outlet record probably does not include water that enters and exits the wetland
during the same storm. Other studies (e.g. Bhaduri, et al., 1997) have shown through chemical load distributions that the
inflow and outflow from a basin are actually two different water masses, except in extreme storm events. The only way
that the same water could appear at the inlet and outlet during the same storm would be if the inflow sheeted over the
water in the basin, arriving at the outlet without significantly mixing with the water stored in the wetland prior to the storm;
or if the storm produced enough water to completely displace the volume of water in the wetland.
Total suspended solids (TSS) concentration data and flow values for the storm are used to calculate TSS loading
values, which depict the effect of the basin in reducing the overall sediment load. Total loads depict the actual physical
amount of sediment entering the wetland and are important for planning activities such as dredging. TSS load values for
the inlet are larger than the outlet (Figure 4). The inlet values start high, dip down, and then increase again. This indicates
that the initial runoff has "first flush" (high load) characteristics, and then the load input rate decreases. A second, lower
peak later in the storm could be the result of the later pulse of higher rainfall intensity (Figure 2). The values of TSS at the
outlet remain fairly uniform throughout the storm. The initial value presumably represents between-storm ambient TSS
loading in standing water in the wetland. During the storm, the increase in flow creates more turbulence, which can stir
up some of the bed sediment, slightly increasing the TSS concentration and, therefore, the load. More importantly, though,
the outlet values are lower than those for the inlet; thus there is a net decrease in TSS loading from the inlet to the outlet
for this particular storm. In one sense, this traditional approach is a valid efficiency measure because the water going out
is compared to that going into the wetland, but in another sense it is a skewed picture because the new inlet water is being
compared with "old" outlet water that arrived in the basin during a previous storm (Bhaduri, et al., 1997).
Multiple storm sequence sampling will provide a better view of overall trap efficiency (TE) than a single storm, just
as a complete storm record is better than a grab sample. Standard grab samples do not always lead to accurate trap
efficiency calculations (Figure 4). If one sample were taken each from the inlet and the outlet at exactly same time, the
data could show a very high trap efficiency (30 minutes), no trap efficiency (200 minutes), or a fairly high trap efficiency
(400 minutes). The overall load TE for this analysis was 84%. This is one of the reasons that this particular study samples
several times after the start of a storm - to bridge the gap between standard grab samples and actual events within the
wetland. Continuous monitoring provides a more complete record of the constructed wetland's activity, more accurately
depicting the trap efficiency of the wetland. From conductivity data, during the monitoring period, 137 kg of dissolved load
entered the basin, and 59 kg left the basin, for a total dissolved solids (TDS) load TE over the storm of 57%. Further
analysis of many storms can be used to determine an overall trap efficiency over longer periods of time. This type of
analysis could be used to determine the effects of different storm intensities, seasonal variations, and increased
urbanization in the area.
Detailed Chemical Scan at the Commercial Site: Between Storm Conditions
Samples for a detailed chemical scan were taken on 17 December 1997 using the sampler's grab sampling
mechanism. At this time, there had not been a precipitation event in a couple of weeks, so these samples represent
between-storm conditions in the wetland. Although these samples were tested for many possible pollutants, only a few
were detected (Table 1).
Parameters which show reductions between the inlet and the outlet were chloride, sulfate, ammonia nitrogen, calcium,
magnesium, sodium, silicon, strontium, and total dissolved solids (TDS). For instance, chloride levels fell from 210 to 160
mg/L, calcium levels fell from 95 to 54mg/L, and strontium levels fell from 0.16mg/Lto below the detection level of 0.10
309
-------
Table 1. Detailed chemical scan of the commercial site constructed wetland. All values are mg/L.
Parameter Inlet Outlet Detection Limit
Chloride 210 160 2.5
Sulfate 49 37 1.3
Nitrogen, Nitrate-Nitrite 0.12 0.11 0.01
Nitrogen, Ammonia 0.21 0.14 0.12
Chemical Oxygen Demand 18 28 10
Aluminum BDL 1.3 0.10
Calcium 95 54 0.10
Iron 0.33 1.7 0.70
Potassium 2.5 2.7 0.10
Magnesium 28 15 0.10
Manganese 0.32 0.24 0.10
Sodium 110 85 0.10
Silicon 5.1 3.0 0.10
Strontium 0.16 BDL 0.10
Total Organic Carbon BDL 4.2 7.0
Total Phosphorus BDL 0.10 0.03
Dissolved Solids 720 490 10
Total Suspended Solids 4 13 1
Notes: Italicized parameters are those which have an outlet value > inlet value.
BDL = below detection limit
mg/L. Also, TDS levels fell from 720 to 490 mg/L. When compared to the values calculated using conductivity data from
the sampler, these values are slightly higher than the values calculated for the 26 October 1997 storm event. The
maximum TDS values calculated for the inlet and the outlet were, respectively, 568 and 365 mg/L, with average values
around 337 and 263 mg/L. The higher between-storm values could be a result of the ability of sediments to dissolve in
the wetland waters. Reductions in values between the inlet and the outlet indicate removal of certain pollutants within the
wetland and also suggest that at the beginning of a storm, the outlet values will be lower than those of water near the inlet.
Because of this, the best TE should be at the start of a storm, which is shown in the 26 October 1997 storm chemical data.
Not all of the detectable parameters were lower at the outlet than at the inlet. The ones that were actually larger at
the outlet than at the inlet were: chemical oxygen demand (COD), aluminum, iron, potassium, total organic carbon (TOC),
total phosphorus (TP), and total suspended solids (TSS). The increase in TSS is interesting, and may be the cause of
increases in adsorbed pollutants. This could be attributable to the lack of growth of plants in the middle of December.
Plants slow flow within the wetland, allowing sediments in the water to settle, and plants have the ability to take into their
roots pollutants carried by the sediments (Pond, 1995). Because of this, as the plants die, they may release the sediments
and pollutants trapped earlier in the year, as well as releasing products of the decay of the organic matter. Aluminum, iron,
potassium, and phosphorus could have been attached to these sediments, especially the finer particles. Findings such
as these agree with previous studies that noted a distinct reduction in the performance of stormwater wetlands in winter
months (Oberts, 1994; Ferlow, 1993). Not only does plant death have an effect, but also the formation of ice on the water
surface can scour the margins and resuspend the sediments and the pollutants that they carry (Oberts, 1994).
Detailed Chemical Scan at the Golf Course Site
First flush samples were collected for detailed chemical analysis during the first pulse of runoff from a storm in
November 1998 and a second storm in June 1999 (Table 2). In November 1998, 14 water quality parameters declined
in terms of a comparison between the urban input (Site 1) and the golf course output (Site 6). Four water quality
parameters improved between the urban input and the water exiting the course during the same storm. This suggests that
the constructed wetlands were not working well soon after initial construction, during the late fall. However, key
parameters such as ammonia and nitrate-nitrite nitrogen and pesticide levels were either decreased as the watercirculated
through the golf course wetlands or were not detectable at either sampling site.
310
-------
A distinctly different pattern of results is apparent in the June 1999 sampling (Table 2). Fifteen water quality parameters
improved between the urban input and the water exiting the course, and only 4 parameters declined. This suggests that
the golf course's created wetland system is functioning well to improve the water quality in the late spring when wetland
plants have become established. Two parameters of particular interest for a golf course are nitrate-nitrite N and ammonia-
N, which were undetectable in water exiting the course, but at 2.1 and 31 ppm, respectively, in water flowing onto the
course.
Table 2. Detailed chemical scan of the golf course site constructed wetland, selected parameters. All values are mg/L.
November 1998 June 1999
Parameter
Simazine
Atrazine
Oil and Grease
Chloride
Sulfate
Nitrogen nitrate-nitrite
Ammonia nitrogen
Chem. 0, Demand
Mercury
Total Organic Carbon
Phosphorus
Dissolved Solids
Suspended Solids
Silver
Aluminum
Arsenic
Calcium
Cadmium
Chromium
Copper
Iron
Potassium
Magnesium
Manganese
Molybdenum
Sodium
Nickel
Lead
Selenium
Silicon
Tin
Titanium
Zinc
Detection
limit
0.10
0.10
5
2.5
2.5
0.01
0.12
10
0.0002
0.03
10
1
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
Site 1
Urban
runoff
BDL
BDL
BDL
8.6
11
1.1
0.23
40
BDL
8.2
0.19
91
17
BDL
0.31
BDL
29
BDL
BDL
BDL
0.51
2.3
7.1
BDL
BDL
4.5
BDL
BDL
BDL
2
BDL
BDL
BDL
Site 6
Created
wetland outlet
BDL
BDL
BDL
22
55
0.06
BDL
37
BDL
10
0.17
270
290
BDL
5.8
BDL
61
BDL
BDL
BDL
4.7
7.8
24
0.21
BDL
6.8
BDL
BDL
BDL
14
BDL
0.14
BDL
increase/
decrease
BDL
BDL
BDL
+156%
+400%
-95%
-52%'
-8%
BDL
+22%
-11%
+197%
+1606%
BDL
+1771%
BDL
+110%
BDL
BDL
BDL
+822%
+239%
+238%
+133%
BDL
+51%
BDL
BDL
BDL
+600%
BDL
+56%*
BDL
Site 1
Urban
runoff
BDL
0.1
BDL
32
18
2.1
31
480
BDL
240
0.32
240
8
BDL
1.8
BDL
40
BDL
BDL
BDL
1.6
2.2
9.9
0.28
BDL
6.5
BDL
BDL
BDL
2.0
BDL
BDL
0.38
Site 6
Created
wetland outlet
BDL
BDL
BDL
20
31
BDL
BDL
25
BDL
1.6
0.08
220
2
BDL
0.16
BDL
34
BDL
BDL
BDL
0.26
0.37
28
BDL
BDL
8.7
BDL
BDL
BDL
4.8
BDL
BDL
BDL
increase/
decrease
BDL
-91 %'
BDL
-38%
+72%
-100%"
-1 00%*
-95%
BDL
-99%
-75%
-8%
-75%
BDL
-91%
BDL
-15%
BDL
BDL
BDL
-84%
-83%
+183%
-64%
BDL
+34%
BDL
BDL
BDL
+140%
BDL
BDL
-74%'
BDL = Below Detection Limit
. where contaminant was BDL, the detection limit was used for % increase/decrease calculations
No unusually high levels of any of a wide array of potential pollutants, including pesticides and metals, were detected
at the golf course sampling sites. However, atrazine was detected in water exiting the neighborhood and entering the golf
course (Site 1). Surprisingly, even from the urban runoff there was no measurable oil and grease. It is reassuring to note
that heavy metals of concern, such as mercury and lead, are below detection limits in all samples.
Conclusions
Constructed wetlands can potentially be used to improve NPS pollution management in urban areas. Although finding
space for constructed wetlands can be a challenge in developed areas, these management tools can be incorporated into
the design of new or renovated commercial and industrial facilities. In some cases, they can be added to recreational
311
-------
facilities such as parks and golf courses. In each of these cases, good initial design and attention to continued water
supply for long-term wetland survival is critical.
The constructed wetland monitoring program in West Lafayette, Indiana, includes both commercial and golf course
constructed wetlands. Selected results presented here illustrate the complexity of developing a program to evaluate
performance of such wetland systems. Traditional grab sampling can provide misleading results compared to continuous
sampling, and it is clear that apparent trap efficiency varies both within storms as well as between seasons. The type of
complete picture of constructed wetland performance that is needed to improve design and enhance understanding of
chemical and biological processes in constructed wetlands can be approached by continuous monitoring through several
years. Initial data suggest that the constructed wetlands studied here are generally performing well to reduce loads and
concentrations of a range of urban NPS pollutants, particularly during spring and summer storm events after wetland
vegetation has become established. However there is also a strong indication that trap efficiencies are much lower, and
in some cases negative, during winter months. The implications of this depend on the context provided by the receiving
area.
Constructed wetlands also provide important benefits beyond water quality control. They provide aesthetic diversity
in urban settings, they represent islands of habitat types that are generally absent or underrepresented in older developed
areas, and they provide important local educational resources in urban areas. Overall, constructed wetlands should be
considered as a potential element of urban retrofit projects, if there are situations where water supply is available to
maintain wetland hydrology.
Acknowledgments
The Purdue constructed wetland monitoring program would not be possible without generous support from the
Showalter Trust, the United States Golf Association, Pete Dye, Inc., and Heritage Environmental. Equally important for
the project has been the willing cooperation of Jim Scott, superintendent of the Birck Boilermaker Golf Complex, Wal*Mart
Inc., and numerous undergraduate and graduate field assistants.
References
Bhaduri, B. L, Harbor, J. M., and Maurice, P. (1997) Chemical load fractionation and trap efficiency of a construction site
storm water management basin. In Environmental & Engineering Geoscience. 3(2): 235-249.
Chang, G. (1994) First flush of stormwater pollutants investigated in Texas. In Watershed Protection Techniques. Thomas
R. Schueler, Editor. Herndon, VA: Center for Watershed Protection. l(2): 88-89. From Chang, G., J. Parrish, and C.
Souer. (1990) The first flush of runoff and its effect on control structure design. Environmental Resource Management
Division of Environmental and Conservation Services. Austin, TX.
Davis, M. L. and Cornwell, D. A. (1991) Introduction to Environmental Engineering. New York: McGraw-Hill, Inc.
Ferlow, D. L. (1993) Stormwater runoff retention and renovation: a back lot function or integral part of the landscape?
In Constructed Wetlands for Water Quality Improvement. Gerald A. Moshiri, Editor. Boca Raton, FL: Lewis Publishers,
Inc. [CRC Press, Inc.] pp. 373-379.
Hicks, D. B. and Stober, Q. J. (1989) Monitoring of constructed wetlands for wastewater. In Constructed Wetlands for
Wastewater Treatment: Municipal, Industrial, and Agricultural. Donald A. Hammer, Editor. Chelsea, Ml: Lewis
Publishers, Inc. pp. 447-455.
Horner, R. R., Skupien, J. J., Livingston, E. H., and Shaver, H. E. (1994) Fundamentals of Urban Runoff Management:
Technical and Institutional Issues. Washington D.C.: Terrene Institute, United States Environmental Protection Agency
Livingston, E. H. (1989) Use of wetlands for urban stormwater treatment. In Constructed Wetlands for Wastewater
Treatment: Municipal, Industrial, and Agricultural. Donald A. Hammer, Editor. Chelsea, Ml: Lewis Publishers, Inc. pp.
253-262.
312
-------
Mindess, S. and Young, J. F. (1981) Concrete. Englewood Cliffs, NJ: Prentice-Hall, Inc.
Mitsch, W. J. and Gosselink, J. G. (1993) Wetlands. New York: Van Nostrand Reinhold.
Oberts, G. (1994) Performance of stormwater ponds and wetlands in winter. In Watershed Protection Techniques.
Thomas R. Schueler, Editor. Herndon, VA: Center for Watershed Protection. l(2): 64-68.
Perry, J. and Vanderklein, E. (1996) Water Quality: Management of a Natural Resource. John Lemmons, Consulting
Editor. Cambridge, MA: Blackwell Science.
Pond, R. (1995) Pollutant dynamics within stormwater wetlands: plant uptake. In Watershed Protection Techniques.
Thomas R. Schueler, Editor. Herndon, VA: Center for Watershed Protection. l(4): 21 O-21 3.
Tatalovich, S.K. (1998). Chemical and Sediment Trap Efficiency of a Stormwater Detention Biofiltration Pond: A Study
of the Wal*Mart Constructed Wetland. Master's Thesis, Civil Engineering, Purdue University.
Witthar, S. R. (1993) Wetland water treatment systems. In Constructed Wetlands for Water Quality Improvement. Gerald
A. Moshiri, Editor. Boca Raton, FL: Lewis Publishers, Inc. [CRC Press, Inc.]. pp. 147-155.
Zumdahl, S. S. (1989) Chemistry. Lexington, MA: D.C. Heath and Company.
313
-------
Advanced Identification (ADID) Techniques Used to Protect
Wetlands and Aquatic Resources in a Rapidly Growing County
Dennis W. Dreher
Northeastern Illinois Planning Commission
Chicago, Illinois 60606
Abstract
McHenry County, Illinois, approximately 40 miles northwest of Chicago, is one of the fastest growing counties in the
state. It also is home to many valued wetland and stream communities that are threatened by the impacts of new
development. Because of this, county government officials, with funding from the U.S. Environmental Protection Agency,
sought the assistance of resource experts from various local, regional, state, and federal agencies. Their task was to
assess the quality of the aquatic resource and to develop strategies for improved wetland protection. The project initially
involved the development of an up-to-date countywide inventory of wetlands, lakes, and streams. This inventory showed
that over 11% of the county was covered with wetlands and waterbodies. These aquatic resources then were evaluated
and rated based on the habitat, water quality, and stormwater storage functions that they provided. While only a small
fraction of the total number of wetlands — about 11% — of the county's wetlands were designated high quality, these
wetlands represented over 60% of the total wetland acreage. Inventory and assessment data were transferred to a
customized CD-ROM mapping tool to provide ready access to project information by resource managers, planners, and
local governments. The project team also developed a protection strategy for aquatic resources that was tied to the
results of the Advanced Identification (ADID) study. In particular, it identified four critical protection components:
improved education, regulations and best management practices, acquisition, and restoration. Though the ADID study
is only recently completed, there are strong indications that this protection strategy is being taken seriously by officials
in the county.
Background
McHenry County, Illinois lies approximately 40 miles northwest of Chicago along the Wisconsin border. Reflecting
a history of glacial activity, McHenry County possesses an abundance of wetland types in a variety of physical settings.
Predominant wetland types include palustrine, lacustrine, and riverine systems. Palustrine wetlands are found in a wide
variety of geographic settingsand terrains in the county and include marshes, bogs, fens, wet prairies, forested wetlands,
and ponds. Lacustrine wetlands are less common. They are found mostly in eastern portions of the county and are
exemplified by the wetlands of the Fox River - Chain 0' Lakes. High quality rivers and streams, and associated riverine
wetlands, are relatively common. In fact, McHenry County has some of the highest-quality stream ecosystems in Illinois,
as exemplified by the Kishwaukee River and its tributaries.
While predominantly rural, McHenry County is one of the fastest urbanizing counties in the state. From 1990 to 1998,
its population grew by 31.5% to 240,945. Population is forecast to grow to nearly 362,000 by 2020, an increase of an
additional 50%. This rapid population growth has raised concerns over possible adverse impacts on the county's
wetlands, lakes, and streams.
Historically, wetland and stream protection measures in McHenry County have included federal regulations, local
government ordinances, and acquisitions by government agencies, primarily the McHenry County Conservation District
and the Illinois Department of Natural Resources. However, with the recent rapid pace of urban development,
unacceptable loss and/or degradation of wetlands and aquatic resources have been observed. Concerned over the
possible environmental effects of rapid growth, the county board invited the U.S. Environmental Protection Agency
Region 5 (EPA) to perform an "advanced identification" (ADID) study of its wetlands and aquatic resources.
314
-------
An ADID project can have several objectives. One objective is to shorten permit processing, while providing
increased predictability to the Corps of Engineers regulations under Section 404 of the Clean Water Act. ADID also
provides information that can be used by state and local governments to aid in zoning, permitting, or land acquisition
decisions. Another objective of ADID is to provide information to agencies, landowners, and private citizens interested
in restoration or acquisition of aquatic sites.
Approach
The ADID study was initiated in 1995, under the coordination of the Northeastern Illinois Planning Commission
(NIPC). The study was a cooperative effort among federal, state, and local agencies to inventory, evaluate, and map
high-quality wetland and aquatic resources in the county. From the federal perspective, the primary purpose of this ADID
study was to designate wetlands or other waters of the United States that are unsuitable for discharge of dredged or fill
material. From the local perspective, the purposes of ADID were to improve the overall protection mechanism for
wetlands via improved local regulation, improved predictability in the permitting process, identification of potential
mitigation/restoration sites, and identification of potential sites for acquisition.
The scope of work for the ADID project included the following tasks:
. form a Technical Advisory Committee and a Planning and Policy Committee;
. develop ADID objectives for McHenry County and a strategy for protection and management;
. identify and map existing wetlands and aquatic resources;
. develop an evaluation methodology for identified functions of wetlands and aquatic resources;
. apply evaluation methodology utilizing Geographic Information System (GIS) technology and field inspection;
. map ADID sites for public review;
. develop a CD-ROM tool that contains both the project data and a customized GIS interface for display, query, and
mapping; and
. produce a final report and brochure and conduct a workshop for local governments, landowners, and consultants.
The Technical Advisory Committee and the Planning and Policy Committee were formed soon after initiation of the
project. There were two general purposes for these committees: 1) provide technical and policy assistance to NIPC and
EPA, and 2) provide a forum for educating local interest groups regarding the value of wetlands and aquatic resources
in the county.
The principal role of the Technical Advisory Committee was to advise project staff on scientific issues, particularly
the development of evaluation methodologies for wetlands, lakes, and streams. Technical committee members provided
expertise in wetland biology, soil science, hydrology, engineering, water quality, and computerized mapping. Technical
committee members contributed substantial time evaluating wetlands, both in the office and the field. The technical
committee consisted of the following invited agencies and organizations:
. U.S. Environmental Protection Agency, Region 5
. U.S. Army Corps of Engineers, Chicago District
. U.S. Department of the Interior, Fish and Wildlife Service, Chicago-Metro Wetlands Office
. USDA, Natural Resources Conservation Service
. Illinois Department of Natural Resources
315
-------
. McHenry County Department of Planning and Development
. McHenry County Conservation District
. McHenry County Soil and Water Conservation District
. Fox Waterway Agency
. Northeastern Illinois Planning Commission
The principal role of the Planning and Policy Committee was to advise project staff on policy, particularly the
determination of wetland functions important to McHenry County. The policy committee also provided advice on the
development of a wetland protection and management strategy. The policy committee included all of the members of
the technical committee as well as members of the following organizations:
. Homebuilders Association of Greater Chicago
. Illinois Audubon Society, McHenry County Chapter
. Illinois Environmental Protection Agency
. Land Foundation of McHenry County
. McHenry County Board
. McHenry County Defenders (a local environmental group)
. McHenry County Farm Bureau
. McHenry County Municipal Association
. McHenry County Realtors Association
• McHenry County Stormwater Committee
. Openlands Project (a regional open space advocate)
Developing an Aquatic Resource Inventory
Detailed inventories of wetlands, lakes, and streams were developed early in the project. Two principal existing
inventories were considered for identifying and mapping wetlands: the National Wetland Inventory (NWI) developed by
the US. Fish and Wildlife Service with the assistance of the Illinois Department of Conservation (1986) in the early
1980s, and an inventory by the Natural Resources Conservation Service (NRCS) that was being completed in McHenry
County just as the ADID project began. While neither inventory was adequate alone (the NWI was becoming dated and
the NRCS inventory, while more recent, and only a data set, focused principally on agricultural areas), in combination
they served as a good starting point. In finalizing the inventory, numerous revisionsand improvements were made based
on reviews of aerials photos, field checks, and the knowledge of local experts.
The resultant inventory identified 2,535 wetlands, including lakes, covering 37,846 acres. The inventory identified
an additional 1,250 farmedwetlands covering 3,839 acres. In total, there were 3,785 wetlands in all categories covering
41,685 acres, or nearly 11% of the county.
Lakes were identified as a subset of wetlands. Specifically, lakes were distinguished based on a criterion of 20 acres
or more of open water. Fifteen such lakes were identified (excluding gravel pits).
316
-------
Type
Wetland
rarmed Wetland
Lake
River
Total
Number
2,518
1,250
15
2
3,785 |
Acreage
33,003
3,839
3,584
1,259
41,685
Percent of
County
Area
8.4
1 .0
0.9
0.3
10.6
Streams were identified and mapped based on an inventory developed by the EPA. EPA's Stream Reach File,
Version 3 (1 :1 00,000 scale), with minor revisions, provided an accurate and complete inventory of county streams. The
inventory included over 570 miles of streams ranging in size from small, unnamed headwaters to large rivers like the Fox
and Kishwaukee.
Evaluation of the Functions and Quality of Aquatic Resources
As the first step in developing a wetland evaluation methodology, members of the policy committee were asked to
identify wetland and aquatic resource functions that were important to McHenry County. After considerable discussion,
the committee recommended that the following functions be considered and evaluated: biological/habitat functions, water
quality mitigation functions, stormwater storage functions, and groundwater functions. These functions then were
evaluated and refined by the technical committee. Ultimately, it was concluded that groundwater functions of wetlands,
while having important water supply implications, could not be evaluated because of insufficient data.
The project team and advisors then proceeded to develop evaluation criteria and methodologies for the following
general categories: biological/habitat functions and water quality/stormwater storage functions. The development of a
methodology for identifying high-functional-quality wetlands in McHenry County relied both on existing wetland evaluation
methodologies and the technical expertise of members of the technical advisory committee. The resultant methodology
builds on a methodology used in nearby Lake County, Illinois (Dreher, et al., 1992) as well as other documented
methodologies, particularly the Wetland Evaluation Technique (WET) manual (Adamus et al., 1987), the Oregon Method
(Roth et al., 1993), and the Minnesota manual (U.S. Army Corps of Engineers, 1988).
The methodology was designed to accomplish two objectives: 1) identify the functions that individual wetlands were
performing, and 2) identify wetlands of such high quality that they merit special consideration for protection strategies.
The evaluation of the identified functions for individual wetlands can be very complex and some of the referenced
methodologies describe fairly elaborate approaches to perform thorough evaluations. However, because of the large
number of wetlands to be considered in McHenry County, it was necessary to adopt a simpler evaluation procedure.
The resultant methodology isfullydocumented in the final project report, "Advanced Identification (ADID) Study, McHenry
County, Illinois" (NIPC et al., 1998). An overview of the evaluation criteria follows.
Biological functions include wildlife habitat, floristic diversity, stream aquatic habitat, and lake aquatic habitat.
Wetlands were considered high qualitytor this function if they met one of several criteria. These criteria included:
. the presence of threatened or endangered plant or animal species;
. designation in the Illinois or McHenry County Natural Areas Inventory (NAI);
. field evaluation as a grade A, B, or C wetland community following NAI methods;
317
-------
. streams with Index of Biotic Integrity (IBI) scores of 41 or greater;
. streams with high quality physical habitat; and
. healthy lake ecosystems with rich/diverse fish and plant communities.
Wafer quality/stormwater storage functions include shoreline and streambank stabilization, sediment and toxicant
retention, nutrient removal and transformation, and stormwater storage and hydrologic stabilization. In order to be
designated high functional value for water quality/stormwater functions, wetlands were required to meet three of the four
following criteria:
. presence of stabilizing vegetation adjacent to an open waterbody or perennial stream;
. surface area larger than five acres and having characteristics indicating the propensity for sediment/toxicant
retention;
. surface area larger than five acres, upstream of a lake or impoundment, and having characteristics indicating the
likelihood of nutrient removal/transformation; or
. surface area larger than five acres, at least 50% outside the floodplain, and having characteristics indicating
significant stormwater retention.
Alternatively, wetlands could be designated high funcf/ona/va/ueforwaterqualityfunctions if they provided individual
water quality functions adjacent to or upstream of wetlands, lakes, or streams that provide high quality habitat.
Individual wetlands and waterbodies were evaluated using a three-step procedure of CIS screening; aerial photo,
map or desk-top evaluation; and field evaluation (as needed). Based on this evaluation, it was determined that 154
wetlands totaling 17,489 acres, or about 42% of the county's entire wetland area, met the criteria for high-quality habitats.
Most of the high-quality wetlands tended to be large parcels, averaging 114 acres in size in comparison to the average
wetland size of 11 acres countywide. An additional 274 wetlands totaling 8,292 acres (average size of 30 acres) met
the criteria for high value for stormwater and water quality functions. Thus, while a relatively small number of wetlands
(about 11%) were designated high quality or high functional value, these wetlands represent over 60% of the total
wetland acreage.
Classification
High Quality
Habitat
High Functional
Value
High Quality Lake
Number
154
274
7
Percent of
all Wetlands
4.0
7.2
0.002
Acreage
17,489
8,292
1.346
Percent of
County Area
4.5
2.1
0.3
Percent of all
Wetland Area
42.0
19.9
3.2
Of the 15 inventoried lakes, seven were determined to be high quality. A total of 572 miles of stream were evaluated
and 170 miles (or nearly 30%) were designated high quality. Interestingly, high-quality stream segments were found on
18 different named streams and rivers scattered throughout the county.
Using ADID for Protection and Restoration
The ultimate measure of success for a project like the McHenry County ADID study is how it contributes to the
protection and restoration of aquatic resources. With this in mind, the project scope included a work element to develop
318
-------
a strategy for protection and management of aquatic resources. With the assistance of the advisory committees, the
project team developed a four-part strategy involving:
. improved education of local government officials, landowners, and the public;
. effective regulations and best management practices;
. expanded acquisition of aquatic sites and buffers; and
. restoration of degraded sites.
This strategy, which is described in detail in the project report (NIPC et al., 1998), is summarized below. Also
described are some recent protection and management activities, although it is still too early to judge the long-term
success of the project.
Improved Education: Educational initiatives are critical to improve awareness of wetlands and aquatic resources
among local citizens, land owners, and elected officials. Improved awareness can enhance local support for protection,
acquisition, and restoration programs.
. A 12-page brochure, McHenry County's Wetlands, Lakes, and Streams, was developed to educate the public and
local officials about the value of wetlands and aquatic resources in their communities. The brochure also
discussed the results of the ADID study and identified additional sources of information and agencies that can
provide help. Over 1000 copies of brochure have been distributed by participating agencies, such as the county
soil and water conservation district.
. Maps and information for all ADID sites were made available on a "user-friendly"CD-ROM. The CD-ROM includes
simplified mapping software developed from a sophisticated CIS tool. The software enables querying and
screening of various wetland characteristics at different geographic scales throughout the county. It also enables
printing out detailed information on individual wetlands. Over 100 copies of the CD-ROM have been provided to
local officials, consultants, and landowners in the county.
. The message of wetland, lake, and stream protection also is being carried to local officials and the public by
county-based environmental groups and consortiums called "ecosystem partnerships" that have been established
for the two main river watersheds in the county (the Fox and the Kishwaukee). ADID will be a useful tool in aiding
the efforts of these organizations.
Effective Regulations: Effective regulations are needed to minimize the effects of new development on aquatic
resources. Specifically, improved regulations are needed to fill in the gaps in existing federal, state, and local regulatory
programs. It was the conclusion of both the ADID team and the McHenry County Comprehensive Stormwater
Managemenf Plan (McHenry County, 1996) that improved regulations are needed to address concerns such as buffers
and setbacks, depressional storage volumes, pretreatment of stormwater runoff, and effective environmental mitigation
for unavoidable disturbances.
. Current federal regulations authorized under Section 404 of the Clean Water Act require a permit for the discharge
of dredged or fill material into wetlands or other waters of the United States. Federal guidelines also authorize the
EPA and the Corps of Engineers to identify in advance of specific permit requests, aquatic sites that will be
considered as areas generally unsuitable for disposal of dredged or fill material. The Chicago District of the Corps
has indicated that it will apply this discretionary authority to high-quality habitat and high- functional value sites in
McHenry County. The Corps also generally will require an individual permit (which allows public input) for
proposed modifications of ADID sites.
. Stream and wetland regulations, based on a model ordinance developed by NIPC, also have been adopted by a
number of local governments in the county. These regulations are intended to complement the federal regulations
by discouraging development in buffers and setbacks adjacent to wetlands, lakes, and streams and requiring pre-
319
-------
treatment of stormwater discharges. The City of Woodstock, the county seat, recently applied its wetland
protection regulations in a residential development review that resulted in an innovative conservation design
around a large wetland. Not only will the wetland be avoided, but the site design calls for clustering of homes and
buffers adjacent to wetland areas. Also, drainage swales and natural landscaping will be incorporated on upland
portions of the site to reduce hydrologic and water quality impacts of the development.
. ADID team members have worked closely with staff and consultants to the McHenry County Stormwater
Committee in the development of a countywide ordinance for new development. It has been recommended that
the countywide ordinance include provisions for stream and wetland protection that complement, but do not
duplicate, federal regulations. While the ordinance adoption process has been challenged by financial constraints
and political changes in the county, it appears likely that significant stream and wetland protections will be added
to existing county and municipal regulations.
Acquisition: Acquisition of important wetlands and stream corridors is one of the best ways to assure their long-term
protection. In fact, recent experience indicates that these areas are becoming high priorities for public land acquisition.
Information developed in the ADID study, particularly the identification of high-quality habitats and high-functional-quality
wetlands, will be valuable to land acquisition agencies, including park districts, the McHenry County Conservation
District, the Illinois Department of Natural Resources, and local land trusts, in assessing acquisition priorities. In a recent
example, the Plan Commission of Nunda Township in east-central McHenry County is developing a comprehensive land
use plan that will utilize ADID maps to identify areas to be preserved as open space.
Restoration: Restoration of degraded wetlands, lakes, and stream corridors, and ongoing management of higher
quality sites, are critical challenges for land management agencies. Management is needed to counteract the effects
of disturbances such as site fragmentation, elimination of fire, invasive species, and hydrologic alterations. Notably, the
McHenry County Conservation District has been a regional leader in restoring degraded streams and wetlands. The
ADID data base will be very useful in identifying appropriate sites to continue this restoration. The availability of GIS data
bases and mapping, particularly in conjunction with other digital data such as soils maps and data on seeps and springs,
will greatly facilitate this objective.
Lessons Learned
ADID was a valuable experience in McHenry County that generally met its identified objectives. In considering ADID
studies in other areas, there are several important lessons one can learn from the McHenry County experience.
1) Engage localgovernment sponsors and keep them informed throughout the project. The McHenry County ADID
began after the county board passed a resolution soliciting EPA's assistance. County staff and elected officials
were invited to participate on advisory committees. When support appeared to waver at critical points in the
process (e.g., staff changes and budget difficulties at the county), the project team reached out by convening
special meetings reminding county officials of the benefits of the project with respect to adopted county objectives.
2) Conduct an open study process involving both traditional supporters of stream and weflandprofecfion efforts and
potential adversaries. Groups ranging from environmental organizations to developers and the agricultural
community were invited to participate on advisory committees where issues and approaches were openly
discussed. When a public meeting was held to present project results, over 200 individuals attended. The vast
majority of those expressing opinions indicated support for ADID objectives and procedures, even though some
had concerns over the ramifications of federal wetland regulations.
3) Utilize the expertise and local know/edge of federal, state, and local resource agencies. While EPA contracted
with NIPC to coordinate the project, staff from numerous resource agencies contributed invaluable expertise in
hydrology, soils, aquatic ecology, and botany. They also contributed countless hours in evaluating field sites.
Scheduling such assistance from multiple agencies resulted in some time delays. However, without these
voluntary contributions, the project could not have been completed.
320
-------
4) Define wet/and and aquatic resource functions from a multi-objective perspective. While there is a tendency
sometimes to focus on just the habitat and recreational values of wetlands, lakes, and streams, it is important to
consider a broader range of benefits to maximize local buy-in to the process. The McHenry County ADID
specifically considered stormwater and water quality functions that were identified as being important in local
plans, such as the McHenry County Comprehensive Stormwater Management Plan.
5) Distribute end-products in user-friend/y formats. While ADID was a highly technical and complex project, efforts
were made to provide products that were readily understandable by local governments, land owners, consultants,
and the public. The product receiving the most interest was the CD-ROM containing ADID data, as well as a
user-friendly CIS-based interface for querying and mapping. The CD-ROM promises to be much more useful
than conventional paper maps.
6) Engage the local press in covering the project. Limited attempts were made to inform the local press during the
course of the study. While there was some resultant news coverage in local newspapers, particularly around the
time of the public meeting, this coverage was not particularly effective in informing the public about the benefits
of wetlands and the importance of the ADID study. Focused efforts, such as targeted press releases, probably
would have improved the frequency and quality of coverage.
Conclusions
The ADID study provides valuable information to advance the protection and restoration of wetlands and aquatic
resources in McHenry County. It can aid residents and organizations desiring to protect high-quality resources or restore
sites that have been degraded. It can inform landowners and developers about an appropriate course of action when
they are considering disturbances in or adjacent to high-quality sites.
While the final ADID products have been available for only a short time, it is apparent that they will greatly facilitate
ongoing efforts to educate county residents and officials, protect streams and wetlands from the effects of new
development, preserve sensitive stream corridors and wetlands as public land, and restore degraded sites. While the
ultimate success of county stream and wetland protection initiatives will depend on the will of landowners and local
government officials, no one will be able to blame wetland loss on inadequate information.
References
Adamus, P., E. Clairain, R. Smith, and R. Young. 1987. Wetland Evaluation Technique (WET). Department of the Army,
Waterways Experiment Station, Corps of Engineers, Vicksburg, Mississippi.
Dreher, D., S. Elston, and C. Schaal. 1992. Advanced Identification (ADID) Study, Lake County, Illinois. For U.S.
Environmental Protection Agency, Region 5.
McHenry County Stormwater Committee. 1996. McHenry County Comprehensive Stormwater Management Plan.
McHenry County, Illinois.
Northeastern Illinois Planning Commission, U.S. Fish and Wildlife Service - Chicago-Metro Wetlands Office, and U.S.
Environmental Protection Agency, Region 5. 1998. Advanced Identification (ADID) Study, McHenry County, Illinois.
Roth, E., R. Olsen, P. Snow, R. Sumner. 1993. Oregon Freshwater Wetland Assessment Methodology. Oregon Division
of State Lands.
U.S. Army Corps of Engineers, St. Paul District. 1988. The Minnesota Wetland Evaluation Methodology for the North
Central United States.
U.S. Department of the Interior, Fish and Wildlife Service, in cooperation with the Illinois Department of Conservation.
1986. National Wetlands Inventory.
321
-------
Local Government Involvement in Mitigation Banking
Lisa T. Morales
Wetlands Division
U.S. Environmental Protection Agency
Washington, D.C.
Abstract
Mitigation banking is a valuable tool that can be used by local officials to achieve wetlands restoration and other local
goals. Mitigation banks can be established by local governments to provide compensation for wetland losses that result
from development projects. There are different strategies that local governments can use to establish a mitigation bank,
depending upon their goals and objectives. The success of a mitigation bank is dependent upon several factors, ranging
from bank location to the availability of funding. The Environmental Protection Agency conducted a survey of local
jurisdictions to identify the different strategies that were utilized for effective development and implementation of
mitigation banking. The findings of the survey are presented in case-studies that characterize the approaches that were
used by local governments to achieve their mitigation banking goals and objectives.
Introduction
Mitigation banking is defined in the Federal Mitigation Banking Guidelines' as "wetland restoration, creation, or
enhancement for the purpose of compensating for unavoidable wetland losses in advance of authorized impacts to
similar resources." Under Section 404 of the Clean Water Act, applicants for permits must first avoid and minimize all
impacts to wetlands and other waters of the United States. If there are still impacts, then applicants must provide
compensatory mitigation through the restoration, creation, and the enhancement of similar type of aquatic resources.
This "sequencing process" under the Section 404(b)(l) guidelines is a central premise of the Section 404 regulatory
program, and mitigation banking can play a role in providing compensatory mitigation for unavoidable wetlands losses.
As a general matter, on-site and in-kind mitigation is preferred under the Section 404 Program. However, in those
circumstances where it is determined, on a case-by-case basis, not to be practicable, then off-site, in-kind mitigation is
acceptable.' Off-site mitigation can be accomplished using a federally-approved mitigation bank. Since the use of
mitigation banking to offset permitted wetlands losses began in earnest in the early 1990's, local governments have been
involved in developing banks to restore and replace lost wetlands functions and values within their jurisdictions. By
simplifying the process for compensating for unavoidable wetlands losses, appropriate use of the banking concept can
improve both permitting efficiency and environmental protection.
1 Federal Guidance for the Establishment, Use and Operation of Mitigation Banks; Notice. Federal Register, Volume 60, No. 228, pages 58605-58614,
November 28, 1995.
2 Memorandum of Agreement between the EPA and the Department of the Army Concerning the Determination of Mitigation Under the Clean Water
Act Section 404(b)(l) Guidelines (February 6, 1990)
322
-------
Mitigation Banking Objectives
• Local needs are best met when there are clear objectives and goals for establishing a mitigation bank. Examples
of objectives for mitigation banking identified by local governments include:
• Use as part of a comprehensive watershed plan that addresses urban development and the need for preservation
and restoration of wetlands
• Provide an incentive for economic growth by streamlining the process for providing compensation of unavoidable
wetlands impacts
. Provide for restoration of degraded wetlands that otherwise might not be improved because of insufficient funding
. Use as part of a multi-objective strategy to manage stormwater, flooding, water quality, etc.
• Compensate for wetland losses from future local agency projects
Types of Mitigation Banks
Once the objectives for the proposed bank have been established by the local agency, the type of bank to meet the
identified objectives must be determined. Local governments can establish a mitigation bank for their individual use or
for credit sale as a commercial venture. The bank can be established solely by a public agency or as a joint venture with
a private entity (e.g., entrepreneurial business). Mitigation bank types need not be mutually exclusive; for example a
commercial mitigation bank may be established through public/private partnerships and still be part of a watershed
management plan.
Commercialbanks: A commercial mitigation bank is one in which the credits are sold to a party other than the bank
sponsor (banker). The banker sells credits within the established service area to permittees who have approval from
the U.S. Army Corps of Engineers (Corps); the agency responsible for issuing wetlands development permits; to
compensate for wetland impacts through a mitigation bank.
Sing/e-user: A single-user mitigation bank is established by a local agency to compensate only for wetland losses
associated with activities conducted by the agency.
Factors Contributing to Mitigation Bank Success
Conditions for successful bank establishment
Local governments which have successfully established a mitigation bank have identified several conditions that
need to be considered in order to successfully meet environmental and/or economic objectives.
• There must be a demand for compensatory mitigation within the local jurisdiction. Demand results from
development pressure in a rapidly growing area where impacts to wetlands are expected to occur. The value of
development in regions with rapid growth increases the willingness of the public agency or developer to pay for
wetland mitigation. Potential bank sponsors should conduct a "needs" analysis to determine the demand for a
mitigation bank in a given area. The analysis will show the extent of potential wetland impacts in the region and
whether mitigation banking is a viable compensation option. Once demand for a bank is decided upon, the size
of the bank can be determined.
323
-------
• There must be a sufficient supply of available sites for restoration, creation, enhancement, or preservation.3 The
availability of appropriate sites will vary by geographic region. For example in Florida, large tracts of degraded
wetlands have been impacted by previous land-use practices and by invasion of exotic plant species. Such large
areas are conducive to restoration and enhancement. In some areas, there are mitigation banks where property
is in high demand, resulting in high land acquisition costs. In other circumstances, finding an appropriate site
may be difficult due to the lack of large wetland expanses. Land that is in public ownership can help lessen land
acquisition cost, but may not have favorable physical attributes that would allow for a mitigation bank to be self-
sustaining over time.
• Finally, regulatory coordination is important. The local agency needs to provide a prospectus to the Mitigation
Banking Review Team (MBRT) established by the local Corps district.4 The prospectus will serve as the mitigation
banking instrument that identifies the objectives and administration of the proposed bank. As pat-t of this
coordination process, the local agency should identify the proposed bank site, the geographic service area,
wetland types suitable for compensation at the site, the debiting and crediting system, performance standards,
monitoring plan, contingency and remedial actions, and provisions for long-term management and maintenance.
4 Proposing a mitigation bank in the context of a regional plan that integrates the bank into a comprehensive
wetland or watershed management strategy may improve the likelihood of acceptance of the prospectus by
the MBRT. A watershed management plan can provide greater certainty about the nature and extent of future
wetland impacts and identify the most appropriate, environmentally beneficial options for offsetting the
anticipated impacts. In this way, the MBRT has a level of assurance that the bank is a part of a broad goal
to maintain or gain wetland functions in a given area or watershed.
Site selection criteria
Local agencies need to carefully consider the sites that are identified for potential mitigation banks. Site selection
is one of the most important criteria affecting the successful establishment of a mitigation bank. The most significant site
selection criteria is the potential that the site can be successfully restored or enhanced in a manner that will provide
appropriate compensation for anticipated unavoidable wetlands losses. Selection of such ecologically important or
desirable sites can further regulatory acceptance.
As identified in the federal guidance on mitigation banking, mitigation sites should be self-sustaining overtime. Sites
with naturally occurring hydrology are preferable to sites that require complex hydraulic engineering features that are
costly to develop, operate, and maintain. Therefore, sites that can be restored without complex improvements should
be the first option when establishing a mitigation bank.
Ideally, site selection would be undertaken in accordance with a watershed management plan under which existing
wetlands have been surveyed to determine which sites are the highest priority for protection, the most suited for
restoration or enhancement, and the most likely to be impacted by development. Through watershed planning, wetland
functions that are lacking in a region can also be used to guide site selection. Additionally, planning at this level can help
determine compensation requirements, because anticipated wetland impacts can also be identified. Two federal
programs, the Special Area Management Plan (SAMP) and the Advance Identification (ADID) Program, can provide
guidance and technical assistance to local sponsoring agencies that meet certain criteria and are interested in
establishing a watershed scale planning effort.
The Federal Mitigation Banking Guidance states that the use of preservation in a mitigation bank may be authorized by the federal agencies when
it is demonstrated that the preserved areas contribute to the functions of the restored, created or enhanced aquatic resource.
* The Mitigation Banking Review Team is established as consistent with the Federal Mitigation Banking Guidance.
324
-------
Mitigation bank site selection can also be targeted to address specific environmental objectives. For example,
wetlands in several counties in Florida were overrun by exotic vegetation. Mitigation banking provided the funds to
improve the wetland habitat through exotic species eradication and revegetation with native plants.
Site selection may also be influenced by a regional demand for mitigation opportunities and the availability of suitable
restoration sites. Land acquisition is often cost-prohibitive for local agencies: land ownership, zoning restrictions, and
allowable uses also are important factors in determining a mitigation bank site. Important functional characteristics of
a potential restoration site include the presence of hydrology, hydric vegetation, and/or hydric soils; size of the site;
historic conditions; and the degraded condition of existing wetland (e.g., exotics, fill, compaction, trash). Experts in
restoration should be consulted regarding specific wetland requirements and regional wetland attributes prior to site
selection.
Funding
The most common funding option used by local governments is the advance-credit sale option addressed in the
Federal Mitigation Banking Guidance.With this funding mechanism, advance sale of a percentage of the total credits is
allowed after certain criteria have been met by the bank sponsor, such as conservation easement, land acquisition,
and/or design plan approval. By selling a limited number of credits in advance, the banker can collect sufficient funds
to begin conducting wetland improvements, then sell more credits as they are certified by the regulatory agency. The
advance sale of credits, however, involves a degree of risk; for example, problems can arise prior to mitigation bank
implementation or completion, leaving the bank sponsor liable for any credits sold.
Local governments have also identified various other strategies for funding the establishment of their mitigation
banks. Some of the options include:
. Completing the project in phases so that initial credit sales will provide funding for the remainder of the project
. "Borrowing" money from other local agency funds, then paying it back using money from credit sales
. Using available federal or state grant money for wetland improvement as seed money to establish the banks;5
. Partnering with a private company to share the costs
• Using combinations of the above mechanisms
Bank Administration and Operation
Several issues related to bank administration and operation pose a challenge to local governments. The geographic
service area, credit ratios, credit valuation, monitoring plans, and long-term maintenance provisions are all issues that
must be addressed by the local sponsor and approved by the signatory agencies. Information on these issues can be
found in the Federal Mitigation Banking Guidance mentioned earlier, as well as in any existing state guidance. A
partnership with a private entity that has experience in mitigation banking is one alternative for addressing bank
administration and operation challenges.
The geographic service area for the mitigation bank should be established and presented in the mitigation banking
planning document. The service area is the area in which credits from a mitigation bank can be used to compensate
for unavoidable wetland losses. This area can be a watershed or a political boundary, such as a county or municipality.
The number of credits produced by a mitigation bank are generally determined by two factors, (1) the number of
acres restored, created, enhanced, or, in exceptional circumstances, preserved, and (2) a quantified evaluation of the
5 EPA has also identified the State Revolving Fund (SRF) as another potential source for funding the establishment of mitigation banks, subject to
approval under each State's SRF regulations.
325
-------
functional value of the wetlands in the bank. As a general matter, the greater the wetland improvement, the more credits
generated. For example, more credits may be associated with restoration and fewer credits associated with
enhancement. The second factor involves a functional assessment of the improved or created wetland, and a method
for converting the functional units to credits.
To determine the appropriate compensation requirements for wetland losses, a mitigation ratio must be developed.
This ratio can be based on lost wetland function and area, the rarity of the affected wetland, and the wetland replacement
kind and function. The ratio will determine the number of credits required from the bank to replace the impacted aquatic
resource. Generally, the mitigation ratio is greater than or equal to one; i.e., at least one credit (acre of restored,
enhanced, or created wetlands) required to replace one acre of lost wetland resource.
As a general matter, the bank sponsor monitors the bank for a period of time (e.g., five years) to determine whether
the bank is functioning at the level required by a previously determined set of performance standards established in the
banking instrument. After the performance standards are met, the local agency may choose to transfer the mitigation
bank to another entity (e.g., land trust) for the long-term maintenance and monitoring. An agency can also use volunteer
labor to offset costs of monitoring and maintenance and to improve public awareness and citizen stewardship for the
project. Among the local bank projects reviewed by EPA in the survey, local sponsors were usually responsible/liable
for the performance of the mitigation site from the beginning, or after an established period of time if the project was a
joint-venture.
Financial assurances for long-term maintenance and contingency plans for the bank most often take the form of
additional fees added onto the cost of a credit, with a fixed amount from the sale of each credit put into an escrow
account. The local agency then draws from the account for necessary maintenance expenditures. In some cases,
completed banks become part of a park system or are turned over to a public agency for long-term management.
Mitigation Bank Approaches with Case-Study Examples
The following case-studies are examples of the different types of mitigation banks that can be established by a local
agency, and highlight specific issues that are of concern to local governments. The banks discussed in the case-studies
are not mutually exclusive, in that a given bank may fit into more than one "bank type" category.
Mitigation Banking in the Context of a Watershed or Wet/and Management Plan
Local governments can use a watershed or wetland management plan as a means of addressing economic
development and environmental concerns. An important aspect of such a plan involves compiling an inventory of
existing wetlands in order to determine both coverage area and functions provided. This information can then be used
to protect important resources, establish areas suitable for development, and determine the best sites for restoration.
Mitigation banking is, therefore, a tool that can be integrated into a watershed plan to help meet resource management
objectives.
Advantages of mitigation banking in the context of a watershed management plan are as follows:
• Sites to be restored and critical sites to be preserved are already identified in the plan
. Market demand for compensatory mitigation can be identified
• The extent and nature of potential impacts have been determined
. Likelihood of restoration project success is improved
• Uncertainty for regulatory agencies is lowered
. Multiple objectives can be attained
326
-------
• Environmental benefits are maximize
West Eugene Mitigation Banking Program
Overview: The Eugene Wetland Mitigation Bank Program (Bank) was established in 1995. The program is operated
by the City of Eugene, Oregon, under a separate fund within the City's financial structure. The goal of the Bank is to
provide a mechanism to fund wetland mitigation projects, carry out the West Eugene Wetlands Plan (Plan) adopted in
1992, and meet other community needs. This program is being conducted in cooperation with the City's wetland partners
(the U.S. Bureau of Land Management [BLM] and the Nature Conservancy [TNC]).6 The objectives that the City seeks
to accomplish with the mitigation banking program include:
• Creating credits in advance of wetland losses
• Meeting the mitigation needs of the community of Eugene
• Achieving community objectives, such as increased flood storage capacity, enhanced water quality, improved
wildlife habitat, and establishment of education and recreation opportunities
. Targeting areas with high prospects for restoration success such as historic wetlands, disturbed agricultural
wetlands, and areas adjacent to waterways
. Communicating the banking program's value to the community
• Operating the bank as part of a national model wetland program in cooperation with the wetland partners
. Establishing a permitting process familiar to businesses, environmental interests, and regulatory agencies
Funding for identifying wetland areas for the bank was provided through EPA's Advanced Identification (ADID)
program. The West Eugene Wetlands Plan integrates wetlands protection and community development needs by
identifying areas best suited for wetland preservation and areas with development potential. Thirteen hundred acres of
wetlands were identified within the plan area, with more than 1,000 acres designated for protection or restoration and
approximately 300 acres of lower value wetland designated as suitable for future fill and development. The Plan calls
for the establishment of a mitigation banking program to compensate for unavoidable losses of wetlands through
restoration and enhancement in conjunction with protection of important wetland resources.
All mitigation bank projects will be located within the Long Tom River watershed. Seven projects were initially
constructed for a total of 56 acres. Three additional projects totaling 60 acres were in planning at the time of the EPA
survey.
In this program, mitigation bank credits can be established in three ways
1. The City or its partners may undertake wetland mitigation work, then seek certification of credits by the Corps and
Oregon Division of State Lands (DSL). The City provides documentation on a site prior to the mitigation work and
after completing improvements to demonstrate an increase in wetland values.
2. The City may create more credits than are needed to compensate for permitted wetland losses as part of
concurrent or advance mitigation work. If the Corps and DSL certify the excess credits, they will be added to the
Mitigation Banking Program's ledger for sale.
3. Uncertified credits may be sold in order to fund initial mitigation work, which are later certified by the Corps and
DSL when the initial hydrological and vegetative work is completed.
6 City of Eugene, Public Works Department. 1997. West Eugene Wetland Mitigation Bank, Annual Report 1996.
327
-------
The City Stormwater Fund provided money for acquisition of site properties. Additional capital for the projects came
from the sale of uncertified credits. No uncertified credit can be sold unless the Corps and DSL have approved a
mitigation plan.
As the bank sponsor, the City is responsible for monitoring and maintaining the bank sites. Monitoring is required
for a minimum of five years and the sites will be inspected by the Corps and DSL on an annual basis. Monitoring goals
are determined on a site-by-site basis. Maintenance will be conducted to ensure that the wetland functions and values
are fully established and functioning. If the site is not meeting the performance criteria, corrective measures will be
taken. Funding for these tasks is included in the credit price. At least an additional 20% will be added to the estimated
credit cost and set aside to be used to monitor, maintain and, where necessary, conduct remedial measures.
Single-user Mitigation Bank
Local governments can establish a single-user bank. In this case, the local government initiates the bank, creates
the credits, and is the principal credit user. Local sponsors with long-range project plans that involve potential impacts
to wetland areas may establish a mitigation bank or banking program in anticipation of the need for compensatory
mitigation. For example, a public works department with road expansion plans may know in advance that there will be
unavoidable impacts and provide the funds necessary to initiate a bank in anticipation of the project.
An advantage of a single-user bank is that long-term planning provides advance knowledge of potential wetland
losses associated with future projects. This reduces the uncertainty surrounding the demand for the bank and eventual
funding. Funding may be available to initiate the bank based on the predicted future use.
Snohomish County Airport Mitigation Banking Program
Overview. The Snohomish County Airport (SCA) needed to address the multiple objectives of their 20-year Master Plan,
the aviation needs of the airport, the economic growth of the region, and environmental protection. The Master Plan
identified three development/construction projects that will occur over five years, each with anticipated wetland impacts.
The objectives that the SCA seeks to accomplish through the banking program include:
. Creating a mitigation alternative for projected airport development
. Replacing or augmenting wetland functions in the watershed
• Replacing habitats that will potentially be lost, including open-water habitat that cannot be created on-site
• Creating recreational and educational opportunities
The SCA analyzed the three watersheds containing the project areas to determine missing wetland functions and
values. Potentially impacted wetlands were then characterized according to acreage, wetland category, and function.
Based on these analyses, two mitigation banks were designed to replace the wetland functions that would be lost by the
airport projects. The SCA banking program creates a "reserve" of mitigation that accommodates Master Plan impacts
for approximately seven years.7
Mitigation banking was an effective alternative for the SCA since the Federal Aviation Administration (FAA) prohibits
the creation of bird attracting habitat within 10,000 feet of a runway. In this circumstance, on-site mitigation was
determined not to be a viable alternative as it would attract waterfowl and migratory birds. The SCA created or enhanced
a total of 32 acres and preserved an additional 23 acres.
Following a format developed by the Washington State Department of Transportation, the SCA Memorandum of
Agreement (MOA) was established. The MOA describes the procedures for selecting, managing, monitoring, and
Snohomish County Airport. 1997. Snohomish County Airport MOA/IM Executive Summary.
328
-------
protecting mitigation banks. Additionally, an Implementation Manual was created by the SCA that presents agency
recommendations for approval of the two specific bank sites. A technical oversight committee reviewed and advised
changes to the MOA, and provides input for credit evaluation and ratio development.
Construction is nearly complete at both sites. One site has a nonprofit organization associated with it called the
Friends ofNarbeck Wetland Sanctuary thai is working to promote citizen stewardship, protection of wildlife and wildlife
habitat, native plant propagation, and environmental research.
Primary funding for the mitigation banks came from airport revenues. In addition, some funds were provided from
the FAA, since the program will be used to mitigate for projects required by FAA safety regulations.
The SCA will be responsible for one of the sites in perpetuity. The other site will be donated to the City of Everett,
Washington, Parks Department after all of the credits have been used. Monitoring will be conducted for five years in the
emergent wetland and ten years in the forested wetland. Both sites have been placed in a conservation easement and
a contingency plan will be executed if performance criteria are not met.
Public/Private Joint Venture
Public/private joint ventures can facilitate mitigation-bankestablishment by providing investment capital and technical
expertise. Mitigation banking is a complex process involving aspects of design, engineering, regulation, project
management, biology, ecology, marketing, and long-term management. Because of this, a mitigation banking joint
venture may be established to share the responsibilities and costs. A local agency role may be bank sponsor, bank
creator, land owner, long-term manager, or all four. The private entity may conduct only the technical work, but can also
assume any of the aforementioned roles. The EPA survey found that the local agencies partnered in joint ventures were
most likely to provide the land for the mitigation bank and the private entity would provide funds and/or technical
expertise.
City of Pembroke Pines and Florida Wetlandsbank, Inc.
Overview: In 1992, the City of Pembroke Pines in Broward County, Florida, entered into a partnership agreement
with Florida Wetlandsbank, Incorporated to restore a heavily degraded site that was already overrun by exotic species
and becoming further deteriorated by all-terrain vehicle use and illegal trash disposal. The City did not have the
resources necessary to restore the site or to establish a mitigation bank. Concurrently, development pressures in the
area were creating a demand for an effective mitigation alternative.' Florida Wetlandsbank agreed to design and
construct a wetland system by eradicating the exotics and replacing them with a mixture of 10 typical Everglade habitats
including cypress stands, emergent marshes, tree islands, and sawgrass prairie. Florida Wetlandsbank also provided
the management for the bank, selling the first credits in 1994. In 1997, restoration was near completion, with 396 credits
sold. The objectives for the Florida Wetlandsbank (the name of the mitigation bank) are to:
. Provide a mitigation alternative for the rapidly urbanizing City of Pembroke Pines
. Provide mitigation at a large site with high potential for ecological success
. Restore native wetland ecosystems in an area that was degraded and infested with exotic species
The mitigation bank provided a number of economic and environmental benefits, including the following.
8 Paul Wattles, Assistant City Manager, City of Pembroke Pines, Florida. Personal communication, March 23, 1998.
329
-------
• The project provided revenue for the City through the collection of franchise fees per acre (paid by Florida
Wetlandsbank, Incorporated after credits were sold).
• Developers used the credits as quickly as they became available.
. The City plans to open part of the area as a park and offer public access and passive recreation opportunities.
. To date, the site is ecologically successful, with little reinfestation of exotic plants.
. Florida Wetlandsbank, Incorporated profited from the sale of credits.
The construction is completed and all credits sold for the original 350-acre site. An additional 100 acres has been
added to the site with construction nearly completed and credit sales pending. The service area for the mitigation bank
was kept at the county level rather than the basin level, which incorporates three counties. The County determined that
mitigation for wetland losses should remain within the county.9 The number of credits available in the mitigation bank
were determined using a modified Wetland Rapid Assessment Procedure (WRAP).
Florida Wetlandsbank, Incorporated provided the capital for initiating the bank. Pre-project sale of credits allowed
restoration work to begin. A trust fund was established to provide for the long-term maintenance of the site, with $1,000
per credit contributed to the fund.
Site monitoring will be conducted by Florida Wetlandsbank, Incorporated from the time that construction is completed
for five years. Quarterly monitoring reports are required by the SFWMD. These reports contain information regarding
the banks performance criteria. A performance bond was placed with the SFWMD for each stage of construction. The
bond will be used in the event that any of the performance criteria are not met. Once the performance criteria for the
site are met, the money will be released to Florida Wetlandsbank.
Commercial Mitigation Banking
A local government can establish a commercial mitigation bank that sells credits to anyone who meets the
requirements to use the bank for compensatory mitigation. The bank discussed here was established on public land,
solely by public agencies, with credits for sale to the public. Optimally, a public commercial mitigation bank is part of a
larger watershed plan.
A major advantage of public commercial banks is that they are in public ownership, which can better provide long-
term management, and the funds generated from the bank can be used to further improve public resources. The
disadvantage is that the up-front capital investment can be difficult for public agencies with limited availability of funds
DuPage County, Illinois
Overview DuPage County is a highly urbanized area located west of Chicago. Land development has negatively
impacted the natural drainage system of the area by eliminating naturally occurring storage, reducing stormwater
infiltration, and increasing the velocity and quantity of runoff. In response, the DuPage County Stormwater Management
Plan (Management Plan) was developed in 1989, to reduce the potential for recurrent and increasing flood damage and
to reduce further environmental degradation associated with development. The ecological assessment conducted during
development of the Management Plan found that wetlands represented a significant portion of the natural watershed
9 Desmond Duke, Project Administrator, Florida Wetlandsbank. Personal communication, June 19, 1998.
330
-------
storage in the county, and are essential for adequate stormwater retention, conveyance, sediment control, and water
quality enhancement."
Individual watershed plans, within the larger DuPage County management plan, were developed to identify wetlands
for protection, enhancement, and restoration. The County attempts to establish mitigation banks in the four main
watersheds within the County. The service area for each bank will be the watershed in which the bank is located. The
County also established an ordinance for stormwater management that requires developers whose development
proposals will affect the function and values of wetlands to consider mitigation banking as an alternative to compensatory
on-site mitigation. The objectives of the DuPage County commercial mitigation banking program are to:
• Manage and mitigate the effects of urbanization on stormwater drainage
. Enhance the quality, quantity, and availability of surface and groundwater resources and prevent further
degradation
. Preserve and enhance existing wetlands, aquatic, and riparian environments
• Encourage restoration of degraded areas
Any investment in a bank must be at least equal to the cost of planning, acquiring lands, constructing, and operating
and maintaining mitigated wetlands of equivalent or greater functional value than those lost to development.
The land for the bank sites was in public ownership. Funding for establishing the bank and initial project work came
from an advanced credit sale. One-third of the credits were allowed for sale prior to work beginning at the site. One site,
Winfield Creek, was in the process of selling credits for project initiation when public opposition to the site halted the
project. The County is now in the process of finding another site but, in the interim, is liable for the credits sold.
The County has five banks that have been approved by the agencies and are in various stages of implementation.
The oldest project, Cricket Creek, is in year two and, to date, is exhibiting successful hydrologic and vegetation
conditions.
Conclusion
The use of mitigation banking by local governments can be an effective tool to restore and protect their community's
valuable wetlands resources. The case-studies presented provide local governments with different mitigation banking
strategies that could be used to address the needs of their community. As demonstrated in the case-studies, the
establishment of sound objectives and goals by the local agency will help determine the type of mitigation bank that will
best meet the local needs. In addition, there are several conditions that local governments must evaluate and consider
for successful bank implementation, including (1) a demand for compensatory mitigation, (2) a sufficient supply of
suitable sites for the bank, and (3) opportunities for working in partnership with the federal and state regulatory and
resource agencies.
Mitigation banks that are established by local governments can address more than just the need for compensatory
mitigation. Wetlands mitigation banks can achieve additional community needs by increasing local flood storage
capacity, improving wildlife habitat, and providing educational and recreational benefits while restoring and enhancing
important wetland resources.
10 DuPage County Stormwater Management Committee. 1989. DuPage County Stormwater Management Plan. DuPage County, Illinois.
331
-------
Massachusetts Stormwater Management Policy/Regulations:
Development, Implementation, and Refinement
Bethany Eisenberg
Vanasse Hangen Brustlin, Incorporated
Watertown, Massachusetts
Abstract
In March of 1996, the Massachusetts Department of Environmental Protection, in conjunction with the Massachusetts
Office of Coastal Zone Management, released the Draft Version of the State's Stormwater Management Policy. The Policy
includes nine specific Stormwater Performance Standards, which are to be met to achieve compliance. The Policy presented
in two volumes: Volume One, the Stormwater Policy Handbook, which contains a description of the policy, its implementation,
and descriptions of the nine individual Stormwater management standards, and Volume Two, the Stormwater Technical
Handbook, which contains more detailed information on Best Management Practices (BMPs), for Stormwater management
(i.e., detention basins, swales, etc.).
The policy is the result of three years of work by a Stormwater Advisory Committee that included representatives from
regulatory offices (EPA, Department of Fisheries and Wildlife, Natural Resources Conservation Service, etc.), engineers and
developers from the private sector, the highway department, and representatives of local conservation commissions. The Policy
is to be implemented as an amendment to the existing Massachusetts Wetlands Protection Act, which is administered at the
local level by local Conservation Commissions.
The nine performance standards are the key components of the policy. General descriptions of the standards include the
following:
1. No new Stormwater conveyances may discharge untreated Stormwater directly to, or cause erosion in wetlands or
waters of the Commonwealth.
2. Post-development peak discharge rates may not exceed pre-development rates.
3. Maximize recharge to groundwater: post-development must be similar to pre-development conditions.
4. Remove 80% of average annual load-post development of Total Suspended Solids (TSS).
5. Use specific BMPs for discharges from areas with higher potential pollutant loads; untreated infiltration prohibited.
6. Use specific BMPs for discharges to critical areas.
7. Redevelopment projects should not meet performance standards to the maximum extent practicable and at a minimum,
improve existing conditions.
8. Erosion and sedimentation controls are required during construction.
9. Operation and Maintenance Plan for Stormwater Management required.
The Policy was introduced in March of 1996 for testing its effectiveness. Two Phases of training were provided over a two-
year period across the state. The first phase focused on introducing the Policy and the second phase focused on detailed
Engineering Companies, and local DPWs and planning departments.
332
-------
Introduction
In March of 1997, the Massachusetts Department of Environmental Protection (MADEP), in conjunction with the
Massachusetts Office of Coastal Zone Management (CZM), released a draft version of the state's Stormwater Management
Policy (herein referred to as the "Policy"). Prior to the development of this policy, the control of peak stormwater discharges from
development sites to prevent flooding and erosion problems was a fairly standard requirement across the country; and was
well-implemented in Massachusetts. However, there were no state-level requirements for stormwater quality treatment,
maintaining groundwater recharge processes, or maintaining stormwater treatment systems. The Policy was developed to
provide standard minimum requirements for stormwater management that could be consistently implemented on development
projects.
The Stormwater Management Policy is currently an amendment to the Wetlands Protection Act which is only applicable
when a project proposes work within the boundary, or buffer zone, of a Wetland Resource Area. Hence, the Policy is not
applicable to all developmental projects.
While the regulatory implementation of the Policy is through local Conservation Commissions, under the State's Wetlands
Protection Act, the performance standards and design guidelines that define the Policy were developed for use by a larger
audience. Development teams (typically the proponent and their engineers, architects, and planners) and the various reviewing
agencies (Conservation Commissions, Planning Boards, DPWs, etc.) were expected to be users of the Policy. The goal was
to provide guidance to ensure that negative impacts from stormwater runoff generated as a result of urban and suburban
development would be minimized without placing an unjustifiable economic burden on developers for new projects, or
preventing redevelopment of existing sites.
The Policy includes nine specific Stormwater Performance Standards for which compliance must be achieved on
development projects. Included in these standards is a requirement to provide groundwater recharge, requirements for ensuring
proper stormwater treatment prior to discharge to waters of the state, and provisions for waiving certain requirements if deemed
infeasible for redevelopment projects. The Policy has been distributed as an Interim Draft to allow for refinements prior to its
final promulgation as state regulations. In orderto fully gage its effectiveness, however, the Interim Policy has been distributed
and implemented as if the regulations were final.
The Policy is presented in two handbooks: Volumes One and Two. Volume One, the Stormwater Policy Handbook,
contains a description of the policy, its implementation, and detailed definitions of the nine individual stormwater management
standards. Volume Two, the Stormwater Technical Handbook, contains detailed information on Best Management Practices
(BMPs), with guidelines for the design of standard stormwater management structures (such as detention basins and water
quality swales).
Development
The Policy is a result of three years of work by a State Stormwater Advisory Committee. In addition to leaders from MADEP
and CZM, this Committee included representatives from such regulatory offices as USEPA, Department of Fisheries and
Wildlife, and the Natural Resources Conservation Service. Also on the committee were engineers and developers from the
private sector (including the author) and representatives from the Massachusetts Highway Department and local Conservation
Commissions.
The goal of the Committee was to provide a cohesive set of performance standards that addressed key issues associated
with stormwater runoff control. The Policy was developed in such a manner that it provides adequate accompanying guidelines
and recommendations, to allow for consistent implementation, while still allowing for flexibility in site-specific designs. Given
the widely varying goals and perspectives of Committee members, the Policy published in 1997 was a result of group
consensus and compromise.
The Committee was divided into two sub-committees, the Policy group, and the Technical sub-committee. The Policy group
was tasked with developing the process for legal implementation and the Technical Committee was responsible for developing
the specific technical requirements of the Stormwater Management Policy.
333
-------
Some of the performance standards introduced completely new requirements for development projects and required
detailed evaluation and discussion during the development stages. Some of the key issues that generated substantial debate
include the following:
• Recharge:~t\\Q loss of recharge to groundwater systems, which provide drinking water supplies and generate baseflow
to streams and rivers, was a state-wide problem. A mechanism for requiring the maintenance of recharge after
development was considered a high priority. The biggest issue relative to this requirement was: How much annual
recharge should be required?
. Water Quality Treatment Volume: The quality and quantity of stormwater runoff from paved and unpaved areas can
vary greatly. It was determined by the Committee that runoff from impervious areas was the highest concern and should
require treatment. The largest decision relative to this concern was: What volume of stormwater runoff should be treated
for water quality? Should it include runoff from both pervious and impervious areas?
. Critical iinniffi.B The Committee felt that certain sensitive environmental areas should have the maximum practicable
protection. Under the Massachusetts Stormwater Management Policy, "Critical areas" are defined as; shellfish growing
areas, public swimming beaches, cold water fisheries, recharge areas for public water supplies and designated
Outstanding Resource Waters (ORWs). ORWs are further defined to include surface drinking water supplies, certified
vernal pools, and state designated Areas of Critical Environmental Concern (ACECs). The issue here was how to ensure
that these areas are provided adequate protection and how to define what adequate protection is.
• Exemptions: Some Committee members expressed concern that stormwater management requirements may be too
costly for small residential projects, or may be a deterrent for initiating redevelopment projects. The issue was what, if
any, projects should be exempt from any or all of the standards?
• Operation/Maintenance: Maintenance of stormwater management practices is critical for their effectiveness. It is often
difficult, if not impossible, to ensure that the operation and maintenance of BMPs will occur as necessary. The issue was:
How can the necessary maintenance of BMPs implemented on the project be ensured?
A brief summary of the decisions on these key issues is as follows:
. Recharge: Annual recharge processes are permanently changed by the introduction of impervious areas to a site. In
order to minimize this impact, it was agreed that the existing annual recharge should be determined and post-
development annual recharge should maintain this to the maximum extent practicable. A preliminary methodology for
determining existing recharge on the site was provided in the Draft Policy.
• Water Quality Treatment Volume:
• For discharges to critical environmental areas (defined in the Policy), the volume of stormwater runoff to be treated
for water quality control is defined as 1 .0 inch of runoff times the total impervious area of the post-development project
site.
• For all other discharges, the volume to be treated is calculated as 0.5 inches of runoff times the total impervious area
of the post-development project site.
These volumes represent total runoff from the smaller, more frequent storms that occur annually, and the initial
volumes of runoff from larger more infrequent storm events. The goal behind this decision was to fully treat the runoff
from the majority of the storm events occurring annually, without approaching values where treatment system sizes
would result in increasing costs with decreasing additional benefit.
• Critical Areas: It was decided, as stated above, that 1 .0 inch of runoff (as opposed to 0.5 inch) must be treated for
water quality if the discharge is to a Critical Resource Area. Also, specific approved BMPs are recommended for use
in particular critical areas. In addition, it was decided that spill prevention/containment methods must be included in
the Stormwater Management System design.
334
-------
. ^mon,«n..H Specific cases were provided exemption from the Policy, including: single family house projects,
residential subdivisions with four or fewer lots that do not discharge to critical areas, and emergency repairs to
highway/roadways or their drainage systems. However, none of these projects are exempt from the standard requiring
sedimentation and erosion control requirements during construction activities.
While redevelopment projects are not exempt, a redevelopment project may comply to the maximum extent possible
if it can be proven that it is not practicable for the project to achieve full compliance.
. Operation/Maintenance: All non-exempt projects require the development of a Stormwater Management System
Operation and Maintenance Plan (O&M Plan). As defined in the Policy, the O&M Plan must contain the: names of the
Stormwater Management System(s) owner and the person(s) responsible for implementing the O&M Plan, a schedule
for inspection and maintenance, and a description of maintenance activities to be performed. Recommendations for
specific maintenance practices and schedules are included in the Policy.
Stormwater Performance Standards
The nine Stormwater Performance Standards most accurately describe the key components of the policy that came out
of the committee deliberations following resolution of the issues described above. The goal of the standards is to protect
groundwater, surface water, and wetland resources from the impacts of Stormwater runoff generated as a result of development
and redevelopment projects. General descriptions of the standards are provided as follows:
1. No new Stormwater conveyances may discharge untreated Stormwater directly to, or cause erosion in wetlands or
waters of the Commonwealth.
2. Post-development peak discharge rates must not exceed predevelopment conditions for the 2-year and 1 0-year storm
events under all conditions. The 100-year event must be analyzed to determine impacts and must be controlled if
necessary.
3. Loss of annual recharge to groundwater should be minimized through the use of infiltration measures, to the maximum
extent practicable. The recharge "requirement" which is to mimic existing annual recharge on sites to the maximum
extent practical, has not been changed. However, a design methodology for estimating existing annual recharge at a
site, and for designing recharge systems has been developed. The methodology uses soil classifications, soil gradation
analyses and specific Massachusetts regional rainfall data as data inputs.
4. Stormwater management systems must be designed to remove 80% of the average annual (post development) load
of total suspended solids (TSS). It is presumed that this standard is met when; (a) suitable nonstructural practices for
source control and pollution prevention are implemented; (b) Stormwater management BMPs are sized to capture the
prescribed runoff volume; and (c) Stormwater management BMPs are maintained and designed as specified in Volume
Two. The Policy provides estimates of the percent TSS removal for individual BMPs when designed in accordance with
the specified guidelines. Water quality treatment volume is 0.5 inches of runoff from impervious areas (1 .0 inch if
discharge is to critical environmental area).
5. Stormwater discharges from areas that are defined as having "higher potential pollutant loads" (as defined in the Policy)
require specific Stormwater BMPs. Infiltration of Stormwater from these areas without pretreatment is prohibited.
6. Specific BMPs must be used for discharges to critical areas and the water quality treatment volume is 1 .0 inch of runoff.
7. Redevelopment projects must meet the performance standards to the maximum extent practicable. It must be clearly
stated why full compliance cannot be achieved and such projects must, at a minimum, improve existing conditions.
8. Erosion and sedimentation controls are required during construction.
335
-------
9. An Operation and Maintenance Plan (O&M Plan) for Stormwater Management is required.
A detailed explanation of each of these Standards is available in Volume One, the Stormwater Policy Handbook.
Implementation
In order to test its effectiveness and identify potential problems, the Policy was introduced in March of 1997, prior to the
promulgation of regulations. Copies of Volumes One and Two were distributed to each Conservation Commission office in the
state, and to other relevant regulatory agencies. Two phases of training were provided across the state over a two-year period:
first to introduce the Policy, and then to focus on detailed case studies and implementation issues. Training was made available
to regulatory agencies, Conservation Commissions, local DPWs, planning departments, and engineering companies. During
the training sessions, the largest turnout was from Conservation Commission representatives and engineering/consulting firms,
with minimal attendance by the other groups invited.
At each of the training sessions, questions from the audience on the Policy were solicited and a list of most Frequently
Asked Questions (FAQs) was developed. The FAQs provided an excellent basis for outlining where additional information
and/or clarification was needed. Based on the number of recurring questions, the Committee decided to prepare a survey to
solicit feedback from potential policy users. The survey was comprised of 23 questions for characterizing the respondent,
determining the usefulness and ease of implementation of the Policy, identifying particular problems in understanding or
implementing performance standards, and determining what type of BMPs were currently being implemented.
The survey was not designed to be a statistically valid data set, but rather to gain a practical working knowledge of what
aspects of the Policy and/or its method of implementation needed to be refined. While this was generally evident from the
FAQs, the survey further substantiated the specific areas where additional work was necessary. Some key findings from the
118 respondents to the survey were as follows:
. The overall sense of the Stormwater Performance Standards was that the Stormwater Policy implementation was good
(63%) and that they consider the Stormwater Handbook to be a "useful resource for designing and reviewing systems
(77%).
. In response to the standards, respondents were generally comfortable implementing peak discharge controls and
sedimentation and erosion controls. This was not surprising since these were existing requirements in most
municipalities that needed to be achieved for most development projects.
. Not surprisingly, new requirements for groundwater discharge, treatment of Stormwater runoff water, and the
preparation of Stormwater Management O&M Plans were the standards for which additional clarification and technical
support were most requested.
Feedback
The Draft Policy was issued prior to formal promulgation so that one to two years of interim implementation could be used
for refinement. The FAQs, user survey, and ongoing feedback from the public defined those areas where refinements were
especially necessary, including the following:
Standard No. 3 Recharge: The recharge requirement clearly created the most confusion, and required additional
technical support. The original brief description included in Volume One was not sufficient for engineers or reviewers to prepare
a comprehensive program for achieving the annual recharge requirement.
As a result, the technical sub-committee has focused on providing a more detailed definition of the recharge requirement
and appointed a group to develop a design methodology for achieving compliance with this standard. The group has developed
a methodology using soil groups and soils analysis and actual rainfall data to determine the existing annual recharge on project
sites. Methodology for designing a recharge system to provide post-development recharge that best mimics pre-development
336
-------
conditions is being developed. This technical is currently in the final stages of development and is expected to be completed
and distributed prior to December 1999.
Standard 4 - 80% TSS Removal: The Total Suspended Solids (TSS) removal requirement was not developed with the
goal of removing only the TSS loads in stormwater runoff. Rather, it was considered an indicator parameter, whereby if 80%
of the TSS is removed, a large portion of the additional pollutants carried in stormwater is also removed. This relationship does
not hold true, however, if the proponent chooses to use only mechanical methods where settling of fines, assimilation of
nutrients, or other biological processes that increase pollutant removal do not occur. For instance, a vegetated infiltrating swale,
or wet pond, that is designed to provide substantial stormwatervelocity reductions may greatly increase fine sediment removal
and may also provide nutrient uptake in addition to gross TSS removal. Astructure that removes solids only, and does not allow
for detention or contact with plants or potential filtering areas does not comply with the goal/intent of the TSS removal standard.
In addition, users were having difficulty calculating TSS removals when numerous BMPs were to be used in a series. The
specific percent removal rates provided in Volume One of the Policy are not additive and, as such, must be calculated as
percents of the pollutant load that they receive.
In response to these issues, further definition of the goal of the TSS removal requirement and recommended practices for
achieving the goal are being developed. A spreadsheet has been developed, which can be easily filled out by hand, to assist
in calculating the percent TSS removal for a project based on the BMPs implemented.
Standard 9 - Operation and Maintenance: This standard has consistently raised the most concern relative to cost and
implementation. Common questions were as follows:
. Who will pay for ongoing operation and maintenance of stormwater BMPs if it is not the town?
. How will the town fund the O&M requirements if they assume responsibility?
. What long-term mechanism is there to ensure that maintenance will be completed?
. What is the frequency of maintenance required for specific BMPs?
While these questions are difficult to answer, for a specific site or on a statewide basis, in terms of required maintenance,
the Committee has responded by preparing operation and maintenance checklists, which may be used by the operator/owner.
These checklists may be submitted to the local Conservation Commissions on an annual basis, if deemed necessary.
Specific maintenance practices and suggested frequencies have also been prepared and are currently being updated, as
new information becomes available. It has also been noted that "one size does not fit all" in terms of required maintenance.
For instance, an infrequently used residential guest parking area clearly does not warrant the same frequency of street
sweeping as a commercial mall parking lot.
The question of financing stormwater management system operation and maintenance activities is also a site- and location-
specific issue. The development of stormwater utilities and management districts is ever increasing and may be the option that
some communities choose. The promulgation of the Policy as regulations will require the project proponent and/or towns and
municipalities to develop a plan for ensuring O&M implementation. There are a variety of ways this may be achieved.
Resolution may be somewhat facilitated in larger municipalities and/or USEPA-designated urban areas that must comply with
the upcoming final USEPA NPDES Phase II Stormwater Regulations.
Summary
The development of the Massachusetts Stormwater Policy, which includes nine performance standards, was an integrated
effort between state and local regulators, policy makers, engineers, developers, and the general public. The group effort and
the implementation of the Policy as a Draft, to which refinements could and have been made, have contributed to the
usefulness of the standards. While the development and implementation of the standards are still in the early stages, the Policy
provides definitive goals for achieving stormwater water quality and quantity control, and addresses annual recharge on
337
-------
development sites. Until the development of these standards, there was no requirement to maximize groundwater recharge
on sites or to mandate the development of the stormwater O&M Plan (unless required under other regulatory programs). These
advances will provide an overall benefit to the natural resources of the Commonwealth of Massachusetts. The Stormwater
Management Policy Volumes I and II may be obtained off the Internet or by request from:
Tom Maguire
Massachusetts Department of Environmental Protection
1 Winter Street
Boston, MA 02108
Phone (617) 292-5602
email: Tom.maguire@state.ma.us
References
Massachusetts Coastal Zone Management, Massachusetts Department of Environmental Protection, Stormwater Management
Volume One: Stormwater Policy Handbook, March 1997.
Massachusetts Coastal Zone Management, Massachusetts Department of Environmental Protection, Stormwater Management
Volume Two: Stormwater Technical Handbook, March 1997.
Massachusetts Coastal Zone Management, Massachusetts Department of Environmental Protection, The Stormwater
Evaluation Survey: Perspectives on the Implementation of the Interim Massachusetts Stormwater Policy, June 1988.
338
-------
Implementation of Michigan's Voluntary
Stormwater Permit-a Community Perspective
Kelly A. Cave, P.E. Director
Watershed Management Division
Wayne County Department of Environment
Detroit, Ml
Dale S. Bryson
Camp Dresser & McKee
Detroit, Ml
Kelly C. Kelly, P.E., Project Engineer
Canton Township, Ml
Jack D. Bails, Vice-President
Public Sector Consultants
Lansing, Ml
Introduction
The Rouge River National Wet Weather Demonstration Project (Rouge Project) has made significant progress in
restoring beneficial uses to a large, urban waterway using a holistic, "bottom up" watershed approach. This project was
initiated in 1992, by the Wayne County (Michigan) Department of the Environment. The purpose of this document is to
present a summary of the activities and progress of the Rouge Project; discuss the watershed approach being utilized in
the Rouge Project, including the use of a general storm water permit; and summarize a community perspective on this
entire effort.
Rouge Project Background and Summary of Progress to Date
The Rouge River National Wet Weather Demonstration Project is a watershed-based effort, substantially sponsored
by the U.S. Environmental Protection Agency (USEPA), to manage wet weather pollution to the Rouge River, a tributary
to the Detroit River in southeast Michigan (See Figure 1). The Rouge River Watershed is largely urbanized, spans
approximately 438 square miles, and is home to over 1.5 million people in 48 communities in 3 counties. The Rouge River
has been designated by the international Joint Commission as a significant source of pollution to the Great Lakes system.
The early focus of the Rouge Project was the control of CSOs in the older urban core portion of the downstream areas
of the watershed. As a finite number of point source CSO discharges could be identified, and responsibility for each
defined, the traditional regulatory approach of issuing National Pollutant Discharge Elimination System (NPDES) permits
mandating corrective action worked relatively well. Additional monitoring of the river showed, however, that other sources
of pollution needed to be controlled before full restoration of the river would be achieved throughout the watershed. These
other sources of pollution include storm water runoff, interflow from abandoned dumps, discharges from illicit connections,
discharges from failed on-site septic systems, and resuspension of contaminated sediment.
The Rouge Project is designed to identify the most efficient and cost-effective controls of wet weather pollution, while
assuring maximum use of watershed resources. A great deal has been accomplished in these efforts. The following
summarizes some of those accomplishments, focusing on CSO controls first. Approximately 50% of the watershed is
served by separate sewer systems, with an additional 20% served by combined sewers (157 overflow points), and the
339
-------
Figure 1. Rouge Subwatersheds & Location of Rouge Watershed in Michigan.
remaining area served by on-site sewage disposal systems. CSO controls are being implemented in phases. Under the
first phase, six communities have separated their sewers and eight communities have constructed or are constructing 10
retention treatment basins. Each of these basins is sized for different design storms and several employ innovative
technologies. A two-year evaluation study of the CSO control basins began on June 1, 1997. The results from the
evaluation study, coupled with efforts to control storm water and other pollution sources in the watershed, will provide the
basis for the second phase CSO control program for the remaining CSO sources in the watershed. The information gained
from the evaluation of design storms and control technologies will be useful to others nationwide to determine cost-effective
CSO controls to meet water quality standards.
The Rouge Project is also evaluating innovative stormwatercontrol and watershed management technologies. Twenty-
five different communities and agencies throughout the watershed are implementing over 100 pilot projects. Categories
of pilot management projects currently underway include wetlands creation and restoration; structural storm water
practices, such as grassed swales and detention ponds; erosion controls; stream bank stabilization; and habitat restoration,
to name a few.
The Rouge Project also discovered that illicit connections and failing septic systems are major sources of pollution
problems in the Detroit urban area. Creative ways to remediate these sources of pollution have been initiated.
A suite of computer models has been developed by the Rouge Project. These models simulate the water quality and
quantity response of the Rouge River during wet weather events for existing and future conditions, and under various CSO
and stormwater runoff management alternatives. This effort has resulted in a very useful public communication tool on
water quality indices tied to actions needed to restore the Rouge River. A comprehensive geographic information system
(GIS) and relational databases were designed and implemented to manage the wealth of data collected under the Project.
In addition, a special data exploration tool, DataView, was developed to support routine analyses of large time-series data
sets. DataView is user-friendly and readily transferable to other locations. Related to DataView is the Rouge Information
Manager, also a user-friendly, readily transferable tool (an "electronic file cabinet") for accessing multi-media information
about the Rouge River restoration effort.
Effective, readily transferable tools have been developed, employed by the Project, and are being shared with other
cities and state agencies. Additional information on the Rouge Project can be obtained from the web site at
h ttp:\www. rougeriver, com,
Evolution of the Watershed Approach
The Rouge River watershed has seven subwatersheds that range in size between 19 and 89 square miles (See Figure
1). Older communities served by combined sewers dominate downstream portions of the Rouge River Watershed, while
340
-------
the headwater areas are typically open space, agricultural land, or low density residential developments that are
undergoing rapid change due to growth pressures. Fully developed areas, typical of the middle portions of the Rouge
Watershed, have separated storm sewers and limited opportunities to address stormwater problems with structural
solutions.
Data gathered by the Rouge Project have shown numerous water quality and designated use problems, including high
bacteria levels and low dissolved oxygen levels during wet weather events in all areas of the watershed. Fish consumption
is prohibited in much of the watershed due to the threat to human health. Six of the seven subwatersheds have moderate
to severe degradation of wildlife habitats, with fish populations suffering severe impairment in three of the subwatersheds.
Aesthetic enjoyment is moderately to severely impaired throughout the watershed. Restrictions to small boat navigation
resulting from logjams, garbage and sedimentation are a moderate to severe impairment in virtually all seven
subwatersheds.
Based upon what was learned, the focus of the Rouge Project became more holistic to consider impacts from all
sources of pollution and use impairments in receiving waters. The historic implementation of water quality management
programs in the United States at the federal and state levels has been to focus on point sources, which are the most
obvious sources of pollution to water bodies. This approach has worked well to control pollution from most point sources
but has also left a patchwork of regulated and unregulated discharges of stormwater and nonpoint source pollution to
surface waters. This patchwork is especially evident in most urbanized areas where multiple local jurisdictions are located
in the same watershed. More subtle sources of pollution, such as stormwater, have emerged as the next priority for
attention. The challenge for the Rouge Project became to develop innovative, coordinated solutions to achieve water
quality objectives that may be: (1) be more cost-effective, (2) implemented in a more timely fashion, and (3) better able
meet local needs.
It has become clear that water resources management must have the support of the general public in order to be
effective and to become self-sustaining. A locally driven watershed approach to pollution management as a means to
achieve management goals is an exciting concept discussed by many, but for which there is limited practical experience.
This is particularly true in urban situations where there are multiple sources of impairment to a water body and stiff
competition for limited local resources to address the pollution sources. The Rouge Project has provided a unique
opportunity for a watershed-wide approach to restoring and protecting an urban river system by using a cooperative, locally
based approach to pollution control.
The Michigan NPDES General Permit for Municipal Stormwater Discharges
As concern expanded to sources of pollution in the upper portion of the watershed above the CSO discharges, and
water quality improvement focused more on watershed-wide approaches, the lack of a defined regulatory framework to
address stormwater pollution and diffuse nonpoint source pollution became a major obstacle to further progress in
improving water quality and restoring beneficial uses to the Rouge River. Beginning in 1995, the Michigan Department of
Environmental Quality (MDEQ), the Rouge Project, and the communities in the Rouge Watershed jointly developed an
innovative, watershed-based NPDES general permit ("General Permit") for municipal stormwater discharges. The permit
was issued on July 31, 1997 (MDEQ, 1997). This collaborative process was outlined in a report entitiled "Adapting
Regulatory Frameworks to Accommodating Watershed Approaches to Storm Water Management" (Fredericks, et al.,
1997).
The MDEQ General Permit and USEPA's draft Phase II stormwater regulations (U.S. Environmental Protection
Agency, 1998) due to be promulgated in September 1999, were developed during the same time frame. Wayne County,
on behalf of the Rouge Project communities, was selected to serve on USEPA's Urban Wet Weather Flows Federal
Advisory Committee, which (among other activities) assisted USEPA with the development of the Draft Phase II
Stormwater Regulations. Participation on the federal advisory group on watershed approaches to stormwater management
was invaluable to the development of the Michigan General Permit, and provides a high likelihood that the General Permit
will be acceptable for implementing the forthcoming federal Phase II Stormwater Regulations in Michigan.
The General Permit incorporates the following elements:
341
-------
General:
- Permit coverage is voluntary until the final EPA Phase II Stormwater Regulations are promulgated
- Public agencies who own, operate, or control stormwater systems are provided the opportunity for coverage
- Watershed size is established by the potential permitees during the application process
- Application and permit process have limited required actions; the focus is to establish desired outcomes.
Requires permittee to develop the following:
- Illicit Discharge Elimination Plan (IDEP) that has the goal of eliminating raw sewage discharges and includes
addressing failing septic systems and improper connections of sanitary sewers to storm drains and open waterways.
- Public Education P/an (PEP) designed to inform residents and businesses about what actions they should take to
protect the river.
- In cooperation with others, a Watershed Management P/an to resolve water quality concerns which includes: short
and long-term goals for the watershed, delineation of actions needed to achieve the goals, estimated benefits and
costs of management options, an opportunity for all stakeholders to participate in the process.
- Storm Water Pollution Prevention Initiative, which includes evaluation and implementation of pollution prevention
and good housekeeping practices and the evaluation and implementation of BMPs to minimize impacts of new
development and redevelopment.
- Monitoring and Reporting P/an, which includes a schedule for revisions to the Watershed Management Plan.
The IDEP and PEP are submitted to MDEQ at the time of application and the implementation of these plans begins
when the MDEQ issues a certificate of coverage to a community/agency. Within six months after the issuance of a
certificate of coverage to a community/agency, the General Permit requires the submission of a public involvement plan
for approval by the state. This plan identifies the approaches that will be used within the hydrologic area to involve
stakeholders in the development of a watershed plan that must be completed within 18 months after the certificate of
coverage is issued. Once a consensus, long-term management plan has been completed, each agency and community
within the watershed must prepare and submit for state approval its own pollution prevention initiative that identifies actions
and schedules to address the pollution concerns identified in the consensus watershed plan. The watershed stormwater
management plans developed by the communities and other public agencies do not require state approval; however, the
individual pollution prevention initiatives emanating from the watershed planning process do require state approval, as the
activities specified in the initiatives become permit requirements upon approval.
Rouge Community/Agencies Approach to Application and Permit Requirements
A total of 43 communities and agencies who own, operate, or control stormwater systems in the Rouge River
Watershed have obtained coverage under Michigan's new watershed-based General Permit for municipal stormwater
discharges. The result is that over 95% of the watershed is covered under this new program. The communities and
agencies requested that, for purposes of the General Permit, the large Rouge watershed be subdivided into the seven
subwatersheds shown in Figure 1. Long-term management plans will, therefore, be developed for each of these
subwatersheds, with coordination of the plans provided by the MDEQ and the Rouge Project staff. The document
"Implementing a Model Watershed Approach Through a State General Storm Water NPDES Permit" (Cave, et al., 1998)
outlines key issues discussed and decisions reached by the communities as they developed their General Permit
applications during 1998. The following section presents additional information regarding the application and permit
requirements recently approved for the communities and agencies in the Rouge Watershed under the Michigan General
Permit.
342
-------
The Michigan General Permit for municipal stormwater discharges is quite flexible and allows those seeking coverage
under the permit to use a wide variety of approaches to meet the public education, illicit connection/illegal discharge, and
public involvement requirements. This flexible framework has allowed communities to experiment with various approaches
that recognize local constraints and seize upon unique opportunities to meet the desired outcomes. While the basic
requirements for what must be in the watershed plan are more detailed, the permitees within a watershed are allowed
considerable freedom in deciding upon their own priorities, remedial actions, and schedules. Pollution prevention initiatives
that are expected to be proposed by the communities will likely involve commitments to continue or expand current
activities, like soil erosion and sedimentation control; implementation of new activities to address priority issues such as
failing septic systems; and implementation of regional projects to reduce the frequency and velocity of storm flows in the
river.
Table 1 and Table 2 outline the variety of public education and illicit discharge detection and elimination approaches
identified by the communities and public agencies in the Rouge Lower 1 Subwatershed (Figure 1).
Across the watershed, communities actively sought ways to cooperatively address illicit connection/illegal discharge
investigations and public education projects. In one subwatershed group, a community with experience in the production
of videos agreed to make a river stewardship video that all other communities within the subwatershed could use on cable
television, or through the distribution of cassette copies to local libraries and/or schools. In the same watershed, another
community offered the use of a consultant to solicit bids for freestanding public information display boards and to develop
stormwater information materials for the boards that could be used by all communities at public gatherings and inside
public facilities. One community obtained the support of the local college to house and provide administrative support for
a well-established non-profit organization, the Friends of the Rouge organization, whose public information activities were
subsequently funded by several subwatershed groups to implement portions of their public education plans.
Advantages of Watershed Approach
The following section presents some of the lessons learned as the communities and agencies in the Rouge Watershed
are beginning to implement the watershed-based, Michigan General Permit for municipal stormwater discharges.
Holistic Solutions/Local Ownership. There are distinct advantages in managing stormwater on a watershed basis. From
the work already completed on the Rouge Project, it is clear that an integrated approach is needed to address all sources
of water pollution and excessive flows in this urbanized watershed. By requiring those agencies and communities with
responsibility for stormwater to work together at the subwatershed level to establish goals and objectives, local agencies
and the state regulatory agency are forced to view solutions holistically. To achieve the desired level of river restoration,
there must be integrated action plans that address not only stormwater but failing on-site sewage disposal systems
(OSDS), CSOs, sanitary sewer overflows (SSOs), and significant nonpoint sources of pollution.
Ideally, a watershed-based regulatory framework should encompass all dischargers so that pollution sources can be
addressed holistically. Practically, it must be recognized that prior NPDES permit programs at the state and federal level
are already in place for municipal and industrial point source waste treatment discharges, and for many industrial and
commercial stormwater discharges. While the Michigan watershed-based stormwater General Permit covers only public
agencies that own, operate, or control stormwater conveyance systems not currently under a Phase I Stormwater Permit,
the required watershed management plan does provide aframeworkfor integrating activities under other permit programs.
In addition, the General Permit gives communities and agencies the flexibility and encouragement to incorporate nonpoint
source controls and pollution prevention activities as part of the required watershed management plan. For example, many
343
-------
Table 1. Illicit Discharge Elimination Activities to be Implemented under Michigan Stormwater General Permit in Communities in Lower 1 Subwatershed,
Rouge River Watershed
Legal Basis
Locating Problem Areas
Finding the
Source
Minimizing Seepage
from Septic Fields and
Sanitary Sewers
SP
1
8
Activity
Existing ordinances for control of illicit connections and/or OSS have
been determined sufficient or community/agency will evaluate existing
ordinances
Will adopt additional ordinances for control of illicit connections and/or
OSDS if determined necessary
Review existing monitoring data to prioritize investigation areas
Plan developed w/County to locate sources of suspicious discharges
previously identified
Develop, modify, implement and/or continue to use complaint system
Procedure to coordinate complaint response/follow up
Develop and/or use GIS for tracking complaints and/or activities
Train field employees for identification & reporting of illicit discharges
Mapping of storm system, jurisdictions and/or locations of outfalls
Systematic dry weather screening of outfalls or manholes
Investigate possibility of systematic screening program
Screen drainage from facilities under jurisdiction
Dye testing when additions made to existing facilities
Establish priority of suspicious outfalls and/or initiate follow up visits
for further analysis of flow
Investigate to find sources of suspicious discharges using upstream
manholes or dve testing or televising
Identify and/or map areas served by OSDS
Determine feasibility of sewer extension/mitigation
Proposals for future sanitary sewer construction will consider existing
OSDS
Develop agreement/cooperate with county for implementing an OSDS
evaluation program
OSDS evaluation prior to sale of property
Continue sanitary sewer maintenance program
Reporting to MDEQ on investigations, violations found & corrective
actions taken
Investigate Funding Mechanism for Stormwater Related Tasks
Canton Township
II
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Plymouth 1
Township 1
X
X
X
X
X
X
X
X
X
X
X
X
X
Van Buren |
Township
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Wayne County
Salem Township 1
X
X
X
X
X
X
X
X
X
X
X
Superior
Township ||
X
X
X
X
X
X
X
X
Ypsilanti
Township |
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
344
-------
Table 2. Public Education Activities to be Implemented Under Michigan Stormwater General Permit by Communities in Lower 1 Subwatershed Rouqe
River Watershed
Activity
Cable TV ads
Clean Sweep program
Coordinate with Master Composters Program
Co-sponsor annual River Day
Co-sponsor educational workshops
Co-Sponsor Healthy Lawn & Garden Workshop
Cosponsor informational outreach workshops
Co-sponsor River Stewards program
Co-sponsor River Watch program
Co-sponsor Rouge River Day
Co-sponsor Rouge Education Project
Co-sponsor Rouge Friendly Neighborhood Program
Co-sponsor Rouge Friendly Business Program
Display maps of community, watersheds & boundaries
Distribute miscellaneous brochures and/or feet sheets
Distribute Rouge Recreational Guide
Distribute Rouge Repair Kit to homeowners
Distribute septic system maintenance packet to homes with OSDS
Distribute storm water general information package to new residents
Heighten visibility & promote school water/resource monitoring
Periodically provide Rouge Information Kiosk system in public buildings
Presentations
Provide articles, Information in community newsletter
Provide fliers/messages in water bills or tax notices
Provide water quality educational information on Website
Public service announcements
Publicize garden hotline
Publicize illicit discharge hotline
Storm drain marking
Tributary signage
Utilize "Our Actions" display at various community events
I
Canton Township
X
X
X
X
X
X
X
X
X
Plymouth 1
Township |
X
X
X
X
X
X
Van Buren I
Township Q
X
X
X
X
X
X
X
X
Salem Township II
X
X
X
X
X
X
X
X
X
X
X
X
Superior |
Township |
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Ypsilanti 1
Township
X
X
X
X
X
X
X
X
X
X
X
communities have initiated pilot projects to evaluate how stormwater best management practices (BMPs) will control
stormwaterflow and prevent pollution. In some cases, these pilot projects have permanently changed the way communities
and/or government agencies manage stormwater. These management practices will be included, as part of a watershed
management plan, and credit will be given to the entities that are performing those functions.
345
-------
Many of the subwatershed units selected in the Rouge River Watershed involve communities that have combined
sewers, separated sewers, and OSDS. Some individual communities have all three within their corporate limits. Once the
communities began to work together at the subwatershed level to establish goals to achieve water quality standards
necessary to restore the river, each found that they had a significant role in the process and that the control of flow in the
rapidly developing headwater areas was as significant as CSO problems in the lower portion of the watershed. Evaluating
the sources of water quality problems and/or the threats to existing uses of the river at the subwatershed level by local
agencies is leading to a better understanding of local constraints, opportunities for innovative solutions, ownership of the
long-term river restoration effort, and interagency cooperation.
Overcoming Institutional/Regulatory Barriers. Local agencies and communities in urbanized areas have a long history
of cooperative efforts to address the delivery of common public services. Recent trends in Michigan, and elsewhere in the
country, to reduce the size and cost of government and limit local taxing power have accelerated efforts at the local level
to integrate or share the cost of a broad range of government services. Local agencies are increasingly seeking ways with
their neighboring jurisdictions to reduce the cost of police and fire protection, solid waste disposal, libraries, recreational
facilities, infrastructure maintenance and repairs, public transit, water supplies, and sewage disposal. Unfortunately, except
in a few isolated instances where a single authority has been created to oversee all aspects of water management, the
legal responsibility for stormwater is widely dispersed among local communities, county health and drain agencies, road
agencies, private developers, and autonomous school districts and public colleges. The creation of a new level of
government in the form of a water management authority with broad powers has been resoundingly rejected in the Rouge
River watershed by local agencies and is likely to receive the same reception in many other urban areas of the country.
State and federal water quality regulatory programs have traditionally focused on large point sources where
responsibility for obtaining and complying with specific permit limits is easy to establish. The regulatory framework to
control water pollution has generally discouraged rather than encouraged cooperative solutions among communities and
has relied upon command and control to achieve results. The complexities involved in addressing wet weather pollution
problems in urban areas, and the widely dispersed accountability for managing stormwater, demands a new regulatory
framework that will encourage cooperation among the locally responsible public agencies to design integrated, holistic
solutions. The watershed approach to stormwater regulation developed in Michigan offers an opportunity to overcome the
institutional and regulatory impediments that have discouraged cooperative local approaches to restoring urban
watersheds.
Institutional arrangements and financing options necessary to implement the General Permit and subwatershed
management plans are among the many elements the local communities in the Rouge Watershed are addressing in their
working groups. As part of the subwatershed planning process, communities and agencies are also identifying issues
which cross subwatershed boundaries. Rouge Project staff and the MDEQ currently provide coordination of individual
subwatershed efforts and assist subwatersheds in developing a comprehensive strategy for addressing watershed-wide
issues. The subwatershed communities are also identifying those activities, such as public education and water quality
monitoring, that may be most cost-effectively performed throughout the entire watershed by a single entity.
Increased Local Accountability and Political Support. Building a watershed restoration project from the bottom-up by
helping local communities and agencies identify problems, set their own priorities for restoration, select their own remedial
measures and propose their own schedules requires a sharing of power among federal, state, and local agencies not
usually found in water pollution control programs. The General Storm Water Permit program in Michigan is voluntary at
this time and it has allowed state regulators the ability to provide flexibility that might not otherwise be available. It has also
increased the accountability of local agencies who no longer have the freedom to blame federal and state officials for the
impositions of requirements, but now are responsible for convincing local elected officials that the programs proposed are
in the best long-term interest of the local residents.
Opportunities for Cost Eff iciencies/lnnovation. As discussed earlier, the Rouge River communities that have obtained
coverage under Michigan's General Storm Water Permit and are working in subwatershed groups have already developed
more cost effective and efficient methods to meet public education requirements through cooperatively developed projects.
Similar joint programs are underway to train local community and agency staff in illicit discharge elimination activities and
in sharing staff and equipment to conduct river and enclosed storm drain surveys. The three county health agencies are
346
-------
developing common approaches to permitting and inspecting OSDS. The county road agencies are working with the state
highway agency to address the design, construction, and maintenance of transportation drainage systems.
The county agencies in the three counties responsible for designated storm drains are working together and with local
communities toward implementing common standards for stormwater management at new developments. County and local
officials have worked together to establish protocols for rapidly developing independent GIS to assure that databases can
be integrated to assist in watershed-wide water quality/quantity management. The economic and political cost for each
community or county agency to develop these approaches has been an impediment in the past. The watershed approach
has enabled these cooperative programs to be established. It is anticipated that the pollution prevention initiatives required
following completion of the watershed management plans would also involve joint projects.
Establishing a broad range of cooperative programs to address stormwater problems across jurisdictional boundaries
is, in of itself, innovative. However, with the development of comprehensive watershed plans, new practical approaches
to implementing total maximum daily load (TMDL) requirements of the Clean Water Act and effectively using water pollution
trading options created at the state level become possible. The Rouge River National Wet Weather Demonstration Project
is funding a pilot program to verify that the watershed management framework under the Michigan Stormwater General
Permit can be used to meet the TMDL requirements, ahead of state schedules (and at potentially lower cost), and the
objectives of the Clean Water Action Plan program. In addition, the pilot program will demonstrate how the General Permit
watershed framework can be used as a basis for the proposed statewide water quality trading program currently under
development.
The top-down, command-and-control approach often requires repeated threats or legal action by state and federal
regulators to ensure compliance with requirements due to lack of political will at the local level. Locally generated
watershed plans containing specific action schedules that have been adopted by elected boards, commissions, and
councils are less likely to be abandoned or require enforcement actions to assure compliance. Peer pressure and citizen
support at the local level should be sufficient incentive in most instances for each local agency to complete their
responsibilities on schedule. Where legal enforcement action is required, the state and federal agencies are more likely
to find support among other local agencies who have met their obligations as outlined in the joint subwatershed plan.
Conclusions and Recommendations from a Community Perspective
Local communities in southeast Michigan and the state regulatory agency are attempting, for the first time, a
consensus, cooperative approach to stormwater management and regulation under the NPDES program. The Michigan
General Permit is a watershed-based, general stormwater permit issued under the National Pollutant Discharge Elimination
System. The permit requires permitees to immediately initiate activities, such as illicit discharge elimination, and to
participate in watershed management planning for a self-determined hydrologic unit. The watershed management plan
forms the basis for implementing watershed goals and objectives, including improved water quality and pollution control.
This new regulatory program implements the watershed approach endorsed by USEPA and others and should facilitate
watershed-based integration of control programs for different pollution sources, such as stormwater CSOs which may be
present within a large, urban watershed. In addition, it is believed that the new watershed-based stormwater program in
Michigan will achieve the objectives of the TMDL program, the Clean Water Action Plan Program, and water quality trading
programs under development across the country. From the perspective of local communities and agencies, this approach
provides optimum flexibility to solve the most pressing problems in their subwatersheds by empowering them to identify
problems, choose from alternative solutions, establish priorities and schedules, and develop common strategies with
neighbors. Communities and others involved in this new program are also addressing issues such as coordination of
subwatershed efforts within larger watersheds.
The Rouge Project (and others) have shown that by holistically addressing all sources of pollution, a cost-effective
action plan can be implemented to address impairments and restore river uses. Storm water issues cannot be corrected
in a vacuum, but must be integrated into an overall solution that addresses the physical, chemical, and biological stressors
in a waterway. Stormwater adversely affects all three and, therefore, must be woven into the fabric of the overall watershed
management plan and watershed control program. Without this integration, stormwater control will become another "add
on" program that misses an opportunity to encourage an integrated program that addresses all sources of ecosystem
347
-------
stress in a cost-effective, prioritized manner. The approach being followed in the Rouge River Watershed should prevent
misplaced efforts and, most importantly, result in a restored Rouge River on a much faster timetable.
A key objective of the Rouge River National Wet Weather Demonstration Project has been to demonstrate alternative
methods to a "command and control," top down regulatory approach for water quality protection and improvement. The
alternative methods sought by the Rouge Project leverage "bottom-up" approaches that put place-based needs in the
forefront and use local initiatives to make progress on water quality restoration. This approach has led to a number of
institutional changes in the watershed that will help sustain the watershed management efforts into the future. From the
perspective of the communities involved, the cooperative, iterative approach being followed appears to be working and
is a welcome change from traditional "command and control" relationships with regulatory agencies.
The Rouge Project approach demonstrates that watersheds can be "managed." When water quality objectives can
only be reached through control of CSO, stormwater, and nonpoint sources, watershed management must involve the
active participation of local units of government. From a community perspective, this local involvement is critical to the
overall success of the Project and to stream restoration. Also, from a community perspective, undertaking a watershed
effort is not a simple matter. Watershed planning and implementation takes a large commitment of time and effort.
The communities involved in the Rouge Project have a sense of overwhelming success with the watershed restoration
efforts to date. Water quality and ecosystem health are improving, and the demonstration techniques have resulted not
only in concrete and steel structures, but in real institutional changes that integrate the work of stormwater and watershed
improvement into the basic institutions of government. Most importantly, the public is able to utilize new canoeing areas
and other river-based amenities, which are now possible due the noticeable improvement in water quality, aesthetics, and
other attributes of the river. It is hoped that this effort, and the work of the Rouge River National Wet Weather
Demonstration Project, will continue to identify and quantify the benefits of cooperative, watershed-based efforts to protect
and restore our nations water resources.
Acknowledgments
This paper represents a summary of select elements from the ongoing efforts of many individuals and organizations
involved in the restoration of the Rouge River in southeast Michigan. The authors also gratefully acknowledge the
assistance of Ms. Sandra Kiser during the preparation of this manuscript.
The Rouge River National Wet Weather Demonstration Project is funded, in part, by the United States Environmental
Protection Agency Grant #X995743-01 through 04 and #0995743-01. The views expressed by individual authors are their
own and do not necessarily reflect those of USEPA. Mention of trade names, products, or services does not convey, and
should not be interpreted as conveying, official EPA approval, endorsement, or recommendation.
References
Cave, K. and J. Bails (1998). "Implementing a Model Watershed Approach Through a State General Storm Water NPDES
Permit". Proceedings of WEFTEC 98.
Fredericks, R., K. Cave, and J. Bails (1997). "Adapting Regulatory Frameworks to Accommodating Watershed Approaches
to Storm Water Management." Proceedings of WEFTEC 97.
Michigan Department of Environmental Quality (July 30,1997), National Pollutant Discharge Elimination System, General
Wastewater Discharge Permit, Storm Water Discharges from Separate Storm Water Drainage Systems, Permit No.
MIG610000.
U.S. Environmental Protection Agency (January 9,1998). "Draft Phase II Federal Storm Water Regulations."40 CFR Part
122 & 123.
348
-------
California's Model Urban Runoff Program (MURP):
Urban Runoff Programs for Small Municipalities
Jennifer Hays, Assistant Engineer
Public Works Department
City of Monterey, California
Cy Oggins, Coastal Program Analyst III
California Coastal Commission
Central Coast District
Santa Cruz, California
Background
Monterey Bay is a crown jewel of the California Coastline and has received special protection under the National
Marine Sanctuaries Act since September 1992, and the California Coastal Act since 1976. The 5,300 square mile Monterey
Bay National Marine Sanctuary includes a number of small coastal communities, and ranges from the City and County of
San Francisco on the north to Cambria on the south. The cities of Monterey and Santa Cruz have long recognized that
protection of the unique marine resources within the Sanctuary is critical to the economic vitality and quality of life of their
communities. Monterey Bay, with its world renowned Monterey Bay Aquarium, rich bird and marine resources, recreational
opportunities that include the Santa Cruz Boardwalk, and commanding vistas has become a major tourist attraction.
The Cities of Monterey and Santa Cruz developed and implemented a Model Urban Runoff Program (MURP) in a
cooperative team effort with the Monterey Bay National Marine Sanctuary, the California Coastal Commission, California
Regional Water Quality Control Board-Central Coast Region, and the Association of Monterey Bay Area Governments
(AMBAG) funded by an EPA 319(h) grant. The MURP was developed to address and support a number of environmental
regulations and initiatives that applied to the Sanctuary and adjacent coastal areas including: the Sanctuary's Water Quality
Protection Program, requirements of the Coastal Zone Act Reauthorization Amendment (CZARA), Coastal Commission
plans and policies, the Regional Board's watershed management initiative and basin plans, the State Water Resources
Control Board's Ocean Plan and Nonpoint Source Pollution Control Program, and EPA's proposed draft Storm Water
Phase II Rule (Storm Water Phase II).
The concept of the MURP originated in the Sanctuary's Water Quality Protection Program for Monterey Bay National
Marine Sanctuary-Action Plan Implementing Solutions to Urban Runoff and a State's Urban Runoff Technical Advisory
Committee Report, developed to address Section 6217 of CZARA.
One of the most important drivers in the development and implementation of the MURP was an initiative of municipal
leaders to address the value that the community places on protecting the local creeks, streams, and wetlands and the
Sanctuary's marine biological resources.
Development of Model Urban Runoff Program
A key objective of the MURP was to produce a document that would assist other communities in the development of
their own urban runoff programs by providing an off-the-shelf "how-to" guidebook building on the experience gained
through the development and implementation of Phase I Storm Water Management Programs. A second objective of the
project was the development of Urban Runoff Management Programs forthe cities of Monterey and Santa Cruz that would
address the community's values and achieve early implementation of and compliance with the various regulatory programs
and initiatives.
Representatives from Phase I municipalities who were responsible for the development of their programs and had
gained experience in implementation were consulted throughout the project. They participated in workshops when the
MURP was presented to Monterey Bay municipalities and provided extensive information and examples of material used
349
-------
in their communities, including what worked and what didn't. Woodward-Clyde Consultants, with its extensive experience
in working with Phase I municipalities, was selected to assist the project team in developing and producing the MURP.
The cities used a conceptual framework for developing urban runoff programs appropriate for their individual needs
following the guidance recommended by the MURP. It included four major steps:
• Assessment - Institutional assessment and assessment of environmental resources and sources of pollutants of
concern
• Development - Program management, institutional arrangements and coordination, legal authority, and fiscal
resources
• Implementation - BMPs and model programs for each of the six control measures proposed in Storm Water Phase
II and for control of commercial and industrial activities
. Evaluation - Progress reporting and evaluation, water quality monitoring, and program update.
Periodic meetings were held during the two-year grant period to share, review, and comment on individual city work
products, review progress, prepare and validate the MURP, and discuss early implementation actions. The Sanctuary was
instrumental in the early implementation effort by providing public information and outreach support and developing public
education materials. The Coastal Commission played an active role in formulating strategies for addressing Sanctuary-wide
water quality and land use concerns and providing support in the development of GIS-based land use maps. Project status
reports were presented to the Monterey Bay Regional Stormwater Management Task Force, AMBAG, city councils, and
Monterey Bay area public works officers.
The Model Urban Runoff Program
The MURP contains these four major steps for the development and implementation of an urban runoff program, and
a corresponding appendix containing additional information, examples, models, references, and contacts for further
information.
The Assessment Phase of the MURP describes the importance of information gathering and research to provide an
early survey of the municipalities, current policies, programs, legal authorities, and fiscal resources applied to control urban
runoff. A similar institutional assessment of existing regional storm water, watershed, and other water quality control
programs is recommended to avoid duplication and to identify potential conflicts, opportunities for coordination, and areas
not previously addressed. This phase also provides guidance and methods for (1) describing a community's water
resources and activities that can be impacted by polluted runoff, (2) mapping the storm drainage system, (3) developing
a relationship of pollutant sources/activities to pollutants of concern, and (4) developing a working map to assist in targeting
problem areas or pollutant sources. Coordination with and building upon existing efforts, including joining Phase I
programs, is encouraged.
The Development Phase of the MURP describes the (1) selection of the program management structure, (2) identifies
individual and departmental responsibilities for management of individual program elements (public education, control
measures, or BMPs) and (3) coordination with other internal and external programs and agencies. The legal authority to
ensure implementation of BMPs and achieve compliance with the MEP standard of the Clean Water Act was developed
through use of a model ordinance. The appendices include examples of language for the amendment of Local Coastal and
General Plans as required by the State of California. Revisions to the California Environmental Quality Act checklist were
recommended to provide planners an additional tool to evaluate impacts of runoff from both new development and re-
development. This Phase includes an estimate of staffing resources to implement each element of an urban runoff
program. It also describes the use of assessment districts, storm water utility fees, and other sources of funding program
implementation.
The Implementation Phase of the MURP describes eight program elements including six required by EPA's Phase II
draft regulations as minimum requirements MURP program elements include (1) public involvement/participation, (2) a
350
-------
public education and outreach program, (3) an illicit connection and discharge detection and elimination program, (4) a
municipal operations control program, (5) a construction site discharge control program, (6) a new
development/redevelopment control program, (7) an optional control program for commercial facilities, and (8) an optional
control program for industrial facilities. The MURP recommends objectives, BMPs, implementation activities, and methods
for program evaluation and documentation. This includes measurable goals for each of the eight program elements. The
appendices contain numerous examples of public participation and education; BMPs for residential areas, food service
operations, municipal operations, construction sites, vehicle service facilities, and shopping centers; sample SWPPPs for
corporation yards and construction sites; and reporting forms.
The Evaluation Phase describes methods for (1) determining whether water quality is improving and whether the
efforts and resources are directed at the right source and pollutants of concern; (2) reporting progress using the BMP
measurable goals, and (3) the developing and implementing of water quality monitoring programs and volunteer monitoring
programs. This phase provides and stresses the need for procedures for modifying and updating the urban runoff program
using the evaluation tools.
Each section of the MURP contains an extensive list of references to assist municipalities in obtaining additional
detailed supporting information on how these programs were developed.
Implementation of the Model Urban Runoff Program
Municipalities in California's major metropolitan areas were encouraged, and in some cases required, to file for
NPDES' permit coverage on an area-wide basis. Numerous smaller municipalities are already regulated by Phase I
requirements. There are now approximately 260 municipalities, with a combined population of 29 million, regulated by
Phase I NPDES permits in California.
In California, 76 incorporated places and counties are proposed to be automatically designated and 38 areas outside
urbanized areas that could be potentially designated under Storm Water Phase II. The Monterey Bay Area has 13
incorporated places and counties that would be automatically designated. The MURP will be of significant benefit to a
number of smaller California municipalities, and particularly in the area covered by this project.
Undertaken as part of a 319(h) Grant, this project was required to conduct an outreach effort to ensure early
implementation of urban runoff programs. Two workshops were held in April of 1998 for planning, public works, building,
parks, public information/education, and general municipal operations staff in the Monterey Bay Area. These workshops,
attended by approximately 120 individuals, covered an introduction to urban runoff pollution, and regulatory requirements.
They featured presentations from individuals experienced in the development and implementation of Phase I storm water
management programs and included four break-out workshops covering MURP development and implementation.
The agencies participating in the development of the MURP have undertaken a number of actions, described in the
following paragraphs, to implement the project recommendations.
City of Monterey
During the development of the MURP, the City of Monterey mapped watersheds, major storm drains, key streams,
creeks and channels. They also identified and mapped automotive servicing facilities, restaurants, several industrial sites,
and pest and weed management activities as potential sources of runoff pollution. Fifteen potential sources or activities
that could contribute primary pollutants of concern were identified. The City has adopted a water quality ordinance and
established a monthly storm water utility fee, currently $3.43, to implement its urban runoff management plan.
In cooperation with the Sanctuary and Coastal Watershed Council, the City has also initiated a citizens participation
program to label storm drain inlets and perform volunteer monitoring (Urban Watch Stormdrain Monitoring Program). This
volunteer effort led to the development of a Restaurant Outreach program to educate employees and eliminate pollution.
The City has commenced implementation of BMPs for new and existing sources, conducted water quality monitoring,
distributed public education material, and is currently working to implement its construction site pollution prevention
program.
351
-------
In cooperation with the Sanctuary, the City has obtained a grant to install and test storm water treatment devices at
the Monterey Harbor and Fisherman's Wharf parking lots to determine the effectiveness of removing oil, sediment, and
trash.
City of Santa Cruz
The City of Santa Cruz developed a computer-generated map of watersheds and land uses to identify potential
pollutant sources. The City adopted a storm water ordinance in April 1998, regulating all water entering the storm drain
system, prohibiting illicit discharges and connections, and requiring implementation of BMPs published by the City. The
City has drafted BMPs for vehicle service facilities, retail shopping areas, residential areas, and food service facilities. The
City Industrial Waste Inspectors will conduct initial inspections of 100 vehicle service facilities in 1999 to determine any
actions which must be taken to comply with the ordinance, with second inspections scheduled to formally determine
compliance. The City hosted an outreach presentation of the program and the proposed BMPs for the business community
during its Pollution Prevention Week.
City of Watsonville
The City of Watsonville began implementing a storm water program in 1991, through its industrial facilities Source
Control Program, and completed a bilingual storm drain stenciling program in 1992. Subsequent to the development of
the MURP, the City has completed a review of existing programs and policies, developed a new storm water ordinance,
started an illicit connection program that has sampled 50% of the City's storm drain outfalls, and established a public
education program in cooperation with the Sanctuary. The City plans to implement a municipal, industrial, and commercial
source control program, a targeted educational outreach program, and a construction and new development program.
Monterey Bay National Marine Sanctuary
The Sanctuary's Water Quality Protection Program addresses a number of water quality issues in addition to urban
runoff and targets nonpoint sources of pollution. The Sanctuary's program supports the cities' urban runoff programs by
developing and distributing educational materials on urban pollution and co-sponsoring teacher training workshops with
the Monterey Bay Aquarium. It also collaborates with the City of Monterey on volunteer monitoring programs and public
education. The Sanctuary has prepared a plan for addressing polluted runoff from agricultural lands and has received
commitments from the California Farm Bureau and six regional Farm Bureaus to form a coalition to address this issue.
The Coalition will focus on educating its members on polluted runoff, establishing landowner committees and pilot projects
in several watersheds, and strengthening farm management practices by developing grower self monitoring and serving
as a liaison with the Sanctuary and the Regional Board.
The Sanctuary and the City of Monterey have a cost-sharing agreement, which funds a Sanctuary employee at half
time in return for the development of a City public education program. This agreement is going into its third year, and has
resulted in the development of public education brochures, posters, exhibits, BMP pamphlets, and the Restaurant Outreach
Program. Current work is focused on the development of a Public Service Announcement, a construction site education
program for developers and inspectors, and signage for Monterey's Harbor.
State of California - Coastal Commission and Regional Water Quality Control Board
Implementation of the MURP has been identified as a priority in the California Nonpoint Source Pollution Control
Program's first 5-Year Implementation Plan, which the State Water Resources Control Board and California Coastal
Commission developed pursuant to the Clean Water Act and Coastal Zone Act Reauthorization Amendments of 1990
(CZARA). Key actions that the State will undertake include the distribution of copies of the MURP Guidebook to California
cities and the providing of technical support and training to cities developing Urban Runoff Management Plans using the
MURP.
352
-------
The Coastal Commission intends to review the experiences of Monterey Bay Area cities in implementing the MURP
and coordinating with the Central Coast Regional Water Quality Control Board to revise the MURP Guidebook as needed.
The document will be made available to other small coastal cities by printed copies, CD-ROM, or Internet web site.
Summary and Conclusions
The Model Urban Runoff Program project, which is funded through a 319(h) Grant has provided small communities
in the Monterey Bay Area an excellent opportunity to develop their own urban runoff programs and to develop and validate
an off-the-shelf" how-to" guidebook on development of urban runoff programs. The MURP will potentially benefit over 100
communities in California that will be required to develop urban runoff programs implementing the six minimum control
measures contained in EPA's draft Storm Water Phase II Rule and the requirements of Section 6217 of the Coastal Zone
Act Reauthorization Act.
References
California Coastal Act of 1976. (P.R.C.§§ 30000 et seq.).
California Coastal Commission. 2nd Edition 1996. Procedural Guidance Manual: Addressing Polluted Runoff in the
California Coastal Zone.
California Storm Water Quality Task Force. March 1993. California Storm Water Best Management Practice Handbooks.
California Regional Water Quality Control Board-Central Coast Region. Water Quality Control Plan,
Monterey Bay National Marine Sanctuary. February 1996. Action Plan I: Implementing Solutions to Urban Runoff. Water
Quality Protection Program for Monterey Bay National Marine Sanctuary.
Monterey Bay National Marine Sanctuary. 1998. Annual Report for the Monterey Bay National Marine Sanctuary.
City of Monterey, City of Santa Cruz, California Coastal Commission, Monterey Bay National Marine Sanctuary,
Association of Monterey Bay Area Governments, Woodward-Clyde, Central Coast Regional Water Quality Control Board.
July 1998. Model Urban Runoff Program, A How-To Guide for Developing Urban Runoff Programs for Small Municipalities.
State Water Resources Control Board and California Coastal Commission. July 1999. Draft California's Nonpoint Source
Pollution Control Program. Volumes I and II.
U.S. Department of Commerce, NOAA, U.S. Environmental Protection Agency-Office of Water. January 1993. Coastal
Nonpoint Pollution Control Program-Program Development and Approval Guidance.
U.S. Environmental Protection Agency. January 9, 1998. National Pollutant Discharge Elimination System-Proposed
Regulations for Revision of the Water Pollution Control Program Addressing Storm Water Discharges: Proposed Rule
U.S. Environmental Protection Agency. January 1993. Guidance Specifying Management Measures for Sources of
Nonpoint Pollution in Coastal Waters.
353
-------
New Stormwater Treatment BMPs: Determining
Acceptability to Local Implementing Agencies
Gary R. Minton
Resource Planning Associates
Seattle, Washington
Paul Bucich
Pierce County Public Works
Tacoma, Washington
Mark Blosser
City of Olympia Public Works
Olympia, Washington
Bill Leif
Snohomish County Surface Water Management
Everett, Washington
Jim Lenhatt
Stormwater Management, Inc
Portland, Oregon
Joseph Simmler
Entrance Engineers
Bellevue, Washington
Steven True
Vottechnics
Federal Way, Washington
Over 200 local governments in the Puget Sound watershed of western Washington require new developments
to install Stormwater treatment systems. The retrofitting of existing developments is also often required. With the guidance
of the State of Washington', each local jurisdiction developed a list of approved treatment technologies, mostly public
domain technologies such as wet ponds and swales, in the mid-1 990's. Having an approved list raises the issue of how
to add new unapproved technologies, in particular the manufactured technologies such as swirl concentrators and drain
inlet inserts. A protocol is needed by which local jurisdictions can determine acceptability of "unapproved" treatment
technologies. This paper presents a protocol* recently developed to assist local jurisdictions in the Puget Sound
watershed.
Protocol Structure
The protocol has four parts:
1. Performance criteria to compare currently unapproved treatment technologies with currently approved treatment
technologies.
2. Types of data to be used in the evaluation and methods of sample collection.
354
-------
3. Factors that are to be considered in the evaluation other than performance.
4. Content of the report provided to the local jurisdiction by the proponent.
Method to Compare Performance
There are many problems inherent in the development of a protocol, not the least of which is that there is no agreed-
upon definition as to what constitutes acceptable performance against which to judge equivalency. An informal "standard"
that is much discussed and that has been adopted by some jurisdictions outside the Puget Sound region is 80% Total
Suspended Solids (TSS) removal over all storms less than a specified event.
A one-number standard of 80% may be inconsistent with two complementary observations about the performance of
stormwater treatment systems. The first observation3 is that removal efficiency of sand filters is directly related to influent
concentration: the higher the influent concentration the greater the efficiency. This relationship may hold for other
treatment technologies. The second observation' in Schueler (1996) proposes that there is a lower limit to the effluent
concentration.
Further, a reasonably strong and direct relationship exists between runoff rate and TSS concentration5. This is
particularly germane to our region with its mild storm intensities. A comparison of data from our region to areas with higher
average rainfall intensities indicates that we tend to have lower TSS concentrations in untreated stormwater. It is therefore
possible that over an aggregate of storms, we cannot achieve 80% TSS removal in the Puget Sound region.
A more appropriate method to compare approved and unapproved technologies may be to relate efficiency to the
influent concentration. This approach allows the pooling of data from sites with different pollutant concentrations.
The protocol presumes that if we are satisfied with the technologies that are currently approved, we should approve
an unapproved technology with similar performance. Therefore, the starting point is to identify the performance of currently
approved technologies: swales, wet vaults, wet ponds, constructed wetlands, and sand filters. The results are presented
in Figures 1 through 3. The data points in the figures are of flow-weighted composite samples from individual storms. Only
data from studies conducted in the maritime climate of the Pacific Northwest are used.
Although this is a large region, from northern California to southwest British Columbia, comprehensive studies have
been conducted only in western Washington. Data considered acceptable are from two wet ponds, three grass swales,
and two sand filters. The protocol' provides detail on the studies that were reviewed. The reasons for using only Pacific
Northwest data differ with the technology. With wet ponds there is concern about the possible effect of differences in
regional climates on effluent quality. For sand filters, possibly because of differing sand chemistry, filters used elsewhere
may be able to remove dissolved phosphorus6and zinc', a capability our sands do not have. Swales studies conducted
elsewhere do not provide the information needed to judge whether they were sized to criteria similar to that used in our
region.
Figure 1 for TSS is used in all comparisons, and is the first performance requirement that must be met. If the receiving
water is nutrient sensitive, Figure 2 is also used. If the water body is of regional significance because it supports salmon,
a central issue in our region, Figure 3 is also used. Zinc was selected to represent all toxics primarily because it is the only
toxic with influent concentrations that are commonly high enough to allow for the evaluation of efficiency.
Each figure has a "Line of Comparative Performance," the origins of which are discussed later. Each line is drawn
so that most of the data points fall above and to the left. This is called the region of acceptable performance. The data
points of the unapproved technology under consideration would be plotted using the same format. If most of the data
points fall above and to the left of the "Line", it can be concluded that the candidate technology is reasonably equivalent.
What constitutes "most" is up to the local jurisdiction. The protocol does not specify a hard rule but offers these
suggestions as to the percentage of data points falling above the "Lines": TSS, 90%; phosphorus, 80%; and zinc, 90%.
Note that low efficiencies generally occur at low influent concentrations.
355
-------
FIGURE 1 TSS Evaluation
PNW data - Individual storms
-80%
80 100 120 140
Influent concentration (mg/L)
FIGURE 2 Phosphorus Evaluation
PNW data - Individual storms
Swale
Wet pond
Sand filter
-80%
150 200 250 300 350
Influent concentration (ug/L)
How were the "lines" derived in Figures 1 through 3? Judgment was used to select a point of "irreducible"
concentration, and to draw the line from this point to the upper right. Regarding "irreducible" concentrations: for TSS Tom
Schueler has proposed4 20 to 40 mg/L, depending on the treatment BMP. A value of 15 mg/L was selected because the
data in Figure 1 suggest that 15 mg/L is attainable. For TP (Figure 2), Schueler4 proposed a concentration of 140 to 330
ug/L, depending on the treatment system. A lower value of 90 ug/L was selected for the same reasons as with TSS.
Schueler4 drew no conclusions with regard to zinc. However, his analysis4 suggests "irreducible" concentrations of 36 ug/L
for ponds and wetlands, so 35 ug/L was selected. It should be noted in Figures 1 through 3 that most of the incidents of
low or negative efficiency occurred at low influent concentrations.
356
-------
FIGURE 3 Zinc Evaluation
PNW data - Individual storms
100%
50 100 150 200 250 300 350
Influent concentration (ug/L)
400
450
500
Attempts were made to identify a line of best fit from which confidence limits could be derived under varying
assumptions such as excluding all data below a particular influent concentration and/or removal efficiency. However, the
relationships were so poor as to make confidence limits meaningless. It was concluded that basing the lines on some sort
of statistical construct would give an air of rigor that is unwarranted at this time. Therefore, the lines were drawn using best
professional judgment. It is expected that the Lines of Comparative Performance will change with time as additional data
are collected.
Figures 1 through 3 can be used for inlet inserts if these pollutants are being considered. If, however, an insert is being
considered for the removal of petroleum hydrocarbons, it must be compared to the effectiveness of oil/water separators.
A graph comparable to Figures 1 through 3 is not provided for oil/water separators because of the lack of data. The
criterion commonly used in the Pacific Northwest is that the concentration of individual samples shall not exceed 15 mg/L.
This protocol could be used to generate data to compare to this criterion.
Data and Data Collection
It is the responsibility of the proponent of the unapproved technology, either the manufacturer or the development
engineer, to obtain the required data. The protocol identifies the minimum requirements. Local jurisdictions are free to
request more data. The protocol specifies the requirements for three sources of data: field studies with real storms, field
studies with artificial storms, and laboratory tests. The protocol calls for discussion of advantages and disadvantages of
each method. It is left to the local jurisdiction to decide the weight to place on each source of data. However, the protocol
recommends that the local jurisdiction not rely solely on laboratory tests, particularly when considering the removal of
dissolved pollutants or oil/grease and related products. It also recommends a size distribution of sediment for the
laboratory tests.
The protocol is very specific with regard to the types and amounts of data that are to be collected. This aspect of the
protocol is summarized in Table 1.
Table 1. Data Requirements
357
-------
Item Requirement
Number of test sites One to three sites: medium density residential, retail commercial, and non-retail commercial.
Number of sampled storms A minimum of 10 per site, total of 30 if one site
Storm depth 50% to the design storm depth (1.35' to 1.70")
Runoff duration 50% to 200% of the mean annual storm duration (7.5 to 30 hours)
Average rainfall intensity 50% to 200% of the mean annual intensity (0.02 to 0.08'Vhour)
Type of samples Flow-weight composite
Minimum aliquots per storm 10
Storm volume sampled Samples are to be taken over not less than 75% of the total volume of each storm
Parameters TSS, pH, total zinc, copper, and cadmium, TP, BAP, and TKN. include dissolved metals and phosphorus if it
is claimed that the technology can remove dissolved constituents. With catch basin inserts where the sole
objective is the removal of petroleum hydrocarbons, measure oil/grease, TPH, TSS, and pH.
Additional At the end of the test period, the sediment accumulated in the treatment system shall be removed, quantified,
and anlyzed. The sediment shall be evaluated for total dry weight, moisture content, particle size distribution,
organic content, and zinc. Use ASTM wet and dry sieve test procedures to analyze the particle size distribution.
Also determine the amount of floatables, i.e. litter, and petroleum products.
The protocol states that for a data point to be used in the analysis of efficiency, the influent concentration of the
parameter shall either be at least ten times its detection limit, or be greater than the "irreducible" concentration, whichever
is greater.
Efficiency is to be calculated three ways.
Method 1: Removal in each sampled storm calculated as:
100 (average influent - average effluent)/average influent
Method #1 is required because it provides the data points to plot figures like Figures 1 through 3.
Method 2: Aggregate removal of the sampled storms as:
100(A-B)/A
Where: A = (influent cone. Storm 1 )(flow Storm l)+(influent cone, of Storm 2)(flow of Storm 2)
+ . . . (influent cone, of Storm N)(flow of Storm N)
B = (effluent cone, of Storm 1 )(flow of Storm 1 )+(effluent cone, of Storm 2)(flow of Storm 2)
+ . . . (effluent of Storm N)(flow of Storm N)
Method #2 is specified because it provides an overall efficiency of removal over the period of the research. If the
amount of sediment that has accumulated in the bottom of the treatment facility has been determined from the separate
lab test (See Section A.4), another calculation can be done to check the above estimate of efficiency. This second
calculation is done as follows: subtract B from A, and then compare this difference to the amount of sediment determined
from the separate laboratory test described in Section A.4. These calculations can also be done for zinc and phosphorus.
Method #3 Efficiency based on geometric mean:
100(A-B)/A
Where: A = geometric mean of all influent samples
B = geometric mean of all effluent samples
Method #3 is specified because it is the most correct method of calculating efficiency, although it has been used
sparingly to-date. All influent and effluent data from multiple sites can be pooled.
358
-------
Consideration of Factors Other than Performance
The protocol suggests that the local jurisdiction consider additional factors in making a decision should the technology
pass the first requirement of acceptable performance. The protocol recommends the following factors, although the local
jurisdiction is free to include other factors.
. Site characteristics
• Constructability
• Reliability
• Robustness
. Receiving water sensitivity
. Groundwater risk
. Operation and maintenance
• Habitat creation
. Thermal effect
. Aesthetics
. Recreational use
• Community acceptance
It is left to the reviewer to place a weight on the relative significance of each factor and to develop a scoring system.
For example, the factors could be categorized and weighted as: "critical/necessary, ""important,"and "desirable." A relative
score, say 1 to 10, could be identified for each factor, and multiplied by the corresponding weight of each of the categories.
Content of the Applicant's Report
The proponent of the technology is responsible for producing a report for the local government conducting the
evaluation. The protocol provides a very detailed list of items that are to be included in the report.
Explanation of the technology, such as:
• How it works, how it removes pollutants
. Where it is currently being used
. Available models
• Treatment and hydraulic capacities of each model
• Documentation of the treatment and hydraulic capacities
• Sediment storage capacities of each model up to the point of recommended maintenance
. Floatables storage capacities up to the point of recommended maintenance
• Sizing methodology
359
-------
• Materials used in the construction of the product
• Recommended maintenance procedures and frequencies
Documentation of the field studies:
• A description of the test site including: total acreage, total impervious acreage, a description of landscaping if
relevant, the acreage draining to the device if it differs from total acreage. A description of the drainage system
including the size of the sumps, and whether the sumps were cleaned prior to or during the test period
. A description of the model used
• Complete drainage calculations showing the calculations to size the treatment device
. All raw data including laboratory reports. All data are to be reported including rejected data with an explanation for
the rejection
. Statement from the analytical laboratory certifying that the appropriate procedures were followed in the preservation
and analysis of the samples
• Calculation of efficiency of each storm by comparing influent and effluent concentrations
• Calculation of the efficiency for all storms by comparing the total aggregate inflow loading of all storms to the total
aggregate outflow loading for all storms
• A graphic of data points showing influent concentration versus efficiency for each storm sampled for TSS, zinc, and
phosphorus. Plot all data, including rejected data, with an explanation for the rejection
• Start and end times of the precipitation and runoff periods of each sampled storm.
. Start and end times of the sampling period of each sampled storm
. Antecedent conditions during the 72 hours prior to each sampled storm
. Total rainfall depth of each sampled storm
• Total runoff volume of each sampled storm
. Runoff volume that occurred during the period of sampling of each sampled storm
. Total rainfall during the period of all of the sampling, i.e. from the first storm sampled through the last storm sampled
• Total runoff that occurred during the period of all of the sampling, i.e. from the first storm sampled through the last
storm sampled
. If artificial storms are used, identify the method and application rates of water and translate those rates into
corresponding rainfall intensities
• Statement of certification signed by the proponent indicating that the protocol was followed
Additional considerations with inserts for drain inlets:
. Data showing the effect of accumulated litter and leaves on performance, flow capacities
• The point in the maintenance cycle that the field tests were run: i.e. were the units tested "fresh," without prior field
exposure or were they in the field for some time
360
-------
Documentation of laboratory studies:
• Description of the composition of the test water
. The size distribution of TSS in the influent and effluent
. The test flow rates
. The performance at each flow rate
• Mass balance of influent, effluent, and collected sediment
Conclusions
The protocol offers a reasonable and defensible approach that provides rationale for the consideration of technologies
that are not currently approved for use in new developments as well as for public projects. The protocol is most suitable
for the maritime climate of the Pacific Northwest. It is anticipated that the protocol will change over time, particularly as the
data base for approved technologies becomes more extensive and as we learn from its use. More performance data are
needed for public domain technologies located in our region, in particular wet vaults and constructed wetlands for which
there are currently no data.
References
1. Washington Department of Ecology, 1992, Stormwater Management Manual for the Puget Sound Watershed.
2. Minton, G., M. Blosser, P. Bucich, B. Leif, J. Lenhart, J. Simmler, and S. True, in press, Protocol for the Acceptance
of New Stormwater Treatment Technologies in the Puget Sound Watershed, APWA Washington Chapter.
3. Bell, W., et al., 1995, "Assessment of the Pollutant Removal Efficiencies of Delaware Sand Filters", Alexandria, Va.
4. Schueler, T., 1996, Irreducible Pollutant Concentrations Discharged From Urban BMPs, Tech Note 75, Watershed
Protection Techniques, 2, 2, 369.
5. Roundtree, L, 1995, The Relationship Between Rainfall Intensity and Urban Stormwater Constituent Levels: A
Comparison of Puget Sound to Other United States Climates Using Pollutographs, Master's Thesis, University of
Washington, Seattle, Washington.
6. City of Austin, 1996, Evaluation of Nonpoint Source Controls, 319 Grant Project.
7. Wellborn, C.T., and J.E. Veenhule, 1987, Effects of Runoff Controls on the Quantity and Quality of Urban Runoff at
Two Locations in Austin, Texas, U.S. Geological Survey, Water - Resources Investigations Report 87-4004.
361
-------
By Any Measure.. .
Thomas R. Adams, P.E., Vaikko P. Allen and Andrea Perley
Vortechnics, Inc.
Portland, Maine
The introduction of a diverse array of stormwater quality management tools in the last few years has created problems
for the growing number of individuals and organizations who would like to compare the performance of these tools.
Comparison is complicated by differences in treatment capacities, targeted pollutants, and treatment approaches. Several
methods of evaluation have emerged in response to the need for verification of theoretical performance predictions: yet
none of these "yardsticks" are appropriate in all situations and results from each are often not readily comparable to results
from other measures.
Complicating the matter further is the confusion regarding what is being compared. In some cases, a technology will
be compared with another technology. In other cases, the technology is compared to a performance standard. (Analogy:
My maple syrup may be better than your maple syrup, but does that make it Grade A maple syrup?)
The confusion stemming from this is greater than meets the eye. For example, many specifiying engineers and
hydrologists want to meet a performance standard of 80% TSS removal on an annual average basis. They go to guidance
manuals and product manufacturers seeking something that will meet the standard. As the selection process develops,
they grapple with cost, maintenance, the availability of land needed, and many other issues under the heading of "cost-
effectiveness" for their clients. But by the time the selection has been made, it has become more a question of who has
the better maple syrup rather than whether or not the selected product meets the standard. That is partly because 80%
TSS removal on an average annual basis is virtually an impossible standard to achieve.
It is not the purpose of this paper to propose the adoption of one standard over another.
It is the purpose of this paper to review the merits of various measures of performance and, more importantly, to
stress that any monitoring program that attempts to measure performance of a stormwater quality management system
should begin with a consideration of how the observations will be reported.
Any monitoring program will consist of:
. Sampling
• Testing
• Reporting
Often, researchers start with the selection of samplers and procede with questions of deployment and maintenance
of the samplers. Only when samples have been collected are the questions of testing, and eventually reporting, given much
thought. We propose that the reporting aspect, even though it is "last" chronologically, should dictate how the testing and
sampling are done. For example, if the report is considered most informative when its focus is the mass of pollutants
removed, as is often the case, there is no need for samplers. To get those results usually requires only the very simple
task of periodically measuring accumulations of precipitated sediments or floatable petroleum products (or other pollutants)
and doing simple arithmetic calculations to determine the mass removed. We know of projects that went to the
extraordinary effort that it takes to procure, set up, operate, and maintain automatic samplers to obtain influent and effluent
concentrations, only to "back into" an estimate of the mass of pollutants removed by a convoluted set of "volume-times-
concentration" calculations. This is a classic case of doing something the hard way, not to mention the very expensive
362
-------
way. So the first rule of thumb for any monitoring project is to decide first how to report the results. Then design the
sampling and testing around the information that is to be reported.
In this paper, we will present the following four "measures" of performance from which to choose in making those
decisions. We consider these measures to be the current "status quo of the art."
• Mass of pollutants removed
• Event mean concentration (EMC)
. Lines of Comparative Performance (Minton, et. al.)
. Plotting efficiency versus operating rate
First, we would like to discuss in some detail some of the broader issues involved in monitoring stormwater treatment
systems and measuring and reporting on their performance.
Setting
The setting is nearly always in the field or in a laboratory. We feel that the many benefits of testing in a laboratory are
generally underestimated by technical professionals and non-professionals alike. What we refer to as the setting tends
to be pre-determined well before a study gets underway. Since the setting tends to influence the important decision of
what performance measure is best, this cursory overview is provided for perspective.
Field testing
The drawback with field testing is that it cannot be replicated very well. Every site is different. Every storm is different.
There are "wet years" and "dry years." There are seasonal variations that can produce easily treated heavy sediment loads
in winter and spring; hard-to-treat loadings, such as pollen and grass clippings, in summer; and moderately treatable
loadings (leaves etc.) in the fall. It is poor science to compare the results of a field test of any treatment system to the
results of a test of the same system at a different time and place.
Still, field testing has tremendous appeal because the stormwater and the sediments are "real."
. Field testing of individual facilities is usually adopted to evaluate the facility's performance in comparison to a
performance standard (such as 80% TSS removal) to see if it "measures up." This setting is the simplest and most
common, and is adaptable to any of the measures. The treatment system can be set up to treat all runoff or it can
bypass flows that exceed the treatment capacity.
• "Side-by-side" field testing of several different facilities is increasingly popular, at least in concept. But, to our
knowledge, this approach has not yet been successfully implemented. A key element is the design of the "flow-
splitter" that takes all of the runoff and "splits" it into an equivalent discharge to each treatment system. The easy
part of the design of the flow-splitter is achieving equal flow rates of wafer to each of the facilities (and even this
"easy part" is not always all that easy). The hard part is getting equal discharge of pollutants to each of the
facilities.
Whether testing an individual system or multiple systems, researchers have the potential to learn something very basic
and very important from field testing that has, to our knowledge, eluded researchers to date. That is the determination of
an appropriate threshold for bypassing peak flows. Consider a site that is estimated by conventional runoff modeling to
discharge stormwater runoff at a rate of 3 CFS in a 1 O-year storm. Consider further two proposals for treatment. One
claims efficiency of 90% and a treatment capacity of 1 CFS. The other claims 80% and a treatment capacity of 3 CFS.
Which treatment option should be selected? If the prevalent standard of 80% is in place, the "safe" choice would seem
to be treating all runoff from a lO-year event with 80% efficiency. But what if price is a factor? Maybe the system with 1
CFS capacity costs less. Even at the same cost, would 90% efficient be preferable to 80%? And with that consideration
363
-------
comes the $64,000 question. Will the system that claims 90% efficiency with a capacity of just 1 CFS even meet the
standard of 80% overall? How much of the pollutant load will bypass the system altogether?
Some proponents of small-flow/high-efficiency technologies have stated that 90% of all storms are less than 1 inch
of total rainfall and, therefore, treating 90% of all rain at 90% efficiency will yield a net annual removal of 81%. This
argument is fundamentally flawed. It assumes that the 90% of rain from small storms carries 90% of the pollutants. This
is simply not the case. The rate of mobilization of virtually all pollutants depends on rainfall intensity, not depth of rain.
Therefore, it is important to treat high-intensity flows resulting from the infrequent event, which tend to carry a
disproportionately high pollutant load.
If 1 inch of rain falls in 24 hours, virtually any system that is reasonably proportioned, designed and, of course,
maintained for the treatment of stormwater will do a good job. Efficiencies of TSS removal should be in the 90% range if
the runoff is fairly dirty with silty-to-fine sandy sediments.
It is questionable, however, as to whether or not all of the runoff would be dirty if the rain that produced the runoff
totaled 1 inch and fell over a 24-hour period. Intuitively, the "last flush" of such a storm would be very clean. But even the
first flush may be very clean in comparison to what it would be if 1 inch of rain fell in 1 hour. This highly variable "dirtiness"
gives rise to another interesting question when trying to measure efficiency. That is the question of how to account for the
inevitable reduction in treatment efficiency when the water to be treated is clean in the first place. No treatment system
can remove what is not there. So it has been argued that some accounting should be made for the fact that there is some
lower limit to the physical treatment that can be provided. Minton's "Lines Of Comparative Performance" (see figure 2.)
take this important consideration into account and are discussed later.
The "Double Whammy" of the "2-Month Storm"
Infrequent, high-intensity storms are important to the effective treatment of stormwater for two reasons:
• Over time, the higher intensity of less-frequent rainfalls, and the resulting higher stormwater runoff velocity, is what
transports most of the sediment off of streets.
• The treatment facility is overloaded by the high flow of water that is transporting the sediment at the same time that
most of the sediment is being transported to it.
Schueler and Shepp (1993) performed monthly observations documenting a random pattern of accumulation and loss
of sediment in a study of 17 different oil/grit separators in Maryland. Overall, the losses of sediment "outnumbered" the
accumulations. In other words, the observed systems lost previously accumulated sediments once every two months.
We have inferred from their work that the "2-month storm" is a reasonable benchmark for stormwater treatment. To be
"measurably" better than the poorly reputed conventional oil/grit separator, a system or a facility should be able to
demonstrate, at a minimum, that it can continue to function in the 2-month storm. If a system is found to lose sediment
in a 2-month storm, it should not be considered any better than conventional technology. Similarly, if a system needs a
bypass to protect it from washing out in 2-month storms, it should be considered only marginally better than conventional
oil/grit separators. Bearing in mind that high flows transport much of the total sediment, treatment systems should be able
to handle more than the 2-month storm without bypass. Otherwise, much of the total sediment load may be discharged
to the receiving waters that the system is supposed to protect.
Clearly the statements of the preceeding paragraph are more of a hypothesis than a statement of fact. One way to
validate or invalidate the hypothesis is described in the following section on side-by-side testing in the field.
Side-By-Side Testing
Testing stormwater facilities "side-by-side" has recently become a very popular idea. The premise is that a well-run
comparison of systems treating "the same stormwater and the same pollutants at the same rates of flow" will go a long
way to reduce the tremendous "scatter" in the data that has been obtained to-date by testing individual systems at different
sites. If two systems are evaluated at different sites, even if the study is carried out by the same researchers using the
364
-------
same protocol, the results will probably not be comparable. Every site is different, and from the point of view of stormwater
treatment, differences that appear slight can actually be significant. We have observed a dozen systems installed on a
single site (a large shopping mall parking lot) which were specified by the same engineer and installed by the same
contractor at more or less the same time and, of course, subjected to the same weather. They exhibited decidedly different
results when we measured sediment accumulations in the systems. The sediment depths ranged from a light dusting to
accumulations of over two feet in less than a year.
So it is important in "side-by-side" testing that there be just one flow stream to the two (or more) systems being tested
and that the flows be split, so that each system gets exactly the same rate of flow and the same pollutant concentration
at all times.
The main benefit of "side-by-side" testing is that it can provide an answer to the question, posed earlier, of whether
it is preferable to have, the arrangement should be as shown in Figure 1.
Influent Sampling
Point 1
80% TSS Removal Efficiency
Treatment Capacity of 6.0 cfs
Effluent Sampling
Point 1
High Flow
Bypass 1
90% TSS Removal Efficiency
Treatment Capacity of 1.0 cfs
Flow
Splitter
Effluent Sampling
Point 2
Influent Sampling
Point 2
High Flow
Bypass 2
Figure 1. Recommended Arrangement For Side-by-Side Field Testing.
By sampling at points 1 and 2, the overall efficiency of the treatment system and bypass can be assessed objectively.
Also, the question of "Which is the better system?" is answered. There are two shortcomings:
. Lack of repeatability. If one system gets 80% efficiency overall and the other gets 70% overall during one year of
testing, there is no assurance whatsoever that either number will be repeated the next year. The test results should
be regarded as indicative of performance. They are certainly not an assurance of performance over time. Such is
the inconsistency, or "noisiness," of stormwater treatment data. A study like this should be conducted over a period
of no less than two years. If the second year's results are reasonably close (in terms of statistical correlation) to the
365
-------
first, it can be considered complete. If not, it would be tempting to average all results. We feel that it would be poor
science to do so, however. With the seasonal variability of stormwater pollutant loadings, one year's results will
produce a single data point. A second year's results will produce a second data point. Many people seem to regard
each storm result as a single data point, but as long as standards continue to be based on "average annual removal
efficiencies," that is simply not the case.
. Variability with other sites. We have already mentioned the differences from one site to the next. The basic premise
of side-by-side testing is to determine relative performance of two or more systems (i.e., which is best). As long as
such a study is limited to this premise, the variability from one site to the next will not be a problem. But we know
from experience that any "study," even the most cursory, tends to be overly generalized. We can only caution
against doing so.
Laboratory Testing
Testing stormwatertreatment systems in a laboratory setting offers some very significant advantages over field testing.
• It is repeatable and demonstrable.
• It is more productive in the sense that decades of rainfall can be simulated in a matter of days.
• It is more economical in terms of labor, sampling equipment, and flow-metering equipment costs.
Laboratory testing achieves these benefits by controlling operating rates, particle sizes, and pollutant loading. When
influent concentrations are very low, removal efficiency will be low; but for concentrations that are generally recognized
as representative for stormwater, all concentrations tend to produce comparable removal efficiencies.
In the lab, a set of tests can be run using one particle size (at representative concentrations) at operating rates from
zero to the system's capacity. At the conclusion of these tests, a curve can be drawn plotting efficiency versus operating
rate on the y-axis and x-axis, respectively. Such a curve typically slopes downward to the right, reflecting reduced
efficiency and higher operating rates. Any point along a constructed curve should be reasonably reproducible when using
the same influent sediment load.
Subsequently, a whole family of other curves can be constructed using different particles. Also, to more closely
simulate "typical" sediment, a graded sediment sample can be developed and tested in the same way.
Laboratory testing should not be considered the "last word" in documentation of a system's performance, but can be
considered a "benchmark" which is very useful in comparing systems operating at flow rates up to their capacity. Some
field testing, where it is feasible, should supplement the work in the lab and, as previously discussed, side-by-side field
testing is the only way to determine the impacts of bypasses on different systems.
The Four Most Common Measures of Performance
1. Mass of Pollutants Removed
This is easily the simplest approach to stormwatertreatment measurements in the field. By measuring the depths of
sediment accumulations in the facility on a periodic basis, it becomes a simple arithmetic exercise to calculate the volume
and mass of sediments removed by the system.
Additional information is made available by this measurement. It may be recalled, from our earlier discussion of the
2-month storm, that Schueler and Shepp used measurements of sediment accumulations to document the poor
performance of conventional oil/grit separators.
Researchers should consider using the same approach for the newer technologies that have come along since their
important work was published. The approach can be made even more informative by correlating observations to such
366
-------
things as activities in the drainage area (e.g., winter sanding, sweeping, a spill, etc.) or meteorological events such as
observed rainfall intensities or precipitation depths.
2. Event Mean Concentration (EMC)
These are sometimes referred to as "flow-weighted" or "flow-based composite" samples. They are nearly always
obtained using automatic samplers, a flow-meter and a flow totalizer that arithmetically converts the flow rate measured
by the flow-meter to flow volume over time and keeps track of the volume.
The sampler receives a signal that causes it to take a sub-sample when a programmed volume of flow is measured.
For example, one sub-sample might be taken every 200 cubic feet of flow through the system. Over the course of the
storm, all sub-samples would be combined into one large sample container from which the concentration will be obtained
that represents the flow- weighted average for the entire storm.
Without a flow meter, the samplers could be set up to take a "time-based composite" sample; i.e., to sample every 30
minutes. Flow-weighted samples are much more representative, as a simple example will show. Consider a volume of
1,000 gallons with a uniform concentration of 300 mg/l flowing at a uniform rate past the sampling point in 15 minutes,
followed by half as much volume (500 gallons) with a 100 mg/l concentration flowing by in the next 15 minutes. The correct
representation of the concentration would be calculated as:
(I.000x300) + (500*100)
1,000 + 500
Flow-weighted sampling will more accurately reflect this. For example, if the sampler were programmed to pull a
sample every 500 gallons, then 2 samples at the higher concentration would be taken and just one at the lower
concentration. The average concentration would be calculated as:
Time-based sampling would, if the programmed time interval were 15 minutes, take one sample with a concentration
of 300 and another with a concentration of 100, and the average would be calculated as:
300 + 100 =200
2
Automatic samplers that can take flow-based composites have become a very valuable tool for sanitary engineers
measuring concentrations of pollutants in wastewater. We believe that they have been too quickly applied to stormwater
monitoring without regard for some of the inherent differences. Waste streams have "highs and lows" of both flow rate
and concentration, but they are not nearly as wide as the variability of stormwater, which can change from flow rates of
zero to a deluge in a matter of minutes and concentrations that can also exhibit a minimum of zero. These "spikes" can
cause very brief periods of negative efficiency if a system is prone to washing out (as stormwater systems were shown
by Schueler and Shepp to do regularly). If a wash-out occurs, it is an important phenomenon to note, but the briefly
elevated concentration in the effluent will be "composited" with the rest of the (presumably lower) effluent samples. This
will reduce the "event-mean-concentration," but will not reveal that a washout has occurred. Noting washouts, and the
flow rate that caused them, is a very important aspect of a stormwater monitoring program; but they are not likely to ever
be revealed by EMC data.
The second drawback of EMC data is that when influent concentrations drop to very low levels that cannot be further
reduced by physical treatment, the efficiency, as measured by EMC's, will be reduced. This tends to obscure the fact that
higher efficiencies can be achieved when they need to be achieved; i.e., when influent concentrations are higher.
3. Minton's "Lines of Comparative Performance"
367
-------
It is widely acknowledged that there is a lower limit to the capabilities of physical treatment systems for stormwater.
This means that it is very unlikely that effluent concentrations would ever be zero. It also means that very low
concentrations would not be significantly reduced.
Minton et. al. has proposed the following mathematical expression to describe this lower limit:
Influent - LowerLimit
Influent
If the lower limit is 20 mg/l, then a plot of this expression is that shown in figure 2.
Performance Standard = (Influent Concentrati9n - Lower Limit) / Influent Concentration
c
-------
Pollutant removal efficiency vs. operating rate for various particle sizes
250 Micron
150 Micron
100 Micron'
75 iMirisnp
63 Micron
20 40 60 > 80
Percent of maximum opeating rate
100
Figure 3. Removal Efficiency versus Operating Rate.
Field data is less likely to fit the relatively tight curves that can be generated in the lab. At the same operating rate,
you may have vastly different influent concentrations, particle gradations, organic content, etc., depending on such factors
as the time of the storm, anticedent dry period, and time of year. Removal efficiency is a function of all of these factors
combined.
We feel that this performance measurement technique and presentation is the most informative. Its repeatability under
controlled conditions makes it ideal for comparing one system to another. Certainly, if one system's performance curve
on 100-micron particles, for example, is higher at all flow rates than another, it could reasonably be judged to be the higher
efficiency system. If the curves are similar at low rates of operation, but either system drops down to zero efficiency at
some higher flow rate, that flow rate should, of course, be considered the peak capacity for that system. This approach
cannot show compliance with any standard for a stated percentage of TSS removal on an annual average basis.
Conclusion
To our knowledge, these four measures represent all of the techniques that have been used to measure the
effectiveness of various stormwater treatment systems.
Measuring sediment accumulations in the field provides a good deal of useful information on mass removals and the
ability to retain (or fail to retain) previously captured pollutants during periods of high flow. This approach costs very little
to implement.
Event-mean-concentrations are the most widely accepted measure, but may not report all efficiencies and will almost
certainly allow any failures to go undetected. This approach requires the use of automatic samplers at considerable cost,
in terms of both time and money.
Minton's Lines of Comparative Efficiency are more fair to the treatment system because they account for the inability
of any system to remove pollutants that are not present (or present in very low concentrations). If EMC data is collected
to plot against the lines, then there are the same drawbacks of cost and automatic samplers allowing failures to go
undetected. Both of those drawbacks can be overcome, but only with a very dedicated effort to take samples manually.
Taking manual samples throughout the duration of all storms is very time-consuming and unpleasant work. For that reason,
it is almost never done.
369
-------
Plotting efficiency versus operating rate, whether in the field or in the laboratory, is arguably the most informative
approach. In the field, automatic samplers are used (with individual samples in individual bottles and not composited), so
there are those costs to consider. In the lab, samplers are needed but the construction of a model treatment facility, and
the pumps and tanks to handle the required flow rates and volumes of water, will more than offset that cost saving.
Since none of these measures provides an ironclad confirmation that the widely prevalent standard of 80% TSS
removal is being met, we submit that a different standard should be adopted by stormwater management jurisdictions.
References
Minton, Gary, M. Blosser, P. Bucich, B. Leif, J Lenhart, J Simmler and S. True, in press, Protocol for the Acceptance of
New Stormwater Treatment Technologies in the Puget Sound Watershed, APWA Washington Chapter.
Schueler, Tom and David Shepp, The Quality of Trapped Sediments and Pool Water Within Oil Grit Separators in
Suburban Maryland, Metropolitan Council of Governments, 1993 (revised).
370
-------
Stormceptor Hydrology and Non-point Source
Pollution Removal Estimates
G. Bryant
Stormceptor Canada
Etobicoke, ON, L3T 3K2, CAN
R. Grant
New England Pipe Wauregan, CT, 06387, USA
D. Weatherbe
Donald G. Weatherbe Associates
Mississauga, ON, L4X 1 H7, CAN
V. Berg
Stormceptor Corporation
Rockville, MD, 20852, USA
Abstract
A model has been developed to estimate total suspended solids (TSS) removal using Stormceptor, an oil/sediment
separator. The model was based on a commonly used, continuous simulation model United States Environmental
Protection Agency Stormwater Management Model (USEPA SWMM) for hydrological processes. The suspended solids
loading was estimated using build-up and wash-off equations. The solids were assumed to be distributed into five particle
sizes for settling calculations. Simulations were conducted using various assumptions of loading and settling velocities to
determine the sensitivity of the model to assumptions. Simulations were also conducted for a diverse range of geographic
areas to determine the sensitivity of the TSS removal rates to regional hydrology. The model was sensitive to the selection
of settling velocities and pollutant loading. The model was less sensitive to changes in hydrology, although significant
changes in hydrology did impact TSS removal estimates.
Keywords: Stormwater; suspended solids; model; hydrology; Stormceptor, separator
Introduction
The Stormceptor is a mechanical water quality separator designed to remove oil and sediment from Stormwater. A key
feature of the design is an internal high flow by-pass to prevent scouring and re-suspension of previously trapped
pollutants. Since the separator is based on treating "the everyday storm," the effectiveness of the separator is dependent
on the distribution of pollution in Stormwater and the frequency and magnitude of Stormwater flows throughout the year.
In 1995, sizing guidelines were derived for the Stormceptor based on field monitoring of sludge accumulation over time
in Toronto, Ontario, Canada. The accumulation data were used to derive estimates of annual total suspended solids (TSS)
removal. Two key assumptions were made in the 1995 analysis to estimate TSS removal: (1) a TSS loading rate of 185
mg/l based on theUSEPA's Nationwide Urban Runoff Program (NURP) median (U.S. Environmental Protection Agency,
1983), and (2) a sludge water content of 75% water. Actual Toronto rainfall data, combined with the NURP TSS
concentration, provided estimates of annual TSS loading. Figure 1 shows the performance relationship derived from the
Toronto monitoring, which forms the basis for the existing sizing guidelines.
Toronto rainfall time-series data (5 minute timestep) were input to a continuous hydrologic simulation model (SWMM
Version 4.3) to determine the percentage of annual runoff treated based on the sizing criteria shown in Figure 1. The
371
-------
10.0
15.0
20.0
25.0
Total Stormceptor Storage (m3/ha)
Figure 1. TSS Removal versus Stormceptor Total Storage (Toronto, 1995).
analysis of Toronto rainfall indicated that 80% - 90% of the annual runoff would be treated if the Stormceptor were sized
according to these 1995 guidelines. This study was initiated to address concerns about the applicability of the Toronto-
based sizing criteria to meteorological conditions in other regions.
Methodology
A computer simulation model was developed based on USEPA's SWMM Version 4.3. Solids build-up, wash-off, and
settling calculations were added to the hydrology code to estimate suspended solids capture by the Stormceptor.
The model accommodates the use of either an EMC (event mean concentration) or build-up/wash-off calculations to
estimate suspended solids loads. The build-up/wash-off model is more theoretically and physically correct. The EMC
method has been shown to provide reasonable estimates of total solids loads (Charbeneau and Barrett, 1998) alone, if
the distribution of the load is not important.
The distribution of pollutant load is important for measures that incorporate a high-flow by-pass (commonly known as
"first flush" measures). Accordingly, preference is given to the build-up/wash-off calculations to correctly distribute the
pollutant load with flow, recognizing the need to optimize the sizing of small-site stormwater quality measures.
In the model, solids build-up and wash-off are both approximated using an exponential distribution. The distribution
of solids build-up is a function of antecedent dry days according to Equation 1 (Sartor and Boyd, 1972).
P, =Pi + (PA-Pi)(1-e-kt)
Where: P, = solids accumulation up to day t (kg)
P = maximum solids build-up (2.4 kg/ha)
A = drainage area (ha)
Pj = initial solids load on the surface (not washed off from the previous storm) (kg)
k = exponential build-up factor (0.4) (days-1)
t = antecedent dry days
(1)
372
-------
The maximum solids build-up (P) load was adjusted to provide similar long-term solids loading rates (124 mg/l) when
compared to the EMC method. An exponential build-up factor (k) of 0.4 was used based on previous literature (SWMM
4.3 Users Manual). A k-value of 0.4 translates into 90% of the maximum solids build-up occurring after 5.66 days. Once
the pollutant build-up reaches the 2.4 kg/ha limit, additional build-up is not allowed (assumed to be wind re-
suspended/driven off the surface). Wash-off is estimated using Equation 2.
P. = Pie-" (2)
Where1 ^ = so''c's remaining on the surface at day t (kg)
Pi = initial solids load (from equation 1) (kg)
k = exponential decay factor (0.2) (mm-1)
V = volume of accumulated runoff from the surface (mm)
The exponential decay factor (k) of 0.2 was based on a review of previous literature that indicates k-values range from
0.03 to 0.55 (Alley, 1981; Charbeneau and Barrett, 1998).
Charbeneau and Barrett (1998) found that the simple wash-off model adequately described observed solids wash-off
in Austin, Texas. Other researchers have cited that the wash-off Equation (2) is reasonable for fine material but may not
be reasonable for larger solids that require a high rainfall intensity for mobilization (Metcalf and Eddy, 1991; Ball and
Abustan, 1995). The SWMM model treats wash-off as a function of the runoff rate to account for mobilization. This
correction is applied indiscriminately to the entire solids load and does not account for the variation in wash-off rate with
particle size. If an "availability" factor is applied to all particle sizes uniformly, the model will underestimate the wash-off
of solids with increasing runoff volume if the majority of particles are fine in size. The approach taken in this study was to
use an availability factor for particles 400 [Jm in size or larger. Smaller particles follow the simple wash-off estimates given
by Equation 2. The larger particles (> 400 |Jm) require greater runoff intensities to induce wash-off according to the
availability factor provided in Equation 3.
A = 0.057 +0.04(r)11
(3>
Where- A = availabilitV factor
r = runoff rate (mm/h)
Equation 3 is based on research by Novotny and Chesters (1997). The runoff rate is used instead of rainfall intensity,
recognizing that the wash-off will lag the rainfall based on the time of concentration. The availability factor varies each
timestep and is only applied to the runoff volume for that timestep, as dictated in Equation 4. The availably factor has an
upper limit of 1.
V = Vi + A(Vt) (4)
Where: V = accumulated runoff volume used in Equation 2 (mm)
V, = accumulated runoff volume prior to current timestep (mm)
A = availability factor (equals 1 for particles smaller than 400 (Jm)
V, = runoff volume for current timestep (mm)
The correction in Equation 4 effectively re-defines the accumulated runoff volume to be the runoff volume sufficient
to mobilize the particles. This methodology requires more accounting in the model but provides a more physically correct
wash-off model.
The separator was treated as a completely stirred tank reactor (CSTR). Alterations to the concentration of solids in
the separator will vary according to Equation 5 (Tchobanoglous and Schroeder, 1987).
CV = QC( - QC, - rcV (5)
373
-------
Where C' = the change in concentration of solids in the tank with time (kg/m3s)
Q = flow rate through the tank (m3/s)
C| = solids concentration in the influent to the tank (kg/m3)
C, = solids concentration in the tank (kg/m3)
V = tank volume (m3)
rc = reduction in solids in the tank (kg/m3s)
For gravity settling devices rccan be estimated using Equation 6.
rc = VSC/D (6)
Where rc = reduction in solids in the tank (kg/m3s)
Vs = settling velocity of solids (m/s)
D = depth of tank (m)
C = concentration of solids in the tank (kg/m3)
Substituting Equation 6 into Equation 5, solving the first-order differential equation and integrating provides the general
form of the non-steady state solution (Equation 7) for the concentration in the tank at time t.
C = QC/(V(VS/D + Q/V))( 1 -e'(Vs/D +Q/V") + C,e-(Vs/D + W)t (7)
Where C = concentration in the tank at time t (kg/m3)
Ci = concentration in the flow influent to the tank (kg/m3)
C, = concentration in the tank at the beginning of the timestep (kg/m3)
Q = flow rate through the tank (m3/s)
V = volume of water in the tank (m3)
Vs = suspended solids settling velocity (m/s)
D = tank depth
t = time
Equation 7 was used to estimate the suspended solids concentration in the tank, and in the discharge from the tank
each timestep. Equation 7 assumes the suspended solids are completely mixed within the tank volume.
During periods without flow (inter-event periods) the solids are not assumed completely mixed at the beginning of each
timestep and the depth of suspended solids in the separator decreases each timestep until all of the solids are removed
or there are subsequent flows into the separator. The concentration of solids in the tank during periods without flow was
calculated using Equation 8.
C = C,(1-Vst/D) (8)
Where: C = solids concentration in the tank (kg/m3)
C, = initial solids concentration in the tank at the beginning of the timestep (kg/m3)
Vs = settling velocity (m/s)
t = timestep (s)
D = depth of solids in the separator (m)
The depth of solids (D) in the separator in Equation 8 decreases each timestep based on the settling velocity until all
of the solids are removed or there are subsequent inflows to the tank.
The model can be used with either hourly or 15 minute rainfall data. Fifteen minute data are preferred, recognizing that
the Stormceptor is only applicable for small drainage areas. Small drainage areas have short times of concentration and
require data with a suitable timestep. Internally, the model performs calculations with a 5 minute timestep.
374
-------
The choice of particle size distribution and settling velocities is a key part of the modeling exercise. Different settling
velocities can be applied to the same particle size distribution, based on the specific gravity of the particles or to account
for the effect of non-ideal settling or flocculation on settling. In this study, a typical stormwater particle size distribution
(USEPA, 1983) was used for analysis (Table 1). The distribution given in Table 1 is commonly accepted by most regulatory
agencies in North America.
The model allows the user to alter the percentages of each size based on site-specific conditions, if required. In most
areas, it is anticipated that the particle size distribution will not vary significantly since it is primarily related to vehicle wear
and atmospheric deposition. There may be certain instances, however, where the native soils contribute loading and the
default distribution needs to be altered. The default percentages were used in our study.
Table 1. USEPA Default Particle Size Distribution
Particle Size (pm)
20
60
130
400
4000
% by Mass
20
20
20
20
20
Settling velocities were then assessed for each of the particle sizes provided in Table 1. Settling velocities were either
calculated or based on empirical literature (USEPA, 1983). The calculation of settling velocities for small particles follows
Stokes' law (Equation 9) since the Reynolds number (Equation 10) is less than 0.3.
Where
Vs = g(p,-pjd2/18u
Vs = settling velocity for particle diameter d (m/s)
g = gravity (m/s2)
ps = density of particles (kg/m3)
pw = density of water (kg/m3)
d = particle diameter (m)
u = viscosity of water (kg/ms)
(9)
N
R = VsdPw/u
(10)
Where
NR = Reynolds number
Vs = settling velocity for particle diameter d (m/s)
pw = density of water (kg/m3)
d = particle diameter (m)
u = viscosity of water (kg/ms)
If the Reynolds number is greater than 0.3, drag on the particles reduces the settling velocity. An iterative solution was
used (solving for the Reynolds number, drag coefficient, and settling velocity until changes in the settling velocity were
insignificant) for particle sizes with the Reynolds numbers. The drag coefficient is given by Equation 11, and the settling
velocity is calculated by Equation 12.
CD = 24/N, +3/(NH0.5) + 0.34
Where CD = drag coefficient
NR = Reynolds number
Vs=(4g(ps-Pw)d/(3CDpw))05
(11)
(12)
375
-------
Where Vs = settling velocity for particle diameter d (m/s)
g = gravity (m/s2)
ps = density of particles (kg/m3)
pw = density of water (kg/m3)
d = particle diameter (m)
CD = drag coefficient
Table 2 provides a comparison of the settling velocities used in this study.
Table 2. Discrete Particle Size Settling Velocities (mm/s)
Particle Size (|jm) S.G. = 1.3 calculated S.G. = 1.8 calculated S.G. = 2.65 calculated USEPA (1983) empirical
20 0.07 0.17 0.36 0.00254
60 0.59 1.57 3.23 0.02540
130 2.50 5.70 11.20 0.12700
400 16.00 37.00 65.00 0.59267
4000 180.00 300.00 450.00 5.50330
S.G. = Specific Gravity
The settling velocities that are based on the empirical USEPA data are 65 to 150 times smaller than the settling
velocities based on a specific gravity of 2.65. A specific gravity of 2.65 is commonly associated with sand-size particles
whereas the fines in stormwater are commonly associated with a lower specific gravity. The use of a higher specific gravity
may be justified, however, if the values are considered representative of the settling velocities of fines in a flocculated or
coagulated state. Research indicates that there is a high potential for coagulation amongst particles (Ball and Abustan,
1995) which will increase settling velocities and TSS removal rates. Furthermore, historical settling velocity calculations
have been based on discrete particle methodologies (vertical settling column tests) that do not account for potential
coagulation. Coagulation would effectively offset the settling velocity columns in Table 2 (i.e., discrete settling velocity for
60 )Jm represents coagulated 20 )Jm particle size).
Numerous field tests on the Stormceptor (Labatiuk, 1996; Henry et al., 1999; Bryant, 1995) have indicated a high
percentage of fines in the Stormceptor. This empirical evidence lends credence to the coagulated settling theory, indicating
that the USEPA discrete particle settling velocities may underestimate actual TSS removal rates. Settling velocities based
on a specific gravity of 1.8 were chosen in this study as the default or benchmark selection. The solids loading was
segmented into the particle size distribution and the concentration of solids in each particle size was tracked individually
during the settling calculations.
Meteorological Data
Rainfall data from the City of Toronto (5 minute timestep, 0.25 mm resolution, 10 years record, 1987-1996) was
agglomerated into 15 minute data for use with the model. Fifteen minute data were obtained for the entire U.S.A. from
Earthlnfo on CD ROM. Stations were selected based on location, period of record, data resolution and completeness
within the period of record. Data were also obtained from CSR Humes for various stations throughout Australia. The rainfall
data were converted into National Climatic Data Center (NCDC) format for input to SWMM.
Fifteen minute data were utilized, recognizing the small time of concentration that would typically be encountered in
most Stormceptor applications. Simulations were also conducted using hourly data to determine the sensitivity of the
results to the precipitation timestep. Numerous hourly stations were available on the Earthlnfo CD for this purpose. The
model uses a 5 minute timestep at all times regardless of the rainfall timestep.
Modeling Parameters
SWMM models catchments and conveyance systems based on input rain, temperature, wind speed, and evaporation
data. Only rain data were used in these analyses. The default SWMM daily evaporation values (2.5 mm/day) were used.
Evaporation data will not be important in this analysis since the catchment area is small (c 10 ha) and has minimal
376
-------
depression storage. The Morton equation was chosen for infiltration. The method of infiltration chosen is unimportant due
to the small amount of pervious area (1%). Table 3 provides a list of the parameters used in the SWMM model.
Table 3. SWMM Area Parameters
Area - ha (ac) variable
Imperviousness 99%
Width - m (ft) variable
Slope 2%
Impervious Depression Storage - mm (in.) 4.7 (0.19)
Pervious Depression Storage - mm (in.) 0.5 (0.02)
Impervious Mannings n 0.015
Pervious Mannings n 0.25
Maximum Infiltration Rate - mm/h (in/hr) 62.5 (2.46)
Minimum Infiltration Rate - mm/h (in/hr) 10 (0.39)
Decay Rate of Infiltration (s"') 0.00055
The width of catchment was assumed to be equal to twice the square root of the area.
Results
EMC versus Build-up/Wash-off
The suspended solids removal results based on the build-up/wash-off model were compared to those based on an
EMC of 124 mg/l (USEPA, 1983) to demonstrate the sensitivity of the model to the different solids loading approaches.
The use of an EMC assumes an equal concentration of suspended solids in all of the stormwater that is conveyed to the
Stormceptor.
Figure 2 shows a comparison of results using an event mean concentration loading and build-up/wash-off loading,
given the default particle size distribution and settling velocities based on a specific gravity of 1.8.
The results in Figure 2 show that the TSS removal rates using the EMC approach are lower by 14% when compared
to the build-up/wash-off method even though the total loads are similar. This is expected due to the by-pass nature of the
Stormceptor. The estimated TSS removals for the existing (1995) sizing guidelines, which are based on an early field study, are lower
than both the EMC and build-up/wash-off estimates for low values (50% TSS removal) of separator storage/drainage area and are higher than the other
estimates for larger values of separator storage/drainage area (80% TSS removal).
The range of TSS removal values based on computer modeling is smaller than the empirical TSS removal rates.
Doubling the size of unit for the same area results in an increase of 30% for TSS removal, based on the current sizing
guidelines, whereas the increase in performance based on the modeling is less dramatic (a 5% to 10% increase in TSS
performance). This finding indicates that the modeling results will be less sensitive to changes in the model size for any
given drainage area.
Selection of Settling Velocities
A comparison was made regarding the choice of settling velocities using Toronto rainfall data and the build-up/wash-off
TSS generation methodology. Figure 3 provides the results of this analysis. The TSS removal estimates using the USEPA
settling velocities are an average of 20% lower than the original TSS removal estimates, 29% lower than the estimates
using the SG=1.3 velocities, and 39% lower than the estimates using the SG=2.65 velocities. These results indicate that
the TSS removal performance results are very sensitive to the selection of settling velocities.
377
-------
Figure 2. TSS Removal vs. Loading Methodologies
on0/
g1 80% -
5f 70% -
o 60% -
» 50% -
co 40% -
CO
^ 30% -
| 20% -
< 10% -
no/.
Ti * -^
•" — — y ^^*^ "
• EMC
• Build-up/Wash-off
—a— Existing Sizing
6.00 8.00 10.00 12.00 14.00 16.00
Storm ceptor Storage (m 3/ha)
Figure 2. TSS Removal vs. Loading Methodologies.
Annual Flo w Treatment
Numerous regulatory agencies design stormwater quality measures using a "design" event. The design event used
generally ranges from the 25 mm storm, or annual storm, to the 25-year storm. The modified SWMM program was used
to calculate the percentage of annual runoff that would be treated (not by-passed) with different by-pass flow rates. This
analysis was conducted using the Toronto rainfall for a drainage area of 2.25 ha. Figure 4 shows that the volume of runoff
that is treated prior to by-pass quickly becomes asymptotic with increasing treatment flow rate. A device that treats 30 Us
prior to by-pass would treat approximately 80% of the annual runoff. A device that treats 70 Us (over 2x higher flow rate)
only treats 10% more runoff (90%). Although the relationship between conveyance (% of annual runoff treated) and TSS
removal is non-linear, Figure 4 shows that high-rate treatment devices are not required for small drainage areas.
The relationship provided in Figure 4 will vary with local meteorological conditions and is inherently accounted for in
the TSS removal modeling.
Regional TSS Removal Performance Analysis
The model was used to compare results from different areas in North America and Australia to determine the effect
of regional hydrology on TSS removal performance. All analyses were conducted using 15 minute rainfall data and based
on the TSS build-up and washoff model and settling velocities for a specific gravity of 1.8.
Table 4 shows the results for various sized Stormceptors with a 2 ha drainage area. The locations of stations listed
in Table 4 were selected to cover a wide geographic area, provide rainfall on a 15 minute timestep with a 0.25 mm
resolution, and provide results representative of large nearby cities. Most data from city airports are recorded hourly, and
therefore were not included in the comparison. The results in Table 4 are plotted in order of decreasing performance
expectations in Figure 5.
Of the 16 stations analyzed, 12 stations provided TSS removal estimates within ±5% of the Toronto values.
Although the majority of stations provided similarTSS removal estimates, there were areas with significant differences.
The performance estimates were lowest for the southeastern United States. This area is well know for its intense seasonal
rainfall distribution. Figure 5 indicates that the TSS removal rates may vary up to 20% under different hydrological
conditions on the same land use/site conditions. The use of local or regional rainfall data is therefore appropriate for design
purposes.
378
-------
90% -i
80% -
_ 70% -
ss
:r 60% -
re
| 50% -
1 46%' •
V) 30%
(/}
*~ 20%
10% -
0% 4
Figure 3. TSS Removal Performance vs. Settling Velocities
* **-r
iff- Z"-
D ^^.^
«*r*i
n ' <
V V
g^* ^ , "a"
>U A
10
0 USEPA
• 1 DSG=1.3
x:3G=1.SG=1.8
A SG=2.6S=2.6
6 8 10 12 14 16
Stormceptor Storage (m3/ha)
Figure 3. TSS Removal Performance vs. Settling Velocities.
Figure 4. Annual Runoff Treatment
(Toronto - 2.25 ha)
20
40 60
Flow Rate (Us)
80
10C
Figure 4. Annual Runoff Treatment.
379
-------
Table 4. Regional Comparison of TSS Removal Performance (2ha)
State/ Province
Colorado
Alberta
British Columbia
California
Massachusetts
Ontario
New South Wales
New York
North Carolina
Queensland
Minnesota
California
Maryland
Missouri
Florida
Texas
Location
Fort Collins
Calgary Forest
Vancouver
Davis
East Brimfield Lake
Toronto
Sydney
Rhinebeck
Cataloochee
Brisbane
Le Sueur
Orange County
College Park
Miller
St. Lucie New Lock
Houston Addicks
Stormceptor Model (CDNIUSA)
300/ 750J 1500/ 3000/ 5000/ 6000/7200
450 900 1800 3600 6000
49% 63% 65% 71% 76% 79%
48% 63% 65% 71% 76% 79%
48% 65% 66% 71% 76% 78%
44% 61% 63% 69% 74% 77%
43% 59% 61% 67% 73% 75%
43% 58% 60% 66% 72% 75%
42% 57% 59% 66% 72% 76%
41% 57% 59% 65% 71% 74%
41% 56% 58% 64% 71% 74%
41% 55% 57% 64% 71% 74%
41% 56% 57% 84% 70% 74%
39% 57% 59% 65% 71% 74%
37% 53% 54% 61% 67% 70%
34% 50% 51% 59% 65% 69%
30% 43% 44% 52% 59% 64%
27% 41% 42% 49% 57% 61%
Figure 5. Regional Comparison of TSS Removal Performance
(2ha)
3001450 750190KB 1500/1800 300013600 5000/6000 6000/7200
Stormceptor Model (CDNIUSA)
Figure 5. Regional Comparison of TSS Removal Performance
Rainfall Timestep
An analysis was conducted to determine the sensitivity of the model to changes in rainfall resolution. Results based
on hourly rainfall data (0.25 mm resolution) were compared to those based on 15 minute rainfall data, to determine the
impact of using the hourly data. Hourly data are more readily available than 15 minute data and most large cities have
airports that collect rainfall on an hourly basis.
The model reads the hourly data as rainfall that falls during the first fifteen minute timestep of each hour. This will
produce higher intensities since the rain is not distributed correctly over the entire hour. The greater intensity is
compensated for, however, by the completeness of the hourly records which translates into a greater number of small
rainfall values.
Four areas were analyzed (Rockville, Maryland; Boston, Massachusetts; Miami, Florida; and Houston, Texas). The
results of this analysis (Figure 6) indicate that the use of hourly data does not significantly alter the TSS removal estimates
380
-------
for units that are designed to remove over 40% of the annual TSS load. Greater discrepancies can be expected at large
ratios of drainage area to separator storage.
90% -
80%
70% -
60% -I
50% -i
40%
30% -
20% -
10% -
o%
Figure 6. TSS Removal versus Rainfall Timestep
(by Stormceptor Model)
-•-MA (15 min)
4 MA (60 min)
A MD(15 min)
—A— MD (60 min)
—•— FL(15min)
-e-FL (60 min)
—•— TX (15 min)
-e-TX (60 min)
3 4
Stotmceptor Model
Figure 6. TSS Removal vs. Rainfall Timestep (by Stormceptor Model).
Conclusions
The TSS removal results were sensitive to the selection of settling velocities for the specified particle distribution.
Differences in TSS removal of up to 40% were obtained, depending on the settling velocities that were evaluated.
Results were also affected by the TSS loading method. The use of an EMC underestimated TSS removal performance
by approximately 15%, when compared to using the build-up and wash-off equations. This difference is expected since
the EMC method increases the load that is by-passed and provides higher loads during higher treated flow rates when the
detention time, and hence settling effectiveness of the unit, is reduced.
The model indicates that high percentages of the annual runoff can be treated with low-flow treatment devices such
as the Stormceptor. The model also predicts that the TSS removal performance is less sensitive to the size of separator
than observed from previous field studies.
Regional hydrology affected the TSS removal estimates provided by the model. Although differences of up to 20%
were observed, significant hydrological differences between the sites were needed to obtain this variance. Most of the
rainfall station locations tested provided TSS removal estimates similar to those of Toronto, where the original sizing
guidelines were developed.
Testing the model with different rainfall timesteps (15 minute versus hourly) indicated that hourly rainfall records can
provide an adequate estimation of performance if the rainfall is collected at adequate resolution (0.25 mm increments).
The modeling indicated that significant TSS removal rates can be achieved using small infrastructure control measures
if the drainage area is limited. The results lend credence to the positive field monitoring results obtained to-date for the
Stormceptor, and to the concept of small storm hydrology being the predominant parameter for urban stormwater quality
design.
381
-------
References
Alley, W., Estimation of Impervious-Area Washoff Parameters, Water Resources Res., 17, 1161, 1981
Ball, J., and Abustan, I, An Investigation of Particle Size Distribution durina Storm Events from an Urban Catchment,
University of New South Wales, 1995
Bryant, G., Misa, F., Weatherbe, D., and Snodgrass, W., Field Monitoring of Stormceptor Performance, 1995
Charbeneau, R., and Barrett, M., Evaluation of methods for estimating stormwater pollutant loads, Water Environment
Research, Volume 70, Number 7, 1998
Henry, D., Liang, W., and Ristic, S., Comparison of Year-Round Performance for Two Types of Oil and Grit Separators.
Draft paper, 1999
Labatiuk, C., Nataly, V., and Bhardwaj, V., Field Evaluation of a Pollution Abatement Device for Stormwater Quality
Improvement, CSCE Environmental Engineering Conference, Edmonton, 1997
Novtony, V., Unit Pollutant Loads. Water Environment & Technology, 1992
Novtony, V., and Olem, H., Water Quality: Prevention, Identification, & Management of Diffuse Pollution. Environmental
Enaineerina Series, Wiley & Sons, 1997
Sartor, J. , and Boyd, G., Water Pollution Aspects of Street Surface Contaminants. EPA-R2-72-081, U.S> Environmental
Protection Agency, Washington, D.C., 1972
Tchobanoglous, G., and Schroeder, E., Water Quality, University of California at Davis, 1987
Tchobanoglous, G., and Burton, L., Wastewater Enoineerina. Treatment. Disposal & Reuse. Water Resources &
Environmental Enaineerina. McGraw-Hill, 1991
U.S. Environmental Protection Agency, Final Report of the Nationwide Urban Runoff Program, Water Planning Division,
Washington, D.C.,1983
U.S. Environmental Protection Agency, Storm Water Management Model, Version 4.3. -User's -Manuali Washington. D.C.,
1988
382
-------
NPDES Phase II Cost Estimates
Andrew J. Reese, P.E.
Ogden Environmental and Energy Services, Inc.
Nashville, TN, 37211
Introduction
The United States Environmental Protection Agency (EPA) has published final rules expanding the existing stormwater
NPDES permitting program to smaller cities and other urban areas throughout the United States. Due both to external
pressures and directives from the current and past administrations, EPA is conscious of attempting to make the current
stormwater NPDES program "cost-effective." For example:
"EPA believes this rule will cost significantly less than the existing 1995 rule that is currently in place, and will result
in significant monetized financial, recreational and health benefits, as well as benefits that EPA has been unable
to monetize, including reduced scouring and erosion of streambeds, improved aesthetic quality of waters, reduced
eutrophication of aquatic systems, benefit to wildlife and endangered and threatened species, tourism benefits,
biodiversity benefits and reduced siting costs of reservoirs." 1
".. . the Agency recognizes the continuing imperative to assure that environmental regulations accomplish statutory
objectives in the least burdensome and most cost-effective fashion. As explained further in this preamble, the form
and substance of NPDES permits to address the sources designated in today's proposal would provide greater
flexibility for the newly covered sources than the existing "standard" NPDES permit."'
While the "benefit" side of the proposed regulations exists in the realm of gross estimates, the "cost" side is also filled
with unknowns. What will the mandated and negotiated stormwater program cost a local community? Are there ways to
reduce costs? What should a local community be doing now to prepare for this regulatory program? This paper seeks
to address these related questions.
The final regulations were published on December 8, 1999 and the changes from the draft regulations are only minor3.
But it is still not possible to say what the regulations will cost everyone in toto. This is so because:
. there is great flexibility inherent in the regulations to create a stormwater quality program tailored to meet an
individual community's needs and situation;
• each permit writer has preferences and "hot buttons" that will color what any particular program will look like; and
• each community setting is different in terms of climate, topography, pollutants of concern, and current condition of
local waters.
1 Federal Register, January 9, 1998 p. 1536
2 ibid. p. 1550
3 Federal Register, December 8, 1999 pp. 68722-68851
383
-------
Basic Approach to Permitting
Under proposed § 123.35(g), an NPDES permitting authority issues a general permit to authorize stormwater
discharges from regulated small municipal separate storm sewer systems. The NPDES permitting authority will also
provide a menu of regionally appropriate and field-tested Best Management Practices (BMPs) that the permitting authority
determines to be "cost-effective."The regulated small municipal separate storm sewersystems could choose to select from
this menu or select other BMPs that they feel are appropriate.
Under Phase II each regulated community will need to develop a set of BMPs under each of six specific program
minimums. These BMPs can be any combination of programs, structures and other controls that, in the agreed opinion
of the permit writer and the regulated community, meet the standard of reducing pollution discharge to waters of the state
to the Maximum Extent Practicable (MEP). In this process, permittees and permit writers would evaluate the proposed
stormwater management controls to determine whether reduction of pollutants to the MEP could be achieved with the
identified BMPs. EPA envisions that this evaluative process would consider such factors as condition of receiving waters,
specific local concerns, and other aspects included in a comprehensive watershed plan.
Under the proposed approach, implementation of BMPs consistent with stormwater management program
requirements at § 122.34 and permit provisions at § 122.33 would constitute compliance with the standard of "reducing
pollutants to the maximum extent practicable." That is, "if you do what you say you will do, you are by definition in
compliance." It is important to note that states implementing their own NPDES programs may develop more stringent
requirements than those proposed in the Federal Register. In fact, we anticipate that many states will require more specific
and rigorous requirements under special circumstances relating to the condition of the receiving water within, and
downstream from, the community. For example, if a certain stream is required to have a Total Maximum Daily Load
(TMDL) or similar study performed on it (for example, a watershed assessment for the purposes of wastewater treatment
plan permitting or expansion), the NPDES stormwater Phase II permit conditions may reflect the allocation of pollutants
to that community.
The steps for a community are: (1) review the conditions of the general permit, (2) develop and submit a Notice of
Intent (NOI) to comply with the general NPDES permit through description of a BMP-based program under each of the six
minimum controls or program areas (see below), (3) negotiate this proposed program with the permit writer, (4) receive
approval of the submittal, and (5) begin implementation of the conditions and programs described in the NOI including
record keeping and submittal of appropriate reports describing attainment of "measurable goals" for each BMP as
described in the NOI.
Current NPDES Phase II Program Cost Estimates
There is naturally much speculation on the actual program elements and costs for a particular stormwater program
developed under Phase II. There have been several attempts at estimating Phase II program costs based on current costs
of "similar" programs.
In the draft regulations, EPA had provided estimates of the probable cost implications of the NPDES Phase II Permit.
These estimates were based on summary information from the permit applications from 21 Phase I cities. Very high and
very low figures were thrown out by EPA in developing these estimates. Figure 1 shows the summary table developed by
EPA.
The range depicted in Figure 1 is from $1.39 to $7.83 per person per year for the first permit five-year period, and
$1.28 to $5.63 for other permit cycles. For a city of 50,000 that is a very wide range of $69,500 to $391,500 annually for
the first permit cycle. This is clearly not helpful in attempting to estimate a specific community's costs.
There is question about the vagueness in the regulatory language, and the high degree of potential flexibility inherent
in briefly described program elements. For example, for the first of the minimum controls the regulatory language states:
384
-------
P?j «Ui«l Ml-
frsF **! j
„ ,1, r ,,.,,., 1... .,,,,, ,,,,'.„..,,,,- „,
.......
il for lir-,1
...
„.,
•$:•!££-
,H ^tH.,,,,,1 (
A Is
-.,
Iti.S
•3a^a^^™..-
i l'L"J."'" ''.'«'»'.'," >.;«'B<.«" T,'i,,,u,,
DAPITA COSTS FOR i
Flra
2nd
Manvuie
( Pei'fTNt Oycl«
f>ubilc eoocftiioo
H-ubtta Involvement
IHiCil Diacruirge D«,E
OooM Oil* SW nunolf Oonirol
Po-wl ConotrucDon SW MH>
PF>/<;IM nf tu*i*ik;ip»l Ops
and 3rd «*«milt CyClB*
Public lnvotv»m«4il
IllkJ.I O>»<^tn.>u» »AC
CooM SltA SW Ruoofl Com ot
r>o«t C&mtlrut.aon SW M»«
PP-/OM of Municipal Os>«
inuitlctpa.ll-
pecied to
(p«r««nt)
30
1OO
ab
4
CT
Lo~.™«
ciipftn coats
SOO^
Otwt
0.0
1 O
o o
o o
0.0*
1,O9
o.r>
«^«K.«
^a^m ^.^fU
so.a-4
a.e*
1 OO
2.OO
034
a.t?
1 O»
*.OB
Figure 1. EPA Cost Estimates for Phase II NPDES Compliance.
1. Public education and outreach on storm water impacts4.. You must implement a public education program to
distribute educational materials to the community or conduct equivalent outreach activities about the impacts of
storm water discharges on water bodies and the steps that can be taken to reduce storm water pollution.
(You may use stormwater educational materials provided by your State, Tribe, EPA, or, subject to the approval of
the local government, environmental or other public interest or trade organizations. The materials or outreach
programs should inform individuals and households about the steps they can take, such as ensuring proper
septicsystem maintenance, limiting the use and runoff of garden chemicals, becoming involved in local stream
restoration activities that are coordinated by youth service and conservation corps and other citizen groups, and
participating in storm drain stenciling, to reduce storm water pollution. In addition, some of the materials or outreach
programs should be directed toward targeted groups of commercial, industrial, and institutional entities likely to
have significant storm water impacts. For example, information to restaurants on the impact of grease clogging
storm drains and to garages on the impact of oil discharges. You are encouraged to tailor your outreach program
to address the viewpoints and concerns of all communities, particularly minority and disadvantaged communities,
as well as children.)
The "regulatory" wording in parentheses is not mandatory but suggested. There is wide room for interpretation of the
intensity and detail necessary to accomplish this minimum control. The devil is always in the details, and there will always
be great variability in what two different programs intend to do to accomplish the same general goals.
NAFSMA (1999a, 1999b) published a survey on potential Phase II program costs responded to by 121 cities and
counties nationally. Ten communities responded with programs that had three or more suggested elements in the first
minimum control: Public Education and Outreach. The annual per capita costs for these ten ranged from $0.04 to $1 .17
- again a wide range.
Of those responding, only one community stated that it had program activity in each of the six minimum control
measure areas and it spent $15.11 per capita annually, well above the EPA estimate (the city has a population of about
25,000). Of the 121 respondents only 26 had programs in at least three (most had only three) of the six mandatory
minimum control areas, and these can be considered far from complete. Figure 2 shows the distribution of costs for these
26 programs. The vertical axis is the annual per capita cost for these elements. The median was $1.44 and the average
was $4.07. The low value was $0.04 and the high was $26.00.
4Federal Register, January 9,1998, p. 1639.
385
-------
s m . 00
524.00 •
$22.00
s m . 00
$18410
$16.OO
914.00
$12.00
s lo.OO
saoo
$aoo
$4.00
$200
$- -1
n
^-"\
_ *. . — -/*" \ Pt
*- n v> t*. o> ^ t£ iPt 5^"
1
|
K
i •
fi
Figure 2. NAFSMA Study Program Costs.
We can speculate that if many of these communities had a fully developed Phase II program, the average costs could
more than double, since each community would be adding both new program areas and upgrading their existing programs
they had to make them comply with the details of the Phase II permit writers requirements.
In the final regulations, USEPAtook a different approach to making estimates of the costs of compliance, using both
the NAFSMA information and past experience with Phase I (EPA, 1999). EPA estimated annual costs for the municipal
programs based on a fixed cost component and a variable cost component. The fixed cost component included costs for
the municipal application, record keeping, and reporting activities. On average, EPA estimated annual costs of $1,525 per
municipality. Variable costs include the costs associated with annual operations for the six minimum measures and are
calculated at a rate of $8.93 annually per household (assuming 2.62 persons per household). The the cost estimating
equation is:
Annual cost = $1,525 + population/2.62*$8.93
Finally, rule of thumb estimates based on the author's experience working in over 100 communities indicate that
comprehensive stormwater programs that include advanced stormwater quality programs cost between $7.00 and $20.00
per capita per year- above the EPA estimates. The quality portion is normally between 20 and 30% of the total average
program cost.
Estimating Costs from Anticipated Programs5
The methods used above do not provide details of the components of the stormwater programs resulting in the costs,
and thus are not very helpful in assisting other communities in their thinking about the regulations. An effort was made
to develop cost estimate ranges based on a direct interpretation of the stormwater regulations as applied to example
communities at each end of the spectrum, in terms of size and intensity of water quality program. This has an advantage
in that it deals directly with the stormwater regulatory requirements and illustrates specific program components so that
we can control and define all details. The following sub-sections will develop two hypothetical permit applications for the
six minimum controls.
The Two Permittees
Permittee one ("Smallville") is a community of 10,000 that is adjacent to a larger city that has obtained a Phase I permit
or that can assist Smallville in many of its permit responsibilities. It is a small bedroom community interested in compliance
with minimum disruption and cost. It does not really have an engineering or planning component of its city staff, but relies
on a city administrator and hired consultants.
5 Based on a presentation made by Andy Reese of Ogden Environmental at the APWA seminar, "Designing and Implementing an Effective Storm
Water Management Program, Denver, 1998.
386
-------
Permittee two ("Midtown") is a larger and more self-contained community with a population of 50,000 located within
an urbanizing county whose total population makes it a designated "urbanized area." It is aggressively annexing growth
areas, and has a thriving economy. It has a City Engineer/Public Works director, road maintenance staff, and other
municipal capabilities and resources. It also has a growing stormwater quantity program and wishes to bring quality
together with quantity in a comprehensive and integrated approach. It wants to take advantage of its GIS database and
capability.
The Programs
We can assume that contained within, or subsequent to obtaining the general permit, the permit writer will publish a
list of regionally appropriate BMPs to be used in permit applications. The general permit will have narrative effluent
limitations which describe goals or narrative standards for each of the minimum controls. Each permittee must then
develop basic program objectives and measurable standards (not included here) under the goals provided by USEPA for
each of the six minimum controls. These measurable standards can be stated in terms of actions taken or results
achieved. It is best to state them in terms of things that can be controlled and which do not have uncontrollable and
unpredictable results.
It is also smart to schedule the programs (the schedule is not demonstrated here) in terms of phases, pilot programs,
demonstration projects, trials, etc., with an evaluation process at some point in the permit. It should then be written into
the NOI that this program will be modified, expanded, curtailed or even abandoned if it is not effective.
Smallville sought to obey only the letter of the law, but did not see many ways to proceed. It had no real stormwater
program, no known water quality problems, and few current responsibilities. This community sought to take advantage
of "big brother" next door in joint programs or education, and to adopt more regionally uniform development regulations
enforced locally. Smallville sought to fund any program needs through budget changes and through economies gained
by taking advantage of regional programs, free information, and expanding duties of existing staff.
Midtown sought to meet the program minimums in a more proactive way focusing on perceived needs within the
community. They took advantage of the strength of existing local programs, a strong economy, a strong environmental
awareness, and outside assistance where available in the form of copied resources and shared efforts. Midtown expanded
its current program using EPA suggestions to build a more comprehensive and meaningful program in several key areas6.
Because they did not have the ability to try to work regionally (the adjacent county had no resources for developing a
stormwater program, but would cooperate as necessary) it needed to build the program alone and to work extra-territorially
as appropriate. Midtown looked at each program to insure the existence of: adequate legal authority, competent technical
approach, dedicated financial resources and appropriate administrative procedures and staffing.
Because program funding became an issue, Midtown sought to establish a stormwater user fee system (often called
a stormwater utility) to provide stable, adequate and equitable funds. The costs and steps of the utility development are
not included here.
Program Objectives
Table 1 develops the basic objectives of each of the programs in each of the six minimum areas. In real life these
objectives would be developed through a series of discussions with staff and, perhaps, a citizen's group, and through early
coordination with the permit writer.
Table 2, which is attached as an Appendix, gives basic cost-estimate information for the two programs. The costs are
approximate and would vary depending on how all costs are accounted for, availability of staff, etc. The intent is to give
ballpark estimates and not to quibble over details. In these estimates all personnel time is costed at $50/hr regardless of
6 NAFSMA has taken an earlier version of the Midtown values, refined them, and developed a minimal and advanced program concept out of this
information. That information can be obtained from NAFSMA by calling 202-218-4122.
387
-------
the source of the labor (in-house or contracted). This corresponds to a fully burdened salary rate plus allocated overhead
costs for a mid-level technical person.
Table 1. Basic Program Objectives
Smallville
Midtown
Public Education and Outreach on Storm Water Impacts
Acquire and mail existing public domain informational brochures
Encourage and facilitate newspaper articles
Educate the few industrial and commercial stakeholders individually
1. Acquire and mail existing and specifically pertinent public domain
informational brochures to the general public
2. Develop a stratified database of stakeholder groups and develop
and execute targeted education programs
3. Develop and implement elementary school education programswith
preexisting curriculum
4. Develop and advertise complaint hotline as a pollution hotline
5. Develop press information and briefings with the objective of having
a quarterly news article
6. Develop and make available a slide show and speakers bureau
Public Involvement/Participation
Develop and implement a citizens advisory group appointed by the mayor
Encourage citizen participation in the neighboring city's programs for
used oil, household hazardous waste, adopt-a-stream, etc. through news
articles in local neighborhood newspaper
1. Develop and implement a stratified and diverse citizens advisory
group/task force
2. Develop a citizen monitoring and/or adopt-a-stream program - may
be partially federally funded
3. Develop a student storm drain stenciling program and student dry
weather screening program (see illicit connections program)
4. Encourage the development of watershed groups for each major
watershed within the jurisdiction (see BMP control)
Illicit Discharge Detection and Elimination
Develop a stormwater major outfall map on USGS base map
Modify slightly and adopt a generic ordinance available from the state or
other organization.
1. Develop a major stormwater system map and inventory on existing
GIS topo. Base mapping
2. Cross-reference map with existing databases on NPDES permit
holders (available from the state) and SARA Title III database to
identify likely source of dry weather pollution
3. Develop an illicit connections and illegal dumping ordinance
including hotspot program
4. Perform initial dry weather screening in several key parts of the city
by student volunteers
5. Develop inspection and enforcement capabilities and resources,
and develop a detection program using city staff and a database of
potential specific locations
6. Advertise hotline and write news articles (see public education)
7. Advertise existing private used oil disposal sites (see public
education)
8. Educate all public employees to recognize and report problems (see
pollution prevention)
9. Develop automotive industry sponsorship of spill prevention,
materials management, and inspection and education programs
(see public education for part of this)
Construction Site Storm Water Runoff Control
1. Modify the adjacent city's sediment and erosion control ordinance to
meet the regulatory minimums
2. Modify plans review and inspection procedures to include program
minimums
3. Train city secretary to collect phone complaints and take appropriate
action on erosion complaints
4. Advertise the complaint line as part of the public education program.
1. Modify existing sediment and erosion control ordinance to
include all the requirements of the regulations
2. Add a BMP section and clear design steps to the drainage
manual
3. Conduct training and familiarization program for developers,
contractors and engineers, as well as in-house training for
inspectors
4. Insure hotline has a formal and defined ability to receive and
properly process erosion complaints
5. Upgrade the erosion control inspection and enforcement
program
388
-------
Post-Construction Storm Water Management in New Development and Redevelopment
1. Modify and adopt the adjacent city's stormwater ordinance regarding
stormwater quantity and quality requirements to require similar controls
and requirements. Add a maintenance requirement for BMPs and
detention designs
2. Transform the inspection process to be able to inspect and enforce the
new ordinance
3. Communicate the new requirements
1. Investigate and seek to institute zoning and policy changes to
encourage density restrictions, transferable development rights,
easier use of PUDs, limitation of impervious areas, conservation
easements, mandatory floodplain dedication, etc.
2. Develop design guidance for the use of structural and non-
structural BMPs
3. Develop and conduct an ongoing training program in the proper
use of BMPs
4. Develop several BMP pilot projects to demonstrate and gain
experience in BMP use
5. Overhaul and develop a comprehensive storm water ordinance
for both water quantity and quality which includes mandatory use
of BMPs and a maintenance requirement
6. Establish inspection program for private BMPs
7. Develop a monitoring program for local surface waters and to
monitor their long term changes
8. Develop master plans for areas facing new development and
establish and enact policy for regional BMP design and
maintenance
9. Develop ways to improve extra-territorial planning and zoning
input
10. Identify key environmentally sensitive areas and take steps to
protect such areas through ordinance, overlay districts, etc.
11. Seek to establish local watershed organizations and
neighborhood adopt-a-stream programs to assist in compliance
and build public support
Pollution Prevention/Good Housekeeping for Municipal Operations
1. Review all current municipal procedures and document ways to reduce
pollution
2. Make changes and document
3. Obtain and distribute materials on ways to reduce pollution as available
and appropriate.
Conduct an outside review of all applicable procedures and
criteria and make recommendations for change, implement
changes
Obtain available information and conduct sensitivity and
familiarization training for all applicable city employees
Seek to control floatables partially through adopt-a-stream
program (see public participation)
Review existing flood control projects to insure advantage is
taken of pollution reduction opportunities in design and operation
Hours are given in most cases. Italicized numbers are one-time costs that are experienced some time in the first permit
period, assumed to fill the year in which they initiate. For ongoing programs, the program initiates beginning in the next
year. The annual costs are the anticipated costs thereafter. I have assumed that all programs initiate in year one for the
total five-year cost estimate. Obviously if a program initiates in a later year there will be savings in annual costs not
incurred until the program initiates. The five-year total is four times the annual cost plus the initial cost — making a total
of five years. Some programs are five-year programs only, ending after the first cycle.
A schedule of tasks and of manpower requirements is not developed in this paper. The costs are given as initial costs
and as ongoing costs (clear from the context of the table). Because not all program elements will be developed and in-
place for the whole permit term, there will be a ramp-up process. Also, most of the program elements will continue to
change and evolve over time, and program costs will also change (up or down) in subsequent permit periods.
Extraordinary volunteer efforts have not been assumed (e.g. writing news articles, manning a hotline, etc.).
It is important to realize that some per capita costs go down for large cities because they have a large fixed component.
For example, it may cost the same to develop a one-page brochure whether the city has 20,000 or 200,000 people in it.
Expenses are based on medium levels of effort wherever appropriate. Detailed expenses (e.g. long distance phone costs)
have not been estimated.
389
-------
Measurable goals have also not been provided in this handout. But for each BMP measure or program it will be
necessary to develop some measurable standard by which to judge success. The standard may be based on internal
activities where it cannot easily be based on external results. For example, sending out brochures three times per year
can be measured. But, the effectiveness of those brochures can only be measured through phone surveys of public
knowledge before and after the brochure was sent, or based on statistics on increased public participation in whatever
program the brochure was about. Neither measure is easy and reliable. And, should a certain percent "effectiveness
increase" be stated as the measurable goal, if it is not achieved the city would, technically, be out of compliance. Better
to make the goal controllable, especially in the first permit cycles when little is known on the effectiveness of certain
(especially non-structural) BMP measures.
In no case have the costs of structural BMPs been estimated or included. Cost estimates are available in several
references including the Center for Watershed Protection (1997) and Northern Virginia Planning District Commission
(1994). The economic benefits of structural BMPs are discussed in EPA (1995).
Monitoring costs are developed for Midtown based on both receiving stream monitoring and some pilot BMP program
monitoring; they are non-existent for Smallville. EPA estimates that about 50% of permittees may incur monitoring costs
in subsequent permit cycles. It is also assumed that there are no TMDL or other types of watershed assessment actions
going on in the watershed which may radically modify the permit conditions, and that there are no regional or state-wide
programs which could simply be adopted by reference for portions of the NPDES minimum requirements.
Summary Results
The summary results of the analysis are presented in Table 3, in terms of cost per capita, for each of the programs in
a manner comparable to the EPA estimates.
The range of results is similar to that experienced by EPA in making its original estimates of the cost of the Phase II
program. The details of this program development can assist a local community in fashioning its own stormwater program
in response to the regulations.
Table 3. Summary Results
Minimum Control
Annual Per-Capita Cost
Small
Midtown
1
2
3
4
5
6
- Public Ed.
- Public Inv.
- Illicit Connections
- Construction
- Post Const.
- Housekeeping
Totals
First 5-year Permit Period
0.39
0.21
0.24
0.20
0.14
0.15
1.33
1.24
0.62
1.77
0.96
5.78
0.59
10.96
Subsequent 5-year Permit Periods
1 - Public Ed.
2 - Public Inv.
3 - Illicit Connections
4 - Construction
5 - Post Const
6 - Housekeeoina
Totals
0.36
0.24
0.10
0.18
0.13
0.10
1.40
0.51
1.16
1.10
1.26
0.20
5.63
390
-------
The Phase II Action Plan
Given the great range in costs for the Phase II program it makes sense to get a jump start on planning for it. Many of
the requirements or potential inter-local arrangements that could be developed take time to implement, more time than is
available if the community waits until the general permit has been finalized and the NOI is due. There are steps that a local
government should take now to prepare itself for the regulations and to position itself to meet compliance in the most cost-
effective manner. These steps can be performed as part of a Phase II action plan:
1. Assess your status
Ask yourself if you are "in," "potentially in," or "out." Find out who else is in your category.
2. Get to know the permit writers
Find out what the permit writers are thinking about the permits, what the general permit will look like, when you will
know more, how they will evaluate those potentially in, what other actions are going on in the state that may impact the
permit, etc. Find out their ideas about what is important in the permit, what their special interests are, do they strongly
support the permit, etc. Plan to establish an ongoing dialog.
3. Assess your surface waters
Find out if there are any ongoing actions which might designate surface waters in your jurisdiction as not meeting water
quality standards. See if there are any planned watershed assessments or TMDL requirements coming in the future.
4. Assess your own program
How much of your own stormwater program looks like the regulations, even with some minor modifications. Can you
get a jump on the requirements through transformation of your current programs?
5. Check out your neighbors
Are there some other programs nearby that might result in savings to you? Can you simply be covered under another
program? Can parts of the requirements be waived because they are already being done by someone else? Can you
plan to be part of a regional permit? Can you split the permit requirements with an adjacent entity and perform them
together at savings to both of you?
6. Get a team together
Once you have answered some of these questions, it is time to pull the action team together. This may include only
your own staff, a multi-disciplinary staff within your own jurisdiction, or a multi-jurisdictional or regional team. Get together
to brainstorm and come up with a proposal to the permit writer which has mutual benefits. Remember, permit writers are
being encouraged to think regionally and on a watershed basis.
7. Develop an action plan
Once you have a team, it is time to have a plan. Begin to formulate what you will need to do to apply for the permit
and to carry it out. What might your program minimums look like? Are there some things you can do now, over several
years, that you cannot afford to do in any one year, or that will take too long to get going if you wait until the permit is upon
you? Can you begin the program transformation process now? What about data collection and mapping? Are there other
uses for any data you will collect which will create synergy?
391
-------
8. Get started
Some things are best started early. But do not jump the gun by committing resources in areas that are not yet
anticipated to be firm. Ask the permit writer for his or her opinion.
References
Center for Watershed Protection, 1997, The Economics of Stormwater BMPs in the Mid-Atlantic Region, August.
The National Association of Flood and Stormwater Management Agencies (NAFSMA), 1999a, Phase II Survey Raw Data
Report, 1299 Pennsylvania Ave. NW, Washington DC, 20004.
The National Association of Flood and Stormwater Management Agencies (NAFSMA), 1999b, Survey of Stormwater Phase
II Communities, 1299 Pennsylvania Ave. NW, Washington DC, 20004.
Northern Virginia Planning District Commission, 1994, Urban Retrofit Techniques: Applicability, Costs, and Cost-
effectiveness, November.
United States Environmental Protection Agency, 1995, Economic Benefits of Runoff Controls, EPA 841-S-95-002,
September.
United States Environmental Protection Agency, 1999, Report to Congress on the Phase II Stormwater Regulations, EPA
833-R99-001, October.
392
-------
Appendix
Table 2. Hypothetical Program Detail and Cost Summary
Smallville
Program Element
Public Education and Outreach on Stonnwater Impacts
Midtown
cost
Program Element
cost
Acquisition of available mailers and information from private
institutions and other governmental entities - 20 hrs
Keep up with available literature - 20 hrs/yr
$1 ,000
$1 000/yr
Acquisition of available mailers and information from private
institutions and other governmental entities - 40 hrs
Keep up with available literature - 50 hrs/yr
$2000
$2500/yr
Coordination with neighborhood or shoppers newspaper to run
articles on pollution sources - 4 hrs
Develop 2 articles per year - 24 hrs/yr
Coordination with the few individual potential sources of pollution
about the program and their needs - 10 hrs
Series of three mailings - stuffers in utility bill
One mailing per year afterward
Responding to information requests - 1/2 hr/wk
$200 Stratified mailing database development for key stakeholder
groups - commercial, automotive, minority, etc. - 100 hrs
$1 ,200/yr Maintenance of database - 1 hr/wk
$500 Obtaining or developing educational materials for the specific
outreach and stakeholders' programs, printing - 30 hr
Updating materials - 1 00 hrs/yr. Mailing 5,00 brochures per year
$3,600 Developing outreach and educational programs - 200 hrs
$1,050/yr Executing programs - updating, mailing, training, presentations -
200 hrs/yr
$1,300/yr Develop elementary and middle school education programs - preexisting
Material/curriculum - free materials - 100 hrs
Ongoing program maintenance - refresher training, 5 schools - 100 hrs/yr
Advertising of hotline - radio spots developed in-house and on public
And other radio service spots and Newspaper ad, 3 times per year -
1 40 hrs - donated spots
Develop white paper and press package - initial, brief - 32 hrs
Develop quarterly press package/briefing - brief press - 24 hrs per +
expenses
Development of a short, scripted stormwater pollution slide show,
Presentation and speakers bureau & initial presentation - 60 hrs
Give presentations - 48 hrs/yr + expenses
General informational brochure development and mailing - once/year
- 60 hrs/yr - 25,000 inserts @ 0.50 per
$5,000
$2,600/yr
$4,000
$7,500/yr
$10,000
$12,000/yr
$5,000
$5,000/yr
$12,000/yr
$1 ,600
$5,000/yr
$3,000
$2,600/yr
$15,500/yr
Continued
393
-------
Table 2. Cont.
Smallville
Midtown
Program Element
Total
Total Cost
Cost (ongoing 5
cost
Initial Cost
Annual Cost
(first 5 years)
-year period)
$5,350
$4,550
$23,500
$22,750
Program Element
Responding to information requests - 2 hrs/wk
Initial Cost
Annual Cost
Total Cost (first 5 years)
Total Cost (ongoing 5-year periods)
cost
$5,200/yr
$30,600
$69,900
$310,200
$349,500
Public Involvement/Participation
Development and implementation of a citizen advisory committee
appointed by the mayor - 2 initial meetings - 14 hrs
Quarterly meetings - 32 hrs/yr
Advertisement of the larger city's stream cleanup program in local
shopper newspapers - news articles, and coordination with them in
all such programs - 1 6 hrs/yr
$700
$1,600/yr
$800/yr
NOTE: italics are initial cost - for first year only
Total
Total Cost
Cost (ongoing 5
Initial Cost
Annual Cost
(first 5 years)
-year period)
$700
$2,400
$10,300
$12,000
Development and implementation of a citizen advisory committee
appointed by the council - 5 initial meetings - 70 hrs + expenses
Bimonthly meetings - 60 hrs/yr
Initial coordination of monitoring program and/or adopt-a-stream
- 60 hrs - equipment purchase
Ongoing^ coordination and equipment, database maintenance
- 100 hrs/yr + expenses
Student storm drain stenciling program development and
implementation - 80 hrs
Annual cost
Watershed group encouragement - presentations, advertising -
50 hrs + expenses
Ongoing coordination, education - 4 groups - 20 hrs. per
Initial Cost
Annual Cost
Total Cost (first 5 years)
Total Cost (ongoing 5-year periods)
$3,700
$3,000/yr
$40,000
$15,500/yr
$6,500
$3,000
$2,800
$4,000/yr
$53,000
$25,500
$155,000
$127,500
Illicit Discharge Detection and Elimination
Collect and plot field information on
5 hrs - contract
system locations
and sizes -
$7,000
Develop system map, perform inventory of major structures - 60
hrs + contract
$150,000
Update map - 60 hrs
$3,000/yr
Adopt ordinance - 20 hrs
Enforcement of ordinance - 20 hrs/yr
$1 ,000
$1 ,000/yr
Database development and GIS programming and mapping -
200 hrs + expenses of $3k
Database maintenance - 100 hrs
$13,000
$5,000/yr
Continued
394
-------
Table 2. Cont
Smallville
Midtown
Program Element
Initial Cost
Annual Cost
Total Cost (first 5 years)
Total Cost (ongoing 5 -year period)
cost
$8000
$1 ,000
$12,000
$5,000
Program Element
Ordinance development with public participation - contract
Initial dry-weather screen in parts of city - student volunteers - 240 hrs
One staff member 1 day/week for inspection and enforcement of
Illicit connection program - + expenses
Development of automotive or other specialty programs - 1 00 hrs +
1 k exp.
Annual implementation of inspection and education - 1 day/wk
Initial Cost
Annual Cost
Total Cost (first 5 years) $433,000
Total Cost (ongoing 5-year periods)
cost
$20,000
$12,000
$28,000/yr
$6,000
$22,000/yr
$201,000
$58,000
$290,000
Construction Site Stormwater Runoff Control
Modify and pass new erosion control ordinance - 40 hrs
Enforcement ordinance in inspection process - 50 hrs/yr
Modify development procedures - 4 hrs
Train secretary to handle calls - 8 hrs
Handle erosion calls - 1 0 hrs/yr
Initial Cost
Annual Cost
Total Cost (first 5 years)
Total Cost (ongoing 5 -year period)
Post-Construction
Modify and get ordinance passed • 40 hrs
Enforce/explain new ordinance provisions - 1/2 hr/wk
$2,000
$2,500/yr
$200
$800
$500/yr
$2,600
$3,000
$14,600
$15,000
Stormwater Management in
$2,000
$1 ,300/yr
Modify existing ordinance - public participation - 60 hrs
Add BMP section to design manual - 140 hrs + printing cost
Conduct training sessions for staff and local development related
persons - 80 hrs
Ongoing biannual training - 32 hrs/yr
Develop hotline procedure for complaints reception - 10 hrs
Hotline @ 150 hrs/yr + expenses
Upgrade erosion control program for more sites and more activities -
one person two days/wk + expenses
Initial Cost
Annual Cost
Total Cost (first 5 years)
Total Cost (ongoing 5-year periods) $275,500
New Development and Redevelopment
Work on major policy changes in land use regulations - contract +
200 hours
Develop design guidance for BMPs - contract
Training program for BMP use - debvelopment - 24 hrs + contract
Annual training - 60 hrs/yr
$3,000
$12,000
$4,000
$1 ,600/yr
$500
$8,500/yr
$45,000/yr
$19,500
$55,100
$239,900
$100,000
$25,000
$3,000
$3,000/yr
Continued
395
-------
Table 2. Cont.
Smallville
Midtown
Program Element
Initial Cost
Annual Cost
Total Cost (first 5 years)
Total Cost (ongoing 5 -year period)
Pollution
Review of all current procedures - modification of procedures - 40 hrs
Obtain
Annual
T
0
T
A
L
T
0
T
A
L
and distribute educational materials - 10 hrs
cost of changed procedures - SWAG
Initial Cost
Annual Cost
Total Cost (first 5 years)
Total Cost (ongoing 5 -year period)
Initial Cost
Annual Cost
Total Cost (first 5 years)
Total Cost (ongoing 5 -year period)
cost
$2,000
$1 ,300
$7,200
Program Element
BMP Pilot projects - federal funding assistance - 5-year program •
contract 5 yrs
Comprehensive stormwater ordinance with public participation -
contract
BMP inspection and enforcement program - one person one day/wk
+ expenses
Data collection program - SWAG
Master planning for new areas for both quality and quantity - 2 mile
Planning zone around -cSyeipr program -40 mi2
Costs of administration of regional BMP program - SWAG
Sensitive area identification program, ordinances and policy enactment-
5-year program - 100 hrs incl. Mapping 5 yrs
Initial Cost
Annual Cost
Total Cost (first 5 years)
Total Cost (ongoing 5-year periods)
Master planning
cost
$200,000
$40,000
$25,800/yr
$30,000/yr
$800,00
5 yrs
$4,000/yr
$25,000
$393,000
$62,800
$644,200
$314,000
Prevention/Good Housekeeping fo Municipal Operations
$2,000
$500
$1,000/yr
$2,500
$1 ,000
$6,500
$5,000
$21 ,500
$13,250
$74,150
$66,250
Review and modification of all applicable procedures and criteria
contract
Site inspections and corrections - 5-year program - $5k/yr
Training for city employees on new procedures - 40 hrs + 10 hrs @
75 persons + expenses
Review flood control projects for retrofit opportunities - contract
Annual cost of changed procedures - SWAG
Initial Cost (without master planning)
Annual Cost
Total Cost (first 5 years without master planning)
Total Cost (ongoing 5-year periods)
Master planning
Initial Cost (without master planning)
Annual Cost
Total Cost (first 5 years without master planning)
Total Cost (ongoing 5-year periods)
Master planning
$25,000
$25,000
5 yrs
$42,000
$15,000
$10,000/yr
$107,000
$10,000
$147,000
$50,000
$800,000
$804,100
$281 ,300
$1 ,929,300
$1 ,406,500
$800,000
396
-------
The Stormwater Utility Concept in the Next Decade
(Forget the Millenium)
Hector J. Cyre, President
Water Resource Associates, inc.
Kirkland, WA
Abstract
In the mid-1 970's, the first stormwater utilities were viewed as novel innovations in a few western states. Today, just
25 years later, more than four hundred cities, counties, and special districts throughout the United States have established
such utilities. The pace is accelerating, and the stormwater utility concept has moved from a novelty to a well-accepted
management and funding approach. What will we see in the next decade?
The stormwater utility has been adapted to fit diverse stormwater management problems and needs across the United
States. Program content, priorities, institutional and organizational structures, and rate methodologies have been tailored
to fit local needs and municipal authority and practices that vary widely. Courts in several states, and even federal courts,
have been engaged in resolving key issues, including but not limited to the legality of utility service fees and the use of
other funding mechanisms.
Major changes in the concept are still emerging today. Stormwater quality has become a concern equal to flood
control in many communities. The National Pollutant Discharge Elimination System Phase II stormwater permits have
spurred a new round of interest in the stormwater utility concept among smaller communities. This is creating a demand
for basic utility concepts suitable for small cities and towns, which will need to be less costly and simpler to implement and
maintain. Concurrently, more large cities, urban and urbanizing counties, regional service agencies such as metropolitan
sewer districts, and consolidated governments are investigating the utility approach. They will require more complex
institutional and funding solutions.
Stormwater management itself is also changing rapidly. Interest and involvement in stormwater management have
broadened. As combined sewer overflow programs, total maximum daily load (TMDL) negotiations, stormwater quality
mandates, coastal zone management measures, and safe drinking water supply issues converge, more wastewater and
even water supply utilities are engaging in stormwater management. Regional resource management programs,
watershed-based master planning, multi-purpose cooperative efforts involving urban forestry and riparian corridor
protection, and use of state revolving loan funds for stormwater quality projects are becoming more common.
Local programs are quickly evolving as well. They have become more comprehensive in scope, more costly, and more
demanding of technical and administrative skills while the pool of resources has grown relatively slowly. Local
governments are accepting responsibility for more components of the stormwater drainage systems or, in some cases,
being forced to take on such responsibilities. Open streams, historic remnants or agricultural ditches and levees, and
detention facilities are being included among the system components actively improved, operated, and maintained by local
stormwater management agencies, A preventive orientation that minimizes problems is replacing reactive measures.
Technology, such as geographical information systems and hydrologic and hydraulic modeling, is more widely available
and more productive in support of stormwater management, even in smaller communities. Public involvement in decisions,
policies, and even the operation of systems is increasing.
This paper examines these and other emerging trends that characterize where stormwater utilities are heading in the
next decade.
397
-------
Pressures Moving us from Draining the Swamp to Stormwater Management
Historically, local drainage flooding, erosion, and water pollution due to stormwater runoff have not been high priorities
for municipal governments. Unless homes, businesses, valuable agricultural land, or public properties have been
devastated by flooding or other "drainage" problems, competing priorities have generally garnered more public concern
and thus more support from elected officials. As a result, stormwater management operations, regulatory measures, and
capital investment were historically ignored or, at best, received inadequate attention and erratic funding. Stormwater
management has been a "stepchild" among municipal programs.
Symptoms of this past disregard are evident in many cities and counties.
. Improvements to stormwater systems in many communities have been limited to site-specific facilities installed by
subdivision and commercial developers.
. Design practices have traditionally emphasized collecting and discharging runoff from each property as quickly as
possible, without regard for downstream consequences.
. Public maintenance of stormwater systems has typically been reactive, and usually limited to road rights-of-way
where uncontrolled stormwater might impact traffic safety, degrade the integrity of road surfaces, or threaten
valuable adjacent properties.
. Maintenance of stormwater systems located outside of road corridors has commonly been left to private property
owners, who are rarely capable of or willing to properly improve, clean, and repair such facilities.
. Municipal governments have usually improved and maintained individual structures or reaches instead of entire
drainage systems, creating a patchwork of pieces having widely varying capacity and reliability.
• Failures of substandard components frequently impair the performance of otherwise adequate parts of the systems
and damage properties near them.
As described by one municipal public works official, this stepchild is also the "sleeping giant" of unmet municipal
infrastructure needs. Long-term stormwater remedial repair costs potentially exceed street and bridge repair needs in
many older cities. Learning the high cost of correcting stormwater management deficiencies through master planning may
have frightened as many local jurisdictions into inaction as it has spurred others. Perhaps the classic example is the
stormwater master plan for Key West, Florida, which (in the early 1990's) identified $78 million in capital needs for that four
square mile island community of less than 30,000 people.
Several factors are now changing local governments' traditional orientation to stormwater management.
. Citizens' service expectations are higher than in the past. In many cities and counties the number of citizen
complaints about stormwater problems exceeds those about potholes in roads.
. Crumbling inlets and silt clogged ditches along roadsides spawn complaints even though they are on public property.
. Individual citizens or neighborhood associations no longer tolerate minor problems like localized flooding and
channel erosion in backyards.
Environmental awareness in general is greater than in the past, and much more attention is being focused on
stormwater impacts on receiving water quality in recent years.
. Stormwater management is now recognized as being part of an effective water resource protection strategy.
. Local concerns about acute threats of water pollution from spills and surreptitious dumping of toxic materials into
stormwater systems are becoming more common.
. Phase II of the National Pollutant Discharge Elimination System (NPDES) stormwater permitting program is
extending the program to smaller communities and those larger urban cities that escaped Phase I due to combined
sewer service area exemptions.
. Programs proposed by local governments in NPDES Phase II permit applications will cost many thousands of dollars
per year in cities, towns, and urban counties.
398
-------
An encompassing, umbrella perspective of water resource management is emerging.
. Solutions to combined sewer overflow (CSO) problems will have to balance optimization of wastewater transport
and treatment facilities against stormwater quantity and quality concerns.
• Several coastal states have instituted restrictive limitations on stormwater runoff to protect fragile estuaries and
offshore waters from stormwater impacts.
. Drinking water supply watershed protection measures have imposed stormwater runoff regulations on developers
independent of local stormwater management control practices.
. The point is becoming clear. Drinking water is water. Wastewater is water. Stormwater is water. Ground water
is water. It is all WATER!
In the face of these pressures, the inadequacies of traditional stormwater management practices and funding are more
widely recognized. More comprehensive and cohesive programs that address both stormwater quantity and quality are
emerging. Clearly, however, the diversity of our communities and their problems and priorities means that no single
solution is appropriate for every county, city, town, and village. Nor can a single funding method or stormwater utility rate
structure fit every situation. Stormwater service fee methodologies can be designed to meet the specific needs of each
community and provide equitable, adequate, and stable funding. The key is to tailor the funding to a clear program
strategy.
"Stormwater Utility" can have Many Meanings
The fact that the simple term "stormwater utility" obscures the various meanings it may encompass, results in many
misunderstandings. The term may imply a funding and accounting method, an organizational approach, a management
concept, or a combination of all these. In reality a "utility" provides an umbrella under which the financial, organizational,
and management approaches of each local stormwater program can be orchestrated to achieve practical and efficient
solutions. Responsibilities may be consolidated and focused. Substantial new funding may be generated. New
technology, different management concepts, and upgraded support systems may be adopted. A comprehensive,
preventive program may be instituted.
Changes in the Approaching Decade
The spectrum of the stormwater utility concept will broaden more in the next 10 years than it has in the 25 years since
the first utilities were established. The definition of "conventional" will change. Smaller towns and even villages will need
to employ simpler variations of the concept. Larger cities, urban counties, consolidated governments, and coordinated
regional approaches will demand more complex institutional, organizational, and funding solutions. The following are a
few of the changes that may occur.
NPDES Phase II Will Impact the Stormwater Utility Concept
The findings of a survey of Phase II cities conducted by the National Association of Stormwater and Flood
Management Agencies (NAFSMA) and published in July, 1999 indicate that 17% of all the respondent communities did
not know how they would obtain funding to meet the stormwater regulations. Nearly half indicated they were not currently
spending money on any of the stormwater program elements mandated by the regulations. Nearly three-quarters did not
have a public information or education program as the regulations mandate. The 54% of respondents that currently fund
programs or activities that fit the Phase II regulations on average spend upwards of $4,000 per square mile, or about $2.79
per capita. The implication seems clear that the NPDES Phase II program poses demands on local governments that may
cause them to look to the stormwater utility concept to meet their stormwater quality program funding requirements.
The typical community that found the stormwater utility concept attractive in the past was a mid-size to larger city
undergoing rapid development. Analyses of their stormwater management needs and programs typically revealed initial
costs of service ranging from $25,000/sq. mile to $50,000/sq. mile annually. Costs per capita were typically $10 to $30
annually, with smaller cities trending toward the higher end of the range. In this context, the NAFSMA survey data is not
399
-------
alarming. It suggests that the Phase II program costs will likely be in the neighborhood of five (5) to fifteen (15) percent
of the typical cost of stormwater management at the outset of utility-based programs.
Given its cost implications, NPDES Phase II makes the stormwater utility concept attractive to a broader variety of
cities, counties, towns, and villages. Many communities that do not suffer flooding or other drainage problems will find the
revenue potential and flexibility of a utility service fee attractive in the face of NPDES permit requirements. This will result
in utility approaches that are outside the current spectrum of our experience. The needs in individual small communities
may be less diverse than in large cities and urban counties, but the range will be cumulatively broader among the
communities involved in NPDES Phase II than in those that have implemented utilities previously.
New institutional arrangements and relationships will have to be devised. The "utility" concept will take on new forms.
Use of interlocal agreements among several local jurisdictions will increase, with responsibilities in some cases
concentrated in one entity capable of providing the range of services required or, conversely, allocated among several
participants.
The limits of existing authorizing legislation in some states will be tested. Many states will need to adopt new
legislation and amendments, giving local governments greater flexibility in dealing with their water resource management
responsibilities. Courts in the various states, and perhaps even federal courts, will be challenged to arrive at some sense
of continuity among the institutional and financial solutions characterized as "stormwater utilities." Whether the court
decisions will enable rather than hinder local governments' efforts to comply with NPDES mandates is a key question.
Organizationally and financially independent stormwater utilities have been common to date. In the next decade more
stormwater utilities will be integrated with other water resource programs: organizationally, through formalized working
relationships, or through financial arrangements. Other resource management agencies and programs, including but not
limited to health departments and growth management authorities, will demand a seat at the stormwater table.
As NPDES permitting is applied to smaller urban areas of less than 100,000 people, more of the regional wastewater
and water utilities already serving those communities will assume stormwater management responsibilities. In some cases
their involvement will be limited to water quality aspects. In others they will address both quantity and quality. Funding
of stormwater management costs will simply be assumed by some of these existing utility agencies without changes in
their rate methodologies. Others will establish independent stormwater cost centers and rate components to track
spending and allocate costs. Some will even modify existing wastewater and/or water rate methodologies to better reflect
the impact of stormwater control on costs of service.
As more small cities and counties seek to establish utilities, stormwaterfunding strategies and rate methodologies will
need to minimize implementation costs, yet be more flexible to accommodate stormwater quality management costs and
unique local needs. The urge to use a "cookbook" solution will cause some to adopt approaches that are poorly suited
to their circumstances. The desire for more precision in service fee rate algorithms will lead to methodologies that give
an illusion of greater refinement without actually achieving it.
The mandated involvement of smaller jurisdictions and more rural communities in stormwater management will spawn
"paper utilities" established solely to generate added revenue. Most of these will be initiated without the foundation of a
solid program strategy. Accountability will become a key issue in some of these communities within a few years. Political
challenges based on accountability issues will cause some of these storm water utilities to be melded into other local
agencies'or programs or even dissolved entirely before they have geared up to address their program priorities.
Despite NPDES storm water permit mandates, locally perceived needs will still predominate in setting priorities.
Flooding will remain a more important local issue than storm water quality. NPDES mandates will influence actual spending
priorities only slightly. Few communities will need to institute a utility service fee just to support their NPDES Phase II
programs, but many will justify it (at least partially) on that basis because it is easy to blame unfunded federal mandates
for new local taxes, assessments, and service fees.
400
-------
The technological resources and expertise required will change from the traditional engineering emphasis to a multi-
disciplinary mix. More natural science and social science skills will be needed. Operational practices will change as new
technology and information management systems enable innovative approaches and result in greater efficiency. Greater
use will be made of outsourcing because of limited personnel resources and the high cost of specialized equipment.
The Stormwater Utility Concept will Impact NPDES Phase II
Local approaches to stormwater management will influence the content of Phase II permits and attainment of NPDES
objectives. Stormwater utilities offer both financial capability and flexibility. Except in rare instances, stormwater utilities
will not be established strictly to address stormwater quality and NPDES permit requirements. Rather, they will have a
broader stormwater management perspective. For many communities this will mean that water quality management
activities to comply with their NPDES permit will be tacked onto other stormwater efforts. NPDES Phase II permitting will,
within a few years, adjust to accommodate this reality in terms of permit mandates, technical and scientific standards, and
reporting requirements.
Related issues ranging from combined sewer overflow strategies to drinking water protection will be melded with Phase
II permit requirements because they have to be. Local stormwater quality management cannot independently meet the
entire range of regulatory expectations operating strictly by reference to NPDES Phase II. Conflicts and primacy battles
will identify inconsistencies and gaps between the issues and programs, and will ultimately filter down to changes in
NPDES Phase II program priorities and the permit requirements imposed on local governments. The unknown is whether
this result in responsibility shifting toward bigger agencies with more resources and a broader perspective or toward local
entities that have the ability to identify and activate locally acceptable solutions.
Watershed-based regulatory programs will overtake jurisdictional-based regulatory programs like NPDES Phase II.
The utility approach will broaden to encompass watersheds through agreements among counties and cities simply because
utility funding has the proven capacity to generate sufficient funding in politically acceptable ways. The transition has
already begun in some areas. Where TMDLs affecting discharges of all sorts into receiving waters are an issue, they will
supercede the six minimum practices identified for NPDES Phase II, making them essentially meaningless. Scientifically
based, public health driven measures to protect drinking water supplies, estuaries, lakes, fisheries, and recreational
beaches will overwhelm the programmatic approach represented by NPDES Phase II.
You May Need a Program to Identify All the Players
Stormwater utilities were first established because no one wanted responsibility for stormwater management. Those
involved were concerned only about the impact of stormwater on their "real" jobs. The utility approach provided a way to
focus responsibility and obtain dedicated, if not always adequate, funding for stormwater management. If there had been
another option that was working, the stormwater utility concept probably would never have emerged.
A key issue in the next decade will be whether stormwater utilities will be major protagonists or bit players among all
those now crowding onto the stage. More established and better-funded water and wastewater utilities now recognize that
stormwater influences their operations directly and, in some cases, dramatically. For example, TMDL-based wastewater
discharge limitations may severely curtail development in some areas. Will local wastewater utility administrators (and
local elected officials) allow independent stormwater management utilities to address stormwater quality when economic
vitality is at risk?
Other interests are becoming involved in stormwater management. Water supply utilities face the requirements of
federal and state legislation regulating sources of supply and treatment. Coastal zone management has recognized that
many priority uses of the shorelines and near-shore areas are dependent on good water quality. Growth management
is an emerging concept, and concurrency of infrastructure improvements with development approvals highlights the issue
of deficient stormwater systems in many communities. Protection of endangered and threatened species, urban forestry,
and riparian corridor protection all have a relationship with stormwater management.
401
-------
The Walls Will ComeTumbling Down (or at least they better)
The proliferation of federal, state, and local water resource (and related) regulations in recent years has created an
environment in which dispersed responsibility for water in various forms and for various purposes is rapidly becoming
unworkable. The institutional barriers that have been created over the past hundred years or so to focus attention, energy,
and responsibility no longer fit the public needs. As watershed-scale studies, planning efforts, and the concept of TMDLs
clearly illustrate, water resources are inextricably bound together regardless of their temporary form, use, and character.
The next decade will see accelerating consolidation of water resource management responsibilities at the local level
of government. This is contrary to the control interests of some individuals and entities, and will not happen silently or
easily. Will cities, counties, and special districts relinquish a little (or a lot) of their control over water resources through
interlocal agreements? Will they accept a regional entity for water supply, wastewatertreatment, stormwater management,
or even water quality? What will be the effect on stormwater utilities?
What are the organizational implications of the coming changes in storm water management? Realistically, local
governments change slowly. Public Works and Street Departments have historically been the lead organizations of storm
water programs, but they rarely have had much involvement in water quality issues. If storm water quality begins to
influence local priorities, it is more likely that water and wastewater utilities will assume storn water management
responsibilities from Public Works and Street Departments than the reverse. Public Works agencies will have to upgrade
their engineering and scientific capability or risk losing their storm water management role to water and wastewater utilities
that are typically well-established, well-funded, and well-understood by the public.
The Ability to Innovate Will Exceed the Need
Most of the early stormwater utilities programs were rather narrowly focused, and the funding mechanisms supporting
them were relatively simple. In recent years, however, there has been a shift toward more sophisticated and complex
approaches to all aspects of stormwater management-from master planning to rate methodology design. Much of the
credit goes to the explosive growth in information processing capability associated with the computer revolution of the past
20 years. It is not clear, however, that much of the added capability to innovate is necessary to meet stormwater
management needs. This is not to suggest that opportunities to improve should be ignored simply because they are based
on increasing capability to do so. The following examples demonstrate how the ability to innovate in stormwater through
technology can run amok, and suggest how it should be managed to the benefit of people and the environment.
There is no substitute for understanding what is really important. One Southeastern United States city invested over
$1 million dollars assembling a highly detailed location inventory of its stormwater systems on a relatively sophisticated
data processing platform. Unfortunately, the need for the inventory was not premised on a clear program strategy, nor
was adequate funding available or established concurrently to support capital improvements or maintenance
enhancements that could be facilitated by the inventory. The local elected officials finally tired of the seemingly mindless
spending on the inventory and refused to discuss program improvements. Today, nearly 10 years later, the inventory has
not been maintained and is out of date, and few improvements have been made in the stormwater management program.
What is technically possible does not always make common sense, and what makes sense is not always technically
possible. A Northeastern city recognized that the stormwater component of its wastewater service fee rate methodology
(one that was based on water meter size and internalized within its water/wastewater rates) was not reasonably allocating
the cost of stormwater services and facilities across the community. Change to a more rational approach was desired,
so a thorough assessment of the range of options was undertaken. A broadly representative advisory committee aided
in the selection process. A relatively simple stormwater rate concept was selected that segregates stormwater funding
from wastewater and water service. It will allocate a portion of the cost of stormwater service on the basis of gross area
and a portion on the basis of impervious area. Once the impact of the change on certain rate payers was recognized,
however, the advisory committee decided that phasing in the new rate over three years was a better idea than making the
change in one step. While the technical support requirements of the phased approach are not especially demanding, the
public information and education challenge is enormous. Not only must the new rate methodology explained to the public;
it and the phase-in concept must be explained every year for three years.
402
-------
Errors by a Few Will Make Life Miserable for Many
As the number of stormwater utilities grows, there is a natural tendency among municipal managers to assume that
the process and results have become standardized, and the experiences of another community can simply be transferred.
In an effort to save money, some cities and counties have established utilities without sufficient foundation and have even
adopted service fee ordinances without the benefit of a cost of service analysis or a rate study. Such misjudgments have
led to some monumental errors that have the potential to erode if not destroy the viability of the utility concept in a region,
a state, or even nationally.
One city recently established a stormwater utility and adopted rates based on internal analyses that did not define a
program, project the cost of service, or estimate the rate base available to generate revenue. As a result, the initial service
fee billing was for nearly three times as much total revenue as the administration had indicated it hoped to raise for
stormwater management. Furthermore, sufficient public information and education had not been conducted prior to the
initial billing, so the public did not understand the purpose of the billing. A lawsuit was filed, and a same judgement on
behalf of the plaintiff has resulted in the servie fees being rescinded and revenues returned with interest.
Expectations Will Advance Faster than Programs
One common experience of the cities and counties that have established stormwater utilities is that public expectations
for the program have exceeded the utility's ability to perform. This means that creating accurate expectations before a
utility is established must be a high priority. One cause for unfulfilled expectations is that stormwater utility revenue
streams are usually insufficient to address all the accumulated problems in a relatively short time. Initial stormwater utility
service fees have typically been less than $3 month for single-family residences.
Perhaps more significant, however, is the fact that most stormwater utilities inherit programs and systems that are not
only deficient, but also do not offer an adequate foundation for a good, more comprehensive, program. Utilities often must
invest one to three years creating the foundation for the program before real results begin to emerge in the form of capital
improvements, remedial repairs, upgraded maintenance, and more effective regulations. Ratepayers tend to have little
patience, however, when they are writing checks regularly to a stormwater utility.
In the context of NPDES Phase II permits, public expectations of improvements in water quality need to reflect the
complexity of water quality issues and the limited ability of local government to quickly alter conditions in receiving waters
through informational and regulatory programs. Attempting to sell a utility to a community as a response to federal water
quality mandates has been unsuccessful in several communities. The public recognizes that stormwater quality, while
important, is still a minor part of the total cost of stormwater management. Unless a comprehensive quantity and quality
control program strategy is apparent, it is difficult to generate support for a stormwater utility.
403
-------
Are Green Lots Worth More Than Brown Lots?
An Economic Incentive for Erosion Control on Residential Developments'
Martha Herzog and Jon Harbor
Department of Earth and Atmospheric Sciences
Purdue University
West Lafayette, Indiana 47937
Keith McClintock
Geauga Soil and Water Conservation District
Burton, Ohio 44021
John Law
Indiana Department of Natural Resources, and Saint Joseph Soil and Water Conservation District
South Bend. Indiana 46614
Kara Bennett
Departments of Statistics
Purdue University
West Lafayette, Indiana 47937
Abstract
Construction sites are major contributors to nonpoint source (NPS) pollution. However, a lack of personnel to enforce
erosion control regulations and limited voluntary compliance means that few developers apply effective erosion control.
New approaches are needed to increase erosion control on construction sites if this source of NPS pollution is to be
significantly reduced. We have tested whether an economic advantage exists for developers who use vegetative cover
for erosion control, independent of advantages gained in addressing environmental or regulatory concerns. Improving
residential lot appearance from muddy brown to green grass may increase the appeal of the lot to buyers. A market survey
shows that homebuyers and Realtors perceive vegetated lots to be worth more than unvegetated lots, and this increased
value exceeds the cost of seeding. Thus, developers can now be encouraged to invest in vegetative cover because of the
potentially high return on the investment.
Introduction
Sediment Pollution and Construction Sites
Nonpoint source (NPS) pollution, produced from diffuse sources such as runoff from agricultural land, construction
sites, and urban surfaces, is now the leading cause of surface waterqualitydegradation in the United States (Novotny and
Chesters, 1989; Federal Register, 1990). In developing areas, construction sites are a major source of NPS pollution
because soil erosion rates are increased dramatically when land is exposed and disturbed by excavation and vehicular
movement (Harbor et al., 1995; Goudie, 1994; Goldman et al., 1986, Fennessey and Jarrett, 1994). In fact, some of the
greatest soil erosion rates ever reported are associated with construction activities (Crawford and Lenat, 1989); erosion
rates on construction sites are typically 2-40,000 times greater than rates under preconstruction conditions (Wolman and
Schick, 1967; Harbor, in press). Sediment contributed to streams by construction sites can exceed that previously
1 This paper is reprinted from the Journal of Soil and Water Conservation, Spring Issue, 2000. We thank the Soil and Water Conservation Society for
their permission to reprint this article.
404
-------
deposited over many decades under pre-development land uses, radically altering stream geomorphology and ecology
(e.g., Wolman and Schick, 1967). The larger-than-normal sediment deposition in waterways frequently exceeds the natural
capacity of the receiving water system to assimilate and equilibrate to the sediment influx (Paterson et al., 1993), causing
rapid channel changes and increased probability of flooding, erosion, and sedimentation problems (Goldman et al., 1986).
In addition to sediment, construction sites generate other pollutants such as pesticides, nitrogen, and phosphorus from
fertilizers, petroleum products such as oil and gas from machinery, soil stabilizers, construction chemicals, and washings
from concrete or bituminous mixing and flushing operations (Koehn and Rispoli, 1982; Lemly, 1982). In some cases these
pollutants are in particulate form or are adsorbed by soil particles and are transported with the suspended sediment in
runoff from construction sites (Paterson et al., 1993; Bhaduri et al., 1997).
Although construction sites generate a wide range of potential pollutants, sediment overshadows all the other
construction site pollutants in total ecological and economic impact on receiving waters (Lemly, 1982). It was estimated
that 15 million tons of sediment were released from urban construction sites to surface waters in or near heavily populated
areas in 1975 (Lemly, 1982). The North Carolina Department of Natural Resources and Community Development has
stated, that "sediment and its effects on stream environments" is the "most widespread water quality problem in North
Carolina" (Crawford and Lenat, 1989). Because construction activities predominantly occur near existing population
centers, the waters that are most seriously degraded are generally those that are most frequently used (Lemly, 1982).
Economic Consequences of Sedimentation
In addition to environmental impacts, enhanced delivery of sediment to off-site areas from construction activities has
significant economic effects (Table 1). These economic impacts result from lakes and streams becoming turbid and filling
with silt, destruction of commercial aquatic species, the need for additional treatment of turbid waterfor industrial use, filling
of harbors and navigation channels, loss of storage capacity of reservoirs, damage to drainage ditches, increased
frequency of flooding, loss of aesthetic value in the environment, and loss of game habitat (Lemly, 1982; Wolman and
Schick, 1967; Koehn and Rispoli, 1982). The economic burden of mitigating these environmental impacts is almost always
placed upon the taxpayer, rather than on the operator of the construction site that is producing high sediment yields
(Harbor, in press). By not paying to prevent the off-site transport of sediment through the use of erosion control measures,
the developer allows sediment from the construction site to reach waterways where the economic and environmental costs
of any impacts are paid by downstream landowners and the community as a whole, and not the developer.
Overall, annual expenditures for in-stream and off-stream impacts due to sedimentation in the United States exceeds
$11.6 billion (Table 1). In-stream effects include impacts while sediment is in a waterway (stream, river, lake, or reservoir).
Off-stream effects can occur before or after sediment reaches a waterway, either in floodwater or in water withdrawn from
waterways to be used for industries, municipalities, or agriculture (Clark, 1985; Clark et al., 1985; Paterson et al., 1993).
Although agricultural areas are far more extensive than construction sites, the mass of sediment per unit volume of runoff
from urban and construction areas is 5 to 20 times greater than that of runoff from agricultural lands (Fennessey and
Jarrett, 1994). In addition, construction sites are usually located in developing or developed areas, where potential impacts
on infrastructure and other water uses are more severe than in rural areas. Estimates of agriculture's contribution to off-
site effects range from 1/3 to 2/3 of the total (Clark, 1985; Clark et al., 1985; Colacicco et al., 1989, Pimentel, et al. 1995).
Thus, urban off-site environmental impacts are probably on the order of $3.9 to $7.8 billion per year (1/3 to 2/3 of the total
off-site effects), and are often borne by off-site landowners and communities. One of the main goals of erosion and
sediment control regulations is to avoid these costs. The problem, however, is that developers have to pay to reduce
erosion yet do not see any immediate return on this investment. Because there is little economic incentive for developers
to control erosion, regulatory and educational approaches have been developed to improve construction site erosion
control, and requests to impose impact fees on developers have increased (Trotti, 1997).
405
-------
Table 1. Off-site damage costs from soil erosion by water in the United States
Cost
Type of Damage (millions in 1997 dollars*)
In-stream damage
Recreational (fishing, boating, swimming) 3,886.0
Water storage facilities (dredging, excavation, construction of sediment
pools) 1,340.7
Navigation (accidents, dredging) 1,088.1
Other in-stream uses (commercial fisheries) 1,748.7
Subtotal in-stream 8,063.5
Off-stream effects
Flood damages (sediment damage to urban and agricultural areas) 1,496.1
Water conveyance facilities (sediment removal of drainage ditches and 388.6
irrigation canals)
Water treatment facilities 194.3
Other-off stream uses (municipal and industrial, steam electric power 1,554.4
plants, irrigation)
Subtotal off-stream 3,633.4
Total water erosion costs 11,696.9*
(Data based on: Clark et al., 1985)
'Conversion using Consumer Price Index from 1980-1997.
* Assuming that effects are the same today as in 1980.
Vegetation and erosion control
The significant ecological and economic impacts of sedimentation provide strong motivation for erosion control. Soil
erosion involves the detachment of soil particles by raindrop impact, wind-blown particle impact, wetting and drying cycles,
freezing and thawing, and runoff, and the transport of detached soil particles by rain splash, wind and runoff
(Ekwue, 1990; Goldman et al., 1986). Climate, topography, vegetative cover, and soil characteristics are the principal
factors that control soil erosion potential. Climate and soil characteristics cannot be readily controlled on a site, and
topography is constrained by pre-existing conditions and the grading plan, leaving surface cover as the most easily
modified variable that controls oil erosion on a site. Increasing vegetative cover on barren areas such as construction sites
is an excellent way to impede soil erosion and decrease sedimentation (Fig. 1).
"Vegetative cover is the most effective form of erosion control...a properly revegetated soil will be protected from
erosion indefinitely without any need for human attention" (Goldman et al., 1986, p. 6.23). Vegetation (especially close
to the ground surface) protects the surface from raindrop impact and reduces the velocity of water flowing over the surface
by increasing surface roughness and disrupting overland flow (Clark et al., 1985; Rogers and Schumm, 1991; Satterlund,
1972). The reduction of water velocity flowing over the surface and the breaking up of soil by plant roots increases the
amount of infiltration, thereby reducing the amount of surface water flow (Clark et al., 1985). Vegetation also depletes
subsurface water between rainfall events, which reduces the amount of runoff during storm events. In fact, vegetative
stabilization on construction sites has been shown to reduce soil loss by 80% (Harbor et al., 1995) to 99% as compared
to bare soil (Koehn and Rispoli, 1982). The cost of reducing soil erosion using vegetative cover depends on the materials
used, but typical temporary seeding on a one-third acre residential lot in the Midwestern US costs from $250 to $325.
Regulations requiring construction site erosion control
In the US, the biological and physical impacts of off-site sedimentation have prompted local, state, and federal
regulations requiring erosion and sediment control for construction sites. The National Pollution Discharge Elimination
System (NPDES) is a national program that issues, monitors and enforces permits for stormwater discharges associated
406
-------
800
600
Sediment
Yield
(kg) 400
200
180
JO 20 30 40
Vegetation Cover (%)
50
Figure 1. Sediment yields for different vegetative cover densities at 30, 60, 120, and 180 minutes of simulated rainfall on a 10% slope (Rogers and
Schumm, 1991).
with industrial activity such as construction underthe Clean Water Act (Federal Register, 1990). State and local regulators,
under the NPDES program, require erosion and sediment control for construction sites with 5 acres or more of land
disturbance (Federal Register, 1990). Because vegetative cover greatly reduces soil erosion, many federal and state
regulations, such as Rule 5 in Indiana and the Model Regulations for Urban Soil Sediment Pollution Control in Ohio,
encourage the use of surface cover as an important element of erosion control on construction sites.
In Indiana, for example, state regulations mandate that sediment should be contained on the construction site and not,
for example, allowed to run onto public or private roadways. Rule 5 requires that if vegetative practices such as seeding
and mulching are used, they must be implemented within seven days of the "last land-disturbing activity" at the site and
that these actions are the responsibility of the person in charge of the construction activity, which usually is the developer
(Indiana Department of Natural Resources, 1992). Similarly, in Ohio, Model Regulations for Urban Soil Sediment Pollution
Control (1980) require that the responsible party for the development stabilize denuded areas with permanent ortemporary
soil stabilization within seven days for any denuded area that has reached its final grade or is to remain dormant for more
than 45 days. The permanent vegetation is not "considered established until ground cover is achieved which., .provides
adequate coverand is mature enough to control soil erosion satisfactorily and to survive adverse weatherconditions"(0hio
Department of Natural Resources, 1980).
Enforcement of erosion control regulations varies significantly among states. For example, in Indiana, the Indiana
Department of Environmental Management (IDEM) controls permitting and enforcement, but local soil and water
conservation districts (SWCD) review and evaluate erosion control plans. At typical staffing levels, SWCDs in developing
areas find it very hard to keep up with the large number of developments they are responsible for. The local SWCDs
inspect the construction sites to establish whetherthe developer is implementing the soil erosion control plan correctly and
to observe whether the possibility of or the actual transport of sediment off-site exists. The SWCD will provide the
developer with written recommendations describing which erosion control measures need to be improved, maintained, or
installed. The developer then has two weeks to comply with the recommendations. If the developer is not found in
compliance with the requirements after recommendations have been made, the SWCD reports the site to the Urban
407
-------
Erosion Control Specialist from the Indiana Department of Natural Resources (IDNR), who has been receiving copies of
all written warnings to the developer. The Urban Erosion Control Specialist will then visit the site and determine whether
the site should be reported to the IDEM. Subsequently, the IDEM determines whether further action, such as levying a
fine against the developer, is warranted. This process can lead to delays of many months between identification of a
problem and regulatory enforcement.
In reality, it takes a great deal of coercion to get developers to promptly apply erosion control measures on their sites.
Developers find applying erosion control measures inconvenient, costly, and time consuming, and are fully aware of the
lack of regulatory personnel to enforce local, state, and federal mandated erosion control (Harbor et al., 1995). Therefore,
developers often do not comply with the regulations and let their sites remain bare (Harbor, in press; Harbor et al., 1995).
When inspected, sites are often either lacking erosion control measures or maintenance of existing control measures is
long overdue. The effort (if any) on the developer's part to maintain or implement the erosion control measures is often
inadequate and is done to appease the local SWCD, rather than with the goal of achieving 'best management' of the site.
Aside from regulation, there is little incentive for a developer to use erosion and sediment control practices. In fact, a
developer who uses erosion control may be at a cost disadvantage compared to other developers who do not, thereby
making construction less profitable (Harbor, in press).
Origin of this study
In a study evaluating the use of rapid seeding and mulching to reduce NPS pollution from construction sites, one
developer commented that he liked seeding because he thought that it made his developments more marketable (Harbor
et al., 1995). The developer soon began to include extensive seeding on his other developments to achieve the same
neat, green looking result. Even though the developer was interested in seeding because he thought it would give him
a competitive edge over other developers, he was voluntarily using vegetative erosion control (Harbor et al., 1995; Harbor,
in press). As similar anecdotal evidence accumulated, it seemed possible that a higher market value for a seeded site
might provide an incentive for voluntary erosion control. If an economic advantage can be established, then it may be
possible to persuade developers to use erosion control on the basis of a profit motive, where regulation and education have
proved ineffective. If widespread voluntary application can be achieved by this means of increased profitability, it will make
it easier to obtain compliance with erosion control programs and reduce the burden on regulators. Furthermore, and most
importantly, the NPS pollution load from construction sites would be reduced.
Methodology
We hypothesize that green, grassed lots are more attractive to buyers and therefore may be valued more highly and
sell faster than bare, dirt lots. There are several ways to test this hypothesis, with the most thorough being a detailed
tracking of the sales prices and sales timing of a large number of randomly selected treated and untreated control lots on
residential constructions sites throughout the US. In the absence of data to perform this type of highly detailed approach,
we undertook a pilot study using photos of treated and untreated lots in a market survey questionnaire aimed at
establishing whether lots with green vegetative cover are valued higher than barren ones by Realtors, developers, and
homebuyers. In the work reported here, however, we do not evaluate whether green, grassed lots sell faster than bare,
dirt lots.
Randomly selected lots in three residential housing developments in Ohio and Indiana were seeded and mulched.
Photographs of these lots were taken prior to seeding and then when the grass was approximately one inch high (Fig. 2).
Lots were photographed from three angles (front left, front center, and front right), and selected photos were used in a lot
valuation survey. The market survey was designed as a broad tool to investigate a wide range of factors which
homebuyers, Realtors, and developers find important when buying/selling a lot in a residential housing development. The
survey included open- and closed-ended questions, and those surveyed were not told the actual purpose of the survey.
The survey is reproduced in Herzog (1997). Included within the wide range of questions in the survey were specific
questions on the importance of lot appearance, and a lot valuation question in which those surveyed were asked to place
prices on lots shown in photographs. Those surveyed were told the lots were in the same neighborhood/subdivision, with
the streets and curbs installed and had the samesewer/septic system, water system, and noise level. They were then
408
-------
Figure 2. Examples of grassed and bare lot photographs used in the survey.
given 10 lot photos and asked to establish prices for each lot, having been told that the average lot value in the
development was $20,000.
Most Realtors and developers were interviewed at their offices in St. Joseph County, Indiana and Geauga County,
Ohio. Potential homebuyers were interviewed either at a neutral location or at their place of work, and included residents
of St. Joseph County and West Lafayette, Indiana, as well as personnel at a chemical engineering facility in Buffalo, New
York which was relocating to Philadelphia, Pennsylvania. The survey typically took 10-20 minutes depending on the
responsiveness of the individual.
After completion of the surveys, comparative statistics were used on the lot valuation data to assess whether there
was any significant difference between "brown" and "green" lot values for Realtors, developers, and homebuyers. Analysis
of variance was initially used to be able to test for the existence of significant interaction between the fixed variables
(respondent group and color), while taking into account variation that occurs in the random variables (eg., subjects). The
assumptions needed to appropriately apply this method, such as normality of the error terms, were found to be satisfied
(Montgomery, 1997).
409
-------
Price Differences for Green and Brown Lots
Results From a Lot Valuation Study
Developers Realtors Homebuyers Everyone
Figure 3. Differences in average prices for green and brown lots between survey groups.
Table 2. Effect of lot treatment on price for three different survey groups
Group
Realtors
Homebuyers
Developers
Green Lot
Mean Price ($),
sample size (n)
20,71 1
(n=155)
20,250
(n=36)
20,469
(n=48)
Brown Lot
Mean Price {$),
sample size (n)
19,967
(n=154)
19,500
(n=36)
20,218
(n=48)
Price
Difference
($)
744
750
251
Test
Statistic
t=4.0085
t= -1 .7957
t= -0.9200
Significance
Level
(p-value)
0.0001
0.0788
0.3609
An important element of the economic analysis of lot greening is the actual cost involved in applying seed.
This can vary widely depending on the method used to apply the seed, and the density of vegetation desired. In
this work, we restrict the analysis to an amount and type of cover intended for erosion control, as opposed to grass
species and density intended for final lawn cover. For this study we used independent contractors to apply seed,
mulch, water and fertilizer by hydroseeding. Other common approaches include use of a hand seeder, and
mulching with straw either by hand or using a blower. During dry seasons in some areas, watering may be
necessary to produce successful germination and early growth. Thus there is a wide range of possible costs of
lot greening. In this study we use the actual cost of hydroseeding for our study sites in Indiana and Ohio, $300 per
lot, although we could have applied seed and mulch by hand for about $100 per lot. Readers may want to contact
their local Soil and Water Conservation District to get estimates of typical costs for their areas.
Results
A total of 478 lot valuations (310 by Realtors, 96 by developers, and 72 by homebuyers) were made.
However, during the survey process, it became apparent that two of the photographed lots were being ranked
either highest or lowest based on their specific background (one with a fire hydrant and another with lush tree
vegetation behind the lot giving an appearance of more privacy than the other lots). Lot valuations based on these
two photos, one green lot and one brown were eliminated prior to the statistical analysis.
410
-------
Initial statistical evaluation of the entire data set focused on determining if the data fit a model in which price
was a function of the overall mean price, effects related to the group, the individual surveyed, the individual lot and
the lot color plus interaction and random error terms. In this model we assume that the effect of the particular lot
and the particular individual are random variables that are independently and normally distributed with a mean of
zero, and also that the error terms are independently and identically distributed as normal random variables with
mean of zero and variance O2. Analysis of variance followed by a normal probability plot of the residuals, and
plotting of the residuals versus the predicted values, demonstrated that the error terms were normally distributed
with constant variance (Montgomery, 1997). This analysis also demonstrated that there appeared to be significant
differences in the variations of prices between groups, which complicates analysis of the data as a combined
group. Thus it was necessary to analyze each group separately, using at test to evaluate the overall effect of color
within each group.
Realtors
The Realtors surveyed gave an average value of $20,711 on the green grassed lots and $19,969 on the brown
dirt lots. The distributions of lot values for green and brown lots were statistically significantly different at a 99.99%
confidence level (Fig. 3, Table 2). As a simple difference between means, the perceived added value for green
lots was $742 per lot.
Narrative questions on the survey revealed additional qualitative insight into Realtors' perceptions of lot value,
and reasons for preferring green lots. One Realtor commented: "I don't like these mud lots." Others said the grass
was more appealing and "easier on the eye," and that the lots look better because they are green. Other Realtors
did not see the importance of seeding and believed that grass should not enter into the decision because it will
be destroyed in the house building process. "Grass makes it look better but means nothing for what's coming."
Overall Realtors perceived that homebuyers would prefer the grassed lots ("I think people like grass,") and
the green lots would sell first because the green grass will remind homebuyers of a yard and allow them to
visualize what a house and yard would look like on the lot. One Realtor stated that the grass/ground cover was
more appealing than dirt, and that homebuyers "wouldn't like the bare ones very much." Another noted that buyers
would be more willing to walk a grassed lot in inclement weather lot because the grass would absorb the moisture
and that buyer would not walk a dirt lot because it would become muddy and puddle. One Realtor stated: the
green lots look "lush and fertile;" some people cannot visualize dirt lots as possibly being lush and fertile. This
Realtor also brought up the concern that a buyer may ask about the drainage if the lot is wet, and if the dirt lot is
dry, caked, and cracked, the buyer will wonder if anything can grow on it.
Homebuyers
The homebuyers surveyed placed an average value of $20,250 for the green lots and $19,500 for the brown
lots. The distributions of lot values for green and brown lots were statistically significantly different at a 92%
confidence level (Fig. 3, Table 2). As a simple difference between means, the perceived added value for green
lots was $750 per lot. The added value of $750 (the greatest added value among the groups surveyed) is
particularly significant because homebuyers are the ones who actually pay for the lots.
Homebuyers stated that grass gives a realistic impression of the future appearance of the lot and it is more
appealing; and that the final product is more difficult to visualize on dirt lots. In general, the homebuyers
acknowledged that the grass looks good and has more appeal, while understanding that the lots would be
disturbed during construction. One homebuyer said that the green look was nicer but that it "wouldn't effect my
decision to buy," because grass was not a "big deal." Even though comments such as these were made, on
average homebuyers valued green lots $750 more than brown lots.
411
-------
Developers
The developers surveyed placed an average value of $20,469 on the green lots and $20,219 on the brown
lots. The distributions of lot values for green and brown lots were statistically significantly different at only a
64% confidence level (Fig. 3, Table 2). Typically this would be viewed as indicating no statistically significant
difference. As a simple difference between means, the perceived value of the green lots was $250 greater
than the brown lots. Clearly, the small difference between the green and brown lots data sets and the
comparatively low significance level indicate that developers perceive little or no difference based on lot color.
During the survey, developers addressed the difference between the grass and dirt lots and stated that it
should not be a factor in lot price. They pointed out that the green lots will become brown lots during
construction and that the homebuyer will put in a yard anyway. Other developers saw that ground cover was
more attractive ("I like the green") and perceived that homebuyers would like the grass. Also, some perceived
that the green look made a development more marketable compared to other developments; one developer
said he "greens up" his developments to make them look more attractive. Another stated that grass makes a
lot look like it has topsoil, and if there are soil concerns, the grass demonstrates that vegetation can be grown
and is holding soil. One developer remarked how ground cover may be important to homebuyers for more
than just appearance. He stated that grass cover is more significant when there is rolling ground because if
there is unseeded soil on an adjacent lot, the soil may erode onto the grassed property to the dismay of the
homeowner.
The Economic Incentive
Although green lots may be priced higher than brown lots, this gross value is only significant if the price
differential exceeds the cost of seeding. The difference in value between grassed and bare lots compared to the
cost of seeding provides a measure of potential net economic benefit to the developer. In terms of a simple net
return on investment, seeding a lot provides potentially excellent return. Homebuyers valued grassed lots $750
more than brown lots, and as it cost $300 to seed a lot in this study, the developer stands to profit by $450 per lot,
which is a 150% return on the initial investment. The ability to more than double an initial investment should be
an attractive and sensible advantage for the developer, if the perceived value difference actually translates into
a sale price difference.
Present Limitations and Future Work
This pilot study is an initial step in developing information that can be used to persuade more developers to
make widespread use of vegetative cover, and other forms of environmental protection. The results of this pilot
study are most relevant in areas where climate conditions allow for relatively easy establishment of temporary
vegetative cover, and are not applicable to arid or semi arid areas. In addition, budget restrictions limited the scale
of the study. Although we collected 478 lot valuations from 62 respondents, a much larger study with respondents
from many areas of the United States would overcome a potential criticism that the current study only represents
conditions in a small portion of the Midwest. Developers are also more likely to notice results based on data
collected within their region, especially if these are coupled with regional demonstration projects. Thus, the next
logical step is to initiate a network of coordinated studies in regions experiencing rapid residential development.
This would provide for analysis on a national as well as a regional level, and for comparisons between regions.
An additional limitation of the results presented here is that they consider only perceived increase in lot values.
In actual sales transactions, buyers may not actually behave in the way they say they would on a survey. As a
linked project, it would be desirable to track actual sales histories (timing and pricing) to provide a more complete
picture of the actual economic impact of lot greening. Future research should include analysis of a large number
of real-world transactions for which lot condition is known. This could be based on a large-scale, long-term study
in which researchers intervene to change lot conditions on selected lots or developments. Alternatively, if some
412
-------
landowners are convinced by the results of this study, the experiment might occur naturally in the marketplace as
the findings of our work are disseminated.
Further extension of the basic concept of examining the direct profitability of environmental protection is also
possible. We were recently contacted by a consultant who had heard of the lot valuation study, and wanted a
similar study performed on the increased value of lots next to ponds on developments. Although ponds are often
built for stormwater control, and also aid in reduction of nonpoint source pollution, they can also have considerable
aesthetic appeal. Thus, it would be potentially very useful to know what the return on investment on a pond is for
a new residential development, both in terms of the increased price of lots adjacent to the pond, as well as the
increase in average price in lots for the development as a whole because of the improved appearance of the
development.
Conclusions
Showing that erosion control may be profitable provides a new way to reach developers who have failed to
act on the logic that erosion control provides environmental protection and is required to comply with local, state
or federal erosion control regulations. Evaluating the cost of environmental damage is not only very difficult, but
also of little direct relevance to a developer who does not directly pay the cost of the damage. Land development
is a business, with profit as a leading motive, so appealing to increased profitability is one potentially effective way
to change behavior.
The pilot study described here indicates that vegetated lots are perceived to be more valuable and more
desirable by Realtors and homebuyers. Realtors perceived that vegetated lots are worth more than barren lots
(by $742). They also perceived that vegetated lots are worth more to homebuyers and that homebuyers would
be willing to pay more for grassed lots. Homebuyers also perceived grassed lots to be more valuable and put the
largest premium ($750) on the lots for all those surveyed. The added lot value is only significant if the price
differential exceeds the cost of seeding ($300), which was the case for Realtors and homebuyers by $417 and
$450, respectively. Developers valued the vegetated lots higher than non-vegetated lots by an average of $250,
but the difference was not statistically significant. Even if it were significant, this price differential is less than the
cost of seeding and indicates that developers perceive that seeding costs are greater than the benefits of
vegetation. This perception of a net cost associated with greening a lot is perhaps why the market has largely
failed to recognize this simple way to increase a property's value. Some developers did recognize the visual
appeal of the grass and believed that a greened development would attract homebuyers more rapidly than a
development that appears unkempt. However, thevaluation study indicates that developers have not aligned their
perception of lot value with that of homebuyers.
An alternative way of interpreting the results is to consider the potential return on the investment in the
vegetative cover. For a $300 investment the developer can receive a return of $750, i.e. a 150% return on
investment. Such a rate of return is difficult to achieve in most conventional investments. Finally, price differential
is not the only economic benefit of lot greening; if lots sell faster because of greening, profits will increase because
of lower financing costs for capital invested in the development process. Further research is needed to clearly
define the value of this potential economic impact associated with lot greening. However, at this stage it is
possible to state that in addition to the environmental benefits, and regulatory requirements associated with using
vegetation for erosion control, there may be significant marketing and thus economic returns associated with lot
greening.
Education concerning the environmental benefits of erosion control, and enforcement of regulations have not
produced widespread, effective use of vegetative cover for erosion control. Because developers generally do not
perceive much incentive to vegetate their developments aside from complying with often-ineffective regulations,
they typically do not. Typically a developer who is using erosion control practices believes s/he is at a cost
disadvantage compared to other developers who are not, thereby making the developer following regulation
believe s/he will be less profitable. Furthermore, the developer does not directly pay for the mitigation of the
environmental impacts caused by the sediment leaving the site; it is the burden of the taxpayer instead. In this
413
-------
study, we have demonstrated that vegetating a development may be a profitable investment. Appealing to the
profit motive will hopefully provide a way to generate widespread use of vegetative cover that also provides erosion
prevention on construction sites.
References
Bhaduri, B., J. M. Harbor, P. Maurice. 1997. Chemical load fractionation and trap efficiency of a construction site
storm water management basin. Environmental and Engineering Geoscience. Ill: 235-249.
Clark, E. H. 1985. The off-site costs of soil erosion. Journal of Soil and Water Conservation 40: 19-22.
Clark, E. H., J. A. Haverkamp, and W. Chapman. 1985. Eroding Soils: The off-farm impacts. The Conservation
Foundation, Washington, D.C. 252 pp.
Colacicco, D., T. Osborn, and K. Alt. 1989. Economic damage from soil erosion. Journal of Soil and Water
Conservation 44: 35-39.
Crawford, J. K. and D. R. Lenat. 1989. Effects of land use on the water quality and biota of three streams in the
Piedmont Province of North Carolina. US Geological Survey Water Resources Investigations Report 89-4007.
Raleigh, N.C. 64 pp.
Ekwue, E. I. 1990. Effect of organic matter on splash detachment and the process involved. Earth Surface
Processes and Landforms. 15: 175-1 81.
Federal Register. 1990. Fed. Reg. 55. November 16. 47990.
Fennessey, L.A. and A. R. Jarrett. 1994. The dirt in a hole: A review of sedimentation basins for urban areas and
construction sites. Journal of Soil and Water Conservation 49: 317-323.
Goudie, Andrew. 1994. Human Impact on the Natural Environment. The MIT Press. Cambridge. 302 pp.
Goldman, S. J., K. Jackson, and T. A. Bursztynsky. 1986. Erosion and Sediment Control Handbook. McGraw Hill
Publishing Company. New York. 454 pp.
Harbor, J. M., J. Snyder, and J. Storer. 1995. Reducing nonpoint source pollution from construction sites using
rapid seeding and mulching. Physical Geography. 16: 371-388.
Harbor, J. M. In press. Engineering geomorphology at the cutting edge of land disturbance: Erosion and sediment
control on construction sites. Geomorphology.
Herzog, M., 1997. Reducing erosion problems from land development: An economic incentive for erosion control.
Unpublished MS Thesis, Department of Earth and Atmospheric Sciences, Purdue University.
Indiana Department of Natural Resources. 1992. Indiana Handbook for Erosion Control in Developing Areas,
Indiana Department of Natural Resources. Indianapolis, Indiana.
Koehn, E. and J. A. Rispoli. 1982. Protecting the Environment During Construction. Journal of the Construction
Division, Proceeding ASCE. 108(CO2): 233-246.
Lemly, D. A. 1982. Erosion control at construction sites on red clay soils. Environmental Management. 6: 343-352.
Montgomery, , D.C., 1997. Design and analysis of experiments. 4th edn. John Wiley & Sons, Inc., New York.
704 p.
414
-------
Novotny, V. and G. Chesters. 1989. Delivery of sediment and pollutants from nonpoint sources: A water quality
perspective. Journal of Soil and Water Conservation 44: 568-576.
Ohio Department of Natural Resources. 1980. Model Regulations for Urban Soil Sediment Pollution Control.
Columbus, Ohio.
Paterson, R. G., M. I. Luger, R. J. Burby, E. J. Kaiser, H. R. Malcolm, and A. C. Beard. 1993. Costs and benefits
of urban erosion and sediment control: The North Carolina experience. Environmental Management. 17: 167-1 78.
Pimentel, D., C. Harvy, P. Resosudarmo, K. Sinclair, D. Kurz, M. McNair, S. Crist, L. Shpritz, R. Saffouri, and R.
Blair. 1995. Environmental and economic costs of soil erosion and conservation benefits. Science. 267: 1117-
1122.
Rogers, R.D. and Schumm, S.A. 1991. The effect of sparse vegetative cover on erosion and sediment yield.
Journal of Hydrology 123: 19-24.
Satterlund, Donald R. 1972. Wildland Watershed Management. The Ronald Press Company. New York. 370 pp.
Trotti, T. 1997. Developer fees don't hack it. Erosion Control. Nov/Dec 1997: 6.
Virginia Department of Conservation and Recreation. 1992. Virginia Erosion and Sediment Control Handbook.
Conservation and Recreation, Richmond, Virginia.
Wolman, M. G. and A. P. Schick. 1967. Effects of Construction on Fluvial Sediment, Urban and Suburban Areas
of Maryland. Water Resources Research 3: 451-464.
415
------- |