&EPA
United States
Environmental Protection
Agency
Center for Environmental
Research Information
Cincinnati, OH 45268
EPA/625/R-95/003
April 1995
Publication
National Conference on
Urban Runoff
Management: Enhancing
Urban Watershed
Management at the Local,
County, and State Levels
March 30 to April 2,1993
The Westin Hotel
Chicago, Illinois
-------
EPA/625/R-95/003
April 1995
Seminar Publication
National Conference on Urban Runoff Management:
Enhancing Urban Watershed Management at the Local, County,
and State Levels
March 30 to April 2, 1993
The Westin Hotel
Chicago, Illinois
Center for Environmental Research Information
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio
-------
Notice
This document has been subjected to the U.S. Environmental Protection Agency's (EPA's) peer and
administrative review and has been approved for publication as an EPA document. The opinions
expressed in each paper, however, are those of the authors and do not necessarily agree with those
of EPA. In addition, mention of trade names or commercial products does not constitute endorse-
ment or recommendation for use.
When an NTIS number is cited in a reference, that document is available from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
703-487-4650
-------
Contents
Page
Watershed Planning and Program Integration 1
Eric H. Livingston
The Evolution of Florida's Stormwater/Watershed Management Program 14
Eric H. Livingston
The State of Delaware Sediment Control and Stormwater Management Program 28
Earl Shaver
Section 6217 Coastal Nonpoint Pollution Control Program: Program Development and
Approval Guidance 32
J. W. Peyton Robertson, Jr.
Compliance With the 1991 South Carolina Stormwater Management and Sediment Reduction Act 37
K. Flint Holbrook and William E. Spearman, III
Florida's Growth Management Program 39
Eric H. Livingston
Stormwater and the Clean Water Act: Municipal Separate Storm Sewers in the Moratorium 47
Kevin Weiss
Municipal Permitting: An Agency Perspective 63
William D. Tate
Municipal Stormwater Permitting: A California Perspective 71
Thomas E. Mum ley
Stormwater Management Ordinance Approaches in Northeastern Illinois 77
Dennis W. Dreher
The Lower Colorado River Authority Nonpoint Source Pollution Control Ordinance 82
Thomas F. Curran
New Development Standards in the Puget Sound Basin 88
Peter B. Birch
Ordinances for the Protection of Surface Water Bodies: Septic Systems, Docks and Other Structures,
Wildlife Corridors, Sensitive Aquatic Habitats, Vegetative Buffer Zones, and Bank/Shoreline Stabilization .... 96
Martin Kelly and Nancy Phillips
-------
Contents (continued)
Page
Urban Runoff Pollution Prevention and Control Planning: San Francisco Bay Experiences 106
Thomas E. Mum ley
Whole Basin Planning: Practical Lessons Learned From North Carolina, Delaware, and Washington 109
Michael L. Bowman and Clayton S. Creager
Application of Urban Targeting and Prioritization Methodology to Butterfield Creek, Cook, and
Will Counties, Illinois 119
Dennis Dreher and Thomas Price
Development of a Comprehensive Urban Nonpoint Pollution Control Program 135
Jennifer M. Smith and Larry S. Coffman
Site Planning From a Watershed Perspective 139
Nancy J. Phillips and Elizabeth T Lewis
Soil Conservation Districts' Role in Site Plan Review 151
Glenn Bowen, Eric H. Buehl, and John M. Garcia, Jr.
The Role of Landscapes in Stormwater Management 165
Steven I. Apfelbaum
The U.S. Environmental Protection Agency's Advanced Identification Process 170
Sue Elston
Wisconsin Smart Program: Starkweather Creek 173
William P. Fitzpatrick
Wolf Lake Erosion Prevention 180
Roger D. Nanney
Incorporating Ecological Concepts and Biological Criteria in the Assessment and Management of
Urban Nonpoint Source Pollution 183
Chris O. Yoder
Overview of Contaminated Sediment Assessment Methods 198
Diane Dennis-Flagler
Linked Watershed/Water-Body Model 202
Martin Kelly, Ronald Giovannelli, Michael Walters, and Tim Wool
AUTO_QI: An Urban Runoff Quality/Quantity Model With a CIS Interface 213
Michael L. Terstriep and Ming T Lee
iv
-------
Contents (continued)
Page
Source Loading and Management Model (SLAMM) 225
Robert Pitt and John Voorhees
Combining CIS and Modeling Tools in the Development of a Stormwater Management Program 244
Chris Rodstrom, Mohammed Lahlou, and Alan Cavacas
Watershed Screening and Targeting Tool (WSTT) 250
Leslie L Shoemaker and Mohammed Lalou
Hydrocarbon Hotspots in the Urban Landscape 259
Thomas Schueler and David Shepp
Design Considerations for Structural Best Management Practices 265
Joseph J. Skupien
Targeting and Selection Methodology for Urban Best Management Practices 274
Peter Mangarella, Eric Strecker, and Gail Boyd
A Catalog of Stormwater Quality Best Management Practices for Heavily Urbanized Watersheds 282
Warren Bell
Postconstruction Responsibilities for Effective Performance of Best Management Practices 293
Joseph J. Skupien
Guidance Specifying Management Measures for Sources of Nonpoint Pollution in Coastal Waters 299
Rod Frederick
Biotechnical Streambank Protection 304
Don Roseboom, Jon Rodsater, Long Duong, Tom Hill, Rich Offenback, Rick Johnson,
John Beardsley, and Rob Hilsabeck
The Use of Wetlands for Stormwater Pollution Control 309
Eric W. Strecker
Constructed Wetlands for Urban Runoff Water Quality Control 327
Richard Homer
Stormwater Pond and Wetland Options for Stormwater Quality Control 341
Thomas R. Schueler
Practical Aspects of Stormwater Pond Design in Sensitive Areas 347
Richard A. Claytor, Jr.
-------
Contents (continued)
Page
Infiltration Practices: The Good, the Bad, and the Ugly 352
Eric H. Livingston
Stormwater Reuse: An Alternative Method of Infiltration 363
Marty Wanielista
Use of Sand Filters as an Urban Stormwater Management Practice 372
Earl Shaver
Application of the Washington, DC, Sand Filter for Urban Runoff Control 375
Hung V. Truong and Mee S. Phua
Stormwater Measures for Bridges: Coastal Nonpoint Source Management in South Carolina 384
H. Stephen Snyder
Controlling Pollutants in Runoff From Industrial Facilities 391
Kevin Weiss
The Role of Education and Training in the Development of the Delaware Sediment and Stormwater
Management Program 403
Frank M. Piorko and H. Earl Shaver
Development and Implementation of an Urban Nonpoint Pollution Education and Information Program 408
Richard Badics
Training for Use of New York's Guidelines for Urban Erosion and Sediment Control 411
Donald W. Lake, Jr.
Field Office Technical Guide: Urban Standards and Specifications 419
Gary N. Parker
Stormwater Outreach at the Federal Level: Challenges and Successes 422
Kimberly O. Hankins
Training for Construction Site Erosion Control and Stormwater Facility Inspection 426
Richard Homer
VI
-------
A cknowledgments
This document is the product of the efforts of many individuals. Gratitude goes to each person
involved in the preparation and review of this document.
Authors
A special thanks goes to the many authors of the papers presented in this document. Their efforts
in preparing papers made this document possible and led to the overall success of the conference.
Peer Reviewers
The following individuals peer reviewed the papers presented in this document:
Frank Browne, F.X. Browne, Inc., Lansdale, PA
Richard Christopher, Greenville County Soil and Water Conservation District, Greenville, SC
Scott Clifton, Chesapeake Bay Local Assistance Department, Richmond, VA
Thomas Davenport, U.S. Environmental Protection Agency (EPA) Region 5, Chicago, IL
John Ferris, RUST Environmental and Infrastructure, Inc., Milwaukee, Wl
Richard Field, EPA Risk Reduction Engineering Laboratory, Edison, NJ
Lisa Jayne Gray, EPA Region 5, Chicago, IL
Marlene Hale, Chesapeake Bay Local Assistance Department, Richmond, VA
Judy Kleiman, EPA Region 5, Chicago, IL
Norm Kowal, EPA Environmental Criteria and Assessment Office, Cincinnati, OH
Jim Kreissl, EPA Center for Environmental Research Information, Cincinnati, OH
Ernesto Lopez, EPA Region 5, Chicago, IL
Daniel Mazur, EPA Region 5, Chicago, IL
Belinda Montgomery, EPA Region 5, Chicago, IL
Timothy Neiheisel, EPA Environmental Monitoring Systems Laboratory, Cincinnati, OH
Thomas O'Conner, EPA Risk Reduction Engineering Laboratory, Edison, NJ
Linda Papa, EPA Environmental Criteria and Assessment Office, Cincinnati, OH
Nancy Phillips, EPA Region 5, Chicago, IL
Randy Revetta, EPA Center for Environmental Research Information, Cincinnati, OH
Don Roberts, EPA Region 5, Chicago, IL
Sue Schock, EPA Center for Environmental Research Information, Cincinnati, OH
Peter Swenson, EPA Region 5, Chicago, IL
Editorial Review and Document Production
Heidi Schultz, Eastern Research Group, Inc., Lexington, MA, directed the editorial review and
production of this document.
VII
-------
Technical Direction and Coordination
Daniel Murray, EPA Office of Research and Development, Center for Environmental Research
Information, Cincinnati, OH, coordinated the preparation of this document and provided technical
direction throughout its development.
Special Thanks
A special thanks goes to Bob Kirschner, Northeast Illinois Planning Commission, Chicago, IL, for
his efforts in producing a high quality, technically sound conference. Without his efforts, the
conference and this document would not have been possible. His talent and efforts are truly
appreciated.
VIM
-------
Introduction
Background
As stormwater and snowmelt flow across the urban
landscape, countless contaminants are carried into our
rivers, lakes, and estuaries. The effects of these con-
taminant discharges on the environment can be severe.
Water quality and sediment characteristics can be de-
graded, threatening the biological integrity of our urban
water bodies. In addition to urban runoff quality, the
quantity of urban stormwater and snowmelt that reaches
urban streams can cause severe physical harm to sen-
sitive ecosystems, including those well beyond urban-
ized areas.
The proper management of urban watersheds is a chal-
lenging and complex task. As urban watersheds are
developed, they produce a site-specific mix of pollutants
that can adversely affect water and sediment quality.
Also, with increased urbanization comes increased im-
permeability, resulting in higher stormwater flows to
streams that can cause streambed and streambank ero-
sion. Urban runoff management is particularly difficult
because government jurisdictions rarely coincide with
watershed boundaries. So, to overcome these institu-
tional obstacles and implement effective urban water-
shed management programs, comprehensive and
coordinated management strategies are needed.
The National Conference on Urban Runoff Management
was held in Chicago, Illinois, from March 30 to April 2,
1993. The purpose of this conference was to bring to-
gether national experts in the field of urban watershed
management to discuss and share ideas and ap-
proaches for effective urban watershed management.
This 4-day conference addressed a wide variety of insti-
tutional and technical issues, from watershed planning
and public information programs to the design and ap-
plication of best management practices.
Purpose
The purpose of this seminar publication is to make
available to a much wider audience the valuable infor-
mation presented at the National Conference on Urban
Runoff Management. This publication comprises 53 pa-
pers that were presented at the conference. The papers
address a broad spectrum of programmatic and techni-
cal topics relating to urban watershed management,
including:
Watershed planning
Stormwater management programs
Regulatory issues
Monitoring, modeling, and environmental assess-
ment
Design and application of best management prac-
tices and controls
Education and information programs
The papers in this publication represent the collective
knowledge and experience of many talented individuals
who have developed and are implementing and support-
ing watershed management programs at the federal,
state, county, and local level. As a result, this document
will be a valuable resource to regulators, watershed
management program personnel, and others interested
in developing and implementing a successful urban wa-
tershed management program.
IX
-------
Watershed Planning and Program Integration
Eric H. Livingston
Florida Department of Environmental Regulation, Tallahassee, Florida
Abstract
Since passage of the Clean Water Act, federal, state,
and local governments together with the private sector
have spent billions of dollars attempting to meet the
act's goals of restoring and maintaining the chemical,
physical, and biological integrity of the nation's waters.
While great progress has been made, especially with
respect to reducing traditional point sources of pollution,
we are faced with a much more complex and difficult
challenge: reducing the pollution associated with our
everyday activities. Facing the environmental chal-
lenges presented by nonpoint sources and stormwater
discharges requires a more comprehensive and inte-
grated approach, especially if we are to maximize the
environmental benefits in a cost-effective manner. This
approach is known as watershed managementthe
integration, on a watershed basis, of the management
of land resources, water resources, social-cultural re-
sources, financial resources, and infrastructure. Imple-
mentation of this approach requires a cooperative
Watershed Management Team effort involving all levels
of government, the private sector, and each citizen.
Besides addressing the need for watershed management,
this paper discusses briefly the many components of a
comprehensive watershed management program. Key
program elements include growth management, land
preservation/purchase, wetlands/floodplains protection,
erosion and sediment control, stormwater management,
wastewater management, watershed prioritization and
targeting, inspections and maintenance, research, pub-
lic education, and dedicated funding sources. Other
papers in this publication review the evolution of Flor-
ida's watershed management program, with emphasis
on successes and failures together with recommenda-
tions to improve the environmental effectiveness of the
program (e.g., "The Evolution of Florida's Stormwater/
Watershed Management Program").
Introduction
When land within a watershed is changed from its natu-
ral state to agricultural land and then to urban land,
many complex interconnected changes occur to the
natural systems within the watershed. These changes
can and do have profound effects on the health of
these systems as well as their inhabitants. As Earl
Shaver describes in his paper, one of the greatest
changes is the alteration of the watershed's hydrology,
especially the infiltrative capacity of the land. Addition-
ally, the everyday activities of humans within the water-
shed add many potential environmental contaminants to
the watershed that can be easily transported by precipi-
tation and runoff.
Managing stormwater and nonpoint sources of pollution
presents many complex challenges to the water resources
manager that are somewhat unique and quite different
from those encountered when managing traditional
point sources of pollution. These challenges include:
Integrating land-use management, because change
in land use creates the stormwater problems.
Educating the public about how everyday activities
contribute to the stormwater/non point source prob-
lem and how they must be part of the solution.
Developing a management framework that is based
on the fact that "we all live downstream" and that
stormwater flows are not constrained by political
boundary lines.
Obtaining the cooperation and coordination of neigh-
boring political entities that exist within a watershed.
Not only managing stormwater from new development
but retrofitting existing "drainage systems" that were
built solely to convey runoff away from developed lands
to the nearest water body as quickly as possible.
-------
Secondly, constraints imposed by current stormwater
treatment technology, such as treatment efficiency, land
needs, and maintenance needs, and by the costs of
assessing and solving existing stormwater/nonpoint
source pollution problems call for a cooperative and
regional framework. Additionally, the proliferation of fed-
eral programs and requirements imposed by federal
legislation, such as the Federal Clean Water Act and the
Coastal Zone Management Act, has caused fragmenta-
tion of efforts and created program "turf wars" and even
conflicts between programs within the U.S. Environ-
mental Protection Agency (EPA). Other federal pro-
grams such as the National Flood Insurance Program,
the Section 205 flood control program, and even agri-
cultural crop subsidy programs directly conflict with
achieving the goals set forth in various environmental
laws and programs. Finally, current environmental man-
agement approaches rely on regulatory efforts that at-
tempt to compensate for adverse effects caused by land
alteration activities on a particular site. Implementing a
watershed management approach helps to overcome
all of these challenges and, just as importantly, allows
inclusion of planning efforts that can prevent problems.
This allows for more extensive use of less expensive
nonstructural management practices.
Watershed Management
"Watershed management" is a flexible framework for
integrating the management of all resources (land, bio-
logical, water, infrastructure, human, economic) within a
watershed. Basically, it is the managing of human
activities so as to cause the least disruption to natural
systems and native flora and fauna. With respect to the
management of stormwater and nonpoint sources, the
crucial factor is the integration of the management of
land use, water/stormwater, and infrastructure. Watershed
management has numerous facets, including planning,
education, regulation, monitoring, and enforcement, that
are performed on a watershed basis.
The watershed management approach discussed in this
paper must be flexible. The size of the watershed to be
managed can be very large (a river basin) or very small
(a subbasin). Selection of watershed size depends on
many factors, including ecological systems in the water-
shed, ground-water hydrologic influences, the type and
scope of resource management problems and goals,
and the level of resources available. Additionally, the
institutional framework for watershed management will
vary greatly depending on the legal framework that has
been established in state law and local ordinances.
Advantages of Watershed Management
As discussed above, solving our nation's stormwa-
ter/nonpoint source problems, especially retrofitting ex-
isting "drainage systems" to reduce the pollutant loads
they discharge to receiving waters, presents many com-
plex challenges. Correcting these problems will be ex-
tremely expensive and technically difficult, and will take
a long time. Accordingly, we need to re-evaluate our
current approach to stormwater management to shift the
emphasis towards more comprehensive, prevention-ori-
ented strategies such as watershed management.
The following comparison illustrates the differences be-
tween the usual piecemeal approach to stormwater man-
agement and a comprehensive watershed approach (1):
The usual approach: For existing urban develop-
ment, the usual approach is to address local storm-
water problems without evaluating the potential for
the runoff control measure to cause adverse effects
in downstream areas. In the case of new urban de-
velopment, stormwater management responsibilities
would be delegated to local land developers, and
each would be responsible for constructing stormwa-
ter management facilities on the development site to
maintain postdevelopment peak discharge rate, vol-
ume, and pollutant loads from the site at predevelop-
ment levels. There would be little or no consideration
of the cumulative effects of the developments with
their individual stormwater systems on either the local
government stormwater infrastructure or the down-
stream lands and waters.
The watershed approach: This option involves de-
veloping a comprehensive watershed plan, known as
the "master plan," to identify the most appropriate
control measures and the optimal locations to control
watershedwide activities. The watershed approach
typically involves combinations of the following:
- Reviewing the watershed and its characteristics to
assess problems and potential solutions.
- Strategically locating a single stormwater manage-
ment facility (a regional system) to control postde-
velopment runoff from several projects within a
basin (or from a fully developed basin or subbasin).
- Providing stormwater conveyance improvements
where necessary upstream from the regional facility.
- Employing nonstructural measures throughout the
watershed, such as acquisition of floodplains, wet-
lands, and natural stormwater depressional stor-
age areas; soundly planned land use; limitations
on the amount of imperviousness; grassed swales
rather than storm sewers; and roof runoff direction
to pervious areas.
While the usual approach to urban stormwater manage-
ment is relatively easy to administer, it offers several
disadvantages. There is a greater risk of negative ef-
fects, particularly in watersheds that cover several juris-
dictions. Insignificant flood protection benefits result
from emphasis on the effects of minor localized flooding.
Ineffective runoff control throughout the watershed is
-------
caused by the failure to evaluate locational differences in
the benefits of stormwater management facilities. Rela-
tively high local costs for facility maintenance are in-
curred, as are unnecessary costs associated with the use
of small-scale structural solutions rather than large-scale
nonstructural solutions, which typically are much cheaper.
Included among the possible negative effects of this
piecemeal approach to stormwater management are the
following:
It may only partially solve the major flooding problem(s).
It may solve flooding problems in the upstream juris-
diction but create flooding problems in downstream
jurisdictions.
Randomly located detention basins may actually in-
crease downstream peak flows.
Maintenance needs and costs associated with nu-
merous onsite runoff controls are very high.
Significant capital and operation/maintenance expen-
ditures may be wasted.
The costs of remedial structural solutions likely will
be much greater than the cost of a proper manage-
ment program.
The watershed master planning approach offers signifi-
cant advantages over the piecemeal approach. It prom-
ises reductions in capital and operation/maintenance
costs and reductions in the risk of downstream flooding
and erosion, particularly in multijurisdictional water-
sheds. It offers better opportunities to manage existing
stormwater problems and the ability to consider and use
nonstructural controls. Other benefits include increased
opportunities for recreational uses of stormwater con-
trols, potential contributions to local land-use planning,
enhanced opportunities for stormwater reuse, and
popularity among land developers.
There are some disadvantages to the watershed ap-
proach:
In advance, local governments must conduct studies
to locate and develop preliminary designs for regional
stormwater management facilities.
Local governments must develop and adhere to a
future land-use plan so that the regional facility is
properly designed to capture runoff from the planned
amount of development and impervious surfaces.
Local governments must finance, design, and con-
struct the regional stormwater management facilities
before most development occurs and provide for re-
imbursement by developers over a buildout period
that can last many years.
In some cases, local governments may have to con-
duct extraordinary maintenance activities for regional
facilities that the public feels are primarily recreational
facilities that merit protection for water quality.
Another advantage of watershed management is that
the resource management goals can be more resource
oriented. Prevention practices and programs to protect
natural systems and beneficial uses of our water bodies
can be stressed. These typically are more cost effective
than trying to restore natural systems after they have
been adversely affected by human activities that occur
within a watershed.
Watershed management allows coordination of infra-
structure improvements with point and nonpoint source
management programs and, most importantly, provides
a vital link between land use and water resources man-
agement.
Watershed Management Framework
There is no single approach or institutional framework
for establishing a watershed management program.
While establishing a watershed management institu-
tional and legal framework would be easiest if we could
start with a clean slate, we cannot. There is an existing
legal framework in each state, county, and city. These
may differ greatly. In some states, there will be a long
list of existing laws, rules, and programs that have been
set up to respond to earlier state needs. In other states,
there will be very few laws, rules, and programs that can
form a foundation for establishing watershed manage-
ment programs. Therefore, one of the keys to opening
the watershed management door is flexibility. In some
cases, the focus will be on enacting new laws. In other
cases, the emphasis will be on revising existing laws
(ordinances) to better integrate and coordinate pro-
grams and objectives.
Another key to establishing a watershed management
framework is patience. Getting state laws or local ordi-
nances enacted or modified is not an easy process. A
long-term game plan must be developed and pursued
with diligence. Each component of a watershed man-
agement program has its own controversies, guarantee-
ing that public debate will be vociferous on many issues.
Therefore, priorities must be established. Typically, pri-
ority setting depends on state resource problems and
needs, public sentiment, and the degree to which an
issue becomes "sexy," thereby receiving coverage by the
news media. In many cases, it may take several years
to get a particular piece of legislation passed or revised.
To succeed, education of elected officials, state agency
managers, and the public must be a priority. Public partici-
pation and support are essential in building a consensus.
Many of the issues that watershed management pro-
grams address are complex and not easily demonstrated.
Managers of stormwater and other nonpoint sources of
pollution, unlike the managers of traditional point sources
-------
of pollution, cannot point to pipes that continuously dis-
charge effluents. Therefore, promoters of watershed
management programs must use multimedia presenta-
tions to not only educate but also to entertain. You must
sell the need for watershed management!
Another key to success is to take advantage of any
opportunities that arise. Unfortunately, these opportuni-
ties often occur after a natural disaster that results in the
loss of property or lives. After Hurricanes Frederick and
Andrew struck South Carolina and South Florida, re-
spectively, considerable public debate arose about
building codes, land uses, and development within sen-
sitive and susceptible coastal areawhether to allow
rebuilding in these areas and whether public programs
such as the National Flood Insurance Program should
subsidize development in such areas. These debates,
especially of the costs and benefits, can be used to help
build support for growth management and land acquisi-
tion programs. Furthermore, flooding (and in a few lo-
cales, water quality problems) can be used to break the
"hydro-illogical cycle" and gain support for stormwater
management programs and local stormwater utilities.
Finally, in building a watershed management frame-
work, one must establish clear goals for the overall
program. Some important goals include:
Providing opportunities for preventive nonstructural
controls in addition to structural controls that can help
to mitigate the impacts of human activities within a
watershed.
Establishing clearly defined, holistic natural resource
management goals.
Setting priorities, both in terms of a long-term legis-
lative agenda and by targeting watersheds.
Encouraging public participation so that everyone
"buys in" and feels that they are part of the solution.
Integrating all available tools and resources into a
coordinated, cost-effective, cooperative approach.
Finding dedicated funding sources outside the main
funding stream (also known as "general revenues")
so that the watershed management programs do not
compete with law enforcement, education, or other
high-priority societal needs.
In developing, selling, establishing, and implementing a
watershed management framework and associated pro-
grams, it is very important to keep in mind "the big Cs
of watershed management" (2):
Comprehensive management of people, land use,
natural resources, water resources, and infrastruc-
ture throughout a watershed.
Continuity of stormwater/watershed management
programs over a long period, which is required to
correct existing problems and prevent future ones.
Cooperation between federal, state, and local gov-
ernments; cities and counties; public and private
sectors; and all citizens.
Communication to educate ourselves and elected of-
ficials about how we are all part of the problem and
how we can and must be part of the solution.
Coordination of stormwater retrofitting to reduce
pollutant loading and of other natural systems resto-
ration activities with other proposed infrastructure
improvements (e.g., road projects) or development/
redevelopment projects to maximize benefits and
cost-effectiveness.
Creativity in best management practice technology,
in funding sources, and in our approach to solving
these complex, costly problems.
Consistency in implementing laws, rules, and pro-
grams nationally and statewide to assure equity and
fairness for everyone.
Cash in large amounts and over a long period to
correct existing problems and prevent future ones.
Commitments solving our current problems and pre-
venting future ones so that we can ensure that our
children have a bright future ("Just Say No To Storm-
water Pollution").
Watershed Management Program
Components
Watershed management involves the integration of
management programs addressing the many differing
human activities that occur within a watershed. This
section discusses briefly many different components or
programs that typically are considered a part of water-
shed management. The following list and discussion of
programs is not all inclusive. Other programs addressing
specific state or regional needs have been implemented
around the country. In developing or implementing pro-
grams, it is important to take advantage of information
and technology transfer clearinghouses and to commu-
nicate with people in other states, cities, and counties
who have implemented similar programs.
Each of the various watershed management programs
includes common aspects such as planning, holistic goals,
science/technical support, implementation (usually with
both regulatory and nonregulatory approaches), and
extensive public participation. Public participation is
needed in all aspects of the program: planning, rule
development and adoption, permitting, and inspec-
tion/enforcement. Programs must also address how to
obtain adequate funding and staffing; how to train staff
-------
and the public, especially the regulated community; how
to ensure inspection and compliance; and how to
ensure long-term operation and maintenance of struc-
tural controls. Finally, programs must be evaluated
regularly to optimize their environmental effectiveness,
cost-effectiveness, and efficiency in providing service.
This requires a commitment to monitoring programs
that can actually ascertain if the program's goals are
being met.
Typically, these programs are implemented following
enactment of a state law that requires a state agency to
set up a program to address a specific concern. Pro-
gram implementation via legislative mandate usually
helps to ensure that a program has adequate legal
authority and staffing/funding support. Some of the pro-
grams discussed can and have been established by the
passage of a rule by a state agency using its general
legislative powers, for example, programs for public
education, pollution prevention, monitoring, and priori-
tizing target watersheds. Given the current scientific
data on the pollutants found in stormwater, erosion and
sediment control and even stormwater treatment pro-
grams can be established using general water pollution
control authorities. These programs are very staff/re-
source intensive, however, requiring legislative approval
of budget requests at a minimum.
Common watershed management programs include
both planning and regulation. It is important to under-
stand the difference between comprehensive planning
and permitting. Both are needed to effectively manage
growth and protect the quality of our environment and
our citizens' quality of life:
Comprehensive planning allows a community to
make decisions about how and where growth will
occur in the future. Comprehensive planning asks, is
this the right location, is this the right time, and is this
the right intensity for the proposed use of the land?
Comprehensive planning seeks to prevent problems
(social, economic, environmental) before develop-
ment occurs.
Permitting, on the other hand, asks only, how can we
do the best job with this development on this particu-
lar site? Permitting is site-specific and seeks only to
mitigate the impacts of the land-use decision. There
always are inherent limitations in any regulatory pro-
gram that comprehensive planning can help to over-
come. Principal among these limitations is the fact
that permitting is piecemeal and does not consider
cumulative effects. Therefore, regulation and permit-
ting cannot substitute for planning.
Watershed planning and management programs must
include two equal components: the land planning
framework and the water planning framework. An exam-
Table 1. The Land Planning Framework Versus the Water
Planning Framework
Land Planning
Water Planning
Land development regulations Water management regulations
Local compliance plans State water management plans
Regional policy plans State water policy
State comprehensive plan State comprehensive plan
pie of the hierarchial relationship of these planning
frameworks is shown in Table 1.
Following is a discussion of many of the program com-
ponents that typically are part of a watershed manage-
ment framework. These can be divided into three
categories:
Land planning and management
Water planning and management
General resources planning and management
Land Planning and Management Program
Components
Land planning and management programs often are
called growth management programs. It is important to
understand the clear distinctions between growth man-
agement, comprehensive planning, and land/environ-
mental regulation:
Growth management looks at broad issues and at
the interrelationship of systemsnatural systems,
infrastructure, land use, and people. It attempts to
assess how well we have been providing for the
needs of our citizens in the past and, when new
people move here, to determine what their needs are
and how they will be provided. Growth management
encompasses comprehensive planning, natural resource
management, public facilities planning, housing, rec-
reation, economic development, and intergovern-
mental coordination.
Comprehensive planning is a governmental process
for inventorying resources, establishing priorities, es-
tablishing a vision of where a community wants to
go, and determining how to get there. It is a system-
atic way of looking at the different components of a
community, county, region, and state.
Regulations are the specific controls applied to dif-
ferent types of development activities to regulate and
minimize their negative impacts. Typically, regula-
tions are administered by all levels of government,
federal, state, and local. At the local level, land de-
velopment regulations are the ordinances that imple-
ment the local comprehensive plan.
-------
State Comprehensive Plan
A state comprehensive plan serves as the base of
both the land and water planning pyramids. A State
Comprehensive Planning Act would establish goals and
policies for each of the plan's various elements and
require the state land planning agency to prepare a
general state comprehensive plan. Elements in a state
comprehensive plan usually include water resources,
natural systems, air quality, coastal and marine re-
sources, land and wildlife resources, waste manage-
ment, public facilities (infrastructure), transportation,
mining, agriculture, education, and economic develop-
ment. If the state's land planning framework includes
"regional planning councils" or "regional council of gov-
ernments," those agencies would be responsible for
developing a regional plan. Both the state and regional
plans would have to be consistent with the goals and
policies set forth in the state comprehensive planning
act. These goals and policies, set by the legislature, are
to provide guidance to state, regional, and local govern-
ments in developing and implementing programs, rules,
or ordinances. Consistency must occur from the base of
the planning pyramid all the way to its apex. To help
ensure consistency and to integrate agency implemen-
tation programs with the law's goals and policies, this
law can require the preparation of state agency func-
tional plans. These plans can form the basis for agency
budget requests, which must be related to the state
comprehensive plan's goals and policies.
Growth Management and Land Development
Regulation
The Local Government Comprehensive Planning Act
(LGCPA), often referred to as the growth management
act, establishes the key piece of the natural resources
jigsaw puzzle: the direct connection between land-use
management and water/natural systems management.
Eight states (Oregon, Florida, New Jersey, Maine, Vermont,
Rhode Island, Georgia, Washington) have implemented
state growth management programs (3). While these pro-
grams have elements in common, each state has differ-
ent implementation requirements. Some states "require"
while other states "recommend" local plans, consistency,
compliance, etc. An LGCPA should at least address the
following components, which are common to each of the
eight existing state growth management programs:
Legislative authority and intent.
Local comprehensive plans: Required? Voluntary?
Schedule? Planning period? Required elements?
Minimum requirements?
Plan implementation: Required? Site planning? Land
development regulations?
Consistency with state goals/policies: Required?
Monitoring? Enforcement?
State review and approval: Required? Which agen-
cies? Administrative process?
Compliance: Monitoring? Incentives? Disincentives?
Citizen enforcement?
Limitations on the number and type of comp plan
amendments: Frequency? Process?
Regular plan updates and implementation appraisals:
Required? Frequency?
Wetlands and Floodplain Protection
Wetlands and floodplains are the "bladder" and "kid-
neys" of a watershed. They provide a wide range of
irreplaceable services at no cost, including maintenance
and improvement of water quality; floodwater convey-
ance and storage; shoreline stabilization; water recharge
and supply; sediment control; aquatic productivity;
spawning and nursery grounds; habitat for shellfish, fish,
waterfowl, endangered species and other wildlife; and
open space and recreation. Unfortunately, we have not
in the past appreciated these benefits. Instead, we
looked on these areas as unproductive, snake-infested
mosquito havens with no socially accepted redeeming
value. Consequently, only about 40 percent of our na-
tion's original 215 million acres of wetlands remain,
largely the result of the conversion of wetlands and
floodplains to agricultural lands.
Although Section 404 of the Federal Clean Water Act
establishes a wetlands program, its effectiveness in
maintaining, protecting, and restoring our nation's wet-
lands is highly questionable. Not only are nationwide
general permits to conduct activities in wetlands rela-
tively easy to obtain, but agricultural and silvicultural
activities are largely exempt. Another problem hindering
the environmental effectiveness of this federal program
is a lack of national consistency. Furthermore, other
federal programs (e.g., Section 205 of the 1948 Flood
Control Act, National Flood Insurance Program) directly
conflict with wetland and water quality protection efforts
by promoting alteration and development of these sen-
sitive lands.
A state wetlands protection act can be an important
addition to a state's watershed management arsenal to
either fill in the gaps of the federal program or to expand
the protection of wetlands and floodplains. In developing
and implementing a state wetland protection program, it
is important to integrate, not duplicate, existing federal
programs. Because the current federal wetlands permit-
ting program is administered by the Army Corps of
Engineers and EPA, typically the state water quality/en-
vironmental management agency is the implementing
agency at the state level. Frequently, the "wetlands
protection act" is simply a new section within a state's
existing environmental laws.
-------
Components that need to be addressed by a state wet-
lands/floodplain protection act include:
Defining "wetland." A wetland should be defined by
three characteristics: the elevation and duration of
flooding, the presence of certain wetland-specific
plants, and hydric soil conditions. The law should
clearly state that wetlands are considered to be "wa-
ters" just like a river, lake, or estuary.
Establishing a standard method to delineate wet-
lands. Wetlands represent the transitional edge be-
tween waters and uplands. Determining where a
wetland ends and the upland begins is neither an
easy nor an uncontroversial undertaking. Wetland
scientists should be allowed to establish combina-
tions of hydrologic, vegetation, and soil indicators and
a process by which to "draw the wetland line."
Requiring consistant statewide application of the wet-
land definition and wetland jurisdictional delineation
method by all levels of government.
Establishing wetland protection/management goals
and policies that can set the basis for wetland regu-
lations and permitting criteria.
Creating goals and policies that foster more cost-ef-
fective pollution prevention approaches by stressing
wetland avoidance rather than mitigation.
Requiring or encouraging regional mitigation banks
rather than onsite mitigation.
Establishing a fair permitting process that ensures
public participation, equity, an appeals process, and
decisions based on scientific/technical merit.
Allowing, with strict pretreatment requirements, the
incorporation of certain wetlands into domestic
wastewater and stormwater management/reuse sys-
tems, provided that the ecological characteristics of
the wetland are protected, restored, or enhanced.
Requiring the annual tracking of wetland losses and
mitigation efforts, successes, and failures.
Providing for assumption, by the state, of the federal
Section 404 wetlands program.
State and Local Land Preservation and
Acquisition
Regulating and restricting the use of private property are
very controversial. The U.S. Supreme Court has ruled
several times, however, that state and local govern-
ments have the legal authority to do so. In fact, it is the
responsibility of government to ensure the health, safety,
and welfare of the public. Restricting what can and
cannot be done on a particular piece of property helps
to maintain property values and to prevent contamina-
tion of air, land, water, and human resources. Care must
be taken, however, to avoid the "taking of property."
One way to help ensure that this goal is met and that
extremely crucial or sensitive lands within a watershed
are preserved is to implement land acquisition pro-
grams.
The federal government has implemented several types
of land acquisition programs that have helped to pre-
serve sensitive lands, protect vital wildlife habitats, and
establish recreational lands, such as our national parks
and national wildlife refuges. Federal budget problems
and intense competition for the limited federal land ac-
quisition funds, however, makes it difficult to gain these
monies to obtain properties, especially those that do not
have national or at least regional significance. Addition-
ally, federal funding programs generally require match-
ing funds from state and/or local governments.
Therefore, the establishment of state and local land
acquisition programs can greatly increase the ability to
purchase and protect sensitive lands and, equally im-
portantly, to capture limited federal funds.
Establishing state or local land acquisition programs
requires extensive citizen participation and support. You
will be asking the public to tax themselves to raise
money to purchase lands, preserve them, and provide
recreational opportunities. You must "sell" the program.
Catchy phrases and acronyms are helpful. Citizens
must see that they or their children will benefit and that
the funds will be spent wisely and cost-effectively. Land
acquisition programs must avoid conflicts of interest and
be administered with great integrity and openness.
A state and local land preservation and acquisition act
should contain the following components and consid-
erations:
Clearly defined program goals and policies. These
will form the foundation for determining what types
of properties will be purchased and how purchasing
priorities will be established. The program's goals
and policies should advocate the preservation and
restoration of lands that contribute nonstructural en-
vironmental benefits. Additional resource management
factors that should be considered in purchasing lands
include open space and recreational and wildlife benefits.
Integrated and coordinated federal, state, local, and
private land preservation and acquisition programs.
This will maximize the ability to leverage funds from
various sources. Establishing interconnected wildlife
corridors and greenways should be a priority.
Extensive participation by citizens, private conserva-
tion groups, and state and local governments to es-
tablish program regulations, administrative procedures,
and, most importantly, land-buying priorities.
The long-term ownership and active land management
of the property once it is purchased. Which agency
-------
will be in charge, an environmental agency? A parks
and recreation agency? A fisheries or wildlife agency?
A private organization (i.e., Nature Conservancy,
Trust for Public Land)? Does a land management
plan need to be developed? How will land manage-
ment be funded?
Dedicated funding sources. Purchasing large quanti-
ties of land and then managing the land, especially
with public access and use, requires significant funds
over a long period. To obtain sufficient funds, it may
be desirable for a state or local government to use
its ability to sell bonds. Bonds can raise large
amounts of money at one time, which can then be
paid off like a mortgage. However, that requires hav-
ing a source of funds that is stable and predictable
over the life of the bond. Fees on real estate trans-
actions (e.g., documentary stamps) and local option
sales taxes have been used extensively around the
country for this purpose.
Water Resources Planning and Management
Programs
In general, the United States is blessed with an abun-
dance of clean water resources. Water generally is
available whenever we want it, in whatever quantity we
desire and at a very low cost. Consequently, less atten-
tion and emphasis have been placed on water re-
sources planning and management, especially from a
holistic approach. In the past, water planning and man-
agement programs were implemented usually to ad-
dress a crisis that had arisen. The continuing growth of
our nation's population, however, continues to exert
ever-growing demands on our vulnerable and limited
water resources. Additionally, the need to begin manag-
ing unconventional pollution sources such as stormwa-
ter and other nonpoint sources requires a re-evaluation
of the way we manage water. Accordingly, water re-
source planning and management programs are receiv-
ing increased attention and evaluation.
Within this subcategory of watershed management pro-
grams, we include water quantity and quality programs
for the protection and management of surface and
ground waters, as well as general environmental protec-
tion programs. All of these programs usually include
both pollution prevention aspects and pollution treat-
ment aspects.
Environmental Protection
Most states have enacted some type of state environ-
mental protection act, typically to control traditional point
sources of pollution. Generally, these laws are patterned
somewhat afterthe federal Clean Water Act. These laws
get revised frequently as either a new state environ-
mental crisis or concern arises or the Clean Water Act
gets amended by Congress. This law is an excellent
example of how, over years, an existing law is revised
to establish or refine existing or new environmental
requirements or programs.
While state environmental protection laws around the
country include many common and similar environ-
mental requirements and mandates, there is also con-
siderable variation among states. A major reason forthis
is that different states approach the same problem dif-
ferently. For example, some states enact separate ero-
sion and sediment control acts and stormwater
management acts. Other states combine these two very
important watershed management components. In
some states, the law governing the siting and use of
onsite wastewater disposal systems is found within a
state's general health code law, while in other states it
is found within the environmental law. These three wa-
tershed management components will be discussed as
separate topics even though their legislative authority
often is integrated into a state's environmental laws.
State environmental protection laws generally contain
such components and considerations as:
Establishment of the state environmental agency, along
with its legal authority and powers and responsibilities.
Establishment of an "environmental regulation com-
mission," generally composed of citizens appointed
by a political body (i.e., governor), which usually
holds public workshops and adopts the state's envi-
ronmental regulations and standards.
Permitting evaluation criteria, permit fees, and admin-
istrative procedures, which typically include a legal,
administrative hearing process to appeal permitting
decisions.
Programs, with adequate legal authority/direction and
resources (staffing and funding), to address general
environmental protection and management of air, land,
and water resources (surface and ground water).
Programs, with adequate legal authority/direction and
resources, to minimize the impacts of specific pollution
sources such as wastewater and industrial discharges,
solid wastes, hazardous wastes, and toxic wastes.
Pollution prevention programs such as "Amnesty
Day," which allows citizens to safely dispose of haz-
ardous or toxic household wastes; used oil recycling
centers; waste reduction and assistance programs
for industry; "Adopt a Road (Stream, Lake, Bay,
Shoreline)" programs; recycling; and "Farmstead As-
sistance" ("Farm*A*Syst") programs.
Programs to restore environmentally damaged lands
and waters, especially critical areas such as wet-
lands, floodplains, steep slopes, and eroding lands.
Programs to monitor the health of the environment
and to assess the effectiveness of watershed man-
-------
agement programs. Monitoring programs need to in-
clude sampling of the water column, sediment, and
biological community. They need to be able to pro-
vide information concerning long-term trends in envi-
ronmental health, as well as the status of the health
of selected water bodies or natural systems.
Water Resources Planning and Management
Many states have enacted a water resources act that
is distinct and separate from the state environmental
protection act, perhaps because the planning and
management of water resources is essential to the con-
tinued survival of life on our planet and because water
is a major determinant of economic development and
quality of life. Water resources planning and manage-
ment must include consideration of both water quantity
(water supply, water allocation, flooding) and water
quality. A state water resources act needs to be fully
integrated with the state environmental protection act. It
must ensure that implementation of programs by both
the state environmental protection agency and state/
regional water resources agency is coordinated, consis-
tent, and complimentary.
A state water resources act creates the framework for
water resources planning and management programs to
be undertaken by state, regional, and local govern-
ments. Using the goals and policies of the state compre-
hensive planning act, the environmental regulation
commission adopts a regulation known as its state water
policy. This rule contains general policy statements ad-
dressing the myriad water resource topics, such as
water supply and conservation, surface water preserva-
tion and management, and natural systems preserva-
tion and management. It provides guidance for the
implementation of all water resource programs and
regulations, whether by a state, regional, or local entity.
The act could establish regional "water(shed) manage-
ment districts" which are set up on the basis of water-
shed boundaries. The districts would conduct regional
watershed planning, help coordinate water manage-
ment efforts undertaken by local agencies to ensure that
watershedwide goals are met cooperatively, and oper-
ate regulatory and research programs.
A state water resources act should include such pro-
gram components and considerations as:
Establishing water(shed) management districts to ad-
minister special regional (watershed) water planning
and management programs. These districts should
provide statutory authorities and be given broad pow-
ers to protect, manage, and restore surface- and
ground-water resources.
Setting the institutional relationships between the
state environmental agency, regional water manage-
ment districts, and local governments. Strong over-
sight of programs, especially regulatory ones, dele-
gated downwards for implementation is essential to
ensure program consistency.
Developing a state water policy to provide guidance
for the implementation of all water programs and
regulations in the state, which should be adopted as
a rule, preferably as part of the state's environmental
regulation code. The state water policy must be
based on and consistent with the goals and policies
in the state planning act. State, regional, and local
water regulations and programs must be consistent
with the state water policy. Ideally, goals and policies
in a local comprehensive plan should also be consis-
tent with the policy.
Providing the districts with dedicated sources of
revenue to ensure long-term, adequate funding of all
necessary water resource management programs.
Sources used in parts of the country include ad va-
lorem assessments (property taxes), fees on water
use, permitting fees, and special assessments.
Supplemental Surface Water and Environmental
Protection Programs
There are several watershed management component
programs that may be established within one of the
above two statutes or which may be established in
statute separately.
Erosion and Sediment Control Act/Program. Land
disturbing activities are among the largest source of
sediments and particle-borne pollutants. Preventing
erosion and minimizing and capturing sediments, espe-
cially from construction sites, are essential parts of any
watershed management framework. Since 1972, over
20 states have enacted erosion and sediment control
laws and programs.
Establishment of an erosion prevention and sediment
control law or program should include the following com-
ponents and considerations:
Clearly defined legal authority, goals/performance
standards, and responsibilities of the implementing
state and/or regional or local agencies.
Assurance that publicly funded projects, especially
highways, must comply with all program require-
ments, and an encouragement for these projects to
serve as models.
Determination of whether utility construction, agricul-
tural, and forestry projects are to be included in the
program.
Agency responsibilities and relationships. Typically,
implementation of an erosion and sediment control
program involves a state agency and a regional/local
agency such as a soil and water conservation district
-------
or a local government. Delegation of the program
from the state to the local agency must involve close
oversight to ensure consistency.
Adequate staffing and other resources to conduct
research on the effectiveness of control measures,
develop scientifically sound rules, and conduct train-
ing and education programs for plan reviewers, in-
spectors, developers, engineers, and site contrac-
tors. A state training and certification program for plan
reviewers, inspectors, and contractors is highly rec-
ommended because it is very unlikely that public
agencies will ever obtain sufficient staffing to conduct
inspections of construction sites on a regular basis.
Mutual integration of the state erosion and sediment
program, the state stormwater management pro-
gram, and the new federal National Pollutant Dis-
charge Elimination System (NPDES) Stormwater
Permitting Program.
Stormwater Management Act/Program. Most states
have implemented some type of stormwater "drainage"
program to ensure that their citizens and their properties
are protected from flooding. In some states, special
"drainage districts" or "drain commissions" have been
established at a regional or local level. Today, however,
we know that stormwater is also one of the major
sources of pollutant loadings to our nation's rivers,
lakes, and estuaries. Stormwater management is evolv-
ing slowly from its "drainage" focus to a much more
comprehensive, multiple-objective program that ad-
dresses stormwater quality and quantity. Stormwater
programs must attempt to prevent or minimize stormwa-
ter problems associated with new land-use activities but
must also develop programs to reduce the pollutant
loading discharged from older "drainage systems." This
latter objective is extremely difficult and expensive to
address. Watershed management approaches are es-
sential. Typically, a state stormwater management pro-
gram begins by addressing the problems associated
with new land uses and then evolves into a more com-
prehensive, watershed-based program to address the
retrofitting of older stormwater systems.
Components and considerations that need to be ad-
dressed by a state stormwater management act/pro-
gram include:
Clearly defined legal authority, goals/performance
standards, and responsibilities of the implementing
state and/or regional or local agencies.
Assurance that publicly funded projects, especially high-
ways, comply with all program requirements, and an
encouragement for these projects to serve as models.
Agency responsibilities and relationships. Typically,
implementation of a stormwater management pro-
gram involves a state agency and a regional/local
agency such as a water(shed) management district,
soil and water conservation district, or a local gov-
ernment. Delegation of the program from the state to
the local agency must involve close oversight to en-
sure consistency.
General goals and minimal treatment performance
standards (on which best management practice de-
sign criteria will be based) based on the state water
policy, and a biological or resource based perform-
ance standard for reducing the pollutant loading from
existing drainage systems.
Adequate staffing for planning, coordinating, permit-
ting, and enforcement, and resources to conduct re-
search on the effectiveness of control measures; to
develop scientifically sound rules; and to conduct
training and education programs for plan reviewers,
inspectors, developers, engineers, and site contrac-
tors.
A state training and certification program for plan
reviewers, inspectors, and contractors. This is highly
recommended, because it is very unlikely that public
agencies will ever obtain sufficient staffing to conduct
inspections of stormwater systems either during con-
struction or afterwards on a regular basis. These pro-
grams can be integrated with similar erosion and
sediment control programs.
Integration of the state stormwater management pro-
gram with the state erosion and sediment control
program and with the new federal NPDES Stormwa-
ter Permitting Program.
A mechanism, such as stormwater operating per-
mits, to ensure that stormwater management systems
are inspected at least annually to determine mainte-
nance needs and that systems are maintained and
operated properly. Ideally, this system is implemented
by a local stormwater utility which provides the owner
of a properly maintained and operated stormwater
system with a stormwater utility fee credit as an eco-
nomic incentive.
Statutory authority for the establishment of dedicated
funding sources for stormwater management pro-
grams at both the state and local level. At the state
level, small fees on concrete, asphalt, fertilizer, or
pesticides might be considered. At the local level,
stormwater utilities are widely used around the coun-
try with great success.
Watershed Prioritization and Targeting Act/Program.
The ever-growing number of water resources problems
along with the financial constraints faced by all levels of
government strongly suggest a need for the establish-
ment of watershed prioritization and targeting programs.
Many states, often as part of the implementation of
stormwater/nonpoint source management programs, have
10
-------
set up such programs (4, 5). Considerations and com-
ponents of a state watershed prioritization and targeting
act/program include:
Clearly identifying which state, regional, and local
agencies will be involved in establishing priority wa-
tersheds. Public participation is essential to ensure
the cooperation and "buy in" of citizens around the
state and within the targeted watershed. Cooperation
and joint ventures with private land conservation
groups need to be encouraged.
Providing guidance on what factors will be consid-
ered in the prioritization process. These may include
requirements such as water bodies being of state-
wide or regional significance or of a certain level of
degradation; the level of local government and citizen
support, especially by those land owners that will
need to install management practices; and the avail-
ability of local matching funds.
Providing a legal mechanism for the adoption of the
"priority list" by the appropriate state, regional, or
local agency. Ensuring that the list is reviewed on a
regular basis and updated or refined as needed.
Providing a dedicated source of funds (state, re-
gional, local) to develop and implement a watershed
management plan within a realistic time schedule.
Onsite Wastewater Management Act/Program. The
nation's rapid population growth and the accompanying
move to the suburbs and even more rural areas has led
to a tremendous proliferation of the use of onsite waste-
water disposal systems (OSDSs). Often considered an
inexpensive alternative to centralized wastewater col-
lection and treatment systems, OSDSs can cause or
contribute to health and environmental resource prob-
lems which are difficult and very expensive to solve. Like
many areas of nonpoint source management, OSDS
programs need to stress prevention but also be able to
correct problems related to the past use and misuse of
these systems. Traditionally, state health departments
rather than state environmental or water resources
agencies have administered OSDS programs. It is in-
creasingly evident, however, that OSDSs are a major
contributor to impairment of aquatic systems.
A state onsite wastewater management act/program should
include the following components and considerations:
Clearly defined legal authority, goals/performance
standards, and responsibilities of the state, regional,
or local entities involved in the implementation of the
program.
Goals and performance standards that not only ad-
dress traditional health concerns but that also require
consideration of the potential environmental effects
of OSDSs.
The adoption of OSDS regulations that govern the
types of OSDS systems (e.g., drainfields, mound sys-
tems, aerobic units), the siting of systems (e.g., water-
table elevation, soil types, setbacks from wetlands/
waters), the design and performance of OSDS (e.g.,
secondary treatment? nitrates <10 mg/L?), determi-
nation of whether surface discharges will be allowed
and under what conditions, inspections during con-
struction and throughout the use of the system, and
maintenance.
Regular inspection (every 2 to 3 years) and mainte-
nance (e.g., pumpout, drainfield) to help ensure that
OSDSs continue to function properly. The estab-
lishment of OSDS management districts, which have
defined service areas, funding sources, and legal
authority, is one mechanism that can be used. An-
other method of ensuring that OSDSs continue to
function properly is to require inspections and up-
grading/maintenance of systems whenever a prop-
erty is sold.
General Resources Planning and
Management Programs
One of the challenges of implementing watershed man-
agement frameworks and programs is their complex,
interwoven nature. Many aspects of watershed man-
agement transcend the simple classification scheme
outlined at the beginning of this section. These include
the need for broad-based natural resource management
programs and for environmental education programs,
especially those integrated into the curriculum of the
K-12 education system. In many states, separate agen-
cies have been established that have responsibility for
the management of land, fish and wildlife, agriculture,
mining, and parks and recreation. Often a state forestry
department is responsible for the acquisition and man-
agement of state forest lands. The activities and pro-
grams of these agencies typically are an essential
component of watershed management. Close coordina-
tion and cooperation between these agencies and the
other "primary" agencies involved in watershed man-
agement are needed to ensure that programs do not
conflict and to maximize the benefits and cost-effective-
ness of all programs.
Additionally, while nearly every natural resources re-
source management agency has some type of environ-
mental education programs, these typically are narrowly
focused, dealing with a particular program. The growing
importance of nontraditional pollution sources such as
stormwater and nonpoint sources requires the develop-
ment and implementation of a broad-based environ-
mental curriculum that begins teaching children in
kindergarten and continues all the way through their
senior year of high school. Each of us must understand
the basic interrelationships of the air, land, and water
11
-------
and how our everyday activities can and do contribute
to the degradation of our natural systems. We must
re-establish the ethic of stewardship, and the best way
to accomplish this is through the education of our youth.
Example State Watershed Management
Initiatives
Several states have developed and implemented some
or many of the watershed management program com-
ponents discussed above. In recent years, states have
begun to try to integrate ongoing programs into a more
comprehensive watershed management framework.
Within this publication can be found papers that de-
scribe or discuss state programs such as Delaware and
Florida, regional programs such as the Puget Sound
(Washington) Management Program and the San Fran-
cisco Bay Program, and local programs such as the
Prince George's County (Maryland) and Summit County
(Ohio) programs.
One of the ways in which existing programs, especially
planning and regulatory programs, can evolve into an
integrated watershed approach is demonstrated by the
ongoing efforts in North Carolina. The North Carolina
Division of Environmental Management (NCDEM) has
developed a plan in which basins, not stream reaches,
are the basic unit of water quality management. The
objectives of North Carolina's Basinwide Water Quality
Management Initiative include (6):
Identify priority problem areas and pollution sources
that merit particular pollutant control, using modifica-
tions of rules (e.g., basin criteria) and increased en-
forcement.
Determine the optimal water quality management
strategy and distribution of assimilative capacity for
each of the 17 major river basins within the state.
Prepare, in cooperation with local governments and
citizens, comprehensive basinwide management plans
that set forth the rationale, approaches, and long-term
management goals and strategies for each basin.
Implement innovative management approaches that
protect the state's surface water quality, encourage
the equitable distribution of assimilative capacity, and
allow for sound economic planning and growth.
The whole-basin initiative is envisioned as a fully inte-
grated approach to water quality assessment and man-
agement. It integrates planning, monitoring, modeling,
point source permitting and control, nonpoint source
control, and enforcement within a basin. NCDEM has
rescheduled its NPDES permit activities so that permit
renewals within a given basin will occur simultaneously
and will be repeated at 5-year intervals.
One of the difficulties in implementing a basin-wide
approach is the setting of priorities, the establishment of
a rotating schedule among the basins, and the correla-
tion of management needs (monitoring, planning, per-
mitting, enforcement) with staff and resource
allocations. North Carolina prioritized and scheduled its
17 basins based on consideration of the nature and
extent of known problems, a basin's importance in terms
of human use, the availability of data, and staff workload
balancing.
For each basin in turn, North Carolina will perform the
15-step process outlined below (6):
1. Compile all existing relevant information on basin
characteristics and water quality.
2. Define the water quality goals and objectives for
water bodies within the basin. Revise as necessary
as more data are obtained.
3. Identify the critical issues (e.g., water supply protec-
tion, shellfish harvesting) and current water quality
problems within the basin. Determine the major fac-
tors and sources (point, nonpoint, habitat degrada-
tion) that contribute to the problems.
4. Prioritize the basin's water quality concerns and
critical issues. Ensure public participation and input
from other government agencies and nongovern-
ment groups.
5. Define the subbasin management units using basin
hydrology, physiographic boundaries, problem ar-
eas, and critical issues.
6. Identify needs for additional information.
7. Collect additional information.
8. Analyze, integrate, and interpret the information col-
lected. Revisit Steps 2 through 5 in light of the new
information.
9. Determine and evaluate the management options
for each management unit in the basin.
10. Select final management approaches for the basin
and targeted subbasins.
11. Complete the draft whole basin management plan.
Perform additional modeling and other analyses to
finalize wasteload allocations.
12. Distribute the draft plan for review and comment,
and conduct public hearings.
13. Revise the plan as appropriate in response to com-
ments. Ensure adoption of the plan by the state's
environmental management commission.
14. Implement the management approaches, including
point and nonpoint source controls.
15. Monitor the program's success and update the plan
every 5 years.
12
-------
RGfGTGnCGS 3. Gale, D.E. 1992. Eight state-sponsored growth management pro-
grams: A comparative analysis. JAPA 58(4):425-439.
1. Camp Dresser and McKee. 1985. Feasibility study for a Roanoke 4. Puget Sound Water Quality Authority. 1991. Puget Sound water
Valley comprehensive stormwater management program. Final quality management plan. Seattle, WA.
report prepared for the Fifth Planning District Commission.
5. Wisconsin Department of Natural Resources. 1986. Nonpoint
source pollution: Where to go with the flow. Madison, Wl.
2. Livingston, E.H., and M.E. McCarron. 1992. Stormwater manage-
ment: A guide for Floridians. Tallahassee, FL: Florida Department 6. U.S. EPA. 1991. The watershed protection approach: An over-
of Environmental Regulation. view. EPA/503/9-92/002. Washington, DC.
13
-------
The Evolution of Florida's Stormwater/Watershed Management Program
Eric H. Livingston
Florida Department of Environmental Regulation,
Tallahassee, Florida
Abstract
Research conducted during the late 1970s as part of the
Section 208 Water Quality Management Program iden-
tified pollutant loading from stormwater discharges as
the major source of water quality degradation in Florida.
This paper reviews the evolution of Florida's stormwater
regulatory program, from its initial emphasis on control-
ling stormwater problems from new development to its
current emphasis on reducing pollutant loading from
existing development. The philosophical and technical
basis for the program are discussed, as are the pro-
gram's major components. The paper emphasizes how
the program is beginning to address the retrofitting of
existing "drainage systems."
Developing and implementing a statewide stormwater
management program requires several key compo-
nents. Research must be undertaken to develop state-
wide rainfall distribution statistics, determine stormwater
pollutant loading characteristics, determine the effec-
tiveness of various stormwater treatment practices, and
identify key design criteria for each type of best man-
agement practice. Education is essential and must be
targeted at many different audiences: design engineers,
state and local government staff and elected officials,
construction personnel, inspectors, maintenance staff,
and citizens. Dedicated funding sources at both state and
local levels are very important, especially if the program
is to achieve the desired environmental benefits and for
retrofitting. Most importantly, integration of stormwater
regulatory programs with other resource management
programs on a watershed basis must occur for maxi-
mum environmental results and cost-effectiveness.
Introduction
Florida is blessed with a multitude of natural systems,
from the longleaf pine-wiregrass hills of the Panhandle
to the sinkhole and sand ridge lakes of the central ridge,
the Everglades "River of Grass," and the coral reefs of
the Keys. Abundant surface-water resources include
over 20 major rivers and estuaries and nearly 8,000 lakes.
Plentiful ground-water aquifers provide over 90 percent
of the state's residents with drinking water. Add the
state's climate and it's easy to see why many consider
the Sunshine State a favored vacation destination and
why the state has experienced phenomenal growth since
the 1970s. Today, Florida is the fourth most populous
state, and its population is still growing rapidly, although
not at the 900 people per day (300,000 per year) rate
that occurred throughout the 1970s and 1980s.
Florida's natural systems, especially its surface- and
ground-water resources, are extremely vulnerable and
easily damaged. This is partially the result of the state's
sandy porous soils, karst geology, and abundant rainfall.
The negative impacts of unplanned growth were seen
as early as the 1930s, when southeast Florida's coastal
water supply was threatened by saltwater intrusion into
the fragile freshwater aquifer that supplied most of the
potable water for the rapidly expanding population. By
the 1970s, it was becoming all too clear that unplanned
land-use, development, and water-use decisions were
altering the state in a manner that, if left unchecked,
could lead to profound, irretrievable loss of the very
natural beauty that brought residents and tourists to
Florida. Extensive destruction of wetlands, bulldozing of
beach and dune systems, continued saltwater intrusion
into freshwater aquifers, and the extensive pollution of
the state's rivers, lakes and estuaries were only some
of the negative impacts of this rapid growth.
Fortunately, Florida's citizens and elected officials be-
came educated about these problems and began devel-
oping programs to protect and manage the state's
natural resources. Florida began serious and compre-
hensive efforts to manage its land and water resources
and its growth as the environmental movement in the
nation and the state gained strength in the early 1970s.
Florida's natural resources management programs have
evolved over a 20-year period. Collectively, the individ-
ual laws and programs enacted during this period can
14
-------
be considered "Florida's Watershed Management Pro-
gram." In many cases, these laws have been integrated
either statutorily with revisions to existing laws or through
the adoption of regulations by various state, regional,
and local agencies.
The evolution of Florida's watershed management pro-
gram typically involves the following sequence: 1) con-
cern about a specific "pollutant" or problem creates a
resource/environmental management program which
usually begins by focusing on "new sources" (site basis);
2) over time, as new sources are controlled and the
effectiveness of the program increases, the focus shifts
to cleaning up "older sources" (watershed or regional
basis); 3) ultimately, the focus shifts to integrating the
program with similar ones to eliminate any duplication
and to improve efficiency and effectiveness.
Florida's Stormwater Program:
The Beginning
Section 208 of the Federal Clean Water Act required the
development of areawide water quality management
plans to control point and nonpoint sources of pollution.
As part of Florida's program conducted during the late
1970s and early 1980s, many investigations were un-
dertaken to assess the impacts of stormwater and the
effectiveness of various best management practices (1).
These studies demonstrated that stormwater, whether
from agriculture, forestry, or urban lands, was the pri-
mary source of pollutant loading to Florida's receiving
waters. Subsequently, it was concluded that the ability
to meet the Clean Water Act objective of fishable and
swimmable waters would require the implementation of
stormwater programs to reduce the delivery of pollutants
from stormwater discharges.
Recognition of this problem, along with the availability
of federal funds, led Florida to draft regulations to control
stormwater in the late 1970s. The first official state
regulation specifically addressing stormwater was
adopted in 1979 as part of Chapter 17-4, Florida Admin-
istrative Code (FAC). Chapter 17-4.248 was the first
attempt to regulate this source of pollution, which, at
the time, was not very well understood. Under Chapter
17-4.248, the Florida Department of Environmental
Regulation (DER) based its decision to order a permit
on a determination of the "insignificance" or "signifi-
cance" of the stormwater discharge. This determination
seems reasonable in concept; however, in practice,
such a decision can be as variable as the personalities
involved. What may appear insignificant to the owner of
a shopping center may actually be a significant pollutant
load into an already overloaded stream.
In adopting Chapter 17-4.248, DER intended that the
rule would be revised when more detailed information
on nonpoint source management became available.
About a year after adoption, DER began reviewing the
results of research being conducted under the 208 pro-
gram. DER also established a stormwater task force
with membership from all segments of the regulated and
environmental communities. A revised stormwater rule,
Chapter 17-25, FAC, was developed over 2 years, in-
volving more than 100 meetings between department
staff and the regulatory interests, and the dissemination
of 29 official rule drafts for review and comment. The
rule was adopted by the state's Environmental Regula-
tion Commission (ERG) and became effective in Febru-
ary 1982. The adopted rule required a stormwater
permit for all new stormwater discharges and for modi-
fications to existing discharges that were modified to
increase flow or pollutant loading.
The stormwater rule had to be implemented within the
framework of the federal Clean Water Act. The act es-
tablishes two types of regulatory requirements to control
pollutant discharges: technology-based effluent limita-
tions, which reflect the best controls available consider-
ing the technical and economic achievability of those
controls; and water quality-based effluent limitations,
which reflect the water quality standards and allowable
pollutant loadings set up by state permit (2). The latter
approach can be developed and implemented through
biomonitoring based on whole effluent toxicity, making it
very applicable to stormwater. Florida's tremendous
growth, the accompanying creation of tens of thousands
of new stormwater discharges, and the lack of data on
stormwater loading toxicity made this approach unim-
plementable, however.
Guidance on the development of stormwater regulatory
programs and the role of water quality criteria has been
issued by the U.S. Environmental Protection Agency
(EPA) (3). The guidance recognizes that best manage-
ment practices (BMPs) are the primary mechanism to
enable the achievement of water quality standards. For
the purposes of this paper, a BMP is a control technique
that is used for a given set of site conditions to achieve
stormwater quality and quantity enhancement at mini-
mal cost. Further, the guidance recommends that state
programs include the following steps:
Design of BMPs based on site-specific conditions;
technical, institutional, and economic feasibility; and
the water quality standards of the receiving waters.
Monitoring to ensure that practices are correctly de-
signed and applied.
Monitoring to determine both the effectiveness of
BMPs in meeting water quality standards and the
appropriateness of water quality criteria in reasonably
ensuring protection of beneficial uses.
Adjustment of BMPs when it is found that water qual-
ity standards are not being protected to a designed
level, and/or evaluation and possible adjustment of
water quality standards.
15
-------
Proper installation and operation of state-approved BMPs
should achieve water quality standards. While water qual-
ity standards are to be used to measure the effectiveness
of BMPs, EPA recognizes that there should be flexibility
in water quality standards to address the impacts of time
and space components of stormwater as well as natu-
rally occurring events. If water quality standards are not
met, then the BMPs should be modified, the discharge
should cease, or, in some cases, reassessment of the
water quality standards should be undertaken.
Rationale for Stormwater Rule Standards
The overriding standards of the stormwater rule are the
state's water quality standards and appropriate regula-
tions established in other DER rules. Therefore, an ap-
plicant for a stormwater discharge permit must provide
reasonable assurance that stormwater discharges will
not violate state water quality standards. Because of the
potential number of discharge facilities and the difficul-
ties of determining the impact of any facility on a water
body or the latter's assimilative capacity, DER decided
that the stormwater rule should be based on design and
performance standards.
The performance standards established a technology-
based effluent limitation against which an applicant can
measure the proposed treatment system. Compliance
with the rule's design criteria created a presumption that
the desired performance standards would be met,
which, in turn, provided a rebuttable presumption that
water quality standards would be met. If an applicant
wants to use BMPs other than those described in the
rule, then a demonstration must be made that the BMP
provides treatment that achieves the desired pollutant
removal performance standard. Actual design and per-
formance standards are based on a number of factors:
Stormwater management goals: Stormwater manage-
ment has multiple objectives, including water quality
protection, flood protection (volume, peak discharge
rate), erosion and sediment control, water conserva-
tion and reuse, aesthetics, and recreation. The basic
goal for new development is to ensure that postde-
velopment peak discharge rate, volume, timing, and
pollutant load do not exceed predevelopment levels.
BMPs are not 100-percent effective, however, in re-
moving stormwater pollutants, while site variations
can also make this goal unachievable at times.
Therefore, for the purposes of stormwater regulatory
programs, DER (water quality) and the state's regional
water management districts (WMDs) (flood control)
have established performance standards based on risk
analysis and implementation feasibility.
Rainfall characteristics: An analysis of long-term rainfall
records was undertaken to determine statistical distri-
bution of various rainfall characteristics such as storm
intensity and duration, precipitation volume, and time
between storms. It was found that nearly 90 percent
of a year's storm events occurring anywhere in Florida
produce a total of 2.54 cm (1 in.) of rainfall or less (4).
Also, 75 percent of the total annual volume of rain falls
in storms of 2.54 cm or less. Finally, the average intere-
vent time between storms is approximately 80 hr (5).
Runoff pollutant loads: The first flush of pollutants re-
fers to the higher concentrations of stormwater pollut-
ants that characteristically occur during the early part
of the storm, with concentrations decaying as the runoff
continues. Concentration peaks and decay functions
vary from site to site depending on land use, the pol-
lutants of interest, and the characteristics of the drain-
age basin. Florida studies (6, 7) indicated that for a
variety of land uses the first 1.27 cm (0.5 in.) of runoff
contained 80 to 95 percent of the total annual loading
of most stormwater pollutants. First flush effects gen-
erally diminish, however, as the size of the drainage
basin increases and the percent impervious area de-
creases because of the unequal distribution of rainfall
over the watershed and the additive phasing of inflows
from numerous small drainages in the larger water-
shed. In fact, as the drainage area increases in size
above 40 ha (100 ac), the annual pollutant load carried
in the first flush drops below 80 percent because of the
diminishing first flush effect.
BMP efficiency and cost data: Numerous studies
conducted in Florida during the Section 208 program
generated information about the pollutant removal
effectiveness of various BMPs and the costs of BMP
construction and operation. Analysis of this informa-
tion revealed that the cost of treatment increased
exponentially after "secondary treatment" (removal of
80 percent of the annual load) (8).
Selection of minimum treatment levels: After review
and analysis of the above information, and after
extensive public participation, DER set a stormwater
treatment objective of removing at least 80 percent
of the average annual pollutant load for stormwater
discharges to Class III (fishable/swimmable) waters.
A 95-percent removal level was set for stormwater
discharges to sensitive waters such as potable sup-
ply waters (Class I), shellfish harvesting waters
(Class II), and Outstanding Florida Waters. DER
believed that these treatment levels would protect
beneficial uses and thereby establish a relationship
between the rule's BMP performance standards and
water quality standards.
BMP Treatment Volumes and Design
Criteria/Guidelines
The current stormwater treatment volumes for various
BMPs are set forth in Table 1. Since adoption of the
stormwater rule in 1982, the design criteria and treat-
ment volumes have been revised several times as new
16
-------
Table 1. BMP Treatment Volumes for Stormwater Discharges to Class III Waters
Swales
Retention
Detention With Filtration
Wet Detention
Wetlands
Infiltration of 80 percent of the runoff generated by a 3-yr/1-hr storm (typically about 5.1 cm [2 in.] of runoff).
Off-line infiltration of the first 1.25 cm (0.5 in.) of runoff or the volume calculated by 1.25 times the percent
imperviousness of the project area, whichever is greater. On-line infiltration must treat an additional 1.25 cm
of runoff above the volume required for off-line treatment.
Filtration of detention volume.
Detention of the first 2.54 cm (1 in. of runoff) or the volume calculated by 2.5 times the percent
imperviousness of the project area, whichever is greater.
Same as for wet detention.
Notes: Discharges to sensitive waters must treat 50 percent more stormwater volume and may require infiltration pretreatment.
Discharges to sinkhole watersheds must treat the first 2 in. of runoff (Suwannee River WMD criterion).
information becomes available about the field effective-
ness of various types of BMPs.
In addition to the stormwater treatment volumes, other
design and performance standards have been set to
ensure that BMPs function optimally to attain the storm-
water treatment goal and other management objectives
(9). These guidelines will be discussed for each of the
BMPs currently being used extensively in Florida.
Swales
Swales are defined by Chapter 403, Florida Statutes
(FS), as manufactured trenches that:
Have a top width-to-depth ratio of the cross section
equal to or greater than 6:1, or side slopes equal to
or greater than 3 ft horizontal to 1 ft vertical.
Contain contiguous areas of standing or flowing water
only following a rainfall event.
Are planted with or have stabilized vegetation suit-
able for soil stabilization, stormwater treatment, and
nutrient uptake.
Are designed to take into account soil erodibility, soil
percolation, slope, slope length, and drainage area
so as to prevent erosion and reduce pollutant con-
centration of any discharge.
Swale treatment of stormwater is accomplished primar-
ily by infiltration of runoff and secondarily by adsorption
and vegetative filtration and uptake (10). Recent inves-
tigations have concluded that Florida soil, slope, and
water table conditions essentially preclude the use of
swales as the sole BMP to treat stormwater (11). There-
fore, the greatest utility of swales is as a pretreatment
BMP within a BMP treatment train stormwater system.
Infiltration pretreatment can be easily accomplished by
using raised storm sewer inlets within the swale, or by
elevating driveway culverts or using swale blocks to
create small retention areas.
Retention
Off-line retention areas, which receive the first flush
volume only while the later runoff is diverted to a flood
control BMP, are the most effective stormwater treat-
ment practice. Treatment is achieved through diversion
and infiltration of the first flush, thereby providing total
pollutant removal for all stormwater that is retained on
site. To reduce operation needs, increase aesthetics,
and reduce the land area needed for stormwater treat-
ment, retention areas should be incorporated into a
site's landscaping and open-space areas. Effectiveness
of retention areas can be increased and ground-water
impacts decreased by:
Infiltrating the stormwater treatment volume within
72 hr or within 24 hr if the retention area is grassed.
Grassing the retention area bottom and side slopes,
which reduces maintenance and maintains soil infil-
tration properties.
Maintaining at least 3 ft between the bottom of the
retention area and seasonal high water tables or
limerock.
In karst-sensitive areas, using several small, shallow
infiltration areas to prevent formation of solution pipe
sinkholes within the system.
Exfiltration trenches typically are used in highly urban-
ized areas where land is unavailable for retention basins.
They consist of a rock-filled trench surrounded by filter
fabric in which a perforated pipe is placed. The stormwater
treatment volume is stored within the pipe and exfiltrates
out of the perforations into the gravel envelope and into
the surrounding soil. Pretreatment with catch basins to
remove sediments and other debris is essential to pre-
vent clogging. To extend longevity and reduce mainte-
nance, exfiltration systems should always be off-line.
Detention With Filtration
Detention with filtration systems were proposed as an
alternative to retention for small projects (less than 8
acres) in those areas of Florida where local conditions,
especially flat topography and high watertables, prevent
infiltrating the stormwater treatment volume. The filters
must consist of 2 ft of natural soil or other suitable
fine-textured granular media that meet certain specifica-
tions, including:
17
-------
Filters must have pore spaces large enough to pro-
vide sufficient flow capacity so that the filter perme-
ability is equal to or greater than the permeability of
the surrounding soil.
The design shall ensure that particles within the filter
do not move.
When sand or other fine-textured material other than
natural soil is used for filtration, the filter material 1)
will be washed (less than 1 percent silt, clay, or or-
ganic matter) unless filter cloth is used to retain such
materials within the filter, 2) will have a uniformity
coefficient between 1.5 and 4.0, and 3) will have an
effective grain size of 0.20 to 0.55 mm in diameter.
Be designed with a safety factor of at least two.
Will recover the treatment volume (bleed down)
within 72 hr.
Filters are placed in the bottom or sides of detention
areas, where the filtered stormwater is collected in an
underdrain pipe and then discharged. Experience has
shown that these filters are very difficult to design and
construct. Operation is also difficult because of low hy-
draulic head, and maintenance is nearly impossible. It
is not a question of if a filter will clog, only when it will
clog. In addition, filters are designed to remove particu-
late pollutants and do not remove dissolved pollutants
such as phosphorus or zinc. Therefore, filtration sys-
tems are not recommended for use except under very
special conditions and where a full-time maintenance
entity such as a local government will assume such
responsibilities.
Wet Detention
Wet detention systems consist of a permanent water
pool, an overlying zone in which the stormwater treat-
ment volume temporarily increases the depth while it is
stored and slowly released, and a shallow littoral zone
(biological filter). In addition to their high pollutant re-
moval efficiencies (12), wet detention systems can also
provide aesthetic and recreational amenities, a source
of fill for the developer, and even "lake front" property,
which brings a premium price.
Wet detention criteria are listed in Table 2. These have
been developed to take full advantage of the biological,
physical, and chemical assimilation processes occurring
within the wet detention system. If the system is de-
signed as a development amenity, the use of pretreat-
ment BMPs integrated into the overall stormwater
management system is highly recommended to prevent
algal blooms or other perturbations that would reduce
the aesthetic value. Raised storm sewers in grassed
areas such as parking lot landscape islands, swale
conveyances, and perimeter swale/berm systems along
Table 2. Wet Detention Guidelines
1. Treatment volume as per Table 1.
2. Treatment volume slowly recovered in no less than 120 hr with
no more than half of the volume discharged within the first
60 hr following the storm:
Volume in the permanent pool should provide a residence
time of at least 14 days.
At least 30 percent of the surface area shall consist of
littoral area with slopes of 6:1 or flatter that is established
with appropriate native aquatic plants selected to maximize
pollutant uptake and aesthetic value.
Littoral zone plants shall have a minimum 80 percent
survival rate and coverage after 2 years. Cattails and other
undesirable plants shall be removed.
The littoral zone is concentrated near the outfall or in a
series of shallow benches ending at the outfall.
Side slopes should be no steeper than 4:1 out to a depth of
2 ft below the level of the permanent pool.
Maximum depth of 8 to 10 ft below the invert of the
discharge structure is recommended. Maximum depth shall
not create aerobic conditions in bottom sediments and
waters.
The flow length between inlets and outlet should be
maximized; a length-to-width ratio of at least 3:1 is
recommended. Diversion barriers such as baffles
An oil and grease skimmer shall be designed into the outlet
structure.
If the system is planned as a "real estate lake,"
pretreatment by infiltration is recommended.
Inlet areas should include a sediment sump.
the detention lake shoreline are techniques that have
been used frequently.
Wetland Treatment
Wetland treatment was authorized by the 1984 Hender-
son Wetlands Protection Act, which allows stormwater
treatment in wetlands that are connected to other state
waters by a constructed ditch or by an intermittent water
course that flows in direct response to rainfall, thereby
causing the water table to rise above ground surface.
Not only does this take advantage of natural treatment
mechanisms but it gives another economic value to
wetlandsan incentive to the developer to use, not
destroy, the wetlandand it revitalizes ditched and
drained wetlands by providing water.
Wetlands may be viewed as nature's kidneysthey
store stormwater, dampen floodwaters, transform pollut-
ants, and even retain pollutants, thereby providing natural
stormwater treatment (13). Care must be taken, how-
ever, to protect the numerous assimilation mechanisms
within the wetland plants and sediments. In addition, the
wetland hydroperiodthe duration that water stays at
various levelsmust be protected or restored because
it determines the form, function, and nature of the
wetland. Therefore, pretreatment practices to attenuate
18
-------
stormwater volume and peak rate and to reduce oil,
grease, and especially sediment are essential. The BMP
treatment train concept must be used to provide pre-
treatment, which normally includes a pretreatment lake
that is constructed adjacent to the wetland.
The following guidelines are presented for incorporating
wetlands into a stormwater management system:
The treatment volume is per Table 1, with the treat-
ment volume slowly recovered in no less than 120 hr
with no more than half of the volume discharged
within the first 60 hr following the storm.
Stormwater must sheet flow evenly through the wet-
land to maximize contact with the wetland plants,
sediments, and microorganisms. Spreader swales,
distribution systems, and level spreaders between
the pretreatment lake and the wetland have been
used extensively.
Swales should be used for stormwater conveyance
throughout the development.
The hydroperiod must be protected or restored.
Treatment volume capacity of the wetland is deter-
mined by the storage volume available between the
normal low and high elevations. These elevations are
determined by site-specific indicators such as lichen
and moss lines, water stain lines, adventitious root
formation, plant community zonation, hydric soils dis-
tribution and rack/debris lines.
Erosion and sediment control during construction is
essential because only a few inches of sediment de-
posited in the wetland will destroy the wetland filter.
Inflow/outflow monitoring, sediment metal levels, and
vegetative transect monitoring are required to help
evaluate the effectiveness of these systems and the
impacts of stormwater additions to wetlands.
Administration of the Stormwater Rule
Under the Florida Water Resources Act of 1972, DER,
a water quality agency, serves as the umbrella adminis-
tering agency delegating authority to five regional
WMDs whose primary functions historically have been
related to water quantity management. Therefore, a
second objective in developing the stormwater rule was
to coordinate the water quality considerations of DER's
stormwater permits with the water quantity aspects of
the districts' surface water management permits.
In addition, the delegation of the stormwater permit-
ting program allows for minor adjustments to stormwater
rule design and performance standards to better reflect
regional conditions. Florida is a very diverse state, with
major variations in soils, geology, topography, and rain-
fall that can directly affect the usability and treatment
effectiveness of a BMP. Such problems can be mini-
mized if districts adopt slightly different design and per-
formance standards which are approved by DER before
implementation.
Both DER's and the districts' stormwater rules essen-
tially require a new development to include a compre-
hensive stormwater management system. The system
should be viewed as a "BMP treatment train" in which a
number of different BMPs are integrated into a compre-
hensive system that provides aesthetic and recreational
amenities in addition to traditional stormwater manage-
ment objectives.
The Challenge Ahead
The implementation of Florida's stormwater treatment
requirements has been very effective in helping to reduce
the stormwater pollutant loading from new development.
As a result, the biggest stormwater management prob-
lem facing Florida is how to reduce pollutant loadings
discharged by older systems, especially local govern-
ment master systems constructed before the storm-
water rule was implemented. These systems were
designed solely for flood protection and rapidly deliver
untreated stormwater directly to rivers, lakes, estuaries,
and sinkholes.
Establishing a stormwater program to retrofit existing sys-
tems presents many technical, institutional, and financial
dilemmas. The unavailability and cost of land in urbanized
areas make conventional BMPs infeasible in most in-
stances. Current state laws and institutional arrangements
promote piecemeal, crisis-solving approaches aimed
at managing stormwater with in political boundaries, yet
stormwater follows watershed boundaries. Land-use plan-
ning and management must be fully integrated into the
stormwater management scheme. Retrofitting is also
prohibitively expensive, and many local governments
are already short of funds. Therefore, solving our exist-
ing urban stormwater problems requires comprehensive,
coordinated, creative approaches and technology.
Following is a brief discussion of some of the essential
elements of a comprehensive long-term effort to reduce
pollutant loadings from older stormwater systems.
Watershed Management
A watershed approach that integrates land-use planning
with the development of stormwater infrastructure is essen-
tial. After all, it is the intensification of land use and the
increase in impervious surfaces within a watershed that
create the stormwater and water resources manage-
ment problems. Consequently, a watershed manage-
ment team effort, involving state and local governments
together with the private sector, is necessary. In fact,
local governments are the primary team member because
they determine zoning and land use, issue building
permits and inspect projects, and have code enforce-
19
-------
ment powers that can help to ensure that stormwater
systems are properly operated and maintained.
Local governments need to identify and map the existing
natural stormwater system: the creeks, wetlands, flood
plains, drainageways, and natural depression areas.
Once mapped, these areas need to be zoned for con-
servation or low-intensity uses compatible with the func-
tions provided by the natural system. The existing
constructed stormwater system must also be mapped,
and essential characteristics such as pipe size, drain-
age areas, and invert elevations must be determined.
This information should then be fully integrated with the
existing and future land-use plan for the watershed and
a master stormwater management plan developed and
implemented. The Growth Management Act of 1985,
which requires all local governments to adopt compre-
hensive plans addressing current and future land use
with infrastructure needs, establishes a base structure
that could promote a watershed management approach.
Treatment Requirements for Older Systems
Numerous problems inherent in a highly urbanized area
prevent the application of new development stormwater
treatment standards from being imposed on older sys-
tems. Instead, a "watershed loading" concept is proposed
which considers the beneficial uses of the receiving
waters and the total stormwater load that can be assimi-
lated by the receiving waters. The actual treatment level
would depend on the watershed's total allowable load-
ing, which is based on citizen desires for certain bene-
ficial uses of the receiving water. The amount of load
reduction needed to restore or maintain the desired
beneficial uses of the receiving waters is known as the
pollutant load reduction goal (PLRG).
Selective Targeting
The extremely high cost of retrofitting older urban storm-
water systems also implies a need for careful evalu-
ation of pollutant reduction goals. A long-term (25 to
40 years) plan based on prioritization of watersheds
such that existing systems are selectively targeted for
modification is needed to ensure that citizens receive
the greatest benefit (pollutant load reduction, flood pro-
tection) for the dollar. The upgrading of older systems
must also be coordinated with other already planned
infrastructure improvements such as road widenings.
An excellent example of this approach is the Orlando
Streetscape Project. While downtown streets were torn
up for this downtown renovation, the existing storm-
water system was modified by the addition of off-line
exfiltration systems to reduce pollution loads to down-
town lakes.
Nonstructural BMPs and source controls also must be
used extensively to reduce stormwater pollution from
already developed areas. Improved street sweepers that
pick up the small particles (<60 microns) that contain
high concentrations of metals and other pollutants could
also prove valuable in reducing stormwater loadings,
especially from downtown business districts where
other BMPs usually are infeasible. Education programs
for the general public and for professionals involved in
stormwater management also are vital. Citizens must
understand how their everyday activities contribute to
stormwater pollution. For example, citizens should not
discard leaves, grass clippings, used motor oil, or other
material into swales or storm sewers. Getting youth and
citizen groups involved in storm sewer stenciling pro-
jects ("Dump No Wastes, Drains To Lake") is an excel-
lent way of reducing dumping of potential pollutants into
these conveyances. Even more importantly, compre-
hensive training and certification programs are needed
for those in the private and public sectors who design,
construct, inspect, operate, and maintain stormwater
management systems.
Funding
The cost of providing needed stormwater infrastructure
improvements to address current and future flooding
and water quality problems is gigantic. Yet local govern-
ments are already struggling financially, and traditional
revenue sources such as property taxes cannot be re-
lied on to pay for stormwater management. Instead, a
dedicated source of revenue based on contributions to
the stormwater problem is needed. The stormwater util-
ity can provide this. The city of Tallahassee implemented
Florida's first stormwater utility in October 1986, and
over 50 local governments have followed this example.
Innovative BMPs
The infeasibility of using traditional BMPs to reduce
stormwater pollutant loads in highly urbanized areas
means that creative and innovative BMPs are needed.
For example, alum injection within storm sewers was
used in Tallahassee to reduce stormwater loadings to
Lake Ella (14). A sonic flow meter measures storm
sewer flow, causing a flow-proportional dose of alumi-
num sulfate to be injected and mixed with the polluted
stormwater. As the alum mixes with the stormwater,
a small floe is produced which attracts suspended
and dissolved pollutants by adsorption and enmesh-
ment into and onto the floe particles. The floe then
settles to the lake's bottom sediments, gradually blan-
keting and incorporating into the sediments and
thereby reducing internal recycling of nutrients and
metals. Other advantages of alum injection include
excellent pollutant reduction (>85 percent) and rela-
tively low construction and operations costs, especially
for the highly urbanized areas. This type of system has
been installed at several locations in Florida with excep-
tional treatment efficiencies.
20
-------
Porous concrete consists of specially formulated mix-
tures of Portland cement, uniform open-graded coarse
aggregate, and water. When properly mixed and in-
stalled, porous concrete surfaces have a high percent-
age of void space which allows rapid percolation of
rainfall and runoff. Porous concrete is being used widely
in Florida, especially for parking lots, and could be an
important BMP to reduce stormwater loadings in highly
urbanized areas. Recent field investigations of porous
concrete parking areas that have been in place for up
to 12 years revealed that the infiltration capacity of the
concrete has not decreased significantly, a major con-
cern (15). Further information about the use, design, and
construction of porous concrete surfaces is available
(15).
Regional stormwater systems that manage stormwater
from several developments or an entire drainage basin
offer many advantages over the piecemeal approach
that relies on small, individual onsite systems. They
provide economies of scale in construction, operation,
and maintenance. Regional systems can also help man-
age stormwater from existing and future land uses and
will be a central part of any retrofitting program. The use
of regional systems is another good reason for a water-
shed management approach that fully integrates land
use and stormwater management.
The Southeast Lakes ProgramA Model
Many of the above elements of a watershedwide mas-
ter stormwater planning approach are being imple-
mented by the city of Orlando. The city has adopted an
excellent local stormwater ordinance and developed a
fine community education program and a prioritized
urban lake management program (16). One of the
most innovative programs is the Southeast Lakes Pro-
ject, which is designed to correct flooding problems and
to reduce stormwater pollutant loads to 15 urban lakes
and 58 drainage wells that currently convey untreated
stormwater to an aquifer. A corrective watershed man-
agement plan was cooperatively developed by the city,
its consultants, DER, and the St. Johns River WMD. The
project was initiated not because of enforcement of
water quality standards but because of a loss of bene-
ficial uses and local citizen desires and perceptions.
Modifications to the existing stormwater systems will be
made over a 10-year period, with treatment require-
ments based on "net environmental improvement" and
total watershed load.
One of the most important aspects of the project is the
use of innovative BMP designs that promote multiple
objectives and take advantage of city-owned properties.
At Al Coith Park, a spreader swale will be built on the
park's perimeter. When it rains, runoff will enter and fill
the swale, overtopping the sidewalk berm and sheet
flow across the grassed parkland where it will percolate
into the ground. At Lake Greenwood, the surrounding
city-owned land is being converted into an urban wet-
land and expanded lake. The wetland and lake is a
complex treatment train that incorporates many BMPs
into a very aesthetically pleasing stormwater system
and park that even includes reuse of stormwater to
irrigate the park and adjacent city-owned cemetery.
Near the Citrus Bowl, a packed-bed wetland filter has
been installed that will treat water from Lake Clear dur-
ing times of no rainfall. In addition to improved stormwa-
ter management, citizens are receiving the added
benefits of recreation and open space. In addition, the
retrofitting project has stimulated redevelopment and
renovation of existing properties, thereby providing citi-
zens with economic benefits as property values rise.
Chronologic Evolution of Florida's
Watershed Management Program
Following is a chronology of the establishment and re-
vision of Florida statutes and programs that are consid-
ered cornerstones of the state's overall watershed
management efforts. As such, this chronology traces the
evolution of Florida's watershed management program.
1970
Chapter 370, FS, created the Coastal Coordinating
Council, which was the first state effort at integrating
state/regional programs in the protection and use of
coastal resources. Initial efforts from 1970 to 1975 fo-
cused on a comprehensive resource-based coastal pro-
tection program.
1972
A package of land and water planning, regulation, and
acquisition programs was created:
Chapter 380, FS: This creates the Developments of
Regional Impact (DRI) and Areas of Critical State Con-
cern (ACSC) land planning and management programs.
Chapter 373, FS: The Florida Water Resources Act
establishes the state's five regional WMDs; desig-
nates the Department of Pollution Control as the
oversight agency for the WMDs; requires the devel-
opment of a state water plan; and allows for the
regulation of the water resource. WMDs financed by
ad valorem property taxing authority of up to 1 mil
($1/$1000 value) which is set in the Florida Consti-
tution. NWFWMD millage capped at 0.05 mil.
Chapter 259, FS: The Land Conservation Act estab-
lishes a program, commonly known as the Environ-
mentally Endangered Lands Program, which
authorizes the state to purchase critical and sensitive
lands; envisioned as a 10-year program investing
$200 million and funded by the sale of state bonds.
21
-------
1973
In Chapter 403, FS, the Florida Environmental Pro-
tection Act renames the Department of Pollution Control
as the Department of Environmental Regulation and
broadens its powers, duties, and programs. This law is
the state's general environmental protection act. It is
amended almost annually as new environmental con-
cerns and needs arise and as existing programs evolve.
1975
Chapter 163, FS, the Local Government Comprehen-
sive Planning Act and the state's first growth manage-
ment legislation, was recommended by the first
Environmental Land Management Study Committee
(ELMS I). The law requires all cities and counties to
prepare comprehensive plans which are submitted for
review to the state's land planning agency, the Depart-
ment of Community Affairs, which in turn sends the
plans to other state agencies for review and comment.
However, the LGCPA contains no "teeth." Local govern-
ments are under no statutory requirement to revise their
plans by incorporating the comments and recommenda-
tions made by the state reviewing agencies. Further-
more, they are not required to pass land development
regulations to implement their plans.
1976
Implementation by EPA and the states of Section 208
of the 1972 Clean Water Act occurs, requiring the
development of Areawide Water Quality Management
Plans. This was the first national program directed at
the assessment and control of nonpoint sources of pol-
lution. In Florida, millions of federal grant dollars allows
the DER and 12 "Designated Area Agencies" to under-
take extensive research on nonpoint source impacts,
sources, controls, control effectiveness, and costs. These
data provide the scientific basis for the development and
implementation in 1982 of a statewide rule that requires
treatment of stormwater for new development and rede-
velopment projects.
1978
Chapter 380, FS, is amended, adding Part II, the Florida
Coastal Management Act, which requires establishment
of a program based on existing statutes and rules to
serve as a basis for receiving federal approval under
the Federal Coastal Zone Management Act of 1972. After
approval of the program by the National Oceanic and
Atmospheric Administration, Office of Coastal Zone Man-
agement, federal grants fund many initiatives to better
protect and manage coastal resources. One particular
initiative establishes an estuarine watershed manage-
ment program with emphasis on sediment mapping.
This project leads to the development of innovative,
reliable coastal sediment sampling, analytical, and as-
sessment techniques.
1979
The first components of the state's Areawide Water
Quality Management Plan, the Agriculture Nonpoint
Source Plan and the Silviculture Nonpoint Source Plan,
are submitted to and approved by EPA. These call for a
non-regulatory approach with a regulatory backstop if
BMPs required by farm conservation plans are not im-
plemented or if the forestry BMPs required by the state's
adopted Silviculture BMP Manual are not followed.
Chapter 17-4.248, FAC, the state's first stormwater rule,
is adopted by the state ERG as a rule of DER. This rule
is intended as a temporary regulation until ongoing re-
search on BMP design and effectiveness is completed.
The rule's adoption is controversial, but data collected
during from 208 program studies conclusively show that
stormwater runoff, especially from urban land uses and
highways, is a "pollutant" and therefore should be con-
trolled. Florida's continuing rapid growth makes it im-
perative that treatment of stormwater, using BMPs, be
required for new stormwater discharges that would be
"a significant source of pollution."
Chapter 253, FS, is amended to establish the Conser-
vation and Recreation Land (CARL) Trust Fund, which
provides additional funding for the purchase of Environ-
mentally Endangered Lands and other lands deemed
appropriate and in the public interest by the Governor
and Cabinet.
1981
Through action taken by the Governor and Cabinet, the
Save Our Coasts land acquisition program is established.
The program proposes to spend $200 million over 10
years to purchase coastal lands such as beaches, shore-
lines, and sensitive areas. Funding is provided by the
sale of state bonds backed by documentary stamps as
authorized in Chapter 375, FS, which sets policy on how
the Land Acquisition Trust Fund is to be administered.
Chapter 373, FS, is amended with the creation of the Save
Our Rivers land acquisition program. Administered by the
WMDs, this program proposes to spend $320 million
over 10 years to purchase wetlands, flood plains, and other
lands necessary for water management, water supply,
and the conservation and protection of water resources.
1982
Chapter 17-25, FAC, is adopted by the ERG after 2 years
of rule adoption workshops and 29 official rule drafts.
The rule is technology based rather than water quality
based, although the state's water quality standards
remain as a backstop should a stormwater discharge be
causing violations. A performance standard of 80 per-
22
-------
cent average annual load reduction is recommended,
based on BMP effectiveness and cost data, to establish
equity with minimum treatment levels for point source
discharges. The rule creates design criteria for various
types of BMPs, including retention, detention with filtra-
tion, and wet detention. The rule creates "general per-
mits" for certain types of BMPs (i.e., retention, detention
with filtration) if they are built to the design criteria.
Implementation of the rule is delegated to the South
Florida WMD, allowing stormwater treatment require-
ments to be merged with stormwater quantity (flood
control) requirements in one permit.
1984
Chapter 403, FS, is revised to create Section IX, which
is known as the Henderson Wetlands Protection Act.
This legislation expands the authority of the DER to
protect wetlands; establishes administrative procedures
to allow landowners to obtain legally binding "wetland
lines"; allows the DER to consider fish and wildlife habi-
tat, endangered species, and historical and archaeologi-
cal resource and other relevant concerns in wetland
permitting; allows the use of certain wetlands for incor-
poration into domestic wastewater and stormwater man-
agement systems; transfers wetland regulation for
agriculture and forestry activities to the WMDs; and
requires the WMDs to protect isolated wetlands and
consider fish and wildlife habitat requirements.
The Southwest Florida Water Management District
(SWFWMD) receives delegation of the stormwater rule.
In the late 1970s and early 1980s, an extensive ap-
praisal of Florida's growth management system was
undertaken, which concluded that the existing system
was not working. Shaped by the Final Report of the
Governor's Task Force on Resource Management (1980)
and the second Environmental Land Management
Study Committee (ELMS II), a totally new blueprint for
managing growth emerged. The ELMS II Committee
recommended a comprehensive package of integrated
state, regional, and local comprehensive planning; reforms
to the DRI law; and coastal protection improvements.
The state legislature responded between 1984 and
1986 by enacting several laws. For example, Chapter
186, FS, the State and Regional Planning Act, mandates
that the Governor's Office prepare a state comprehen-
sive plan and present it to the 1985 state legislature. It
also requires the preparation of regional plans by the
state's 11 regional planning councils and provides them
with $500,000 for plan preparation.
1985
Chapter 187, FS, the State Comprehensive Plan, origi-
nally is envisioned to be a leadership documentthe
foundation of the entire planning processwith strong,
measurable, and strategic goals that will set the course
for Florida's growth over the next 10 years. Each state
agency is to prepare an agency functional plan, based
on the state plan, on which its budget appropriations will
be made. Unfortunately, one of the most important ele-
ments of the state plan, the development and adoption
of a capital plan and budget, is never prepared. How-
ever, the plan contains important goals and policies in
25 different areas, including water resources, coastal
and marine resources, natural systems and recreation,
air quality, waste management, land use, mining, agri-
culture, public facilities, and transportation.
Important and relevant goals include:
Ensure the availability of an adequate water supply.
Maintain the functions of natural systems.
Maintain and enhance existing surface- and ground-
water quality.
Important and relevant policies include:
Eliminate the discharge of inadequately treated
wastewater and stormwater.
Protect natural systems in lieu of structural alterna-
tives, and restore modified systems.
Promote water conservation and the use and reuse
of treated wastewater and stormwater.
Establish minimum flows and levels for surface wa-
ters to ensure protection of natural systems.
1985 to 1986
Chapter 163, FS, is amended with enactment of the
Local Government Comprehensive Planning and Land
Development Regulation Act of 1985. This law requires
all local governments to prepare local comprehensive
plans and implementing regulations, which must be con-
sistent with the goals and policies of the state and
regional plans. Numerous state and regional agencies
review the local plans and submit their objections, rec-
ommendations, and comments to the Department of
Community Affairs for transmittal to the local govern-
ment. This time the local plans must be revised to
incorporate the objections, recommendations, and com-
ments. Furthermore, local governments face sanctions
from the state that could result in the loss of state
funding if adopted local plans are not consistent with the
state and regional plans.
Florida's revised growth management system is built
around three key requirements: consistency, concur-
rency, and compactness. The consistency requirement
establishes the "integrated policy framework," whereby
the goals and policies of the state plan frame a system
of vertical consistency. State agency functional plans
and Regional Planning Council regional plans must be
consistent with the goals and policies of the state
23
-------
plan while local plans are required to be consistent
with the goals and policies of the state and appropriate
regional plan. Local land development regulations
(LDRs) must also be consistent with the local plans
goals and policies. Horizontal consistency at the local
level also is required to ensure that the plans of neigh-
boring local governments are compatible. Consistency
is the strong cord that holds the growth management
system together.
Concurrency is the most powerful policy requirement
built into the growth management system. It requires
state and local governments to abandon their long-
standing policy of deficit financing growth by implement-
ing a "pay as you grow system." Once local plans and
LDRs are adopted, a local government may approve
development only if the public facilities and services
(infrastructure) needed to accommodate the impact of
the proposed development can be in place concurrent
with the impacts of the development. Public facilities and
services subject to the concurrency requirements are
roads, stormwater management, solid waste, potable
water, wastewater, parks and recreation, and, if applica-
ble, mass transit. Level of service standards acceptable
to the community must be established for each type of
public facility.
Compact urban development goals and policies are built
into the State Comprehensive Plan and into regional
plans. Such policies as separating rural and urban land
uses, discouraging urban sprawl, encouraging urban
in-fill development, making maximal use of existing in-
frastructure, and encouraging compact urban develop-
ment form the basis for this requirement.
1986
Chapter 403.0893, FS, is created as the only surviving
section of a stormwater management bill that was de-
veloped over a 10-month period. The bill was an attempt
to put into statute a cost-effective, timely process to
retrofit existing drainage systems to reduce the pollutant
loadings discharged to water bodies. Only the section
creating explicit legislative authority for local govern-
ments to establish stormwater utilities or special storm-
water management benefit areas is enacted.
The St. Johns River WMD adopts Chapter 40C-42,
FAC, and the Suwannee River WMD adopts Chapter
40B-4, FAC. Adoption of these two stormwater manage-
ment regulations and the addition of staff to imple-
ment these programs allows DER to delegate
administration of its stormwater treatment rule to these
WMDs, which, in turn, allows DER's stormwater qual-
ity permit to be combined with the districts' stormwater
quantity permit.
1987
Chapter 373, FS, is revised to add a new section, the
Surface Water Improvement and Management (SWIM)
Act, which establishes six state priority water bodies. It
directs the WMDs, under DER supervision, to prepare
a priority water body list and develop and adopt compre-
hensive watershed management plans to preserve or
restore the water bodies. It provides $15 million from
general revenue sources and requires a match from the
WMDs. The act does not establish a dedicated funding
source, making the program dependent on uncertain
annual appropriations from the legislature.
1988
Chapter 17-43, FAC, the SWIM rule, is adopted by the
ERG. It sets forth factors to consider in the selection of
priority water bodies, specifies the format for SWIM
plans to ensure some consistency, and establishes ad-
ministrative processes for the development and adop-
tion of SWIM plans by the WMDs and for their submittal
to DER for review and approval.
The State Nonpoint Source Assessment and Manage-
ment Plan, prepared pursuant to Section 319 of the
federal Clean Water Act, is submitted to EPA and ap-
proved. This qualifies the state for Section 319 nonpoint
source implementation grants, which are used for BMP
demonstration projects and to refine existing nonpoint
source management programs. The delineation of the
state's ecoregions (based on river systems), selection
of ecoregion reference sites, and modification of EPA's
Rapid Bioassessment Protocols and metrics for use in
Florida are initiated.
1989
Chapters 373 and 403, FS, are revised as part of the
1989 stormwater legislation. The legislation clarifies the
stormwater program's multiple goals and objectives;
sets forth the program's institutional framework, which
involves a partnership between DER, the WMDs, and
local governments; defines the responsibilities of each
entity; addresses the need for the treatment of agricul-
tural runoff by amending Chapter 187, FS, to add a
policy in the Agriculture Element to "eliminate the dis-
charge of inadequately treated agricultural wastewater
and stormwater"; further promotes the watershed ap-
proach being used by the SWIM program; attempts to
integrate the stormwater program, SWIM program, and
local comprehensive planning program (but does not
succeed); establishes State Water Policy, an existing
but little-used DER rule, as the primary implementation
guidance document for stormwater and all water re-
sources management programs; and creates the State
Stormwater Demonstration Grant Fund as an incentive
to local governments to implement stormwater utilities
and provides $2 million.
24
-------
1990
Chapter 17-40, FAC, State Water Policy, undergoes a
total revision and reorganization so that it can be used
as guidance by all entities implementing water resource
management programs and regulations. Section 17-
40.420 is created and includes the goals, policies, and
institutional framework for the state's stormwater man-
agement program.
DER is designated as the lead agency with responsi-
bility for setting goals for the program, for providing
overall program guidance, for overseeing implementa-
tion of the program by the WMDs, and for coordinating
with EPA, especially with the advent of the new National
Pollutant Discharge Elimination System stormwater per-
mitting program.
WMDs are the chief administrators of the stormwater
regulatory program (quantity and quality); they are re-
sponsible for preparing SWIM watershed management
plans, which include the establishment of stormwater
PLRGs; they provide technical assistance to local gov-
ernments, especially with respect to basin planning and
the development of stormwater master plans.
Local governments are the frontlines in the stormwa-
ter/watershed management program because they de-
termine land use and provide stormwater and other
infrastructure. They are encouraged, but not required, to
set up stormwater utilities to provide a dedicated funding
source for their stormwater program. Their stormwater
responsibilities include preparation of a stormwater
master plan to address needs imposed by existing land
uses and those needs to be created by future growth;
operation and maintenance activities; capital improve-
ments of infrastructure; and public education. They are
encouraged to set up an operating permit system
wherein stormwater systems are inspected annually to
ensure that needed maintenance is performed.
Important goals include:
Preventing stormwater problems from land-use
changes and restoring degraded water bodies by re-
ducing the pollution contributions from older storm-
water systems.
Retaining sediment on site during construction.
Trying to ensure that the stormwater peak discharge
rate, volume, and pollutant loading are no greater
after a site is developed than before.
Important minimum treatment performance standards
include:
80 percent average annual load reduction for new
stormwater discharges to most water bodies.
95 percent average annual load reduction for new
stormwater discharges to Outstanding Florida Wa-
ters, which are a special class of exceptionally high-
quality water bodies.
Reducing, on a watershed basis, the pollutant loading
from older stormwater systems as needed to protect,
maintain, or restore the beneficial uses of the receiv-
ing water body.
Chapter 375, FS, is amended with the creation of Pres-
ervation 2000, a 10-year land acquisition program with
a goal of spending $300 million per year. The legislation
divided available annual funding among seven programs:
CARL, Save Our Rivers (SOR), Florida Communities
Trust, State Parks, State Forests, State Wildlife Areas,
and Rails to Trails. The program is funded the first year
by state bonds backed by an increase in the documen-
tary stamp fee. Unfortunately, a long-term dedicated
funding source is not identified, making the program
subject to annual legislative appropriations. Between
1972 and 1991, the state's land acquisition programs
have invested over $1.5 billion to buy over 1.2 million
acres. Equally important, as a result of the state land
acquisition programs, 14 Florida counties have created
local programs that currently commit up to $600 million
for land conservation. Revenue sources for these local
land acquisition programs include local option sales tax,
impact fees, added property taxes, and local bonds.
1991
Chapter 40C-42, FAC, is completely revised by the St.
Johns River WMD to modify the design criteria for
stormwater treatment BMPs so that they will achieve the
minimum treatment levels set in State Water Policy.
Stormwater reuse becomes essential for developments
discharging to Outstanding Florida Waters.
Chapter 40C-44, FAC, is adopted by the St. Johns River
WMD to regulate certain agricultural pumped dis-
charges (formerly regulated as industrial wastewater)
and establishes design and performance criteria for
these agricultural stormwater management systems.
The SWFWMD initiates development of an agricultural
stormwater management program for certain types of
agricultural activities including row crops and citrus. The
program includes regulatory incentives to obtain techni-
cal assistance from U.S. Department of Agriculture, Soil
Conservation Service, or other qualified individuals to
prepare and implement a farm-specific resource man-
agement plan that contains certain required BMPs.
1992
DER and the WMDs, in response to increasing de-
mands on the state's waters and the increasing number
of water quantity and quality problems, begin the devel-
opment of district water management plans, collectively
these district plans, together with the DER's plan, will
create the state water management plan. These plans
25
-------
are based on the goals and policies set in State Water
Policy and in the state comprehensive plan. For each of
four major areas (water supply, water quality, flood pro-
tection, natural systems protection), four key planning
steps will occur:
Resource assessment to identify current or antici-
pated problems.
Examination of options.
Declaration of policy.
Designation of implementation strategies.
Section 314 Federal Clean Lake Program Lake Assess-
ment Grant is obtained to initiate the delineation of lake
ecoregions, select lake ecoregion reference sites, and
test/validate lake bioassessment sampling protocols
and metrics.
1993
Chapters 373 and 403, FS, are revised extensively as
part of the DER/Department of Natural Resources
merger to create the Department of Environmental Pro-
tection (DEP) and as a part of the Environmental Permit
Streamlining bill. The goals of the streamlining bill are
to eliminate duplication, especially in permitting; in-
crease administrative and environmental effectiveness
by increasing delegation of programs from DEP to the
WMDs; and ensure greater program consistency and
integration. Key specific actions of the bill include:
Moving the "Wetlands Protection Act" from Chapter
403 to Chapter 373, FS, thereby delegating the wet-
land resource permits to the WMDs except for certain
projects that require other types of DEP permits.
Merging the existing surface water/stormwater man-
agement permit with the wetland resource permit to
create an environmental resource permit.
Redefining wetlands based on their hydrology, vege-
tation, and soils, and requiring the development of a
single wetland delineation method that will be used
by the DEP, WMDs, and local governments.
Recommendations of the third Environmental Lands
Management Study Committee (ELMS III) are enacted
into law (with a 180-page act), thereby amending sev-
eral state laws. The act seeks to strengthen the state
planning process by:
Requiring the Governor to biannually review and ana-
lyze the state comprehensive plan and recommend
any necessary revisions.
Requiring the Governor to prepare a new growth
management portion of the state comprehensive
plan. This is to provide a more detailed and strategic
state policy guidance for state, regional, and local
governments in implementing the state plan. It is to
identify urban growth centers; set strategies to protect
identified areas of state and regional environmental
importance; and provide guidelines for determining
where urban growth is appropriate and should be en-
couraged. The growth management document must be
adopted by the legislature. However, to what extent
local comprehensive plans, state agency strategic plans,
and regional policy plans must be consistent with the
state plan is unknownto be recommended by the
Governor and adopted as law by the 1994 legislature.
The act also provided greater flexibility and less require-
ments in local comprehensive plans for small cities (^5,000)
and counties (<50,000); streamlined the plan amendment
process by limiting the types of revisions requiring state
review and approval; strengthened the local plan evalu-
ation and appraisal process; terminated or made optional
the development of regional impact (DRI) process in cer-
tain areas and revised the DRI process; and authorized
local option gas tax of up to 5 cents.
Discussion and Recommendations
Florida has established a wide variety of laws, regula-
tions, 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 resources. There is no doubt that these programs
have been effective in helping to reduce adverse im-
pacts on natural resources resulting from the state's
rapid and continuing growth over the past 20 years.
Even with the implementation of these programs, how-
ever, many of Florida's natural resources have been
severely strained or degraded. Some of these adverse
effects can be attributed to activities that occurred be-
fore the implementation of modern watershed manage-
ment 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 20 years. These in-
clude 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 al-
ready degraded areas? Following is a list of program
deficiencies and recommendations to correct them:
While the statutes enacted by the legislature may be
helpful, insufficient resources have been provided to
the governmental entities that are to implement the
programs. The state's reliance on sales tax as it pri-
mary means of raising "general revenues" means that
26
-------
state revenues are tied closely to economic condi-
tions. Relying on such sources during a recession,
especially when population growth is still occurring,
means that the state budget is nearly always in crisis.
Dedicated sources of funding are needed if water-
shed management programs are going to compete
for limited state resources and have adequate re-
sources to actually achieve their intended benefits.
The statutes and programs are not fully integrated,
leaving gaps in both land planning and water plan-
ning programs. In particular, there is a need to better
integrate water and land planning and regulatory pro-
grams. The local government growth management
program needs to be more closely connected to state
and regional water management programs. The re-
quirements set forth in State Water Policy and in the
district/state water management plans need to be
used by local governments in their land-use planning
programs. These local plans need to be consistent
among all state, regional, and local programs.
Greater emphasis needs to be placed on ensuring
the long-term maintenance and operation of storm-
water management systems. Because these systems
are a part of the local infrastructure, local govern-
ments need to take a more active role in this area.
Establishing stormwater operation permits as part of
a stormwater utility funded program is an excellent
way of providing an economic incentive to a land
owner to maintain and operate an onsite stormwater
management system properly.
Greater emphasis needs to be placed on erosion and
sediment control on construction sites and on utility
installation projects. A major deficiency is ensuring
the regular inspection of erosion prevention and sedi-
ment control practices. Implementation of a training
and certification program for inspectors and contrac-
tor supervisors, similar to the Certified Construction
Reviewer Program in Delaware, is needed.
Retrofitting existing drainage systems to reduce their
pollutant loading is one of the biggest, most difficult,
and most expensive challenges the state has ever
faced. One of the major problems in meeting this
challenge is the need to develop new stormwater
treatment techniques that are not land intensive.
Funding of demonstration projects and for research
of new techniques in needed.
While Floridians are among the most educated citi-
zens in the country with respect to water resources
and stormwater management issues, more education
is needed to help gain citizen support for watershed
management programs. The state's environmental
education program needs to focus on establishing a
comprehensive natural resources management cur-
riculum that begins in kindergarten and continues all
the way through high school. Additionally, because of
the large number of people who are moving to Florida,
especially retirees, continuous education programs are
needed to educate these people about the vulnerabil-
ity and importance of Florida's natural resources.
References
1. Livingston, E.H. 1984. A summary of activities conducted under
the Florida Section 208 water quality management planning pro-
gram, February 1978September 1984. Final report submitted
to the U.S. Environmental Protection Agency.
2. U.S. EPA. 1983. Water quality standards handbook. NTIS
PB92231851. Washington, DC: Office of Water Regulations and
Standards.
3. U.S. EPA. 1987. Nonpoint source controls and water quality
standards. In: Water quality standards handbook. Washington,
DC: Office of Water, pp. 2-25.
4. Anderson, D.E. 1982. Evaluation of swale design. M.S. thesis.
University of Central Florida, College of Engineering, Orlando, FL.
5. Wanielista, M.P., Y.A. Yousef, G.M. Harper, T.R. Lineback, and L.
Dansereau. 1991. Precipitation, interevent dry periods, and reuse
design curves for selected areas of Florida. Final report submitted
to the Florida Department of Environmental Regulation, Talla-
hassee, FL.
6. Wanielista, M.P., and E.E. Shannon. 1977. Stormwater manage-
ment practices evaluations. Report submitted to the East Central
Florida Regional Planning Council, Orlando, FL.
7. Miller, R.A. 1985. Percentage entrainment of constituent loads in
urban runoff, South Florida. U.S. Geological Survey WRI Report
84-4329.
8. Wanielista, M.P., Y.A. Yousef, B.L. Golding, and C.L. Cassagnol.
1982. Stormwater management manual. Prepared for Florida
Department of Environmental Regulation, Tallahassee, FL.
9. Livingston, E.H., J.C. Cox, M.E. McCarron, and PA. Sanzone.
1988. The Florida development manual: A guide to sound land
and water management. Tallahassee, FL: Florida Department of
Environmental Regulation.
10. Yousef, Y.A., M.P. Wanielista, H.H. Harper, D.B. Pearce, and R.D.
Tolbert. 1985. Removal of highway contaminants by roadside
swales. Report FL-ER-30-85. Submitted to Florida Department of
Transportation, Tallahassee, FL.
11. Wanielista, M.P, Y.A. Yousef, L.M. VanDeGraaff, and S.H.
Rehmann-Koo. 1985. Enhanced erosion and sediment control
using swale blocks. Report FL-ER-35-87. Submitted to Florida
Department of Transportation, Tallahassee, FL.
12. U.S. EPA. 1983. Results of the nationwide urban runoff program.
Final report.
13. Richardson, C.J. 1988. Freshwater wetlands: Transformers, fil-
ters or sinks? FOREM 11(2):3-9. Duke University School of
Forestry and Environmental Studies.
14. Harper, H.H., M.P. Murphy, and E.H. Livingston. 1986. Inactiva-
tion and precipitation of urban runoff entering Lake Ella by alum
injection in storm sewers. In: Proceedings of the North American
Lake Management Society International Symposium, Portland,
OR (November).
15. Florida Concrete and Products Association. 1988. Pervious pave-
ment manual. Orlando, FL.
16. Zeno, D.W, and C.N. Palmer. 1986. Stormwater management in
Orlando, Florida. In: Urban runoff quality: Impact and quality
enhancement technology. Engineering Foundation Conference,
Henniker, NH (June).
27
-------
The State of Delaware Sediment Control and Stormwater Management Program
Earl Shaver
Delaware Department of Natural Resources and Environmental Control,
Dover, Delaware
Institutional Philosophy
Before submitting proposed legislation regarding storm-
water management or sediment control, representatives
of the State of Delaware Department of Natural Re-
sources and Environmental Control (DNREC) con-
ducted an extensive educational program to document
the serious nature of water quantity and quality prob-
lems that exist statewide. This problem documentation
was successful in that elected officials, affected indus-
tries, and the general public acknowledged the need for
a comprehensive approach to sediment control and
stormwater management. The statewide legislation was
unanimously approved in four committees and on the
floor of both the state senate and the house of repre-
sentatives. The local conservation districts were instru-
mental in their support of the legislation. In addition, the
regulations detailing the legislative requirements were
approved with no negative comments after an extensive
educational process and with the assistance of a regu-
latory advisory committee.
A basic premise of the program is that sediment control
during construction and stormwater quantity and water
quality control postconstruction are all components of
an overall stormwater management program that func-
tions from the time that construction is initiated through
the lifespan of the constructed project (Figure 1). Pro-
gram implementation was initiated on July 1, 1991, and
the initial emphasis of the program is to prevent existing
flooding or water quality issues from worsening. The
intent is to limit further degradation until more compre-
hensive, watershed-specific approaches, as detailed in
the state legislation and regulations, can be adopted.
Program Structure
The structure of the sediment and stormwater manage-
ment program is based on the premise that ultimate
program responsibility must rest with the state. In the
case of Delaware, the state agency responsible for pro-
gram implementation is DNREC. DNREC is the ultimate
approval authority. Local conservation districts and ju-
risdictions, however, may request delegation of four
program components:
Sediment control and stormwater management plan
approval.
Inspection during construction.
Postconstruction inspection of permanent stormwater
facilities.
Education and training.
The sediment control and stormwater management plan
review and approval process must be completed before
any building or grading permits are issued. Criteria for
plan review and approval are contained in state regula-
tions, and design aids and handbooks have been devel-
oped or approved by DNREC. One important distinction
of the Delaware program is that the delegated local
agency handles day-to-day inspection responsibilities.
Stormwater
Management
During
Construction
Erosion and
Sediment Control
After Construction
Permanent
Stormwater
Controls
\
Quantity
Quality
Figure 1. Stormwater management.
28
-------
Projects for which site compliance cannot be achieved
are transferred to the state, where progressive, aggres-
sive enforcement is carried out. State enforcement op-
tions include civil and criminal penalty provisions.
Control Practices
Site control practices (Figures 2 and 3) are grouped into
two categories: temporary practices during construction
and permanent practices for postconstruction runoff.
Sediment control practices, designed for temporary site
control, must comply with the Delaware Erosion and
Sediment Control Handbook. This handbook details nu-
merous practices that are available for use depending
on applicability. The plan review process ensures that
the sediment control practices are located appropriately.
In addition to the traditional structural controls that the
handbook contains, the regulations have several re-
quirements that are important to providing overall site
control. Site stabilization must be accomplished if the
disturbed areas are not being actively worked for a
period in excess of 14 days. In addition, unless modified
for a specific type of project, no more than 20 acres
may be disturbed at any one time to facilitate phasing
of a project.
The regulations specifically require that water quality
must achieve an equivalent removal efficiency of 80
percent for suspended solids. From a permanent storm-
water management standpoint, initial consideration for
control must be a pond that has a permanent pool of
water. These wet ponds also have an extended deten-
tion requirement placed on them in addition to peak flow
control of larger storms. Ponds having a normal pool are
preferred over either normally dry extended detention
ponds or infiltration practices due to their documented
performance records and the ability of wet ponds to
reduce downstream nutrient loadings. Wet ponds, if
properly designed, also can be an amenity to the com-
munity where they are placed. A major emphasis is
being placed on constructed wetlands as a primary
stormwater treatment system in upland areas. The Dela-
ware program does not encourage the use of existing
wetlands for stormwater treatment.
Another option for site control is the use of infiltration
practices. These practices are allowed but not encour-
aged due to their potential for clogging and concern over
Inflow
Settling
Velocity
(Gravity)
Dewatering
Riser.
Sediment Velocity
Toward
Trash Rack
and
Antivortex
Pond Embankment
Crest of Emergency
Spillway
Cutoff
Trench
Figure 2. Sediment pond (to be converted to permanent stormwater management facility).
Antiseep
Collar
Freeboard
Perforated Riser
Encased in Gravel Jacket
Figure 3. Extended detention pond.
29
-------
ground-water pollution. Experience in other jurisdictions
has demonstrated the potential that infiltration practices
have for clogging. Where infiltration practices are used,
upslope and downslope impacts in the event of clogging
are carefully considered during the plan review process.
Infiltration of stormwater runoff is a necessary compo-
nent of an overall stormwater management program, but
critical safeguards relating to filtering of stormwater and
ground-water pollution concerns must be considered
before design approval.
Filtration of runoff also must be a program component
either as a stand-alone practice or in conjunction with
other practices, primarily infiltration. Common filtration
practices generally rely on vegetative filtering of runoff
over filter strips or through swale systems. On highly
impervious sites, vegetative filters often are not feasible;
in these situations, a sand filter design may be appro-
priate for initial water quality treatment (Figure 4). Sev-
eral variations in sand filter designs may be applicable
from site to site, but defined design criteria must be
followed if the system is to be effective at pollutant
removal.
Unique Features
Several features of the Delaware program are unique.
The regulations clearly require that stormwater manage-
ment practices achieve an 80-percent reduction in sus-
pended solids load after a site has been developed. The
only other state to present a similar performance criteria
is Florida. The 80-percent figure was selected based on
a review of documented stormwater practice perform-
ances around the country. That level of performance can
be achieved with present technology application. Long-
term removal rates in excess of 80 percent may require
extraordinary measures such as water reuse, which
may be required on a local basis but which is not prac-
tical from a statewide perspective.
The concept of delegation of program components is
fairly unique with respect to program implementation. In
Delaware, each aspect of program implementation may
be delegated, with DNREC acting as a safety net in the
event that a conservation district or a local government
fails to adequately implement an aspect of the program.
The initial concept of delegation was developed in Mary-
land for inspection of sediment control; the concept was
expanded in the Delaware law and regulations to en-
compass all aspects of program implementation. The
actual interaction of state and local program implemen-
ters has quickly become a partnership effort, with the
state providing technical expertise and educational
training while the conservation districts and local gov-
ernments provide for actual program implementation.
A major way in which the Delaware program is unique
is in the use of privately provided inspectors (Certified
Construction Reviewers). The land developer on larger
projects (over 50 acres in size or where the state or
delegated inspection agency requires) must provide
sediment control and stormwater inspectors to assist the
appropriate governmental inspection agency. These in-
spectors must attend and pass a DNREC course on
inspection, inspect active construction sites at least
once a week, and submit an inspection report to the
developer/contractor and the inspection agency on their
findings and recommendations. The inspection agency
still must periodically inspect the site to ensure the
adequacy of site controls, but the designated inspector
reduces the frequency of inspection for the inspection
agency. Failure to accurately record site conditions or
failure to notify either the contractor/developer or inspec-
tion agency of site deficiencies may jeopardize the desig-
Cover Grates
Trapped Solids
18 In. of Sand
Sedimentation Chamber
(Heavy Sediments,
Organics, Debris)
Filtration Chamber
Screen Covered
With Filter Fabric
Figure 4. Sand filter design.
30
-------
nated inspector's certification, which could be grounds
for enforcement action against the contractor/developer.
Another important concept that is becoming increasingly
popular among states implementing sediment control
programs is the requirement that contractors must have
a responsible individual(s) certified as having attended
a DNREC course for sediment control and stormwater
management. The Delaware course lasts approximately
4 hours and attempts to acquaint contractors with the
importance of good site erosion and sediment control
and stormwater management, as well as with their re-
sponsibilities under the law. The contractor certification
program is extremely popular with contractors and re-
duces the "we-they" problems that often exist in regula-
tory programs.
Evolution
The program discussed above represents the initial
phase of program implementation in Delaware. The next
step relates to addressing stormwater management
from a watershed perspective. The sediment and storm-
water regulations contain a Designated Watershed con-
cept that allows for the design and construction of
practices on a watershed basis that, when coupled with
land-use planning, wetland restoration, and other non-
structural practices, reduces existing flooding problems
or improves existing water quality. The expectation is
that one watershed will be designated in each county to
serve as a model for other watersheds. These water-
sheds will be studied from a hydrologic, water quality,
and stream habitat and diversity standpoint, and alter-
native land uses and stormwater controls will be consid-
ered along with their impact on water quality. Based on
the results of the watershed study, a recommended
approach for watershed protection will be developed in
conjunction with local government officials that presents
a blueprint for future resource protection in these Des-
ignated Watersheds.
Funding is another area that must be addressed if the
initial program is to be expanded. The state law and
regulations provide a framework for expanding tradi-
tional funding mechanisms with more innovative types
of funding. The regulations contain significant informa-
tion on the consideration of stormwater utilities (user
fees) as an alternative to permit fees or general funding.
The stormwater utility is expected to accompany the
Designated Watershed concept as a mechanism to fund
the watershed studies, planning, design, implementa-
tion of practices, and the maintenance of completed
stormwater management structures.
One area that has not been satisfactorily addressed at
this time is the maintenance of residential stormwater
management structures. Commercial stormwater man-
agement structure maintenance is not expected to pre-
sent a significant problem, because one entity is
generally responsible for overall site maintenance; resi-
dential stormwater management structure maintenance,
however, is not so easily assured. At this time, residen-
tial maintenance is generally the responsibility of a com-
munity association, but eventually that responsibility
must become a public responsibility if maintenance is to
be assured. If that shift of responsibility is to occur, a
dedicated funding source, such as a stormwater utility,
will have to be implemented.
The issue of land use and its relationship to water
quantity and water quality needs to evolve if resource
protection is to be accomplished. Significant effort will
be expended in educating local government officials on
the importance of wetlands, open space, greenways,
cluster development, and other options to conventional
"cookie cutter" zoning. The Designated Watershed ap-
proach will provide specific details on the benefits of
alternative land-use approaches and their impacts on
water quality and aquatic resources.
An effective stormwater management program must be
multifaceted in its approach and implementation. It must
cross conventional lines that are based on an erroneous
assumption that total resource protection can be pro-
vided through the implementation of structural controls
that are considered only after entire site utilization has
been maximized. Land-use limitations, dedicated open
space, vegetated buffer areas, and reduced impervious
areas are all components of an overall resource protec-
tion strategy. The implementation of a structural control
strategy alone will only reduce the rate of resource decline.
That type of program needs to be implemented as a first
step, but programs should recognize the need for contin-
ued evolution for true resource protection to occur.
31
-------
Section 6217 Coastal Nonpoint Pollution Control Program:
Program Development and Approval Guidance
J.W. Peyton Robertson, Jr.
National Oceanic and Atmospheric Administration,
Office of Ocean and Coastal Resource Management, Washington, DC
Abstract
In recognition of the fact that over half of the nation's
population lives in coastal areas and that nonpoint
source pollution remains a significant limiting factor in
attaining coastal water quality goals, Congress enacted
Section 6217 of the Coastal Zone Act Reauthorization
Amendments of 1990 (CZARA). Section 6217 estab-
lishes a requirement that states with federally approved
coastal zone management programs develop and im-
plement coastal nonpoint pollution control programs to
address nonpoint sources affecting coastal waters.
These coastal nonpoint programs are to be imple-
mented through changes to state nonpoint source pol-
lution programs approved by the U.S. Environmental
Protection Agency (EPA) under Section 319 of the Clean
Water Act and through changes to state coastal zone
management programs approved by the National Oce-
anic and Atmospheric Administration (NOAA) under
Section 306 of the Coastal Zone Management Act. The
central purpose of Section 6217 is to strengthen the
links between federal and state coastal zone and water
quality management programs and thereby enhance
state and local efforts to manage land uses that affect
coastal water quality. States are to achieve this by im-
plementing 1) management measures in conformity with
guidance published by EPA under Section 6217(g) of
CZARA, referred to as the (g) guidance or the manage-
ment measures guidance, and 2) additional manage-
ment measures developed by states where necessary
to achieve and maintain water quality standards.
In addition to the (g) guidance, NOAA and EPA have
jointly produced program development and approval
guidance that outlines the requirements for state coastal
nonpoint programs. The program guidance outlines the
process by which states will develop their programs and
submit them for approval. It also includes the criteria by
which EPA and NOAA will evaluate state coastal non-
point programs.
This paper provides an overview of the program develop-
ment and approval guidance by briefly describing the ele-
ments of the program development process and the
necessary components for an approvable state program.
Included in this description are coastal zone boundary
modification recommendations; identification of nonpoint
sources to be addressed; implementation of management
measures; additional management measures/critical
areas; enforceable policies and mechanisms; program
coordination, public participation, and technical assis-
tance; and the program approval process.
Overview
As part of the Coastal Zone Act Reauthorization Amend-
ments of 1990 (CZARA), Congress enacted a new Section
6217, entitled "Protecting Coastal Waters." This new sec-
tion requires states with federally approved coastal zone
management programs to develop and implement coastal
nonpoint pollution control programs (referred to here as
coastal nonpoint programs).1 These coastal nonpoint pro-
grams are to build and expand upon existing efforts to
control nonpoint pollution by state coastal zone manage-
ment and nonpoint source control agencies.
Section 6217(g) of the statute requires the U.S. Envi-
ronmental Protection Agency (EPA), in consultation
with the National Oceanic and Atmospheric Administra-
tion (NOAA), the U.S. Fish and Wildlife Service, and other
federal agencies, to publish and periodically update
"guidance for specifying management measures for
sources of nonpoint pollution in coastal waters." This
1The term "state" refers to states, territories, and commonwealths
having coastal management programs approved under Section 306
of the Coastal Zone Management Act. There are currently 29 such
programs.
32
-------
technical guidance, or (g) guidance, was published on
January 19, 1993. A companion guidance document,
entitled Coastal Nonpoint Pollution Control Program:
Program Development and Approval Guidance, was
also released on the same date. Though the program
guidance was not required by the statute, NOAA and
EPA developed the guidance in an effort to identify
clearly the necessary elements for an approvable state
coastal nonpoint program.
The statute sets out a two-tiered process for implement-
ing management measures. First, states are to imple-
ment technology-based management measures
throughout the Section 6217 management area. Sec-
ond, states must implement additional management
measures where water quality standards are not at-
tained or maintained. The states are to determine these
additional measures. The program guidance further ex-
plains the justification necessary to exclude any non-
point source category or subcategory from the first tier
of a state coastal nonpoint program and sets out the
components each state program should include. The
program guidance provides for a threshold review proc-
ess that allows states to work with NOAA and EPA to
evaluate their existing nonpoint programs and identify
gaps that need to be filled. Finally, the program guidance
establishes a process for submitting programs to NOAA
and EPA for approval and a schedule for program de-
velopment, approval, and implementation.
The focus of this paper is the "nuts and bolts" of each
state coastal nonpoint program. Each program will vary
due to unique differences in both state physiographic
features and government structure. Even so, the basic
components of a state coastal nonpoint program need
to include those elements identified in the statute and
discussed in the program guidance.
Statutory Requirements
Section 6217 requires that several elements be included
in each state coastal nonpoint program in order to re-
ceive NOAA and EPA approval. These basic statutory
requirements, excerpted from the program guidance,
appear below. State programs must:
Be closely coordinated with existing state and local
water quality plans and programs developed pursuant
to Sections 208, 303, 319 and 320 of the Clean Water
Act, and with state coastal zone management programs.
Provide for the implementation, at a minimum, of
management measures in conformity with the guid-
ance published under Section 6217(g) to protect
coastal waters generally.
Provide for the implementation and continuing revi-
sion from time to time of additional management
measures that are necessary to attain and maintain
applicable water quality standards and protect des-
ignated uses with respect to:
- Land uses that, individually or cumulatively, may
cause or contribute significantly to a degradation of
1) coastal waters not presently attaining or maintain-
ing applicable water quality standards or protecting
designated uses or 2) coastal waters that are threat-
ened by reasonably foreseeable increases in pollu-
tion loadings from new or expanding sources.
- Critical coastal areas adjacent to coastal waters
that are failing to attain or maintain water quality
standards or that are threatened by reasonably
foreseeable increases in pollutant loadings.
Provide for technical and other assistance to local
governments and the public to implement additional
management measures.
Provide opportunities for public participation in all as-
pects of the program.
Establish mechanisms to improve coordination be-
tween state agencies and between state and local
officials responsible for land-use programs and per-
mitting, water quality permitting and enforcement,
habitat protection, and public health and safety.
Propose to modify state coastal zone boundaries as
the state determines is necessary to implement
NOAA recommendations under Section 6217(e),
which are based on findings that modifications to the
inland boundary of a state coastal zone are neces-
sary to more effectively manage land and water uses
to protect coastal waters.
Program Development
The Section 6217 Management Area
The statute requires that NOAA conduct a review of
each state's existing coastal zone boundary to deter-
mine whether or not the area encompassed by the
boundary includes the land and water uses that have
"significant" impacts on the state's coastal waters. The
impact of land and water uses on coastal waters is
considered both "individually and cumulatively." In
cases where NOAA finds that modifications to the inland
boundary of a state's existing coastal zone are neces-
sary to more effectively manage land and water uses,
NOAA is required to recommend a modification to the
existing coastal zone. Although expressed in terms of a
recommendation that a state modify its coastal zone
boundary, NOAA's recommendation also defines what
NOAA and EPA believe should be the geographic scope
of that state's coastal nonpoint program, i.e., the "6217
management area."
NOAA conducted a review of each state's coastal zone
boundary, using existing national data to evaluate land
33
-------
and water uses within the state. The national data in-
cluded information on such parameters as population,
land area, harvested crop land, and soil loss from crop
land. Information was compiled for each state and sum-
marized in a draft document entitled National Summary:
State Characterization Reports.
In evaluating indicators of nonpoint source pollution,
NOAA analyzed data for areas within the state's existing
coastal zone and for areas within and outside of coastal
watersheds. NOAA used the smallest U.S. Geological
Survey mapping unit as a definition of the coastal wa-
tershed. In cases where indicators suggested that non-
point pollution beyond the coastal watershed might have
a significant impact on coastal waters, NOAA assessed
the need to further extend the boundary to encompass
these land and water uses. The area finally recom-
mended by NOAA for inclusion (both the land area
encompassed by the existing coastal zone boundary
and any area landward of the existing boundary) consti-
tutes the 6217 management area.
NOAA recently provided recommendations to states for
modifying their existing coastal zone boundaries. These
boundary recommendations generally conform with the
state coastal watershed boundaries, except in cases
where indicators of nonpoint pollution beyond the
coastal watershed appear significant. In such cases,
NOAA recommends that an additional area landward of
the coastal watershed be included in the 6217 manage-
ment area. In addition to the boundary recommenda-
tions, NOAA issued a set of draft criteria that states may
use in developing their response to the boundary modi-
fication recommendation. The final boundary determina-
tion will be accomplished through the state response to
the NOAA recommendation and a public review and
comment process at the state level. States have the
option of either extending their existing coastal zone
boundary inland or exercising other state authorities
within the 6217 management area.
Identification of Nonpoint Sources To Be
Addressed
The basic premise of Section 6217 is that technology-
based controls should be implemented for all nonpoint
sources that, either individually or cumulatively, have
significant impacts on coastal waters. There need not
be a demonstration that an individual source has an
impact on water quality. In this sense, Section 6217 is
akin to the technology-based approach of the point source
program under the Clean Water Act. For program ap-
proval, states are to implement management measures
throughout the 6217 management area for all nonpoint
source categories (e.g., agriculture) and subcategories
(e.g., confined animal facilities) identified in the manage-
ment measures guidance. States also may include man-
agement measures for other sources (e.g., mining) not
identified in the guidance if the state determines such
measures are necessary to protect coastal waters generally.
The program guidance provides for exclusions of non-
point source categories and subcategories under cer-
tain circumstances. If the state can demonstrate that the
source is neither present nor anticipated in the 6217
management area, the source may be excluded. States
also may exclude sources that do not, individually or
cumulatively, present significant adverse effects to living
coastal resources or human health. It should be noted
that the burden of proof is on the state to demonstrate
that the application of the management measures to
the remaining sources will protect coastal waters gener-
ally. In other words, if a state wishes to exclude a
particular nonpoint source category from management
measures implementation, the state must demonstrate
that the nonpoint category does not (and is not reasonably
expected to) present significant adverse effects to living
coastal resources or human health.
For either type of exclusion, the state must provide docu-
mentation of the rationale and data used to justify the
exclusion. The program guidance includes certain factors
that may be considered in exclusions. They are as follows:
Pollutant loadings or estimates of loadings from the
sources.
Intensity of land use.
Ecological and human health risk associated with the
source.
NOAA and EPA will review the information provided by
the state to determine if the category or subcategory
may be excluded from the coastal nonpoint program.
Implementation of the (g) Management
Measures
State programs need to provide detailed information on
how each of the management measures will be imple-
mented. The program guidance includes a description
of the information to be included in the coastal nonpoint
program for each nonpoint category and subcategory.
This information includes the scope, structure, and cov-
erage of the state program; the designated lead agency
and supporting agencies that will implement the pro-
gram; a program implementation schedule with mile-
stones; enforceable policies and mechanisms to ensure
management measure implementation; interagency co-
ordination mechanisms; a process to identify practices
to implement the management measures; operation,
maintenance, and inspection procedures to ensure con-
tinuing performance of the measures; and monitoring
activities to evaluate the effectiveness of the measures.
States may already have programs in place that can be
incorporated into the coastal nonpoint program. States
need to provide information on how these existing pro-
34
-------
grams can be used to implement the management
measures and identify where necessary changes will be
made. For example, a state may have a program that
requires local ordinances for erosion and sediment con-
trol. Because the program guidance requires "enforce-
able policies and mechanisms" at the state level, the
state would have to show some means of ensuring local
implementation of erosion and sediment control. This
could be in the form of backup state enforcement or
some other state oversight of local programs.
Where states do not have existing programs to address
a given nonpoint category or subcategory, they will have
to develop new authorities and programs to ensure
implementation of the management measures. This
may include developing new state authority. Both exist-
ing and new programs need to be incorporated into the
coastal nonpoint program.
Additional Management Measures/Critical
Areas
The program guidance requires states to implement
additional management measures undertwo conditions:
Where coastal water quality remains impaired even
after implementation of the (g) measures.
In areas whose function is critical to water quality.
States must first identify waters that are threatened or
impaired as a result of nonpoint pollution impacts. Land
adjacent to these waters plays a particularly important
role in attaining or maintaining water quality. There may
be situations where new and expanding land uses could
result in further impacts to threatened or impaired wa-
ters from nonpoint sources, beyond those controlled by
the (g) measures. The purpose of additional manage-
ment measures in this case is pollution prevention to
avoid water quality problems that might otherwise develop.
Additional management measures also are required for
coastal waters that are not attaining or maintaining ap-
plicable state water quality standards or protecting des-
ignated uses. There are two instances where states will
need to implement additional management measures
due to water quality impairments. First, if a state has
identified waters that are failing to meet water quality
standards and determines that existing pollution preven-
tion activities and/orthe implementation of the (g) meas-
ures will not be adequate to achieve water quality
standards, the state will have to implement additional
measures for those waters at the time of program ap-
proval. The second is following implementation of the
(g) measures and monitoring to evaluate effectiveness
of the (g) measures. If a state determines that water
quality impairments (as a result of nonpoint sources)
exist even after implementation of the (g) measures, the
state will have to implement additional management
measures.
Enforceable Policies and Mechanisms
Besides the provisions for state coastal nonpoint pro-
grams found in Section 6217, CZARA also amended
Section 306 of the Coastal Zone Management Act
(CZMA) to require that (before approving a coastal zone
management program) NOAA finds "... the manage-
ment program contains enforceable policies and
mechanisms to implement the applicable requirements
of the coastal nonpoint pollution control program of the
state required by Section 6217 .. ." (Section 306(d)16).
The CZMA also includes a definition of "enforceable
policy": "[t]he term 'enforceable policy' means state
policies which are legally binding through constitutional
provisions, laws, regulations, land use plans, ordi-
nances, or judicial or administrative decisions, by which
a state exerts control over private and public land and
water uses and natural resources in the coastal zone."
The program guidance outlines a variety of both regu-
latory and nonregulatory approaches that a state may
design to meet the requirement for enforceable poli-
cies and mechanisms. Examples of regulatory ap-
proaches include permit programs, local zoning
requirements, and state laws. Nonregulatory ap-
proaches could include economic incentives (such as
cost-share programs) or disincentives (such as taxes
or user fees). Nonregulatory approaches must be
backed by enforceable state authority to ensure man-
agement measure implementation.
Several existing state programs to control nonpoint
sources are backed by state laws. In other cases, state
requirements are delegated to local authorities for im-
plementation or rely on state funds, which provide cost-
share monies for implementing practices. For a state
coastal nonpoint program to be approvable, the state
needs to demonstrate that these programs are ulti-
mately subject to state enforcement authority. An exam-
ple of how this might work for a cost-share program that
is currently voluntary is for the state to back up the
voluntary program with a "bad actor" provision in state
law. In cases where participation in the voluntary pro-
gram does not result in implementation of the manage-
ment measures, the state would have the ability to
penalize the "bad actors" or those who failed to take
advantage of the voluntary opportunity.
Traditional regulatory approaches could offer more di-
rect state oversight of management measures imple-
mentation. A state could issue general permits for
specific source categories that include certain criteria
that must be met by all those who meet the category
definition. Conditions on the general permit would allow
tailoring of requirements for site-specific circumstances.
Issuance of individual permits (such as those issued by
many states for septic systems) could also be used for
a specific entity.
35
-------
Program Coordination, Public Participation,
and Technical Assistance
The program guidance requires several other program
elements, including provision for administrative coordina-
tion, public participation, and technical assistance. These
elements are critical to successful implementation of
coastal nonpoint programs because they provide neces-
sary linkages between state, regional, and local govern-
ments; between government agencies and the public; and
between government agencies and affected user groups.
Such linkages ensure the involvement of a variety of
players and, if well developed, build strong support for
programs from the grass-roots level to the state capitol.
Administrative coordination is inherent in the involve-
ment of state coastal zone management agencies and
state water quality agencies as equal partners in the
development of coastal nonpoint programs. These ties
need to be further enhanced through the involvement of
other state agencies (such as state forestry, state agri-
culture, and state health departments) and with local
governments who will be instrumental in implementing
programs at the ground level. Such relationships can be
further defined and solidified through memoranda of
agreement, joint permitting processes, cross training of
staff, and interagency committees.
Public participation is an integral part of the coastal
nonpoint program because public support is necessary
to ensure effective program development and imple-
mentation. The program guidance requires that states
must provide opportunities for public participation in all
aspects of the coastal nonpoint program. Specifically,
each state needs to demonstrate that its program has
undergone public review and comment before submittal
to NOAA and EPA for approval.
Technical assistance is particularly important in provid-
ing regional and local governments with needed direc-
tion on how to implement the provisions of state coastal
nonpoint programs. The statute outlines a variety of
technical assistance areas, including "assistance in de-
veloping ordinances and regulations, technical guid-
ance, and modeling to predict and assess the
effectiveness of such measures, training, financial in-
centives, demonstration projects, and other innovations
to protect coastal water quality and designated uses."
Technical assistance also will be necessary for affected
user groups and the public. The program guidance also
includes assurances that NOAA and EPA will continue
to provide technical assistance to states as they develop
and implement their programs.
Program Submission and Approval
States have 30 months from the publication of the final (g)
guidance to develop their coastal nonpoint programs.
The final (g) guidance document was published on
January 19, 1993, giving states until July 19, 1995, to
submit their programs (see timeline below). During this
period, states have opportunities to meet with NOAA
and EPA and discuss their progress on program devel-
opment. The program guidance establishes a threshold
review process whereby NOAA and EPA conduct an
initial review of a state's program to address key issues
and decision points. Threshold review is voluntary but
provides an opportunity for states to identify gaps in their
programs early in the process, giving a better idea of
what to expect when the program is finally submitted for
approval. It also helps focus limited resources where
they can be used in the most efficient and effective
manner.
In addition to threshold review, the program guidance
sets out a conditional approval provision for state pro-
grams that are submitted without all of the necessary
elements for final approval. NOAA and EPA recognize
(under limited circumstances) that a state may submit a
program for which all necessary enforceable policies
and mechanisms are in place but that the state may
need additional time to develop state, regional, or local
authorities to implement the state requirements. Under
such circumstances, NOAA and EPA may grant condi-
tional approval of a state program for a period of 1 year.
Final approval of the program would depend on the
state's ability to demonstrate that all necessary enforce-
able policies and mechanisms are in place. A conditional
approval will not affect the date by which states must
achieve full implementation of the (g) measures. Full
implementation still must proceed and be completed
within 3 years of the first federal approval action,
whether that approval is conditional or not.
Summary
Table 1 presents a timeline for coastal nonpoint program
development, approval, and implementation.
Table 1. Coastal Nonpoint Program Development, Approval,
and Implementation
Date
Process
January 1993 Final (g) measures and program approval
guidance issued
January 1993 Coastal nonpoint program development:
threshold review (optional), formal/informal
July 1995 States submit final Section 6217 coastal nonpoint
programs
January 1996 EPA/NOAA complete review of state programs
(program approval)
January 1996 State begins implementation of (g) measures
January 1999 Full implementation of (g) measures
January 2001 Completion of 2-year monitoring period
January 2004 Full implementation of additional management
measures
36
-------
Compliance With the
1991 South Carolina Stormwater Management and Sediment Reduction Act
K. Flint Holbrook and William E. Spearman, III
South Carolina Land Resources Conservation Commission,
Columbia, South Carolina
Abstract
The 1991 Stormwater Management and Sediment Re-
duction Act is comprehensive legislation intended to
address the management of Stormwater runoff from a
watershed perspective. The Act establishes a statewide
program making requirements consistent across politi-
cal boundaries. It gives local governments several op-
tions to address specific problems though the creation
of Stormwater utilities or designated watersheds. Con-
siderations are made for citizen complaints and input
into program development and operation.
Introduction
Stormwater management and sediment reduction is an
integral part of nonpoint source pollution control.
Amendments to the federal Clean Water Act in recent
years have emphasized Stormwater management and
sediment control as basic parts of National Pollutant
Discharge Elimination System (NPDES) permitting.
Several states recognized erosion and sediment control
as a major problem in the early 1970s. States had used
different approaches, ranging from comprehensive
statewide regulatory legislation (e.g., North Carolina) to
the voluntary approach of enabling legislation to allow
local governments to enact ordinances to regulate ero-
sion and sediment control on the local level. Tradition-
ally, Stormwater management was not part of enabling
legislation or statewide programs.
In the early to mid 1980s, some states began to incor-
porate Stormwater management into these programs.
The Clean Water Act amendments strengthened the
case for attaching the Stormwater management issue
to the erosion and sediment control programs. To date,
several states have implemented combined programs.
South Carolina passed enabling legislation in 1971 to
allow local governments to pass ordinances to regu-
late erosion and sediment control. This approach met
with very little success; only 22 local ordinances were
passed in 22 years. In 1983, the Erosion and Sediment
Reduction Act was passed to regulate state-owned
lands. This act was to set an example for local pro-
grams. The act exempted the South Carolina Depart-
ment of Highways and Public Transportation by
requiring them to establish a program of their own.
In 1991, the South Carolina General Assembly recog-
nized the increasing problems from years of misman-
agement of Stormwater runoff. On May 27, 1991,
Governor Carroll Campbell signed the 1991 Stormwater
Management and Sediment Reduction Act. Pursuant
Regulation 72-300 became effective June 26, 1992.
Requirements of the Act
The 1991 act sets minimum standards for program
development for control of sediment and water quan-
tity statewide. The act allows local governments to es-
tablish Stormwater utilities and designated watersheds.
It also mandates a statewide regulatory program for
Stormwater management and sediment reduction.
The intent is to delegate program components to local
governments or conservation districts. There are four
components to the program: plan review, inspection,
enforcement, and education and training. Criteria for
delegation of each component is set forth in the regu-
lations. Any or all of the components may be dele-
gated. The delegation is valid for 3 years. The South
Carolina Land Resources Conservation Commission
provides oversight of the local program to ensure its
proper operation. In the event that delegation is not
requested, the commission operates the program
within that jurisdiction or until a local entity requests
delegation. The local government has first right of
refusal to request delegation. If the local government
chooses not to request delegation, the local conser-
vation district may request the delegation.
37
-------
The commission retains jurisdiction of certain activi-
ties to the exclusion of all others. The commission will
permit activities by persons with eminent domain, the
federal government, and all local governments.
Requirements for Individual Site
Development
Minimum standards are established for individual site
development. There are important dates that should be
recognized when determining specific requirements for
site development. The effective date of the act was May
27, 1992. The effective date of Regulation 72-300 was
June 26, 1992. All sites with land-disturbing activities
that affect 5 acres or more and that began on or after
October 1, 1992, are required to permit through this
program regardless of local program status. Beginning
July 1, 1993, any land-disturbing activity starting on or
after that date in the fifteen most populated counties as
listed in Section 72-303 must permit through the pro-
gram. Additional counties are phased in for 1994 and
1995. Size limits have been set for land disturbances
from 0 to 2 acres as a reporting requirement following
guidance in 72-307(H). Permits for land disturbances of
2 to 5 acres are required under the guidelines of 72-
307(1). Land disturbances greater than 5 acres must
follow Section 72-307.
Site-Specific Requirements
The site-specific requirements have some general simi-
larities to the federal Clean Water Act requirements for
construction. One of the major differences addresses
the quantity of water released. These regulations are
broken into different parts according to the stage of the
land-disturbing activity.
Postconstruction requirements include both quantitative
and qualitative controls. For quantity control, post-
development release rates for the 2-year/24-hour and
10-year/24-hour design storms are controlled to the
2-year/24-hour and 10-year/24-hour predeveloped re-
lease rates. Quality controls for the first flush are imple-
mented where ponds are the proposed method of
control. A wet pond requires capture of the first half inch
of runoff volume from the impervious areas site. This
flow can be mixed with the clean permanent pool
volume and discharged over 24 hours. A dry pond
requires that the first 1 inch of runoff volume from
impervious areas is captured and released over 24
hours. The first flush must be separated from the
additional flow into the dry basin.
Where ponds are not the proposed method of control,
nonstructural controls are required. Riparian vegetation
strips, grass waterways, sand filters, and other meas-
ures to meet postconstruction water quality concerns
are acceptable alternatives.
During construction, the requirement is qualitative, deal-
ing exclusively with control of offsite discharge of sedi-
ment. A performance standard of 80 percent removal
(total suspended solids in versus total suspended solids
out) or an efficiency of an effluent standard of 0.5 mL/L
peak effluent settable solid concentration, whichever is
most lenient, must be achieved. Sites with 10 disturbed
acres draining to a single point are required to have a
sediment basin. Otherwise, a combination of structural
and nonstructural practices may be used. There is no
sampling requirement to prove compliance with these
standards. Plans are developed using modeling tech-
niques to predict performance of this standard for the
10-year/24-hour design storm.
A construction sequence, one of the most important require-
ments, is required as part of the overall plan. The sequence,
which is developed by the project designer, contains all site
activities, from installing tree protection to final landscaping
and paving. Close compliance with the construction is re-
quired. The contractor must follow this sequence, with modi-
fications allowed for unforeseen circumstances; however,
the sequence is not normally modified.
Inspection and Enforcement
Site inspection is of primary importance to operations of
this program. Without inspection, the program is
doomed to failure. Weekly unannounced site inspec-
tions are made on each site. Further, a set of approved
plans is required to be held on site.
Enforcement provisions in the act provide for fines of up
to $1,000 per day. Also, stop-work orders may be is-
sued. These enforcement provisions are used when
violations occur and cooperation is not received to cor-
rect the problem. There are no criminal penalties asso-
ciated with violations of this act.
Enforcement actions require that the owner be notified by
certified mail of any violation. Land-disturbing activities
commencing without a permit are subject to an immediate
stop-work order. Violations are cited in the inspection re-
port, with a copy given to the designated day-to-day con-
tact and a copy mailed to the owner. If corrective action is
not taken within the specified time frame, a certified letter
is mailed to the owner. This letter outlines the corrective
action required and the penalties to be assessed.
Citizen Complaint Process
A citizen may file a complaint concerning any portion of
program operation or site-specific regulation. The com-
plaint is filed with the implementing agency for action. If
satisfaction is not achieved, a hearing may be re-
quested. This hearing must follow procedures listed in
the South Carolina Administrative Procedures Act. If
satisfaction is not achieved in this hearing, the complaint
may be appealed in the court system.
38
-------
Florida's Growth Management Program
Eric H. Livingston
Florida Department of Environmental Regulation, Tallahassee, Florida
Abstract
Between 1970 and 1990, Florida's population nearly
doubled, from 6,791,418 to 12,937,926. Recognizing
that this rapid growthup to 900 people per daycould
overwhelm the state's social, economic, and environ-
mental resources, the Florida legislature twice passed
growth management acts. This paper reviews the his-
tory of growth management in Florida, with emphasis on
the differences between the 1975 and 1986 legislation.
The state's current growth management program and
process is described, focusing on the institutional frame-
work and the relationship to the state's water quality
management program. The role of various state and
regional resource management agencies in the review
and approval of local government comprehensive plans
and the implementing land development regulations is
discussed, including specific areas of Florida's growth
management program that are essential to the manage-
ment of water resources. The paper also presents ex-
amples of goals within the State Comprehensive Plan
that can form the foundation forwatershed management
and the maintenance and restoration of water re-
sources. Lessons learned in the implementation of Flor-
ida's growth management program are reviewed, with
recommendations made to improve the program's envi-
ronmental effectiveness.
Introduction
Florida's citizens and political leaders accepted the no-
tion that the strong and sustained growth that Florida
enjoyed after World War II was an unmixed blessing that
would ensure economic health with no negative effects.
It was assumed that growth not only paid for itself but
also produced surplus revenues for state and local gov-
ernments. Florida's public policy toward growth during
the 1950s and 1960s could best be described as "Build
now, worry later."
During this period, Florida grew at a phenomenal rate
with the population rising from 2,771,305 in 1950 to
6,791,418 in 1970 and to 12,937,926 in 1990. Today,
Florida is the fourth most populous state and is still
growing rapidly, although not at the rate of 900 people
per day (300,000 per year) that occurred throughout the
1970s and 1980s.
The negative impacts of unplanned growth were seen
as early as the 1930s, when southeast Florida's coastal
water supply was threatened by saltwater intrusion into
the fragile freshwater aquifer that supplied most of the
potable water for the rapidly expanding population. By
the 1970s, it was becoming all too clear that unplanned
land use and development decisions were altering the
state in a manner that, if left unchecked, could lead to
profound, irretrievable loss of the very natural beauty
that brought residents and tourists to Florida. Extensive
destruction of wetlands, bulldozing of beach and dune
systems, continued saltwater intrusion into freshwater
aquifers, and the extensive pollution of the state's rivers,
lakes, and estuaries were only some of the negative
impacts of this rapid growth.
What Is Growth Management?
Florida is one of eight states to have implemented a growth
management program (1). Understanding Florida's growth
management system requires a clear understanding of
the distinctions between growth management, comprehen-
sive planning, and land/environmental regulations:
Growth management looks at broad issues and at
the interrelationship of systems: natural systems, in-
frastructure, land use, and people. It attempts to as-
sess how well we have provided for the needs of our
citizens in the past and on how to determine and
provide for the needs of new citizens. Growth man-
agement encompasses comprehensive planning,
natural resource management, public facilities plan-
ning, housing, recreation, economic development,
and intergovernmental coordination.
Comprehensive planning is a governmental process
for inventorying resources, establishing priorities, es-
tablishing a vision of where a community wants to
39
-------
go, and determining how to get there. It is a system-
atic way of looking at the different components of a
community, county, region, and state.
Regulations are the specific controls applied to dif-
ferent types of development activities to regulate and
minimize their negative impacts. Typically, regulations
are administered by all levels of government, federal,
state, and local. At the local level, land development
regulations are the ordinances that implement the
local comprehensive plan.
Comprehensive Planning Versus
Regulation
Comprehensive planning allows a community to make
decisions about how and where future growth will occur.
Comprehensive planning asks, Is this the right location?
Is this the right time? Is this the right intensity for the
proposed use of the land? Comprehensive planning
seeks to prevent problems (social, economic, environ-
mental) before development occurs.
Permitting, on the other hand, asks only, How can we
do the best job with this development on this particular
site? Permitting is site-specific and seeks only to miti-
gate the impacts of the land-use decision. Limitations
are always inherent in any regulatory program, and
comprehensive planning can help to overcome them.
Principal among these limitations is the fact that permit-
ting is piecemeal and does not consider cumulative
effects. Therefore, regulation and permitting cannot sub-
stitute for planning. Both are needed to manage growth
effectively and to protect quality of life.
Growth Management in Florida, Chapter 1
Florida began serious and comprehensive efforts to
manage its growth as the environmental movement in
the nation and the state gained strength. In 1972, the
Florida legislature enacted the first modern package of
land and water planning, regulation, and acquisition pro-
grams. This package included:
Chapter 373, Florida Statutes (F.S.), establishing the
state's five regional water management districts, re-
quiring the development of a state water plan, and
allowing for the regulation of the water resource.
Chapter 403, F.S., establishing the state's Depart-
ment of Environmental Regulation and its powers
and duties.
Chapter 259, F.S., establishing the Environmentally
Endangered Lands program, which authorized the
state to purchase critical and sensitive lands.
Chapter 380, F.S., creating the Developments of Re-
gional Impact (DRI) and Areas of Critical State Con-
cern (ACSC) programs.
In 1975, at the recommendation of the first Environ-
mental Land Management Study Committee (ELMS I),
the Legislature enacted the state's first growth manage-
ment legislation. Chapter 163, F.S., the Local Govern-
ment Comprehensive Planning Act (LGCPA), required
all cities and counties to prepare a comprehensive plan.
These plans were submitted for review to the state's
land planning agency, the Department of Community
Affairs (DCA), which in turn sent the plans to other state
agencies for review and comment.
Despite the legislature's good intentions, the growth
management legislation passed in the 1970s contained
fatal flaws. First, the LGCPA contained no "teeth." Local
governments were under no statutory requirement to
revise their plans by incorporating the comments and
recommendations that the state agencies involved in the
review of the local comprehensive plans had made.
Furthermore, they were not required to pass land devel-
opment regulations to implement their plans. Most im-
portantly, state and local officials never recognized that
substantial new funding would have to be provided to
make the program work. Funding was essential for the
mandated planning, for supporting the costs of infra-
structure, and for implementing strategies to manage
growth. Finally, the law did not require local govern-
ments to ensure that public facilities and services kept
up with the demands imposed by population growth. As
Florida's population continued to boom in the 1980s, this
failure to connect the costs of growth with land-use
decisions and population increases resulted in billions
of dollars of backlog in public facilities and services,
increased strain on existing facilities, and an ever-in-
creasing deficit in the quality of life for Floridians.
Growth Management in Florida, Chapter 2
In the late 1970s and early 1980s, an extensive ap-
praisal of Florida's growth management system was
undertaken; the appraisal concluded that the existing
system was not working. Shaped by the Final Report of
the Governor's Task Force on Resource Management
(1980) and the second Environmental Land Manage-
ment Study Committee (ELMS II), a totally new blueprint
for managing growth emerged. The ELMS II recom-
mended a comprehensive package of integrated state,
regional, and local comprehensive planning, reforms to
the DRI law, and coastal protection improvements. The
legislature responded by enacting the following growth
management framework:
The State and Regional Planning Act of 1984 (Chap-
ter 186, F.S.) mandated that the Governor's Office
prepare a state comprehensive plan and present it to
the 1985 legislature. It also required the preparation
of regional plans by the state's 11 regional planning
councils and provided $500,000 for plan preparation.
40
-------
The 1985 State Comprehensive Plan (Chapter 187,
F.S.) originally was envisioned to be a leadership
documentthe foundation of the entire planning
processwith strong, measurable, and strategic
goals that would set the course for Florida's growth
over the next 10 years. Each state agency was to
prepare an agency functional plan, based on the
State Comprehensive Plan, upon which its budget
appropriations would be made. Unfortunately, one of
the most important elements of the State Planthe
development and adoption of a capital plan and
budgetwas never prepared.
The Local Government Comprehensive Planning and
Land Development Regulation Act of 1985 (Chapter
163, F.S.) required all local governments to prepare
local comprehensive plans and implement regula-
tions consistent with the goals and policies of the
state and regional plans. Numerous state and re-
gional agencies reviewed the local plans and submit-
ted their objections, recommendations, and com-
ments to the Department of Community Affairs for
transmittal to the local government. This time, the
local plans had to be revised to incorporate the ob-
jections, recommendations, and comments. Further-
more, local governments faced sanctions from the
state that could result in the loss of state funding if
adopted local plans were not consistent with the state
and regional plans.
Florida's revised growth management system is built
around three key requirements: consistency, concur-
rency, and compactness:
The consistency requirement established the "inte-
grated policy framework," whereby the goals and poli-
cies of the State Plan framed a system of vertical
consistency. State agency functional plans and re-
gional planning council regional plans had to be con-
sistent with the goals and policies of the State Plan,
while local plans had to be consistent with the goals
and policies of the state and appropriate regional
plan. Furthermore, the individual elements of each
local plan must be internally consistent, a require-
ment that has the power to make local plans into
coherent, meaningful, balanced documents for guid-
ing the future of a community. Local land develop-
ment regulations (LDRs) must also be consistent with
the local plan's goals and policies. Horizontal consis-
tency at the local level also is required to ensure that
the plans of neighboring local governments are com-
patible. Consistency is the strong cord that holds the
growth management system together.
Concurrency is the most powerful policy requirement
built into the growth management system. It requires
state and local governments to abandon their long-
standing policy of deficit financing growth by imple-
menting a "pay as you grow system." Once local
plans and LDRs are adopted, a local government
may approve a development only if the public facilities
and services (infrastructure) needed to accommodate
the impact of the proposed development can be in place
concurrent with the impacts of the development.
Public facilities and services subject to the concur-
rency requirements are roads, stormwater management,
solid waste, potable water, wastewater, parks and
recreation, and, if applicable, mass transit.
Compact urban development goals and policies are
built into the State Comprehensive Plan and into
regional plans. Policies such as separating rural and
urban land uses, discouraging urban sprawl, encour-
aging urban in-fill development, making maximum use
of existing infrastructure, and encouraging compact
urban development form the basis for this requirement.
Synopsis of the 1985 Growth
Management Process
Content of Local Comprehensive Plans (2)
The plans are prepared in accordance with the minimum
requirements set forth in Rule 9J-5, Florida Administra-
tive Code (FAC), "Minimum Criteria for Review of Local
Government Comprehensive Plans and Determination
of Compliance."
Who Prepares the Plan?
The local government may designate itself as the local
planning agency (LPA) or designate a LPAby ordinance
to prepare the plan and recommend it to the local gov-
ernment for adoption. Procedures assuring maximum
public input and participation must be implemented by
the local government and the LPA.
What Is Included in the Plan?
Plans shall consist of materials, written or graphic,
including maps, as are appropriate for the prescrip-
tion of goals, objectives, principles, guidelines, and
standards for the orderly and balanced future economic,
social, physical, environmental, and fiscal develop-
ment of the area. The plan must contain the nine required
elements and, if the local government population ex-
ceeds 50,000, a Mass Transit Element and an Aviation
and Port Element.
What Are the Required Plan Elements?
These elements must be internally consistent and eco-
nomically feasible. Each element consists of data analy-
sis along with the setting of goals and policies to achieve
desired results. The elements include:
1. Capital Improvements Element, which must con-
sider the projected need and location of public facili-
ties over the next 5 years:
41
-------
a) This element must contain a component with
principles for construction of new public facilities
or for increasing capacity of existing facilities.
b) A component must also be provided outlining
principles for correcting existing public facility
deficiencies.
c) The element must set forth standards to ensure
availability and adequacy of public facilities.
d) It must establish the acceptable levels of service
for all facilities.
2. Future Land Use Element, which must include a future
land use map. The map and policies of this element
must be based on studies, data, and surveys that
determine the projected population changes, show
the distribution and amount of land for each land
use type (e.g., residential, commercial, industrial)
needed to accommodate the growth, show the
availability of public services, address renewal of
blighted areas, and eliminate nonconforming uses.
3. Traffic Circulation Element, showing existing and
proposed transportation routes needed to achieve
the desired level of service based on future popula-
tion and land uses.
4. Public Services/Facilities Element, which estab-
lishes the level of service for wastewater, solid
waste, stormwater, and potable water. An analysis
must be undertaken to determine whether existing
facilities are providing current residents with the
desired level of service, and whether these facilities
can meet the demands for service created by pro-
jected future development; to identify any existing or
future service deficiencies; to determine strategies
and schedules for correcting these deficiencies; and
to insert these needed infrastructure improvements
into the Capital Improvements Element.
5. Conservation Element, to provide principles and
guidelines for the conservation, use, and protec-
tion of natural resources, including air, water, re-
charge areas, wetlands, estuarine marshes, soils,
beaches, floodplains, rivers, bays, lakes, wildlife
and marine habitat, and other natural and environ-
mental resources.
6. Recreation and Open Space Element, which must
establish a level of service for recreational facilities,
set forth how these will be met as the population
grows, and ensure public access to beaches.
7. Housing Element, with standards and principles to
be followed to ensure the provision of housing for
existing residents and provide for future growth. It
must also include provisions for adequate sites of
future housing for low and moderate income per-
sons, for mobile homes, and for group homes.
8. Coastal Management Element, which must be pre-
pared by those jurisdictions having a coastline. This
element is to set forth policies to maintain, restore,
and enhance the overall quality of the coastal zone
environment, including wildlife; to protect human life
against the effects of natural disasters; and to limit
public expenditures that subsidize development in
high-hazard coastal areas.
9. Intergovernmental Coordination Element, to coordi-
nate the plan with those of adjacent local govern-
ments, school boards, special districts, etc.
The Plan Adoption and Review Process
Local plans are submitted to the DCA at a rate of 10 to
15 per month in accordance with the schedule and dates
set out in Rule 9J-5, FAC.
The local government sends the proposed plan to DCA
for review and written comment. DCA in turn sends
copies to other state agencies for review and comment
within 45 days. Within 45 days after receiving comments
from these other agencies, the DCA issues an Objec-
tions, Recommendations, and Comments (ORC) Re-
port, which summarizes the comments received from all
of the reviewing agencies. The local government has 60
days to revise the plan, hold a public hearing, and
formally adopt it.
Upon adopting the revised plan, the local government
sends the adopted plan to DCA. DCA has 45 days to
review and issue a legal Notice of Intent to find the plan
"in compliance" or "not in compliance." The term "in
compliance" means consistent with the State Compre-
hensive Plan, the Regional Plan, and Rule 9J-5, which
sets forth minimum criteria.
If the local plan is found to be not in compliance, the
following process occurs:
A formal Chapter 120, F.S., Administrative Hearing is
held, at which the local government can show by a
preponderance of evidence that the plan is in com-
pliance. A Final Order upholding or overturning DCAs
determination of compliance is sent to the Governor
and Cabinet.
If the plan is not in compliance, the Governor and
Cabinet can either specify remedial actions to bring the
plan into compliance or impose sanctions on the local
government, resulting in the loss of state revenue
sharing funds, loss of state funds for road improve-
ments, and loss of eligibility for some grant programs.
If the local plan is found to be in compliance:
A legal notice of intent is published in a local newspaper.
Within 21 days, any affected party may file a petition
for a formal Chapter 120 hearing to appeal DCAs
compliance decision.
42
-------
After the hearing, a final order is issued that either
upholds or overturns the DCA compliance determina-
tion. If overturned, the Governor and Cabinet again can
either specify remedial actions or impose sanctions.
Plan Adoption and Approval Status
As of August 1993, a total of 186 local comprehen-
sive plans were in compliance, while 30 were not in com-
pliance. Another 212 plans had been brought into
compliance through a negotiated compliance agreement
between the DCA and the local government, and 29
plans that were not in compliance have a pending com-
pliance agreement that has not been signed (3). Of the
259 local comprehensive plans determined to be not in
compliance, the compliance issues that caused the find-
ings to be made are summarized in Table 1 (4).
The Plan Amendment Process
Chapter 163 limits amendments to an adopted compre-
hensive plan to only twice a year. These amendments
must be adopted following the same procedure as when
the plan was first adopted. The plan amendment review
process is similar to the original plan review process,
involving the following steps:
1. The land owner submits a request for plan amend-
ment to the local government. Usually this must
include certain data and information to help the local
government determine the potential impacts of the
proposed amendment.
2. The local government holds a public hearing to
determine whether to adopt the proposed plan
amendment.
3. Proposed plan amendments are submitted to the
DCA for review to ensure consistency with state and
regional plans and with Rule 9J-5. DCA transmits
the amendment to other state agencies for their
review and comment within 30 days. DCA has a
total of 45 days to review the amendments; incorpo-
rate comments, objections and recommendations
Table 1. Compliance Issues
Compliance Issue
Number
Percentage
Natural resource protection
Level of service standard
Land use
Concurrency management system
Affordable housing
Financial feasibility
Coastal management
Intergovernmental coordination
Land development regulation
198
183
163
128
89
84
59
56
21
76
71
63
49
34
32
23
22
8
from other state agencies; and send the ORC Re-
port to the local government.
4. The local government conducts a public hearing
where it can adopt, adopt with modifications, or not
adopt the amendment.
Implementing the Plan: Adopting Land
Development Regulations
A key feature of the 1985 growth management legisla-
tion is the requirement that local governments adopt
LDRs within 1 year after submission of the revised plan
to DCA for formal review. LDRs are defined in Chapter
163, F.S., as "ordinances enacted . . . for the regulation
of any aspect of development." They are an exercise of
the general governmental police power for the protec-
tion of the public health, safety, and welfare. LDRs must
address, at a minimum, the following areas:
Subdivisions.
Implementation of land-use categories included in the
land-use element and map (zoning), along with regu-
lations to ensure the compatibility of adjacent land
uses and to provide for open space.
Protection of potable water wellfields.
Stormwater management (quantity and quality).
Protection of environmentally sensitive land.
Signage.
Public facilities and services to meet or exceed the
established level of service standards.
Onsite vehicular and pedestrian traffic flow and parking.
The LDRs must be adopted by ordinance, and the adop-
tion process must comply with the notice and public
hearing process set forth in Florida law. Finally, the LDRs
must be combined into a single land development code.
Unlike local plans, LDRs do not undergo comprehensive
state review and approval. The DCA may review and
take action on individual LDRs under only two circum-
stances. The first is for "completeness review," in which
the DCA must have reasonable grounds to believe that
a local government has totally failed to adopt any of the
required LDRs. "Reasonable grounds" means that DCA
has received a letter(s) from a party or parties stating facts
that show the local government has failed to adopt one
or more of the required LDRs. DCA can then require a
local government to submit its LDRs for review. DCA
then enters into a period of review and consultation with
the local government to determine whether the local
government has complied with statutory requirements.
If DCA determines that a local government has failed to
adopt one or more required LDRs, it notifies the local
government within 30 days. The local government then
must adopt the LDRs and submit them to DCA. If the local
43
-------
government fails to adopt the LDRs, DCA institutes action
in circuit court to require adoption of the required LDRs.
The second type of state review is to assure that the
LDRs "implement and are consistent with the local com-
prehensive plan." This review looks more closely at the
actual content and substance of the ordinances. This
review can only be initiated by a "substantially affected
person" (citizen), however, and it cannot be initiated by
the DCA. A consistency challenge must occur within 12
months afterthe final adoption of the LDR. The substan-
tially affected person must petition DCA to initiate a
Chapter 120 administrative hearing. If DCA reviews the
information in the petition and determines that the LDRs
are not consistent with the plan, then DCA requests an
administrative hearing. If DCA reviews the information
in the petition and determines that the LDRs are consis-
tent with the plan, then the affected party can request
an administrative hearing. If the Final Order from the
administrative hearing finds the LDR is inconsistent,
then the Governor and Cabinet determine what types of
sanctions will be imposed on the local government.
Comprehensive Plans and the Protection of
Natural Resources
Amain purpose of the comprehensive planning program
is to maintain, restore, and protect Florida's very valu-
able, vulnerable natural resources. The goals and poli-
cies set forth in the State Comprehensive Plan along
with the requirements in Rule 9J-5, which set forth spe-
cific objectives and policies that must be included in
each plan element, provide the basis for the protection
of natural resources.
Within the State Comprehensive Plan, goals and poli-
cies that specifically address minimizing impacts of vari-
ous activities on natural resources and the general
conservation, protection, and proper use and manage-
ment of natural resources are found within the Water
Resources, Coastal/Marine Resources, Natural Sys-
tems and Recreation Lands, Air Quality, Waste Materi-
als, Land Use, Mining, Agriculture, Public Facilities,
Conservation, and Transportation Elements. The follow-
ing are examples of these goals and policies.
For the Water Resources Element, the goal is to "as-
sure the availability of an adequate supply of water. . .
and . . . maintain the functions of natural systems and
the overall present level of surface and ground-water
quality. Florida shall improve and restore the quality of
waters not presently meeting water quality standards."
Policies include:
Protect and use natural water systems in lieu of struc-
tural alternatives, and restore modified systems.
Establish minimum seasonal flows and levels for sur-
face waters to ensure protection of natural resources,
especially marine, estuarine, and aquatic ecosys-
tems.
Discourage the channelization, diversion, or dam-
ming of natural riverine systems.
Encourage the development of a strict floodplain man-
agement program to preserve hydrologically significant
wetlands and other natural floodplain features.
Protect surface and ground-water quality and quantity.
Eliminate the discharge of inadequately treated waste-
water and stormwater runoff into waters of the state.
Coastal/Marine Resources policies include:
Accelerate public acquisition of coastal and beach-
front land to protect coastal and marine resources.
Avoid spending state funds that subsidize develop-
ment in high-hazard coastal areas.
Protect coastal and marine resources and dune sys-
tems from the adverse impacts of development.
For the Natural Systems and Recreational Lands Ele-
ment, the goal is to protect and acquire unique natural
habitats and ecosystems and to restore degraded natu-
ral systems. Policies include:
Protect and restore the ecological functions of wet-
lands systems to ensure their long-term environ-
mental, economic, and recreational value.
Promote restoration of the Everglades system and of
the hydrological and ecological functions of degraded
or disrupted surface waters.
Implement a comprehensive planning, management,
and acquisition program to ensure the integrity of
Florida's river systems.
Agriculture policies include:
Eliminate the discharge of inadequately treated agri-
cultural wastewater and stormwater runoff to surface
waters.
Conserve soil resources to prevent sedimentation of
state waters.
Rule 9J-5 contains many minimum requirements for
goals, objectives, and policies that are directly related to
the conservation, protection, and proper use and man-
agement of natural resources. The following are some
examples.
Public Facilities policies include:
Correct existing facility deficiencies and coordinate
the extension of, or increases in the capacity of, fa-
cilities to meet future needs.
Maximize the use of existing facilities to discourage
urban sprawl.
44
-------
Regulate land use and development to protect the
functions of natural stormwater features and natural
ground-water aquifer recharge areas.
Conservation policies include:
Conserve, appropriately use, and protect the quantity
and quality of water, minerals, soils, native vegetative
communities, fisheries, wildlife, and wildlife habitat.
Protect air quality, native vegetative communities,
and water quality.
Protection and conservation of the natural functions
of soils, fisheries, wildlife habitats, surface waters,
ground waters, and beaches and shorelines.
Growth Management in Florida, Chapter 3
After several years of living with and implementing the
1985 growth management law, numerous issues were
arising that suggested that the program needed fine
tuning. On one side were people who thought that the
program and process were hindering economic devel-
opment, stepping on private property rights, and becom-
ing cumbersome administratively. Others felt that the
program was not adequately protecting social, eco-
nomic, and environmental resources. In 1991, the third
Environmental Land Management Study Committee
(ELMS III) was formed to provide recommendations to
the 1993 legislature on ways to further improve and
refine Florida's growth management laws. The Commit-
tee's report included the following conclusion (5):
Florida's growth management process is not in a state
of disrepair, but it needs some immediate attention.
More importantly, it needs executive leadership to
protect the substantial investment that has been
made so that it will not be lost, or worse, become a
liability. Decisions that are made over the next 12 to
18 months will determine whether our efforts will be
able to deliver the promises made. The tools for
managing future growth and change are in place.
The challenge is whether these tools and our lead-
ership can respond when asked to perform.
The Committee's Final Report and Recommendations
formed the basis fora new planning and growth manage-
ment act which passed by overwhelming margins in both
the house and the senate in the closing days of the 1993
session. Among the provisions of the 180-page law are
some major changes relating to state planning, regional
planning, the DRI process, local planning and concur-
rency, and infrastructure funding as explained below (6).
State Planning
One of the biggest criticisms of Florida's growth manage-
ment system is the lack of strong leadership at the state
level. The State Comprehensive Plan originally was envi-
sioned as a leadership document with strong, measurable,
and strategic goals that would set a course for the state's
growth and guide the development and implementation of
state programs. State agency and program budgeting
decisions, however, never were changed to incorporate
the State Plan's requirements. Furthermore, key compo-
nents of the State Planthe capital plan and budget
never were developed or adopted. These omissions
have resulted in a lack of a cohesive, integrated, com-
prehensive vision of Florida's future as well as a lack of
financial resources to implement the program and to
correct existing infrastructure deficiencies.
The 1993 Growth Management Act strengthens the
state planning process in two ways. First, it requires the
Governor's Office to review and analyze the State Com-
prehensive Plan biannually and submit a written report
recommending revisions or explaining why no revisions
are necessary. Second, the act requires that a new
Growth Management Element be prepared and submit-
ted to the 1994 legislature. The element must be strate-
gic in nature; provide guidance for state, regional, and
local actions necessary to implement the State Plan;
identify metropolitan and urban growth centers; estab-
lish strategies to protect identified areas of state and
regional environmental significance; and provide guide-
lines for determining where urban growth is appropriate
and should be encouraged.
Regional Planning
The 1993 Growth Management Act greatly changes the
role and powers of the regional planning councils. The
regional planning councils are charged with planning
and coordinating intergovernmental solutions to multi-
jurisdictional growth-related problems, with no regula-
tory authority. Regional policy plans will now be required
to address only affordable housing, economic develop-
ment, emergency preparedness, regionally significant
natural resources, and regional transportation, and
these plans will no longer be a basis for determining the
consistency of local plans.
The DRI Process
The act provides for the termination of the DRI process
in large jurisdictions (counties greater than 100,000
population) when they adopt specific intergovernmental
coordination mechanisms. The law also greatly revises
the DRI process in those counties and cities that retain
the process. Fewer projects will be considered DRIs, the
regional planning councils will be allowed to address
only state and regional resources or facilities, and the
review process is expedited for projects that are consis-
tent with the local comprehensive plan.
Local Planning
The act makes several very substantial changes in the
local planning process, especially with respect to the
45
-------
plan amendment review process, sanctions, intergov-
ernmental coordination, and evaluation and appraisal
reports. The plan amendment review process is stream-
lined, with DCA issuing an ORC Report for a proposed
amendment only if a regional planning council, affected
person, or local government requests it or if DCA de-
cides to conduct such a review. All adopted plan amend-
ments will be reviewed by DCA for compliance with state
laws. The law greatly changes and strengthens the
evaluation and review reporting requirements. The DCA
is directed to adopt a rule establishing a phased sched-
ule for the submittal of evaluation and appraisal reports
no later than 6 years after local plan adoption and then
every 5 years thereafter.
Concurrency and Infrastructure Funding
The act codifies DCAs existing concurrency management
rule and policies, thereby providing specific legislative
guidance on this critical component of the planning
process. To avoid conflicts with other state planning goals,
the act authorizes local governments to provide an ex-
ception from transportation concurrency requirements in
areas designated for urban in-fill development, urban
redevelopment areas, existing urban service areas, or
certain downtown revitalization areas. The act author-
izes local governments to adopt a "pay and go" system
for transportation concurrency if the local plan includes
a financially feasible capital improvement plan to up-
grade transportation facilities and establishes an impact
fee or other system requiring the developer to pay its fair
share of needed transportation facilities. Unfortunately,
while ELMS III recommended a 10-cent statewide gas
tax increase to provide infrastructure funding, the legis-
lature only authorized local governments to increase the
local option gas tax by up to 5 cents.
Recommendations
Based on experience with Florida's growth management
programs over the past 15 years, the following recom-
mendations are made to streamline the process and
enhance protection of Florida's natural resources.
The program and its requirements must recognize the
inherently different growth management needs of highly
urbanized areas or rapidly growing areas and separate
them from the planning needs of rural areas, especially
those with very slow growth rates. Flexibility, with con-
sistency, is the key.
Rural local governments, especially in those areas ex-
periencing growth, have the most to gain from compre-
hensive planning. Hopefully, they can avoid the
mistakes that have been made in central and southern
Florida where unplanned growth adversely affected so-
cial, economic, and environmental resources. Rural lo-
cal governments, however, need extensive technical
assistance and funding to develop and implement sound
comprehensive plans.
Probably the greatest hindrance to solving Florida's
existing growth management problems and prevent-
ing future growth from exacerbating them is the imple-
mentation, at both state and local levels, of dedicated
funding sources. At the state level, the Growth Man-
agement Program, the Surface Water Improvement and
Management Program, the State Stormwater Demon-
stration Grant Program, and the Preservation 2000
Land Acquisition Program are underfunded and depend
on annual legislative appropriations. Dedicated funding
sources such as increases in documentary stamp taxes
or the placement of small fees on products such as
concrete, asphalt, fertilizer, pesticides, and water use
or even electric bills could generate sufficient funding
levels to ensure that these programs succeed. At the
local level, impact fees, gasoline taxes, and the estab-
lishment of stormwater utilities (already implemented by
over 50 local governments) are essential if funds suffi-
cient to pay for needed infrastructure improvements are
to be raised.
The state's land planning and water planning frame-
works need to be better integrated. In particular, the
Department of Environmental Regulation and the five
regional water management districts need to be the lead
agencies involved with water management issues.
Greater consistency and integration is needed between
local comprehensive plans and requirements set forth in
State Water Policy, Chapter 17-40, FAC. Currently, local
comprehensive plans only are required to "consider"
State Water Policy rather than to be "consistent with."
References
1. Gale, D.E. 1992. Eight state-sponsored growth management
programs: A comparative analysis. JAPA 58(4):425-439.
2. Gluckman, D., and C. Gluckman. 1987. Citizen's handbook to
the Local Government Comprehensive Planning Act. Handbook
prepared for the Florida Audubon Society, Maitland, FL.
3. Department of Community Affairs. 1993. Compliance determina-
tions for local government comprehensive plans. Tallahassee, FL.
4. Department of Community Affairs. 1993. Analysis of issues re-
sulting in a finding of not in compliance for local government
comprehensive plans. Tallahassee, FL.
5. ELMS III Committee. 1993. Final report and recommendations.
Tallahassee, FL.
6. Pelham, T 1993. The ELMS III legislation: Revising Florida's
Growth Management Act. Florida Administrative Law Section
Newsletter 16(3):11-16.
46
-------
Stormwater and the Clean Water Act:
Municipal Separate Storm Sewers in the Moratorium
Kevin Weiss
Office of Wastewater Enforcement and Compliance, Office of Water,
U.S. Environmental Protection Agency, Washington, DC
Abstract
Urban stormwater and related pollutant sources have
been shown to be major sources of water quality impair-
ment. Section 402(p)(6) of the Clean Water Act requires
the U.S. Environmental Protection Agency to identify
additional stormwater sources to be regulated to protect
water quality under Phase II of the National Pollutant
Discharge Elimination System (NPDES) program. Miti-
gating water quality impairment associated with urban
runoff requires comprehensive efforts with special em-
phasis on comprehensive approaches to stormwater
management for new development. Municipal govern-
ments in urbanized areas appear to be critical institu-
tions for making many of the day-to-day decisions
necessary to address problems associated with storm-
water, including measures to minimize the risks to water
resources associated with stormwater from areas un-
dergoing urbanization. In addition, municipalities have
the police power needed to implement some compo-
nents of stormwater programs and the ability to collect
funds to be used in program implementation. This paper
looks at the use of NPDES permits for discharges from
municipal separate storm sewers systems in urbanized
areas as a tool for defining the federal/state/municipal
relationship for addressing stormwater management.
Environmental Background
Urban stormwater discharges have been shown to be a
major cause of impairment of surface water resources.
The National Water Quality Inventory 1990 Report to
Congress provides a general assessment of surface
water quality based on biennial reports submitted by the
states under Section 305(b) of the Clean Water Act
(CWA). The report indicates that of the rivers, lakes, and
estuaries that the states assessed, roughly 60 to 70
percent are supporting the uses for which they were
designated. Urban lands, however, only account for 2
percent of lands in the United States (1). The report
indicates that urban runoff is a major source of impair-
ment for 53 percent of impaired estuary acres, 36 per-
cent of impaired ocean coastal miles, 29 percent of
impaired lake acres, 6 percent of impaired Great Lake
shoreline, and 9.6 percent of impaired river miles. The
report also indicates that combined sewer overflows,
which are a mixture of urban runoff, sanitary sewage,
and industrial process discharges, are sources of im-
pairment for 4 percent of impaired estuary acres, 3.6
percent of impaired ocean coastal miles, 7.5 percent of
impaired Great Lakes shoreline, and 2.8 percent of im-
paired river miles. Urban runoff affects receiving waters
in or near urban population centers and therefore may
limit the uses and values of the waters closest to the
most people.
Surface water resources are affected by two charac-
teristics of urban runoff: 1) elevated pollution concentra-
tions and loadings and 2) changes in flow patterns that
accompany urbanization. The nature of the receiving
water determines whether increased pollutant loadings
or changes to natural flow patterns or a combination of
both are causes of impairment. For example, slower
moving rivers, streams, lakes, and estuaries can be
more sensitive to increased pollutant loadings than to
changes in flow patterns. Conversely, faster moving
streams, such as those found in hilly or mountainous
areas, can flush pollutants but may be sensitive to dra-
matic changes in flow patterns. A good comparison of
these impacts is provided by Pitt, who compares im-
pacts in Coyote Creek (San Jose, California), a stream
with relatively slow flows, with impacts in Kelsey and
Bear Creeks (Bellevue, Washington), streams with high
flows and good flushing capabilities (2, 3).
Sources of Pollutants in Urban
Stormwater
Pollutants discharged from municipal separate storm
sewer systems originate from a variety of diffuse
47
-------
sources. EPA has identified four major classes of
sources that contribute pollutants to discharges from
municipal separate storm sewer systems (4):
Nonstormwater sources
Residential and commercial sources
Industrial sources
Construction activities
Nonstormwater Sources
Although separate storm sewers are primarily designed
to remove runoff from storm events, materials other than
stormwater find their way into and are ultimately dis-
charged from separate storm sewers. For example, in
Sacramento, California, less than half the water dis-
charged from the stormwater drainage system was di-
rectly attributed to precipitation (5). Nonstormwater
discharges to storm sewers come from a variety of
sources, including:
Illicit connections and cross connections from indus-
trial, commercial, and sanitary sewage sources.
Improper disposal of wastes, wastewaters, and litter.
Spills.
Leaking sanitary sewage systems.
Malfunctioning septic tanks.
Infiltration of ground water contaminated by a variety of
sources including leaking underground storage tanks.
Wash waters, lawn irrigation, and other drainage
sources.
For a more complete description of nonstormwater dis-
charges to storm sewers, see U.S. EPA (6).
Table 1 provides a summary of several studies involving
problems with nonstormwater discharges. These case
studies illustrate the wide range of pollutants that can
enter storm sewers from nonstormwater discharges,
Table 1. Summary of Nonstormwater Discharge Problems
Study Site Comments
Jones Falls Watershed,
Baltimore City and
County, MD
Tulsa, OK
Washtenaw County, Ml
Fort Worth, TX
During the NURP study of the Jones Falls Watershed, 15 illicit connections were discovered in portions of the
watershed. The illicit connections were grouped into four types: direct discharges from residences; leakage
from cracked or broken sewer lines; decades-old overflows from the sanitary sewer; and sanitary sewage
pumping station malfunctions. Elevated levels of pathogens, TSS, ammonia, TKN, total nitrogen, COD, and
TOC were identified.
A physical inspection was conducted of 120,000 ft of storm sewer 48 in. and larger serving a drainage area of
approximately 12 square miles. Thirty-five potential nonstormwater discharges were observed. Twenty-three of
these were observed and/or suspected sanitary sewer connections, four were potable water discharges, and
eight were of unknown origin. In addition, 12,900 ft of sanitary sewer was laid within the storm sewer, where
the storm sewer served as a conduit. Most illicit connections were associated with development that occurred
before 1970. Other documented observations were structural defects (900 ft of pipe showed signs of structural
defects), pipe cross through (176 total), and debris buildup.
Inspection of 1,067 businesses, homes, and other buildings was conducted, with 154 of the buildings (14%)
identified as having illicit connections, including connections in restaurants, dormitories, car washes, and auto
repair facilities. About 60% of the automobile-related businesses inspected had illicit discharges. A majority of
the illicit connections discovered had been approved connections when installed. Pollutants that were detected
included heavy metals, nutrients, TSS, oil and grease, radiator fluids, and solvents.
Twenty-four outfalls in a 10-mile radius were targeted for end-of-pipe observations. The success of the
program was judged by a decline in the number of undesirable features at the target outfalls from an average
of 44 undesirable observations per month in 1986 (522 total) to an average of 21 undesirable observations
per month in 1988. The Fort Worth investigation indicated problems associated with allowing septic tanks,
self-management of liquid waste by industry, and construction of municipal overflow bypasses from the
sanitary sewer to the storm drains. These problems were attributed to the inability of the publicly owned
treatment works to expand as rapidly as urban growth occurred. During a 30-month period, problems detected
included 133 hazardous spills, 125 incidents related to industrial activity, 265 sanitary sewer line breaks, and
21 bypass connections of the sanitary sewer to the storm sewer. Highlighted cases included a 20-gal/min flow
from a cracked sanitary sewer from a bean processing plant to a storm drain and an illicit connection of a
sanitary sewer line from a 12-story office building to a storm sewer. Most industrial pollution enters the storm
sewer system from illegal dumping, storm runoff, accidental spills, and direct discharges. Metals were not
detected in dry-weather discharges but were found in significant levels in receiving water sediment. City
officials state that the high metal concentrations in sediment are consistent with otherwise unexplained serious
reported fish kills.
48
-------
Table 1. Summary of Nonstormwater Discharge Problems
Study Site Comments
Seattle, WA
Upper Mystic Lake, NY
Bellevue, WA
Ann Arbor, Ml
Medford, OR
Toronto, Ontario
Grays Harbor, WA
Seward, NY
Norfolk Naval Station,
VA
Sacramento, CA
Hazardous waste case
studies
The city of Seattle has detected improper disposal and illicit connections from industrial sites by investigating
sediment in storm sewers. One storm drain outfall representing a major source of lead to the Duwamish River
was traced back to a former smelter that crushed batteries to recover lead. Lead concentrations in the
sediment were high enough to allow the city to send it to an operating smelter to be refined. Another storm
drain contained high levels of creosote, pentachlorophenol, copper, arsenic, and PCBs, which (except for the
PCBs) were traced back to a wood treatment facility. Contaminated sediments removed from the storm drain
(30 yd3) contained 145 Ib of contaminants. Sediments removed from storm drains in another industrial area
contained very high levels of PCBs (about 1 Ib PCBs/70 yd3 sediment).
The NURP study for the Mystic Lake Watershed project identified contamination of stormwater runoff and
subsequently surface water contamination of surface waters by sanitary discharges as a major problem in the
watershed that contributed large quantities of phosphorus, certain metals, and bacteria. Interactions at 19
manholes that served both sanitary and storm sewer lines were identified as the major contributor of pollutants.
The NURP report for Bellevue recorded 50 voluntary citizen reports of illegal dumping and other
nonstormwater discharges during a 27-month period. The incidents reported were varied and resulted in at
least two significant fish kills. Of the citizen reports, 25% involved improper disposal of used oil to the storm
sewer. Other reports involved spills; illicit connections of floor drains, septic tank pipes, and a car wash;
chemical dumping; and concrete trucks rinsing out into catchbasins or streams.
Studies in 1963, 1978, and 1979 found that discharges from the Allen Creek storm drain contained significant
quantities of fecal coliform, fecal streptococci, solids, nitrates, and metals. Of the 160 businesses dye-tested,
61 (38%) were found to have improper storm drain connections. Chemical pollutants including detergents, oil,
grease, radiator wastes, and solvents were causing water quality problems. Monitoring of the storm drainage
system during storm events indicated a decrease in the concentration of 32 of 37 chemicals monitored after
the improper connections were removed.
Fecal coliform tests at storm drain outfalls in city parks were used to detect four leaking sewer lines, which
either were located above the storm lines or saturated the ground with effluent, which entered the nearby
storm drains, an agricultural equipment wash rack, and a house with sanitary lines plumbed to the storm
drain. In addition, in one of the oldest sections of town a large storm drain bored in the early 1900s also
contained the sanitary sewer line. Under manholes, the sanitary line was only a trough. Even minor clogs or
breaks resulted in a spillover of effluent in the storm drain below.
Dry weather sampling of discharges from 625 storm drains in the Humber River Watershed. About 10% of the
outfalls were considered significant sources of nutrients, phenols, and/or metals, while 30 of the outfalls had
fecal coliform levels >10,000/100 mL. Investigations identified 93 industrial and sanitary sewage illicit
connections. Problems included residential connections of sanitary sewage to the storm sewers and yard
runoff from a meat packing plant to a storm drain.
Dry weather sampling of 29 outfalls of separate storm drains indicated that discharges from six of the outfalls
had abnormally high pollutant levels with suspected illicit connections. The area under consideration had
originally been served by combined sewers. Earlier efforts to separate the system had been incomplete, with
some residences discharging sanitary sewage to the storm drain.
Sewage from septic tanks with clogged drainfields in clay soils flowed into open storm sewers. The open
storm sewers posed health risks to neighborhood children and lowered property values.
The Norfolk Naval Shipyard was originally built in 1767 and has had numerous additions since that time. It
has an extensive network of underground pipes that includes both separate storm sewers and
sanitary/industrial sewers. In response to a lawsuit, officials at the shipyard conducted dye-testing of sanitary
facilities throughout the shipyard, which led to the identification and elimination of 25 cross connections of
sanitary and industrial waste to the separate storm sewer system.
The city of Sacramento is currently undertaking a project to identify pollutant discharges and illegal
connections into the stormwater drainage systems. Recent studies identified acute toxicity in stormwater, and
revealed that less than half the water discharged from the drainage system was not directly attributable to
precipitation. Mass loading estimates of copper, lead, and zinc discharged by the drainage system were
several times higher than the estimated pollutant loads of these metals from the Sacramento Regional
Treatment Plan secondary effluent.
Cases of onsite waste disposal where pollutants were added to runoff that eventually ended up in drainage
systems and other cases where a generator dumped wastes directly down a drain were common. Of the 36
cases of illegal dumping investigated in a GAO report, 14 cases investigated involved disposal of hazardous
waste directly to, or with drainage to, a storm sewer, flood control structure, or the side of a road. An
additional 10 sites involved disposal to the ground, landfills (other than those receiving hazardous wastes),
and trash bins, which can then result in added pollutants to subsequent stormwater discharges.
49
-------
including pathogens, metals, nutrients, oil and grease,
metals, phenols, and solvents. Removal of these non-
stormwater pollutant sources often provides opportuni-
ties for dramatic improvement in the quality of
discharges from separate storm sewers.
Residential and Commercial Runoff
Residential and commercial activities are the predomi-
nate land uses in most urbanized areas (UAs), typically
occupying between 55 to 85 percent of the total area.
Major pollutants associated with residential and com-
mercial runoff include heavy metals, oxygen demanding
materials, bacteria, nutrients, floatables, organics, pes-
ticides, polynuclear aromatic hydrocarbons (PAHs), and
other toxic organic pollutants.
From 1978 through 1983, the U.S. Environmental Pro-
tection Agency (EPA) provided funding and guidance to
the Nationwide Urban Runoff Program (NURP) to study
the nature of runoff from commercial and residential
areas. The NURP study provides insight into what can
be considered background levels of pollutants in runoff
from residential and commercial land uses. Sites used
in the NURP study were carefully selected so that they
were not affected by pollutant contributions from con-
struction sites, industrial activities, or illicit connections.
Data from several sites had to be eliminated from the
study because of elevated pollutant loads associated
with these sources.
Data collected in NURP indicated that on an annual
loadings basis, suspended solids in discharges from
separate storm sewers draining runoff from residential
and commercial areas are approximately an order of
magnitude or more greater than in effluent from sewage
treatment plants receiving secondary treatment. In ad-
dition, the study indicated that annual loadings of chemi-
cal oxygen demand (COD) is comparable in magnitude
with effluent from sewage treatment plants receiving
secondary treatment.
Table 2 compares annual pollutant loadings for three
metalszinc, lead, and copperfrom urban runoff from
the Metropolitan Washington UA, from a sewage treat-
ment plant that provides advanced treatment and that
serves about 2 million people (the Blue Plains sewage
treatment plant), and from major industrial process
wastewater discharges located in Maryland and Vir-
ginia.
When analyzing annual loadings associated with urban
runoff, it is important to recognize that discharges of urban
runoff are highly intermittent, and that the short-term load-
ings associated with individual events will be high and
may have shockloading effects on receiving water.
Pollutant loadings for urban stormwater are based on
the "Simple Method" developed by the Washington Met-
ropolitan Council of Governments (7). Pollutant concen-
trations used in this model were based on those
published in U.S. EPA (8). The values for lead were
reduced by 75 percent to account for assumed reduc-
tions due to reductions in the use of lead in gasoline.
Pollutant loadings for direct dischargers in the Toxics
Release Inventory are as reported in Cameron (9). The
Toxics Release Inventory contains data on toxic chemi-
cal releases by industrial facilities that use 10,000 Ib or
more of specified toxic chemicals and does not include
all releases from all industrial facilities in a state.
Industrial Runoff
A number of studies indicate that runoff from industrial
land uses has relatively poorer water quality than other
general land uses (8, 10-13). In general, a greater vari-
ety and larger amounts of toxic materials can be used,
produced, stored, or transported in industrial areas. In-
dustrial activities that can provide a significant source of
pollutants to stormwater from industrial sites include
loading and unloading, outdoor storage, outdoor proc-
esses, illicit connections or management practices, and
waste disposal practices. In addition, many heavy indus-
trial areas have a large degree of imperviousness, which
results in high volumes of runoff. Atmospheric deposition
and spills and leaks associated with material transport
can contribute to significant levels of toxic constituents
in runoff to areas surrounding or in close proximity to
heavy industrial activity.
Table 2. Annual Pollutant Loadings (in Pounds) in Stormwater From Selected Pollutant Sources
Urban Stormwater From
Pollutant Metropolitan Washington
Blue Plains POTW3
All MD and VA Direct Industrial Discharges in
1987 Toxic Release Inventory
Zinc
Lead
Copper
Nitrogen
Phosphorus
BODS
480,000
132,600
113,000
30,000,000
1 ,200,000
9,500,000
137,000
5,500
21 ,000
12,000,000
113,000
1 ,400,000
132,000
31 ,300
127,000
Not available
Not available
Not available
aBlue Plains POTW loadings estimates based on EPA Permit Compliance System (PCS) data for 1989.
50
-------
Runoff From Construction Activities
The amount of sediment in stormwater discharges from
construction sites can vary considerably, depending on
whether the discharges are uncontrolled or whether ef-
fective management practices are implemented at the
construction site. Sediment loads from uncontrolled or
inadequately controlled construction sites have been
reported to be on the order of 35 to 45 tons/acre/year.
Sediment loads from uncontrolled construction sites are
typically 10 to 20 times that of agricultural lands, with
sediment loads as high as 100 times that of agricultural
lands and typically 1,000 to 2,000 times that of forest lands.
Over a short period, construction sites can contribute
more sediment to streams than was previously depos-
ited over several decades.
Changes to Flow Patterns: Physical
Impacts
Urbanization can result in dramatic changes to the natu-
ral flow patterns of urban streams and wetlands. In
undeveloped watersheds, most rainfall infiltrates into the
ground and recharges ground-water supplies. Urbaniza-
tion alters the natural vegetation and natural infiltration
characteristics of a watershed, which results in much
higher peak flows and reduced base flows in urban
streams. Increased peak flows can result in stream bank
erosion, streambed scour, flooding, channelization, and
elimination or alteration of habitat (14). Increases in
peak flows can also create the need to modify stream
channels through a variety of engineered structures,
such as retaining walls, rip-rap, and channel dredging.
Increased imperviousness and loss of wetlands and
natural flow channels also decrease the amount of rain-
water available for ground-water recharge. Reduced ground
water levels reduce base flows in streams during dry
weather periods, which impairs the aquatic habitat, impairs
riparian wetlands, and makes receiving streams more
sensitive to other pollutant inputs and sedimentation.
Development Patterns
In the United States, population patterns typically do not
follow the political boundaries of municipalities. Prior to
1950, many large core cities annexed additional fringe
areas as populations of the urban center increased. The
trend of core cities increasing in area through annexa-
tion has largely stopped in most major UAs. In most
states, smaller "suburban" local governments surrounding
the core city are retained or created.1 Thus, today most
urban centers are composed of a large core city sur-
rounded by several smaller "suburban" municipalities.
1The patterns and functions of local governments in suburban fringe
areas vary from state to state. In some states, such as Maryland,
Virginia, Florida, and California, and, to a lesser degree, a number of
southern states and Texas, large urban populations outside of core
Every 10 years, the Bureau of Census defines UAs to
characterize the population and development patterns
of large urban centers of 50,000 or more. UAs are
composed of a central city (or cities) with a surrounding
closely settled area. The population of the entire UA
must be greater than 50,000 persons. The closely set-
tled area outside of the city, the urban fringe, must have
a population density generally greater than 1,000 per-
sons per square mile (just over 1.5 persons per acre) to
be included. The boundaries of UAs are based on popu-
lation patterns, not political boundaries; therefore, they
do not include significant portions of rural land.
The Bureau of Census has defined 396 UAs in the
United States based on the 1990 Census. These UAs
have a combined population of 158.3 million, or 63.6
percent of the nation's total population;2 however, these
areas only account for 1.5 to 2 percent of the land
surface of the country. Most increases in population
occur in urban fringe or suburban municipalities rather
than in core cities.3
Clean Water Act Requirements
In 1972, the CWA was amended to provide that the
discharge of any pollutants to waters of the United
States from a point source is unlawful, except where the
discharge is authorized by an NPDES permit. The term
"point source" is broadly defined to include "any discern-
ible, confined and discrete conveyance, including but
not limited to any pipe, ditch, [or] channel,... from which
pollutants are or may be discharged." (Congress has
specifically exempted agricultural stormwater dis-
charges and return flows from irrigated agriculture from
the definition of point source.) Although the definition of
point source is very broad, prior to 1987, efforts under
the NPDES program to control water pollution have
focused on controlling pollutants in discharges from
cities are in unincorporated portions of counties. In these cases, the
county government conducts the major functions of local government.
However, in most States, including New England, mid-Atlantic, Great
Lake, midwestern, and most western states, the primary form of local
government for many municipal functions is not a county but either an
incorporated place or a minor civil division. (These terms are defined
in Table 3.)
2The Census Bureau defines urban populations more broadly than
UAs. Urban populations include the populations of UAs and any other
dense population of 2,500 or more people. The 1990 Census indi-
cates that 28.8 million people who lived outside of UAs were classed
as urban populations. The Bureau of Census classified populations
that are not classified as urban (including UAs) as rural. The 1990
Census indicates that 61.6 million people were classified as living in
rural areas.
3The 1990 Census indicates that the total population of the United
States increased by 22.1 million between 1980 and 1990. Of this
growth, 86 percent (19 million) was in Census-designated UAs. Cities
with a population of 100,000 or more accounted for 22 percent of this
growth (4.9 million), while suburban areas surrounding these areas
grew by 11.5 million (52 percent of the national total). Another 12
percent of the national growth (2.6 million) occurred in UAs that did
not have a core city of 100,000 or more.
51
-------
publicly owned treatment works (POTWs) and industrial
process wastewaters. The major exception to this are
the 10 effluent limitation guidelines that EPA has issued
for stormwater discharges: cement manufacturing (40
CFR 411), feedlots (40 CFR 412), fertilizer manufactur-
ing (40 CFR 418), petroleum refining (40 CFR 419),
phosphate manufacturing (40 CFR 422), steam electric
(40 CFR 423), coal mining (40 CFR 434), mineral mining
and processing (40 CFR 436), ore mining and dressing
(40 CFR 440), and asphalt emulsion (40 CFR 443).
As part of the Water Quality Act of 1987, Congress
added Section 402(p) to the CWA to require EPA to
develop a comprehensive, phased program for regu-
lated stormwater discharges under the NPDES pro-
gram. Under the first phase of the post-1987 program,
EPA is to develop requirements for:
Stormwater discharges associated with industrial
activity.
Discharges from large municipal separate storm sewer
systems (systems serving a population of 250,000 or
more) and medium municipal separate storm sewer
systems (systems serving a population of 100,000 to
250,000).
Discharges that are designated by EPA or an
NPDES-approved state as needing an NPDES per-
mit because the discharge contributes to a violation
of a water quality standard or is a significant contribu-
tor of pollutants to waters of the United States.
Section 402(p)(1) of the CWA creates a temporary
moratorium on the requirement that point source dis-
charges of pollutants to U.S. waters must be authorized
by an NPDES permit for other stormwater discharges.4
Under the moratorium, EPA is prohibited from issuing
NPDES permits for discharges composed entirely of
stormwater that are not specifically exempted from the
moratorium (the discharges listed above to be ad-
dressed during the first phase of the program) prior to
October 1, 1994.5 Before this time, EPA, in consultation
with the states, is required to conduct two studies on
stormwater discharges. The first study is to identify
those stormwater discharges or classes of stormwater
discharges for which permits are not required prior to
October 1,1994, and to determine, to the maximum extent
practicable, the nature and extent of pollutants in such
discharges. The second study is to establish procedures
The Conference Report for the 1987 amendments to the CWA pro-
vides that after the moratorium ends on October 1,1994, "all municipal
separate storm sewers are subject to the requirements of Sections
301 and 402" (emphasis added) (15).
5The 1987 amendments to the CWA originally provided that the mora-
torium on other stormwater discharges (Water Resources Develop-
ment Act) expire on October 1, 1992. Under the amendments, EPA
was required to issue additional regulations to address these sources.
and methods to control stormwater discharges to the
extent necessary to mitigate impacts on water quality.
Based on the two studies, EPA is required to issue
regulations by no later than October 1, 1993, that des-
ignate additional stormwater discharges to be regulated
to protect water quality and establish a comprehensive
program to regulate such designated sources. The pro-
gram must, at a minimum:
Establish priorities.
Establish requirements for state stormwater manage-
ment programs.
Establish expeditious deadlines.
The program may include performance standards,
guidelines, guidance, management practices, and treat-
ment requirements, as appropriate.
The 1987 amendments to the CWA made significant
changes to the permit requirements for discharges from
municipal separate storm sewers. Section 402(p)(3)(B)
of the CWA provides that NPDES permits for such dis-
charges:
May be issued on a system- or jurisdictionwide basis.
Shall include a requirement to effectively prohibit non-
stormwater discharges into storm sewers.
Shall require controls to reduce the discharge of pol-
lutants to the maximum extent practicable, including
management practices, control techniques and sys-
tem, design and engineering methods, and such
other provisions as the Director determines appropri-
ate for the control of such pollutants.
Initial Implementation
On November 16,1990, EPA published the initial NPDES
regulations under Section 402(p) of the CWA (see 55 FR
47990). The November 16, 1990, regulations:
Defined the initial scope of the program by defining
the terms "stormwater discharge associated with in-
dustrial activity" and large and medium "municipal
separate storm sewer systems."
Established permit application requirements.
Established deadlines.
The regulatory definitions of large and medium munici-
pal separate storm sewer systems specifically identified
173 incorporated cities and 47 counties, and allowed for
additional designations of adjacent municipalities on a
case-by-case basis. EPA estimates that 400 additional
municipalities with a combined population of about 16
million people have been designated by EPA and author-
ized NPDES states, and that 23 cities with a population
of 100,000 or more (and a combined population of 8.6
million people) have been excluded from stormwater
52
-------
requirements due to large populations served by com-
bined sewer systems.
The November 16, 1990, regulations were based on
1980 Census data. Data from the 1990 Census indi-
cates that 30 additional cities have a population of more
than 100,000, and five of the cities listed in the Novem-
ber 16,1990, regulations no longer have a population of
100,000 or more. In addition, the 1990 Census indicates
that 12 additional counties have an unincorporated, ur-
banized population of 100,000, and two counties listed
in the November 16, 1990, regulations no longer have
an unincorporated, urbanized population of 100,000.
The November 16, 1990, regulations also established
requirements for a comprehensive, two-part permit ap-
plication for discharges from large and medium munici-
pal separate storm sewer systems. The major objectives
of the permit application requirements are to ensure that
municipalities develop comprehensive municipal storm-
water management programs that address water quality,
and to begin to implement these programs.
The permit application requirements for discharges from
municipal separate storm sewer systems represent a
new approach to addressing pollutant sources underthe
NPDES program. NPDES permit application require-
ments for other types of discharges traditionally focused
on sampling end-of-pipe discharges. Permit applications
for discharges from municipal separate storm sewer
systems place a lesser emphasis on discharge sampling
for a number of reasons, including the large number of
discharge points commonly associated with municipal
systems and the recognition that many municipalities
were only initiating efforts to reduce pollutants in storm-
water discharges at the time (see 55 FR 47990). Munici-
palities are required to submit comprehensive
applications providing information that: 1) identifies ma-
jor sources of pollution to the system, 2) characterizes
pollutants in system discharges, 3) describes existing
and proposed municipal stormwater management pro-
grams, and 4) describes the administrative and legal
aspects of the municipal stormwater management pro-
gram.
Perhaps the most important aspect of the permit appli-
cation requirements is that they lay out the framework
for municipalities to propose comprehensive municipal
stormwater management programs. When developing
permit conditions, permit writers will consider the man-
agement programs that are proposed as part of the
permit applications. The municipal stormwater manage-
ment programs envisioned by the November 16, 1990,
regulations address the four following areas:
Measures to reduce pollutants in runoff from residen-
tial and commercial areas: A major focus of this pro-
gram component is controlling pollutants in
stormwater from new development where stormwater
controls are generally more cost effective and mu-
nicipalities do not have to incur costs directly. Retrofit-
ting controls for existing development can also be
considered where practicable. Another focus is vege-
tation maintenance and snow removal activities for
roads. Other source control measures, such as trans-
portation plans, can be required where practicable.
Measures to reduce pollutants in runoff from indus-
trial facilities: EPA anticipates that a large percentage
of stormwater discharges associated with industrial
activity discharge through municipal separate storm
sewer systems. The Agency intends to coordinate
requirements in permits for stormwater discharges
associated with industrial activity with efforts to de-
velop municipal stormwater management programs
in permits for discharges from municipal separate
storm sewer systems serving a population of 100,000
or more. Under this coordinated effort, municipal per-
mittees will have a major role in implementing pro-
grams to control pollutants from stormwater
associated with industrial activity that discharges
through their municipal separate storm sewers. For
example, municipal operators can assist EPA and
authorized NPDES states in identifying priority storm-
water discharges associated with industrial activity;
reviewing and evaluating stormwater pollution pre-
vention plans developed by industrial facilities pursu-
ant to NPDES permit requirements; and complying
with requirements. (See 56 FR 40972 for a more
complete description of the relationship EPA intends
to develop between federal, state, and local govern-
ments for controlling pollutants in stormwater from
industrial sources.)
Measures to reduce pollutants in runoff from con-
struction sites: Many municipalities currently have
sediment and erosion requirements for construction
activities. These programs, however, often are not
adequately implemented or enforced. NPDES permit
conditions for municipalities are expected to focus on
ensuring adequate municipal implementation and en-
forcement of their controls. (See 57 FR 41206 and
Metropolitan Washington Council of Governments
[17].)
Measures to detect and control nonstormwater dis-
charges to the storm sewer system: Nonstormwater
discharges to separate storm sewer systems are a
major pollutant source in many municipalities. EPA
anticipates that permits will require municipalities to
continue field screening efforts started during the per-
mit application phase of the program and to under-
take other efforts to detect and control nonstormwater
discharges.
For a more complete description of the components of
a municipal stormwater management program, see
Guidance Manual for the Preparation of Part 2 of the
53
-------
NPDES Permit Applications for Discharges From Mu-
nicipal Separate Storm Sewer Systems (16).
The November 16, 1990, regulations take two very dif-
ferent approaches to defining the roles of different levels
of government. With respect to permits for large and
medium municipal systems, the efforts of the NPDES
permit authority (EPA or an authorized NPDES state)
are directed toward ensuring that municipalities develop
and implement stormwater management programs to
control pollutants to the maximum extent practicable.
Under these requirements, the NPDES program can
define the role of municipalities in a flexible manner that
allows local governments to assist in identifying priority
pollutant sources within the municipality and to develop
and implement appropriate controls for such discharges.
With respect to permits for stormwater discharges asso-
ciated with industrial activity, the NPDES permit author-
ity has a direct role in regulating individual industrial
sites.
Moratorium Sources: Why Municipalities?
Section 402(p)(6) of the CWA requires EPA to issue
regulations that designate additional stormwater dis-
charges to be regulated to protect water quality and that
establish a comprehensive program to regulate such
designated sources. EPA can generally take two differ-
ent approaches to identifying classes of discharges to
be regulated by NPDES permits: 1) to require munici-
palities to develop systemwide stormwater manage-
ment programs, or 2) to require NPDES permit coverage
for targeted commercial and residential facilities. When
evaluating whether to address selected municipalities
in the regulatory program required under Section
402(p)(6), the following factors should be considered:
There are institutional considerations.
Some existing municipal functions can be modified
to address stormwater concerns in a cost-effective
manner.
Municipal participation is necessary for regional or
systemwide stormwater management programs.
There are pollutant load considerations.
Issuing permits to municipalities allows for municipal
programs that incorporate innovative controls, such
as market-based incentives and pollutant trading.
Municipalities are in the best position to address high
risk sources, including new development, and to pro-
tect priority resources and watersheds.
Some municipal activities are significant pollutant
sources.
Municipalities can ensure maintenance of structural
controls and implementation of nonstructural measures.
Institutional Considerations
Municipalities contain the institutions that are critical for
surface water resource protection programs. Urban
stormwater management has been, is, and will continue
to be primarily the responsibility of local governments
(18). Municipalities install or oversee the installation of
storm sewer systems to provide drainage for lands used
for residential, commercial, and industrial activities as
well as roads and highways. Municipalities can provide
the institutional framework necessary to implement
many components of an effective stormwater manage-
ment program.
Components of a comprehensive stormwater manage-
ment program that only municipalities can effectively
address include land use planning, detailed oversight of
new development, maintenance of roads, retrofitting
controls in areas of existing development, and operation
and maintenance of municipal storm drains. Municipali-
ties can provide the detailed planning necessary to im-
plement watershed and other risk-based approaches.
The role of municipalities under the NPDES program is
to make stormwater management programs work. This
involves overseeing day-to-day program operations,
identifying local priorities and pollutant sources, devel-
oping detailed program requirements, conducting site
inspections and evaluations, monitoring activities, as-
sessing impacts to surface water resources, initiating
compliance efforts, and ensuring effective outreach. Mu-
nicipal activities can be funded by a variety of mecha-
nisms, including general revenues, developer fees,
flood control assessments, and stormwater utilities.
Raising funds at the municipal level can provide a mu-
nicipalitywide source of funds that can then be directed
at priority projects. Thus, comprehensive programs can
be implemented in a phased manner over a long period.
In addition, such an approach takes advantage of pol-
lutant trading concepts by directing resources from
many sources to priority sources.
The role of the federal government and authorized
NPDES states under the NPDES municipal stormwater
program is to ensure that regulated entities implement
pollution control measures. In the municipal stormwater
area, this means providing oversight to guide the direc-
tion of municipal programs and providing technical as-
sistance. Oversight activities include issuing permits
that establish the framework for municipal stormwater
control programs and taking targeted enforcement ac-
tions, for example, when municipalities fail to develop
and implement a program. In addition, the NPDES
authority must work in partnership with municipalities to
ensure that, where appropriate, priority pollutant
sources that municipalities may have difficulty control-
ling, such as certain federal or state facilities, are directly
issued NPDES permits for their stormwater discharges.
As Thomas Mumley, Associate Water Resource Control
54
-------
Engineer at the San Francisco Regional Water Quality
Control Board (19) states:
Successful control of urban runoff will require a
carrot, a stick, and ... the implementation of com-
mon-sense, cost-effective, environmentally benefi-
cial measures. . . . We need incentives to change
our ways ... we now have a big stick to drive these
needed efforts, in the form of the NPDES stormwa-
ter regulations [for municipalities] which require the
implementation of these measures. Fortunately, the
current regulations promote flexibility6 and don't im-
pose a lot of bureaucratic red tape, and therein lies
the carrot.
Expanding the Mission of Existing Municipal
Programs
Municipalities typically operate programs whose primary
mission is to address a set of concerns other than
stormwater or water quality. Expansion of the mission
of these existing municipal programs to address storm-
water concerns can be much more cost effective than
initiating entirely new programs. Municipal functions that
can be adapted to assist in providing stormwater man-
agement benefits include oversight of new develop-
ment, pretreatment program implementation, fire safety
inspections, flood control, trash collection, management
of municipal lands, and road maintenance. Municipal
lands, for example, can provide retrofit opportunities for
a number of reasons. The use of municipal lands for
retrofits typically does not require additional property
purchases. In addition, the use of municipal lands en-
sures opportunities to provide future maintenance and
security in preservation of the retrofit control. (See
Washington State Department of Ecology [20] for spe-
cial stormwater management practices for public build-
ings and streets; vehicle and equipment maintenance
shops; maintenance of open space areas; maintenance
of public stormwater facilities; maintenance of roadside
vegetation and ditches; maintenance of public utility
corridors; water and sewer districts and departments;
and port districts.)
In addition, many municipal activities and programs
can be significant sources of pollutants, such as road
maintenance, road construction, siting and operating
flood control devices, maintenance of municipal vehicles,
municipal landfills, and airports.7 Expanding the mission
of these programs can assist in the development of a
Concerns have been raised regarding the requirements under the
current Clean Water Act that NPDES permits for municipal separate
storm sewers, in addition to mandating the reduction of pollutants to
the maximum extent practicable, must ensure compliance with water
quality standards. The water quality standards issue is not discussed
in this paper.
7Some municipal activities are considered to be industrial activities
under the NPDES program. Section 1068 of the Intermodial Surface
Transportation Efficiency Act of 1991 placed stormwater discharges
pervasive municipal ethic regarding stormwater man-
agement that ensures effective use of municipal re-
sources and mitigates the effects of municipal activities
that can affect water resources.
Regional or Systemwide Programs
Urban stormwater is a diffuse source of pollution. The
impacts of stormwater on receiving waters generally
cannot be attributed to individual sources or discharge
points; rather, the cumulative effects of many discharges
from widespread areas of urban development in a wa-
tershed are of major concern. Often, approaches that
consider watershed characteristics are necessary for
success.
Control of urban stormwater is critical from a regional
perspective, which addresses the entire UA. The lack of
regional orsystemwide planning is often cited as a major
reason for incomplete and unsuccessful stormwater
control efforts and for the inability to protect downstream
areas from stormwater from upstream development. A
comprehensive stormwater management program can-
not rely solely on addressing individual sources within
large UAs.
A regional approach can also bring together financial
resources, planning, and scientific expertise not other-
wise available for individual municipalities, thereby in-
creasing the likelihood for success. Regional entities
that can play an important role in planning, implement-
ing, and evaluating stormwater programs include flood
control districts, stormwater or drainage districts, coun-
ties, and Councils of Governments.
Pollutant Load Considerations
UAs comprise a mixture of different land uses. For gen-
eral planning purposes, most UAs are distributed as
follows: residential, 50 to 70 percent; commercial, 10 to
20 percent; industrial, 10 to 15 percent; open area, 10
to 15 percent (13). Concentrations of pollutants in storm-
water from nonindustrial areas can be assumed to be
roughly the same for different land use types, but the
degree of imperviousness plays an important role in
determining pollutant loads (8). This is because many
diffuse sources of pollutants to urban stormwater oper-
ate in different land use areas, and areawide sources
are important. While commercial and industrial land
uses generally have a higher level of imperviousness
than some types of residential development, a large
amount of residential area will result in residential land
use being a major pollutant source to stormwater. For
example, a study of the Santa Clara Valley found that
the volume of stormwater flows from residential and
associated with industrial activity owned or operated by a municipality
with a population of less than 100,000 in the moratorium from NPDES
permit requirements.
55
-------
commercial land uses in the Valley was 10 times greater
than the volume of flow from industrial uses. The loading
of metals in stormwater flows from residential and com-
mercial lands was estimated to be 5 to 30 times greater
than from industrial lands (11).
A program that only addresses industrial stormwater
flows is limited because it only addresses a fraction of
the total urban stormwater flows. Similarly, programs
to address illicit connections to storm sewers should
address municipal sources. Municipalities have respon-
sibilities associated with several important classes of
illicit connections, including sanitary collection systems
(ownership of collection system), improper connections
between sanitary and storm sewer systems, and im-
proper connections from residential or commercial ar-
eas. For example, investigations in Houston, Texas,
indicated that most of the city's problems associated
with nonstormwater discharges to the separate storm
sewer system were associated with broken wastewater
collection system lines discharging to its stormwater
collection system (21).
In general, municipal programs should include legal
authority to address the majority of stormwater sources
into their municipal system. However, this does not
mean that a municipality should have to ensure that
every existing residential, commercial, or industrial site
within its jurisdiction actively controls its stormwater.
Rather, municipalities should develop programs that re-
sult in the implementation of practicable controls for
high-priority sources that maximize cost-effectiveness
by considering possible sources and conditions within
the jurisdiction. In addition, EPA must be a partner in
efforts to control selected priority sources, such as in-
dustrial, federal, and state facilities. For example, some
municipalities have indicated that practical problems are
associated with controlling stormwater from federal and
state facilities. In such cases, a partnership between the
municipality and the NPDES authority may be appropri-
ate where the municipality identifies high-risk state and
federal facilities for the NPDES authority to consider
issuing an NPDES permit directly.
In addition, the Agency should lead national efforts to
directly reduce some pollutant sources or find product
substitutes. For example, federal requirements under
the Clean Air Act have resulted in significant decreases
in the use of lead in gasoline, which in turn have resulted
in decreases in lead concentrations in urban runoff.
Other areas of national regulation and/or pollution pre-
vention efforts that have been suggested are reduction
in the amount of zinc in tires, reductions in the amount
of copper in brake pads, and lower emission standards
for particulate emissions for diesel engines (11).
Flexibility in Selecting Measures
Municipal stormwater management programs should be
comprehensive efforts that address a wide range of
innovative measures in addition to traditional command-
and-control requirements. Federal or state permitting
programs generally have limited flexibility to directly
implement many types of innovative control strategies
in a widespread manner. Requiring municipalities to
obtain NPDES permits fortheir municipal systems could
create a regulatory framework that could support mu-
nicipalities' use of innovative controls, such as market-
based incentives.
For example, municipalities can fund stormwater pro-
grams with a utility rate system that accounts for the
impervious area at a site, which is roughly proportional
to the amount of stormwater generated at the site. A
survey of 54 stormwater agencies with stormwater utili-
ties located in 19 states indicated that 70 percent of the
agencies surveyed based their utility on the amount of
impervious area at a site, while an additional 17 percent
based their utility on the product of area times an inten-
sity of development, which can approximate impervious
area (22). Such a rate system can also considerwhether
stormwater controls are provided at a site. These ap-
proaches create market-based incentives for reducing
site imperviousness (thereby reducing stormwater vol-
umes and pollutant loads) and for installing and operat-
ing stormwater measures. (See U.S. EPA for a list of 21
municipal stormwater utilities that provide credits for
onsite stormwater management [23].)
Municipalities have a wide range of tools for ensuring
stormwater control measures occur with new develop-
ment. For example, municipalities can have zoning pro-
visions that establish setbacks for buffer zones, limit the
amount of impervious area, require maintaining mini-
mum amounts of open space, and encourage cluster
development. Municipalities can also develop watershed
management plans that provide for preservation of flood-
plains, wetlands, shoreline, and other critical areas. In
addition, during the building plan approval process, mu-
nicipalities can designate, through deed modification or
other means, an entity or individual who is responsible
for maintaining the stormwater management systems of
a new development. Controls on siting, installing, and
maintaining septic systems and for ensuring proper
sanitary sewer connections can reduce pollutant dis-
charges from municipal separate storm sewer systems.
Other innovative approaches to stormwater manage-
ment include used oil and/or household hazardous
waste municipal collection programs. Municipalities can
conduct portions of public outreach programs in a more
cost-effective way than other levels of government. For
example, municipalities can stencil catchbasins to mini-
mize improper dumping of materials and send informa-
tional flyers with water or sewer bills.
56
-------
Another approach is for a municipality to use pollutant
trading concepts to select cost-effective controls. One
example of pollutant trading is for a municipality to allow
a developer to contribute to an offsite regional stormwa-
ter measure where onsite measures are not feasible.
Other pollutant trading concepts are discussed in Santa
Clara Valley Nonpoint Source Pollution Control Program
(11) and U.S. EPA (24). It should be noted that some
concerns have been raised regarding trading structural
controls for nonstructural controls where opportunities
to install structural controls can be lost and the contin-
ued implementation of nonstructural controls cannot be
assured.
Municipalities can also incorporate voluntary compo-
nents into their municipal stormwater management pro-
grams, such as adopt-a-highway litter programs or
adopt-a-stream programs. In addition, the development
of stormwater programs at the municipal level can en-
courage high levels of public input from local groups.
Flexibility To Address High-Risk Sources and
To Protect Priority Resources and Watersheds
Controlling pollutants in stormwater involves addressing
many and diffuse pollutant sources. The nature of the
problem calls for focusing on priority sources and em-
phasizing controls in priority watersheds. Municipalities
are in the best position to evaluate local conditions and
to determine local priorities for implementing and over-
seeing control strategies and measures that ensure the
water quality impacts of land use activities in its jurisdic-
tion are mitigated. This is particularly true when evalu-
ating the risks of new development.
Urbanization is a gradual process that spans decades and
occurs over a wide region. It is composed of hundreds of
individual developments that take place over much shorter
time frames. The true scope of water resource degrada-
tion associated with urbanization may not fully manifest
at the watershed scale for many years. This presents the
challenge of evaluating the impact of individual develop-
ment proposals over the long term at the watershed
scale (25) and planning appropriately. Such detailed
planning can only occur on the municipal level.8 Detailed
efforts to plan and oversee new development could not
(and should not) be undertaken at the federal level.
Municipalities typically have planning processes and ad-
ministrative systems in place to address some aspects of
new development. When municipalities plan for new
development, the total development of the area can be
considered. This can provide a much more comprehen-
sive basis for planning than when developers plan at the
EPA has recognized that many local governments typically require
sediment and erosion plans, grading plans and/or stormwater man-
agement plans that are significantly more detailed and are accompa-
nied by a more rigorous review process than those required under
EPA-issued general permits (57 FR 41196).
site level. Municipalities can accomplish these tasks
with a much greater sensitivity to local conditions and in
a more equitable and reasonable manner. In addition,
municipalities can develop watershed plans that con-
sider the tradeoffs associated with the placement of
onsite controls and regional stormwater management
approaches. Some municipalities advocate stormwater
control strategies that use a mix of regional controls and
onsite controls that reflects watershed hydrology. Ad-
vantages of this approach are said to include better
control of peak flows; reduced impacts to streams and
riparian wetlands; improved pollutant removal efficien-
cies; lower costs; a significantly higher likelihood of
adequate maintenance; and recreational amenity values
(26).
The ability of EPA or NPDES states to conduct such
detailed planning is limited. For example, EPA indicated
that a consideration of possible water quality impacts
associated with the timing of releases from onsite storm-
water management measures involves a complex array
of variables, including the nature and locations of other
activities within a watershed, and is generally beyond
the scope of the Agency's NPDES general permits for
stormwater from construction activities (see 57 FR
41202). Municipal consideration of mitigation measures
for numerous smaller projects in a watershed may better
maintain the integrity of an aquatic ecosystem.
A goal of the stormwater program should be that munici-
palities have planning procedures to identify and ad-
dress the potential impacts of development on water
resources. NPDES permits for municipal separate storm
sewer systems can assist in reaching this goal by en-
suring that municipalities consider the impact of storm-
water on surface waters. Traditionally, the major
objective of installing separate storm sewers has been
to remove as much stormwater runoff from developed
lands as soon as possible. To achieve this goal, local
governments have constructed thousands of miles of
curb, gutter, road side ditches, and other storm sewers
to convey stormwaters as quickly and as efficiently as
possible to the nearest stream (18). Efforts often focus
on channelization projects that attempt to make streams
more "efficient" at conveying waters downstream. Ex-
tensive channelization projects and other stream "im-
provements," such as concrete-lined walls or heavy
riprap, can destroy the habitat value of streams.
A few communities have developed programs where
stormwater is managed for multiple purposes, including
controlling water quantity (to avoid flooding and stream
scour and to maintain stream flows during dry weather
by recharging ground water during storms) and improving
water quality. A range of alternative stormwater control
measures and facilities can be implemented to serve
multiple purposes effectively. The natural cycles and
processes that occur before land development are used
57
-------
as a guide for managing stormwater after development
has occurred, and natural flow patterns and rates of
discharge are retained through special stormwater con-
trol facilities and measures. Natural processes are in-
corporated into the design of many "soft" engineered
systems, including vegetated buffers, greenways,
revegetation of stormwater systems, wetland creation or
retention for stormwater management, and onsite reten-
tion, detention, or infiltration systems. Policies emerging
from these programs include:
Reducing peak flows and improving stormwater qual-
ity through onsite retention.
Reducing the volume of stormwater leaving the site
using natural infiltration.
Releasing stormwater from onsite facilities at a rate
similar to the predevelopment runoff rate.
Managing for smaller storm events as well as those
larger storm events that can cause major floods.
Protecting wetlands and floodplains as natural storm-
water storage areas.
Making stormwater facilities amenities of the devel-
opment (such as retaining natural drainage channels
or providing attractive landscaping for stormwater
management ponds) and encouraging open space
and recreational uses.
Developing programs that relate erosion and sedi-
ment controls during construction with stormwater
management after construction is completed.
The implementation of this approach typically involves
somewhat higher costs for development plan review by
local governments but lower costs for stormwater facility
construction, and results in lower social costs.
Maintenance of Controls
The installation of structural controls (e.g., wet ponds,
infiltration devices) during the construction phase of new
development is often cited as a key component to a
successful stormwater program. To continue to operate,
these devices need to be maintained every 5 to 15
years. Lack of maintenance is often cited as a leading
cause of failure of stormwater management devices.
While NPDES permits for stormwater discharges from
construction activities disturbing more than 5 acres can
require the installation of stormwater measures during
the construction phase of a project, permit coverage for
residential and commercial sites ends when the site is
stabilized. Therefore, NPDES permits for stormwater
discharged from construction sites may not be able to
ensure the continued maintenance of these sites. Mu-
nicipalities are in a better position to require or conduct
maintenance activities for these devices. For example,
municipalities can require maintenance of stormwater
management devices through deed modification prior to
site development or through ordinances.
Moratorium Sources: Which
Municipalities?
Public commentors on previous NPDES stormwater
rulemakings have identified a number of principles that
are critical to successful implementation of NPDES re-
quirements for a stormwater regulatory program (55 FR
48039):
Municipalities should be regulated as equitably as
possible.
Major sources of pollutants must be addressed
through control, treatment, or prevention.
The approach must be administratively realistic and
achievable.
New development should be addressed.
Programs must be coordinated or developed on a
regional basis to avoid fragmentation or balkanized
programs and to support watershed approaches.
Regional approaches are necessary to address inter-
related discharges into the municipal separate storm
sewer system.
Municipalities associated with Census-designated UAs
or a subset thereof appear to meet most of the criteria
in a way that makes them candidates for consideration
for Phase II stormwater requirements. Additional munici-
pal candidates for Phase II requirements are pockets of
high growth levels outside of Census-designated UAs
and areas with large seasonal activities (e.g., some
tourist towns) that are not classified as part of a Census-
designated UA because of small year-round popula-
tions.
Equitable Treatment/Major Pollutant Sources
Currently, NPDES requirements for discharges from
municipal separate storm sewer systems focus on core
cities, and generally do not address UAs surrounding
core cities in a comprehensive manner. The regulations
do address 47 counties that were selected because they
had significant populations in unincorporated, urbanized
portions of the county. In most UAs, however, areas
surrounding core cities are broken into incorporated
areas and/or minor civil divisions with populations of
less than 100,000. These areas are not addressed by
current NPDES requirements even though they may be
in a heavily populated county. For example, 400 coun-
ties have a population of greater than 100,000 but are
not addressed by the current NPDES regulations.
At least three factors are important to consider when
determining whether municipalities are being regulated
as equitable as possible: 1) demographic patterns asso-
58
-------
elated with per capita income; 2) the pollutant sources
that are being addressed; and 3) the ability to control
major pollutant sources. Some states have also advo-
cated national NPDES requirements to ensure national
consistency and to prevent economic disincentives that
make it difficult for states and municipalities to imple-
ment progressive stormwater management programs
(57 FR41205).
The per capita income of suburban fringe areas is
typically significantly higher than the per capita income
of core cities. A 1991 report by the National League of
Cities indicates that the per capita incomes of residents
in the largest cities is only on average 59 percent of
the per capita incomes in the surrounding suburbs.
The magnitude of these income disparities was cited as
a clear indicator of the disparities in tax bases. The
report also suggested that continued demographic shifts
are expected to increase these differences (27). In ad-
dition, municipal governments associated with core
cities often provide a greater range of services than
surrounding areas, resulting in higher per capita munici-
pal government costs.
As discussed above, the pollutant sources associated
with urban stormwater are diffuse in nature and are
associated with widespread areas of development. Cen-
sus data from 1990 indicate that approximately 46 per-
cent of the total area and 35 percent of the total
population of UAs containing a city with a population of
100,000 or more are located outside of the core city in
suburban fringe areas.9 As a rough approximation, sub-
urban fringe areas are generating as much stormwater
pollution as core cities with a population of 100,000 or
more. Failure to address suburban fringe areas outside
of these cities would severely limit the ability of the core
city to protect receiving waters.
The equity issue is also related to the types of controls
that are available to municipalities. Older, densely de-
veloped core cities have limited opportunities to control
pollutants in their stormwater (8). Areas with substantial
new growth, however, including many suburban fringe
areas, have greater opportunities to ensure appropriate
stormwater management and mitigate impacts to receiv-
ing waters associated with new growth.
Between 1970 and 1980, the population of incorporated
cities with a population of 100,000 or more (those with
municipal separate storm sewer systems addressed by
NPDES regulations before October 1,1992) increased by
only 0.6 million, with much of this increase associated with
the addition of the populations of 17 cities that had popu-
lations of 100,000 or more forthe first time. The land area
In the United States, most people served by combined sewers are
located in cities with a population of 100,000 or more (57 FR 41349).
Thus, the percentage of urbanized population served by separate
storm sewers in suburban fringe areas is higher than indicated above.
of most of these cities remained the same, while the
populations of many large cities decreased.
Most growth in UAs occurs in areas that were not re-
quired to obtain an NPDES permit for their stormwater
discharge before October 1, 1992. Between 1970 and
1980, the population of UAs outside of cities with a
population of 100,000 or more increased 30 times more
(an increase of 18.9 million) than the population of these
cities. This growth resulted from both increases in popu-
lation densities of existing urban lands and by the ur-
banization of previously rural lands. Factors such as
lower costs of land, commercial space, and residential
housing continue to cause urban sprawl even in UAs
that are not experiencing population growth.
Equity and pollutant source considerations would ap-
pear at least to require that NPDES requirements be
extended to cover suburban fringe municipalities in Cen-
sus-designated UAs in which one or more large or me-
dium municipal separate storm sewer systems are
already subject to NPDES requirements. Municipalities
with a large or medium municipal system should not be
held solely responsible for implementing NPDES storm-
water requirements when stormwater from suburban
municipalities limits the opportunities of the core cities
to effectively protect water resources.
Perhaps a more equitable approach would be to ex-
pand NPDES requirements to cover municipalities as-
sociated with Census-designated UAs of a specified
size (e.g., 100,000 or 50,000). This approach would
ensure that urban centers of similar size and the
largest sources of urban runoff would be subject to
program requirements.
Administratively Achievable/New Development
In core cities, urban streams are typically already heav-
ily degraded, with limited opportunities for full restora-
tion. Significant opportunities exist in suburban fringe
areas, however, to conduct new development in a way
that mitigates impacts on water resources. A basic prin-
ciple of stormwater controls is that developing controls
for new development is much more cost effective (8) and
institutionally feasible than retrofitting old development.
EPA has also indicated that, where properly planned,
stormwater controls can increase the property values
and satisfy consumer aesthetic needs (56 FR 40989).
Municipalities often oversee the development process.
They usually have some form of approval or permit
program in place. Developers have incentives to comply,
because enforcement can be stringent (e.g., stop-work
orders), and the developer usually wants to have a
workable relationship with the municipality to ensure
that future projects proceed smoothly. In addition, the
costs of the controls are not borne by the municipality
directly but rather by the developer. Several states with
59
-------
progressive stormwater management programs have
initially focused on new development (e.g., Maryland,
Florida, and Delaware). This is unlike the approach
taken in the 1987 amendments to the CWA, which in-
itially focused on core cities with little or no growth and
temporarily excluded suburban municipalities. The No-
vember 16, 1990, EPA regulations addressed 47 coun-
ties and 173 cities. The counties that were addressed
where in a handful of states, primarily Maryland, Vir-
ginia, Florida, and California. While the Agency was able
to address suburban growth in these states, in most
parts of the country the regulations only address core
cities and exclude suburban development.
Perhaps the biggest challenge associated with Phase II
NPDES stormwater requirements for municipalities is
the potentially large number of small municipalities
that should be addressed. Census-designated UAs
offer advantages over broader classifications of metro-
politan areas, such as Standard Metropolitan Statistical
Areas (SMSAs),10 in that UAs do not include significant
amounts of rural areas or small urban municipalities
that are isolated from larger urban centers. In many
parts of the country, however, suburban urban fringe
areas are broken into a significant number of small
municipal entities (see Table 3). In developing Phase II
requirements for municipalities, EPA could consider pro-
moting regional approaches, developing tiered require-
ments for different sizes of municipalities, and limiting
requirements or providing exemptions for very small
municipalities. For example, the Agency could consider
focusing requirements for small municipalities on a few
key program components, such as new development,
municipal activities that affect stormwater quality (e.g.,
road building and maintenance), illicit connections, and
public education.
Regional Approaches
As discussed above, regional approaches to stormwater
management offer a number of advantages, including
providing municipalities with the opportunity to pool
resources and to address stormwater management
with a more holistic watershed approach. Successful pro-
grams must face the challenge that municipalities do not
follow watershed boundaries. Currently, the NPDES mu-
nicipal stormwater program principally focuses on core cit-
Unlike Census-designated urbanized areas, SMSAs, which are
identified by the Office of Management and Budget, are based on
county boundaries and can contain significant rural areas. Urbanized
areas are defined to describe population densities. An urbanized area
consists of the contiguous builtup territory around each larger city and
thus corresponds generally to the core of the SMSA. SMSAs are
defined to describe a large population nucleus and adjacent commu-
nities that have a high degree of economic and social integration with
that nucleus. This designation has been developed for use by federal
agencies in the production, analysis, and publication of data on met-
ropolitan areas (28).
ies with a population of 100,000 or more.11 If suburban
municipalities fail to develop adequate stormwater pro-
grams, the ability of core cities adequately to protect the
receiving waters of the core city will be limited. As Tucker
(18) states,
Dealing with drainage across jurisdictional lines is
important.... The ability to look at urban stormwater
management from a regional or metropolitan wide
perspective is important. The larger drainageways
typically flow from one jurisdiction to another and
what happens in one entity can impact others. Plan-
ning should be approached on a basinwide basis
and not stop at jurisdictional boundaries. . . . Once
the Phase II regulations for NPDES permits for mu-
nicipal separate storm sewers become a reality,
more metropolitan areas will seriously consider re-
gional approaches to stormwater management.
Conclusion
Urban stormwater discharges have been shown to be a
major source of water quality impairment. Section
402(p)(6) of the CWA requires EPA to identify additional
stormwater sources to be regulated to protect water
quality. In UAs, pollutants associated with stormwater
come from many sources distributed throughout the
area of urban development. Commercial and residential
areas appear to be significant sources of pollutants,
along with certain municipal activities. Municipal govern-
ments in UAs must play a significant role in developing
and implementing programs that effectively address pri-
ority pollutant sources within their jurisdictions. Munici-
pal governments have the critical institutional framework
for making the day-to-day decisions to address these
problems, to minimize or prevent the risk associated
with stormwater from areas undergoing urbanization,
and to collect the majority of funds necessary to imple-
ment the comprehensive programs needed to address
urban stormwater management. The condition of a wa-
terbody is a reflection of watershed management and
land use characteristics. To ensure that the waterbody
is protected and maintained, citizens must be empow-
ered to work together to that end.
References
1. U.S. Bureau of the Census. 1990. 1990 Census of population.
2. Pitt, R. 1993. Effects of urban runoff on aquatic biota. In: Handbook
of ecotoxicology. Lewis Publishers.
The NPDES storm water program also currently addresses unincor-
porated portions of 47 counties. However, most large counties, includ-
ing those in many heavily urbanized areas of the country, are currently
not subject to NPDES stormwater requirements. Those counties cur-
rently addressed by the NPDES storm water program have large
populations in unincorporated areas and only represent a few states,
notably, California, Florida, Maryland, and Virginia.
60
-------
Table 3. Municipalities Associated With Census-Designated UAs Based on 1990 Census Data3
No. of
Incorporated No. of No. of
Class of UA No. of UAs Places'3 MCDsc Countiesd
All UAs 396
250,000 or more 103
100,000-250,000 121
50,000-100,000 172
Phase I municipalities Parts
UA with large or 137
medium MS4
3,624
2,672
490
462
of 137 621
2,147
3 Examples of Census-designated UAs and associated 1990
Brunswick, GA
Ithaca, NY
San Luis Obispo, CA
Lafayette-West Lafayette, IN
Sioux Falls, SD
Jacksonville, NC
Pensacola, FL
Sacramento, CA
San Antonio, TX
50,066
50,132e
50,305
100,103
100,843
101,297
253,558
1 ,097,005
1,129,154
1 ,655 703
1 ,022 358
349 185
284 258
0 70
665 280
populations:
Ogden, UT 259,147
Albuquerque, NM 497120
Albany-Schenectady-Troy 509106
Akron, OH 527,863
Oklahoma City, OK 784 425
Salt Lake City, UT 78g 447
New Orleans, LA 1 040,226
Shreveport, LA 256,489
Total Population
(millions)
158.3
127.5
18.9
11.9
76.2
116.8
b Incorporated places include incorporated cities, towns, villages, and boroughs.
c Minor civil divisions (MCDs) include unincorporated towns and townships in 20 states.
d County equivalents include counties, parishes in Louisiana, and boroughs in Alaska. Some double counting of
counties occurred as portions of several UAs may be in one county. (For example, portions of the Washington UA,
Baltimore UA, and Annapolis UA are in Ann Arundel County, Maryland.)
e The Ithaca, New York, population does not include student population at Cornell University.
3. Pitt, R., and R. Field. 1991. Biological effects of urban runoff
discharges. In: Effects of urban runoff on receiving systems: An
interdisciplinary analysis of impact, monitoring, and management.
Engineering Foundation Conference, Mt. Crested Butte, CO. New
York, NY: ASCE.
4. U.S. EPA. 1990. 55 Federal Register 47990. November 16.
5. Montoya, B. 1987. Urban runoff discharge from Sacramento,
California. Report No. 87-1 SPSS. California Regional Water Con-
trol Board, Central Valley Region.
6. U.S. EPA. 1993. Investigation of inappropriate pollutant entries
into storm drainage systems: A user's guide. EPA/600/R-92/238
(January).
7. Schueler, T 1987. Controlling urban runoff: A practical manual
for planning and designing urban best management practices.
Washington Metropolitan Council of Government.
8. U.S. EPA. 1983. Results of the Nationwide Urban Runoff Pro-
gram, Vol. 1. Final report.
9. Cameron, D. 1989. NRDC's poison runoff index for the Washing-
ton metropolitan region (November).
10. Pit, R. 1992. Stormwater, baseflow, and snowmelt pollutant con-
tributions from an industrial area. Presented at the 65th Annual
Conference, Water Environment Federation, New Orleans, LA
(September).
11. Santa Clara Valley Nonpoint Source Control Program. 1992.
Source identification and control report (December).
12. Ontario Ministry of the Environment. 1986. Toronto area watershed
management strategy study: Number River Pilot Watershed Pro-
ject (June).
13. U.S. EPA, Region 5. 1990. Urban targeting and BMP selection:
An information and guidance manual for state nonpoint source
program staff engineers and managers (November).
14. U.S. EPA. 1992. Environmental impacts of stormwater discharges: A
national profile. EPA/841/R-92/001 (June).
15. Congressional Record H10576, Vol. 132. Conference Report. Oc-
tober 15, 1986.
16. U.S. EPA. 1992. Guidance manual for the preparation of part 2 of
the NPDES permit applications for discharges from municipal
separate stormwater sewer systems. EPA/833/B-92/002 (Novem-
ber).
17. Metropolitan Washington Council of Governments. 1990. Per-
formance of current sediment control measures at Maryland con-
struction sites (January).
18. Tucker, L.S. 1991. Current programs and practices in stormwater
management. In: Water and the City: The next century.
19. Mumley, T. 1991. EPA Journal. November/December.
20. Washington State Department of Ecology. 1992. Stormwater man-
agement manual for the Puget Sound Basin, Vol. 1. Minimum tech-
nical requirements (February).
61
-------
21. Glanton, T., M. Garrett, and B. Goloby. 1992. The illicit connec-
tion-is it the problem? Water Environ. Tech. (September).
22. Benson, R. 1992. Financing stormwater utility user feeswhere
are we now? Water Environ. Tech. (September).
23. U.S. EPA. 1992. Stormwater utilities: Innovative financing for
stormwater management. Draft final report. Office of Policy, Plan-
ning, and Evaluation (March).
24. U.S. Department of Commerce/U.S. EPA. 1993. Coastal nonpoint
pollution control program: Program development and approval
guidance (January).
25. Anacostia Watershed Restoration Team. 1992. Developing effec-
tive BMP systems for urban watersheds. Metropolitan Council of
Governments.
26. Prince William County. 1993. Prince William County Comprehen-
sive Storm Water Management/Planning and Demonstration Wa-
tershed Project. Prince William County, MD.
27. National League of Cities. 1991. City fiscal distress: Structural,
demographic, and institutional causes.
28. U.S. Bureau of the Census. 1983. Number of inhabitants, United
States summary (April).
62
-------
Municipal Permitting: An Agency Perspective
William D. Tate
Office of Wastewater Enforcement and Compliance, Office of Water,
U.S. Environmental Protection Agency, Washington, DC
This paper presents the U.S. Environmental Protection
Agency's (EPA's) perspective regarding the municipal
side of the National Pollutant Discharge Elimination Sys-
tem (NPDES) stormwater program. It begins by briefly
providing some background information on the storm-
water program. It then highlights an EPA review of costs
that municipal separate storm sewer systems (MS4s)
have incurred or anticipate incurring during the next 5
years. After discussing the types of programs that MS4s
proposed in their Part 2 applications, the paper con-
cludes by presenting the current status of the permitting
process.
Background
The Water Quality Act (WQA) of 1987 added Section
402(p) to the Clean Water Act (CWA). In Section 402(p),
MS4s serving a population of 100,000 or more must
obtain an NPDES permit for their stormwater dis-
charges. Section 402(p)(3)(A) specifically provides that
permits for these discharges:
May be issued on a system- or jurisdictionwide basis.
Shall include a requirement to effectively prohibit non-
stormwater discharges into storm sewers.
Shall require controls to reduce the discharge of pol-
lutants to the maximum extent practicable; controls
may include management practices, techniques, sys-
tem design and engineering methods, and such other
provisions as the Administrator or the state deter-
mines appropriate for control of such pollutants.
NPDES permits historically have imposed end-of-pipe
controls on industrial and publicly owned treatment
works discharges. The legislative history of the WQA,
however, indicates that Congress does not consider
end-of-pipe controls to be necessarily appropriate for
stormwater discharges from MS4s. Consequently, in
the November 16, 1990, Federal Register, EPA pub-
lished a final rule intended to reflect the unique nature
of discharges from MS4s. The final rule establishes
permit application requirements and application deadlines
for all MS4s covered under Phase I of the stormwater
program. For MS4s required to obtain a stormwater
permit, EPA established a two-part permit application
process. The Part 1 application primarily focuses on a
municipality's existing stormwater management activi-
ties and includes the following components:
General information
Discharge characterization
Existing legal authority
Existing stormwater management programs
Source identification
Existing fiscal resources
The Part 2 application requires additional information
that builds on the information submitted with the Part 1
application. Rather than emphasizing current stormwa-
ter management activities, however, the Part 2 applica-
tion focuses on what future stormwater management
activities an MS4 will adopt. Major components of the
Part 2 application are similar to those identified above;
however, their level of detail is much greater.
Some of the major highlights of the stormwater program
involve:
Obtaining the adequate legal authority to implement
an MS4's stormwater management program.
Developing estimates of annual pollutant loadings
and a schedule to submit seasonal pollutant loadings
estimates.
Developing a monitoring program to run throughout
the permit term.
Developing a site-specific and comprehensive storm-
water management program.
Conducting an assessment of the effectiveness of
stormwater controls.
63
-------
Conducting a fiscal analysis of the costs to imple-
ment the applicant's proposed stormwater manage-
ment program.
The cornerstone of the stormwater program is the re-
quirement that MS4s must develop site-specific and
comprehensive stormwater management programs.
MS4s should employ all program requirements identi-
fied in the final rule. Given their geographical, clima-
tological, and physical differences, however, MS4s
can exercise discretion when establishing priorities for
their site-specific stormwater management programs.
For example, an MS4 in a densely populated urban
corridor is not reasonably expected to have the same
program priorities as an MS4 servicing an area ex-
periencing rapid development. Later, the paper pre-
sents a few different approaches and types of programs
that various MS4s are proposing. First, however, is a
brief discussion of the present status of the MS4
permitting process.
Present Status of the MS4 Permitting
Process
Effects of the 1990 Decennial Census
In the November 16, 1990, Federal Register, EPA iden-
tified 219 municipalities required to seek coverage un-
der an NPDES stormwater permit. Appendices F and H
of 40 CFR 122 identified 73 of these municipalities as
large MS4s. Similarly, Appendices G and I of 40 CFR
122 identified 146 municipalities as medium MS4s. EPA
based these 219 identifications on the definition of a
municipal separate storm sewer system, which incorpo-
rates population data from the latest Decennial Census.
In this case, the 1980 Census helped identify the 219
MS4s. Recently, however, the results of the 1990 De-
cennial Census have become available and, conse-
quently, affect more municipalities. EPA is currently
drafting a Federal Register notice (FRN) that identifies
42 additional municipalities (30 cities and 12 counties)
that now meet the definition of a medium MS4 based on
the results of the 1990 Census. Sixty percent of the new
cities now required to seek NPDES permits are in the
state of California, while 33 percent of the new counties
are located in the state of Florida.
In contrast to the number of newly identified MS4s, the
1990 Census found that five cities and two counties
dropped in population to below 100,000. Although these
municipalities no longer satisfy the definition of a me-
dium MS4, two counties and one city still participate in
the stormwater program.
Next, the paper discusses municipalities that the ap-
pendices of 40 CFR 122 did not originally identify
but that nevertheless have been designated as
Phase I sources.
Designated MS4s
Section 402(p)(2)(E) and 40 CFR 122.21 (b)(4)(iii) and
(7)(iii) provide that permitting agencies may use their
authority in designating municipalities that operate
separate storm sewer systems and serve populations of
less than 100,000 as regulated MS4s. EPA has com-
piled some preliminary information on the number of
these municipalities, some of which are volunteering to
participate in the program. Based on the best informa-
tion available to date, it appears that states and EPA
regions designated small municipalities as regulated
MS4s primarily because they share common water-
sheds or are interconnected with a nearby regulated
MS4. In at least two states, EPA observed that all incor-
porated cities below a population of 100,000 were des-
ignated if they are within the boundary of a regulated
MS4 (county); therefore, these municipalities must sub-
mit a stormwater permit application. EPA is currently
trying to determine what permit application deadlines
have been established for these designated MS4s and
whether they are participating as coapplicants with a
regulated MS4 or are filing as single applicants.
Table 1 summarizes some preliminary data on the num-
ber of cities, counties, and special districts that have
either been designated or who are voluntarily participat-
ing in the program as Phase I stormwater sources.
EPA considers the figures presented in Table 1 prelimi-
nary because additional information is still pending
from three Regional Water Quality Control Boards
(RWQCBs). Some general observations, however, are
noteworthy. First, 65 percent of the designated cities in
Region 4 are located in the state of Florida. In the case
of the 47 designated special districts, 26 are state de-
partments of transportation, 11 are flood control districts,
Table 1. Summary of MS4 Designations by EPA Region
EPA
Region
1
2
3
4
5
6
7
8
9a
10
Total
Designated
Cities
0
0
13
236
1
0
1
1
127
1
380
Designated
Counties
0
0
5
9
0
0
0
0
7
1
22
Special
Districts
0
0
2
6
8
6
2
2
14
7
47
Additional information pending three RWQCBs in the state of California.
64
-------
four are state universities, three are port authorities, and
three represent a group of water control districts.
Effects of Combined Sewer Overflow
Exclusions
The NPDES stormwater regulations allow municipalities
to deduct the population served by combined sewer
systems from the total population served by the MS4. To
date, this provision has exempted 29 municipalities as
Phase 1 sources. An additional eight large MS4s have
been reclassified as medium MS4s. Table 2 provides a
breakdown of combined sewer overflow (CSO) exclu-
sions by EPA region.
Current Permit Applications
As noted earlier, the NPDES stormwater regulations
require MS4s to submit a two-part permit application.
Tahlo ^ nrnwiHoc the lotoct infnrmatinn awailahla nn the
Table 3. Summary of Part 1 and Part
2 Submissions by EPA
Region
EPA
Region
1
2
3
4
5
6
7
8
9a
10
Total
Medium
MS4s.
Parti
3
0
10
24
12
7
7
3
2
6
74
Medium
MS4s.
Part 2
0
0
0
0
0
0
0
2
0
0
4
Large
MS4s.
Part 1
0
5
11
20
5
9
3
3
4
4
64
Large
MS4s,
Part 2
0
5
10
15
5
7
1
1
3
2
49
number of submissions of Part 1 and Part 2 applications.
This table specifically excludes permit application sub-
missions for the states of California and Nevada.
The next section of this paper summarizes the results of
a recent EPA effort to document costs that MS4s have
incurred or are expected to incur over a 5-year period.
The information represents the most specific information
EPA has received to date on stormwater costs associ-
ated with the stormwater program.
Review of MS4 Program Cost Data
EPA recently conducted an analysis of Part 2 applica-
tions in an effort to gain a better understanding of costs
associated with implementing the municipal effort of the
stormwater program. EPA is currently completing a re-
view that documents the costs that 20 MS4s expect to
incur or have incurred as a result of implementing their
Table 2. Summary of CSO Exclusions by EPA Region
EPA
Region
1
2
3
4
5
6
7
8
9
10
Total
Medium
MS4s
5
7
2
0
6
0
0
0
0
1
21
Medium
to Large
MS4s
1
4
0
0
0
0
2
0
0
1
8
Large
MS4s
0
2
1
0
2
0
2
0
1
0
8
Total
6
13
3
0
8
0
4
0
1
2
37
3 California RWQCBs have issued permits for 130 applicants.
Information is still pending from three RWQCBs. The state of
Nevada has issued final permits for its regulated MS4s. Permit
application submission figures for EPA Region 9 reflect those
applications that are currently under review.
stormwater management programs. These costs are
based on fiscal information provided in Part 2 permit
applications. The primary purpose of this effort is to
assist EPA's Office of Water in determining the cost
burden that results from developing and implementing
programs in response to the NPDES stormwater regu-
lations. To that end, EPA has developed a preliminary
draft estimate for the total annual per capita cost to
develop and implement the stormwater management
program over a 5-year period. Some background infor-
mation on the analysis may provide a basis for better
understanding the results.
Applications Reviewed
EPA selected the Part 2 applications for this analysis
from among those that had been submitted to permitting
agencies by the November 16, 1992, deadline. EPA
selected municipalities located throughout the country
to obtain a more realistic representation of the cost data.
Thus, eight MS4s are located in the eastern part of the
United States, seven in the central part, and five in the
west. Selected municipalities also fall within eight of the
nine Rainfall Zones of the United States. The 20 munici-
palities reviewed are:
Aurora, Colorado
Baltimore, Maryland
Charlotte, North Carolina
Dallas, Texas
Denver, Colorado
65
-------
Fairfax County, Virginia
Harris County, Texas
Honolulu, Hawaii
Houston, Texas
King County, Washington
Lakewood, Colorado
Norfolk, Virginia
Philadelphia, Pennsylvania
Phoenix, Arizona
Prince Georges County, Maryland
Seattle, Washington
Tampa, Florida
Tucson, Arizona
Tulsa, Oklahoma
Virginia Beach, Virginia
Based on the 1990 Decennial Census, the combined
populations of these MS4s totaled over 11.3 million.
Fifteen percent of these MS4s have populations ex-
ceeding 1 million, 75 percent have populations between
250,000 and 1 million, and 10 percent have populations
of less than 250,000. With the exception of Aurora and
Lakewood, Colorado, all of these MS4s were previously
identified as large MS4s in the November 16, 1991,
Federal Register.
Grouping of Cost Data
This analysis broke down the actual and estimated costs
that MS4s reported in their applications into the follow-
ing eight major program components:
Public education
Monitoring
Commercial and residential
Construction
Industrial facilities
Maintenance of controls
Improper discharges
Miscellaneous
EPA selected these categories because they generally
reflect the variety of costs reported in the applications
and are largely consistent with the categories outlined
in the permit application regulations. Each of these eight
major categories were further subdivided into specific
program components. An underlying objective of this
effort was to determine the additional financial burden
the stormwater program imposed on municipalities.
Whenever possible, therefore, a breakout between new
and existing program costs was made for each reviewed
application.
Limitations
At this point, it is crucial to note some of the limitations
associated with this analysis. First and foremost are
limitations with the sample. Applications selected repre-
sented mostly large MS4s; therefore, EPA cannot be
certain that these results are fully representative of costs
that medium MS4s would report. Nearly 68 percent of
the regulated MS4s were not required to have submitted
their Part 2 applications at the time EPA conducted this
analysis. Consequently, this limits the availability of Part
2 applications that the analysis could have included.
One other important consideration with regard to the
sample selection is that the results may be overstated
in instances where MS4s are subject to more stringent
local and regional controls or other environmental initia-
tives for stormwater management.
The second limitation is that, in many instances, MS4s
did not include the cost of projects normally included
in a capital improvement program (CIP). Although
these projects often pertain to flood control, future CIP
projects typically will have features that also address
stormwater quality. Therefore, although providing the
additional benefit of improved stormwater quality may
be in response to the stormwater program, the analysis
results do not typically reflect these associated costs.
In contrast, EPA did not attempt to exclude significant
costs that MS4s reported for programs unreasonably
attributed to the stormwater program, even though they
probably would have existed regardless of the storm-
water program.
The third limitation reflects the difficulty in making direct
comparisons between applicants. The regulations pro-
vide flexibility to the MS4s with regard to proposing
stormwater management programs that reduce or elimi-
nate the contribution of pollutants in stormwater dis-
charges to the maximum extent practicable. The diverse
approaches to stormwater management that MS4s have
proposed reflect this flexibility. MS4s also used a variety
of methods to report annual cost data.
Inconsistencies that existed within individual applica-
tions account forthe fourth limitation. In many instances,
the text describing a proposed stormwater management
program component often did not correlate with the cost
information provided. For example, the application may
have indicated that an existing program would cover an
activity, but the fiscal analysis section of the application
did not provide the costs associated with the existing
program. Often, MS4s reported that an existing storm-
water management program was "absorbing" a new
66
-------
proposed program. The MS4s, however, provided no
separate fiscal data in the application.
Finally, the results of this analysis suggest that in a
number of instances MS4s both overreported and un-
derreported costs. EPA did not attempt to exclude any
reported costs from this analysis. Consequently, EPA is
attempting only to document average costs.
Results
Of the 20 MS4 applications reviewed, the average an-
nual reported cost for both new and existing programs
ranged from $211,000 or $0.76 per capita (Tampa Bay,
Florida) to $98 million or $190.85 per capita (Seattle,
Washington). Table 4 highlights the ranges of average
annual costs that municipalities reported.
Using population data from the 1990 Census, EPA cal-
culated a preliminary average annual per capita cost for
both new and existing programs of $23.91. Based on
information reported by MS4s, it appears that costs for
new programs or initiatives typically ranged from 10 to
15 percent of the average annual cost. As noted earlier,
EPA reviewed Part 2 applications mostly from large
MS4s. As medium MS4 applications become available,
EPA anticipates examining cost data from some of these
applications as well.
Programs the Part 2 Applications
Proposed
Having reviewed some of the cost data, this paper will
now present more specific details and examples of the
types of stormwater management programs proposed in
a number of Part 2 permit applications. The discus-
sions's structure follows the organization of the Part 2
application (e.g., adequate legal authority, source iden-
tification, characterization data, and management pro-
grams). The discussion's scope is confined to some
observations on a sample of eight Part 2 applications.
Legal Authority
According to the stormwater regulations, municipalities
must demonstrate that they possess the adequate legal
authority to implement their stormwater management
activities when they submit their Part 2 applications. In
Table 4. Ranges of Average Annual Costs Reported by
Municipalities
Average Annual Costs
Number of Municipalities
Less than $1,000,000
$1,000,000 to $5,000,000
$5,000,000 to $10,000,000
Greater than $10,000,000
the Part 2 guidance manual, EPA acknowledges that
this is not always possible if an MS4 lacks the enabling
legislative authority to develop the necessary ordinances.
In these cases, applicants need to provide a schedule
as to when adequate legal authority will be obtained.
Six municipalities stated that they had obtained the
adequate legal authority to carry out the requirements
of the stormwater regulations. One municipality antici-
pated having necessary legal authority by the spring of
1993, and one anticipated having the authority within 2
years. As a general note, municipalities reported existing
ordinances that addressed most of the legal authority
requirements of the regulations, especially with regard
to controlling improper discharges, illegal dumping, and
erosion and sediment control provisions. The comprehen-
sive nature of the stormwater regulations, however, re-
quired most municipalities to establish new ordinances
or update existing ones, particularly for obtaining the
necessary authority to conduct monitoring and surveil-
lance of stormwater discharges from private sources.
Several municipalities provided detailed excerpts or, in
some cases, the complete text of their comprehensive
stormwater ordinances. For example, Seattle, Washing-
ton, and Prince Georges County, Maryland, provided the
text of their grading, erosion, and control ordinances,
while King County, Washington, provided the text of both
its water quality ordinance and its pesticide regulation.
Ordinances of both Seattle, Washington, and Prince
Georges County, Maryland, addressed the require-
ments of the stormwater regulations in addition to other
local or regional initiatives, such as the Puget Sound
Water Quality Management Plan and the Chesapeake
Bay Preservation Act, respectively.
Source Identification
The principle requirement of the source identification
component of the Part 2 application is to identify any
previously unknown major outfalls and to compile an
industrial inventory. The industrial inventory must then
be organized on a watershed basis. Perhaps one of the
biggest challenges of the permit application is identifying
all major outfalls that comprise the storm sewer system.
Several MS4s reported using the analytical capabilities
of their geographic information systems (GISs) to iden-
tify potential locations of outfalls not previously identified
in the Part 1 application. A few applicants specifically
noted that this was a particularly effective approach.
Although a CIS is not a requirement of the stormwater
regulations, EPA recognizes that GISs are well suited
for many of the activities associated with stormwater
management. Out of the eight applications reviewed, at
least six reported having CIS capability, while one appli-
cant anticipated having CIS capability in the near future.
67
-------
Characterization Data
The characterization data portion of the Part 2 application
requires an MS4 to submit the results of wet weather
sampling with the application. More specifically, appli-
cants must submit sampling data for five to 10 outfalls
from at least three representative storm events. EPA has
not had an opportunity to conduct a detailed analysis of
this information. Some general observations, however,
follow.
First, although many of the applicants reported complet-
ing their wet weather sampling requirements, they typi-
cally expressed similar difficulties in doing so. MS4s
often noted that they had to sample several more than
the requisite minimum of three storm events to obtain
the number of requisite samples. In one instance, an
applicant reported that it took a total of 18 storm events
to obtain the requisite number of samples. Applicants
also frequently cited that they had to discard samples
because a particular storm's duration and rainfall accu-
mulation did not meet the requirements of a repre-
sentative storm event. Other problems commonly cited
included sampling during storm events with frequent
starts/stops and the logistics of mobilizing sampling
crews at the onset of a storm event. The unpredictability
of storm events and the logistics associated with wet
weather sampling prompted at least four of the eight
MS4s to use automatic samplers.
In at least one instance, an MS4 obtained approval to
use available historical data to satisfy the majority of
their sampling requirements. In this case, the applicant
needed to sample one additional storm event at two
sampling sites. Applicants often cited that concentra-
tion data compared well with the results of the NURP
study. In general, the eight MS4s reported that the
results of the analysis of composite samples exhibited
characteristic concentrations for metals such as cop-
per, cadmium, zinc, and lead. The sampling data also
suggest that the concentration of organic contami-
nants often fell below detection levels for composite
samples. Individual grab samples, however, detected
many organic contaminants.
The second major component of this portion of the appli-
cation requires the municipalities to estimate annual
pollutant loadings. EPA allows MS4s the flexibility of
selecting an appropriate method to estimate pollutant
loadings. A majority of the eight applicants elected to use
computer models such as SWMM, P8, and the COM
Nonpoint Source model to estimate annual loadings. Afew
applicants elected to use the simple method developed
by the Metropolitan Washington Council of Governments.
EPA expected that computing pollutant loadings would
satisfy at least two objectives. First, loading estimates
would raise the level of awareness within municipalities
of the relative magnitude of pollutant loadings associated
with stormwater discharges. Second, the estimates could
be used as part of a screening process when estab-
lishing priorities for stormwater management activities.
One applicant specifically noted using loading estimates
in this manner. Some applicants noted that these esti-
mates had limited value and that other means of repre-
senting sampling data would be more appropriate.
The Part 2 application requires applicants to maintain an
ongoing monitoring program for the duration of the term
of the permit. An approach proposed by the city of
Baltimore, Maryland, warrants special mention. Balti-
more proposed a comprehensive and phased approach
to monitoring which consists of four major components:
Dry weather stormwater outfall monitoring
Pollutant source tracking
Long-term trend monitoring
Stormwater runoff monitoring
The city identified the following six major goals to its
monitoring program:
Dry weather screening: This entails developing a
"water quality dry weather flow" database to assist in
isolating watersheds that may require further investi-
gation as potential sites of illicit connections.
Dry weather source tracking: This entails conducting
investigations to detect and eliminate sources of dry
weather flows.
Toxicity testing: A pilot toxicity testing program would
evaluate the impact of pollutants on a receiving water
ecosystem due to unknown contaminants and syner-
gistic effects.
Stream ecosystem database: A database that de-
scribes the biological integrity of the receiving
streams could assist in analyzing long-term trends,
prioritizing management practices, and assessing the
effectiveness of management programs.
Stormwater runoff and best management practice
(BMP) assessments: This effort could characterize
stormwater runoff quality and assess the effective-
ness of BMPs that may be used in the future.
Receiving stream water quality database: This en-
tails establishing dry and wet weather flow water
quality databases for major stream systems that can
be used for conducting long-term assessments and
determining the effectiveness of watershed manage-
ment programs.
The city's proposal to establish a stream ecosystem
database is particularly noteworthy because it would
provide the city with a baseline of its existing biological
community (e.g., benthic macroinvertebrate population
and diversity). It would also provide a basis from which
68
-------
to conduct a long-term assessment of the effectiveness
of watershed management activities. More importantly,
it would allow the opportunity to gain a greater under-
standing of the effects of stormwater discharges on a
specific aquatic habitat. Finally, the city is closely coor-
dinating its monitoring program with several subwater-
shed studies to determine the effectiveness of certain
BMPs in protecting receiving water quality, including
aquatic habitat.
Management Programs
Of course, the cornerstone of the two-part permit appli-
cation is the requirement that MS4s develop site-spe-
cific and comprehensive stormwater management
programs. Each applicant must address four major
areas in its application:
A description of structural and source control meas-
ures to reduce pollutants in runoff from residential
and commercial areas.
A description of procedures to detect and remove illicit
connections and a program to control improper disposal.
A description of structural and source control measures
to reduce pollutants in runoff from industrial areas.
A description of programs to maintain structural and
nonstructural BMPs to reduce pollutants from con-
struction sites.
In most instances, applicants elect to follow the applica-
tion format established in the November 16, 1990, Fed-
eral Register to describe their management programs.
From an initial review of eight applications, it appears
that many MS4s are proposing approaches that entail
phasing in components of their programs over the permit
term. Applicants not only cited economic reasons forthis
approach but also the desire to ensure that a particular
BMP is effective before it is implemented on a system-
wide basis. For example, several applicants reported
initiating studies to determine what factors significantly
influence the performance of a specific structural control
before its use on a systemwide basis. Pending the
results of these studies, applicants proposed modifying
their watershed management programs accordingly.
While a phased approach may be reasonable in some
instances, there are cases where the permitting author-
ity may not consider it appropriate.
In one of the reviewed cases, an applicant proposed a
phased approached to its illicit connections program.
Although EPA acknowledges the effort necessary to
detect and isolate the source of an illicit connection, a
phased approach appears to overlook the immediate
benefits of a fully implemented illicit connections pro-
gram. This is especially true for municipalities in densely
populated urban corridors that have both separate and
combined sewer systems.
Implementing a comprehensive stormwater management
program is a complex effort that requires the participa-
tion of numerous inter- and intragovernmental agencies.
Before implementing a program, a municipality needs to
establish program priorities. It may be helpful at this
point to briefly illustrate one applicant's approach to
establishing criteria for prioritizing basins for watershed
management activities.
In 1987, King County, Washington, completed a "Basin
Reconnaissance Program" that provided the information
necessary to establish an initial basin planning prioriti-
zation scheme. The county provided a complete set of
the results of this effort with its Part 1 application. King
County established four major prioritization categories
with commensurate criteria for each category. The major
categories and criteria are as follows:
Existing problems
- Landslides
- Erosion/Sediment
- Flooding
Future problems
- Unincorporated land in King County
- Subdivision/Plat activities
- Population growth
- Permitted residential units
Existing resources
- Stream habitat
- In-stream resources
- Wetland value
- Wetland storage potential
- Water quality potential
Urgency/Timeliness
- Other Agency interest
- Opportunity to integrate with other programs
For all 37 basins identified, King County assigns a nu-
merical rating to each criterion and a composite score
for each major category, then establishes a total basin
numerical rating. After completing basin prioritization
ranking, the county proceeds with a six-step basin plan-
ning process. The first step is the formation of a basin
plan team consisting of a project manager, biologists,
geologists, water quality specialists, engineers, re-
source planners, mapping and CIS technicians, and
graphics support. In the next step, the team collects data
that include information on rainfall, flow levels, geologi-
cal makeup, geomorphology, habitat complexity and di-
versity, fish utilization, and water quality. The basin plan
team may spend up to 2 years compiling data.
The third and fourth steps entail computer modeling of
a basin's hydrology and predicting the effects of alterna-
69
-------
tive land-use activities. The results of the modeling ef-
forts assist in developing a current and future conditions
report that documents existing conditions and provides
an analysis of future trends.
The fifth step entails drafting a basin plan and conduct-
ing public meetings and hearings. After necessary modi-
fication, the team finalizes the draft plan and submits it
to the King County Council for approval. Following ap-
proval, the King County Surface Water Management
(SWM) Division is responsible for implementing the ba-
sin plan. King County SWM anticipated completing 12
of its 37 basin plans by the end of 1992.
The King County basin planning program reflects a
resource-intensive effort and a commitment to reducing
the deleterious effects of stormwater discharges. Mu-
nicipalities that are essentially new to stormwater man-
agement may find elements of King County's program
not only innovative and informative but also adaptable
to their needs.
MS4s proposed some general observations about par-
ticular program components. First, a majority of the
applications placed a heavy emphasis on minimizing
future problems associated with stormwater manage-
ment, specifically in the area of long-term planning for
future development. In several instances, MS4s re-
ported that they had either completed or initiated the
development of stormwater management master plans
for major watersheds.
Also, MS4s are increasingly requiring approval of ero-
sion and sediment control plans before approving a site
plan or allowing construction to begin. Similarly, many
MS4s require permanent BMPs (privately financed),
such as installation of retention/detention basins for all
new developments over a certain size area. MS4s also
frequently reported that inspections programs had been
or are being established to ensure maintenance of pub-
licly and privately owned BMPs over their useful life. In
at least one instance, an MS4 provides an economic
incentive to install BMPs by establishing a BMP credit-
ing system for non-single-family residences.
A couple of applicants also reported a substantial commit-
ment to preserving open space. In one case, a munici-
pality reported that it is pursuing a "Greenways" program
that could potentially preserve 16,000 acres as open
space. To date, 400 acres have been preserved. Simi-
larly, one county has established a stream valley park
system. All major streams in the county are to become
part of the park system. In this instance, the county has
imposed an additional requirement: new development
must provide for buffer zones or easements.
Over the long term, approaches like these may minimize
the need to construct costly structural controls to remove
pollutants from stormwater discharges. Moreover, this
preventative approach to stormwater management
can potentially reduce the significant costs that some
municipalities are incurring to restore degraded stream
corridors and wetlands. EPA recognizes that this is a
contentious issue. It is encouraging to note, however,
the emphasis municipal applicants are placing on
community involvement and public outreach programs.
The "adopt-a-stream" program and other similar com-
munity-based environmental programs, such as
household hazardous waste collection, routinely ap-
peared in Part 2 applications.
Paraphrasing one applicant's comment, the goals of a
stormwater management program cannot be fully
achieved unless there is participation and consensus
among those who are affected. Otherwise, past prac-
tices will continue to have a detrimental influence on
valuable water resources within our communities.
Current EPA Activities in the Area of MS4
Permitting
Several EPA regions and state permitting authorities
have supported the formation of an MS4 steering com-
mittee to look at specific issues pertinent to MS4 per-
mits. The steering committee is looking at program
components and permits that may be suitable as model
programs or model permits. It also will assist in deter-
mining how to incorporate core elements of a stormwa-
ter program into an MS4 permit. Lastly, the steering
committee will be exploring alternative mechanisms of
exchanging information on stormwater management.
The committee will coordinate this particular effort with
ongoing outreach activities at EPA.
EPA also is conducting a municipal assessment project
(MAP) that continues to examine the progress of the
municipal permitting process. This entails compiling in-
formation on the status of both permit applications and
permit development. Whenever possible, EPA will sug-
gest future improvements or enhancements to the MS4
permitting process. EPA is continuing to compile infor-
mation on MS4s designated by state permitting agen-
cies and EPA regions. Other objectives of the MAP
include examining the Part 2 applications in more detail
to identify programs as potential model candidates.
As the permitting process moves from the develop-
ment of permit applications to permit development, EPA
anticipates distributing information on the progress of
permit development to permitting authorities. Hopefully,
this approach will benefit all those participating in the
permitting process.
70
-------
Municipal Stormwater Permitting: A California Perspective
Thomas E. Mumley
California Regional Water Quality Control Board, San Francisco Bay Region,
Oakland, California
Abstract
The California Regional Water Quality Control Board,
San Francisco Bay Region (Regional Board), began a
program for control of stormwater discharges from ur-
ban areas in 1987. The initial focus of the program has
been on the municipalities in Santa Clara and
Alameda counties. An areawide approach was pro-
moted in which all the cities in each county, the
county, and the county flood control agency worked
collectively. The Santa Clara and Alameda programs
were issued municipal stormwater National Pollutant
Discharge Elimination System (NPDES) permits in
June 1990 and August 1991, respectively. These ef-
forts have focused on implementation of stormwater
management programs rather than on the NPDES
permit itself. Essentially, the permit serves as an en-
forceable mechanism requiring implementation of the
programs developed by the municipalities and ap-
proved by the Regional Board.
The municipal stormwater management programs all
involve similar elements, including public informa-
tion/participation, elimination of illegal discharges, pub-
lic agency activities, control of industrial/commercial
stormwater discharges, new development manage-
ment, stormwater treatment, program evaluation, and
monitoring. The process of developing these programs
has uncovered several issues and problems, mostly
nontechnical, which could potentially impede successful
implementation. On the other hand, workable solutions
to most of these problems have also been identified. The
essential ingredient of the process that has enabled
progress has been a cooperative, proactive relationship
between the Regional Board and municipalities. Con-
tinuation of this process is expected to result in a real-
istic and meaningful municipal stormwater NPDES
permit program.
Background
The California Regional Water Quality Control Board,
San Francisco Bay Region (Regional Board), is the
state water pollution control agency responsible for pro-
tection of San Francisco Bay and its tributaries. San
Francisco Bay is a highly urbanized estuary and as such
receives significant loads of pollutants through dis-
charges of urban runoff. The responsibilities of the Re-
gional Board include water quality control planning,
control of nonpoint sources of pollution, and issuance
and enforcement of NPDES permits. Using its authori-
ties, the Regional Board began a program for control of
stormwater discharges from urban areas in 1987. The
initial focus of the program was on the most highly
urbanized areas, which include the municipalities in
Santa Clara and Alameda counties. An areawide ap-
proach was promoted in which all the cities in each
county, the county, and the county flood control agency
worked collectively.
Santa Clara and Alameda counties developed their pro-
grams through a strategic planning process (1). The
process followed a series of steps that involved estab-
lishing program goals and framework; compiling existing
information; assessing water quality problems through
collection and analysis of data and modeling of pollutant
loads; identifying, screening, and selecting appropriate
control measures; and establishing a plan for implemen-
tation. This planning process lead to development of a
comprehensive stormwater management plan by each
program (2, 3). In addition, institutional arrangements,
legal authorities, and fiscal resources for implementa-
tion were addressed.
The efforts of the Regional Board and the Santa Clara
and Alameda municipalities were well underway when
the stormwater National Pollutant Discharge Elimination
System (NPDES) permit regulations were promulgated
71
-------
in November 1990. The Regional Board found the infor-
mation that the planning process followed by the two
areawide programs provided was equivalent to federal
permit application requirements. Consequently, the Re-
gional Board issued municipal stormwater NPDES per-
mits to the Santa Clara and Alameda programs in June
1990 and August 1991, respectively, which required
implementation of their stormwater management plans.
Issuance of these "early" permits served to recognize
the accomplishments of the two programs and to pro-
vide a focus on implementation actions while avoiding
the time delays and costs associated with the promul-
gated application requirements. We also have focused
attention on the adequacy and effectiveness of the
stormwater management plans rather than the permits.
Essentially, the permit serves as an enforceable mecha-
nism requiring implementation of the programs devel-
oped by the municipalities and approved by the
Regional Board.
The efforts of the Santa Clara and Alameda municipali-
ties have provided a meaningful framework for and the
essential elements of an effective stormwater manage-
ment program. A similar approach is being followed by
municipalities in the other urban areas of the San Fran-
cisco Bay region. The process of developing these pro-
grams has uncovered several issues and problems,
mostly nontechnical, which could potentially impede
successful implementation. On the other hand, work-
able solutions to most of these problems have also been
identified. The following discussion provides a status
report of the San Francisco Bay programs, a description
of the elements of the stormwater management pro-
grams, and insight into the problems encountered and
their solutions.
San Francisco Bay Region Municipal
Stormwater Programs
In the San Francisco Bay region, nearly all municipali-
ties in urban areas have stormwater management pro-
grams and NPDES permits under way or under
development. The Regional Board has encouraged, rec-
ognized, or required areawide programs in which all
municipalities within a watershed or municipal systems
that interconnect are managed under one program. In
addition, municipal flood management agencies are in-
cluded as co-permittees. The California Transportation
Department (Caltrans) is required to implement a storm-
water management program for all storm drain systems
within the region. The municipal stormwater programs
in the San Francisco Bay region are listed below.
Santa Clara Valley Nonpoint Source Pollution Control
Program, including the county and all cities:
- Population approximately 1,500,000
- NPDES permit issued June 1990
Alameda County Urban Runoff Clean Water Program
including the county and all cities:
- Population approximately 1,250,000
- NPDES permit issued October 1991
Contra Costa Cities, County, District Stormwater Pol-
lution Control Program including the county and all
cities:
- Population approximately 800,000
- Part 1 Application submitted May 1992
- Part 2 Application due May 1993
San Mateo County Urban Runoff Clean Water Pro-
gram, including the county and all cities:
- Population approximately 650,000 (no city nor the
county has population more than 100,000)
- Combined Parts 1 and 2 Application due May 1993
Caltrans, including all operation, maintenance, and
construction activities:
- Incomplete application submitted July 1992
- Complete application due May 1993
City of Vallejo:
- Population more than 100,000 (as of 1990 Cen-
sus)
- Part 1 Application due March 1993
- Part 2 Application due March 1994
Cities of Fairfield and Suisun City Joint Program:
- Population more than 100,000
- Part 1 Application due March 1993
- Part 2 Application due March 1994
Municipal Stormwater Program Elements
The municipal stormwater management programs all
involve similar elements except for Caltrans, which will
not be discussed here. These include public informa-
tion/participation, elimination of illegal discharges, pub-
lic agency activities, control of industrial/commercial
stormwater discharges, new development manage-
ment, stormwater treatment, program evaluation, and
monitoring. The activities associated with each of these
essential program components are presented below.
Public Information/Participation
This element is considered the most important early
action and is the cornerstone of effective pollution pre-
vention. Its objectives are to inform the public, commer-
cial entities, and industries about the proper use and
disposal of materials and waste and to correct practices
of stormwater runoff pollution control. Activities include
development of general and focused information mate-
rials and public service announcements. Participation
72
-------
activities include citizen monitoring programs, stenciling
of storm drain inlets with no dumping signs, and organ-
ized creek cleanups.
Elimination of Illegal Discharges
Elimination of illicit connections to the storm drain sys-
tem and the prevention of illegal dumping are other
essential early action elements. The objective is to en-
sure that only stormwater or otherwise authorized dis-
charges enter storm drains. Activities include inspection
of storm drain outfalls, surveillance of storm drain sys-
tems, and enforcement actions.
Public Agency A ctivities
Many public agency activities affect stormwater pollu-
tion. Some activities prevent or remove stormwater pol-
lution, while other activities are sources of pollution.
The objective of this element is to ensure that routine
municipal operations and maintenance activities are
initiated or improved to reduce the likelihood that pol-
lutants are discharged to the storm drain system.
Activities include street sweeping; maintenance of
storm drain inlets, lines and channels, and catch ba-
sins; corporation yard management; and recycling
programs. Coordination of road maintenance and
flood control activities with the stormwater manage-
ment program is also included.
Control of Industrial/Commercial Stormwater
Discharges
Industrial and commercial sources may contribute a
substantial pollutant loading to a municipal storm drain
system. The objective of this element is to identify and
effectively control industrial and commercial sources of
concern. Activities include compiling a list of industrial
and commercial sources, identifying appropriate pollu-
tion prevention and control measures, and inspecting
facilities. The focus is not only on facilities associated
with industrial activity as defined in the stormwater regu-
lations but on any facility that conducts industrial activi-
ties, as well as commercial facilities such as automotive
operations and restaurants. This effort is expected to
complement federal and state industrial stormwater per-
mitting efforts.
New Development Management
Areas of new development and redevelopment offer the
greatest potential for implementation of the most effec-
tive pollution prevention and control measures. The ob-
jective of this element is to reduce the likelihood of
pollutants entering the storm drain system from areas of
new development and significant redevelopment, both
during and after construction. Activities include review
of existing local permitting procedures and modification
of the procedures to identify and assign appropriate site
design, erosion control, and permanent stormwater con-
trol measures.
Stormwater Treatment
The initial focus of the stormwater management pro-
grams is on pollution prevention and source control.
Treatment of stormwater is expected to be a costly
alternative. There may be opportunities, however, for
installation or retrofitting of structural controls. The
objectives of this element are to study the various
treatment alternatives available, to test the feasibility
of conducting the activities, and to determine the ef-
fectiveness of the treatment through pilot-scale pro-
jects. Initial focus has been on existing wetland
systems, flood control detention basins, and treat-
ment of parking lot runoff.
Program Evaluation
Stormwater management programs are expected to
change as they mature. Consequently, they should
have built-in flexibility to allow for changes in priori-
ties, needs, or levels of awareness. The objective of
this element is to provide a comprehensive annual
evaluation and report of program effectiveness. Meas-
ures of effectiveness include quantitative monitoring
to assess the effectiveness of specific control meas-
ures and detailed accounting of program accomplish-
ments and funds and staff hours expended. The
annual report provides an overall evaluation of the
program and sets forth plans and schedules for the
upcoming year. The annual report is considered a
program's self audit and provides a mechanism to
propose modifications to the stormwater management
plan in response to program accomplishments or fail-
ures. The annual report also serves as the key regu-
latory tool for providing accountability and public
review in accordance with the NPDES permit.
Monitoring
Monitoring is an essential component of any pollution
control program. The objectives are to obtain quantita-
tive information to measure program progress and ef-
fectiveness, to identify sources of pollutants, and to
document reduction in pollutant loads. The success of
a monitoring program can be measured by the ability to
make more informed decisions on a program's direction
and effectiveness. Monitoring activities include baseline
monitoring of storm drain discharges and receiving wa-
ters and focused special studies to identify sources of
pollutants and to evaluate the effectiveness of specific
control measures. Types of monitoring include water
column measurements, sediment measurements, and
nonsampling and analysis measurements, such as
number of outfalls inspected or amount of material re-
moved by maintenance. Toxicity identification evalu-
73
-------
ations are an integral component of monitoring pro-
grams in the San Francisco Bay area.
Municipal Stormwater Program Problems
The process of developing these programs has uncov-
ered several issues and problems, mostly nontechnical
in nature, that could potentially impede successful im-
plementation. The first step towards avoiding or solving
these problems is understanding what they are and how
they may affect a program. The following discussion
provides insight into the more common problems.
Internal Agency Coordination
Municipalities are public agencies, often with multiple
departments serving different functions, that are an in-
tegral part of stormwater management. The missions
and actions of separate departments are often carried
out without coordination with other departments. Com-
mitments or actions by planning department personnel
that are not coordinated with public works result in prob-
lems. All affected departments must participate in devel-
opment of a stormwater management program. The
stormwater program plan also must clearly identify the
roles and levels of participation of all involved depart-
ments.
External Agency Coordination
In addition to coordination within a municipality, commu-
nication and coordination is necessary between adja-
cent cities, the county, and regional organizations such
as flood control and wastewater treatment agencies.
Historically, there may have been little need for coordi-
nation, or problems encountered by other programs may
have created barriers. As with the internal agency issue
noted above, all affected agencies must participate in
the program development process and clearly under-
stand their implementation responsibilities.
Resistance by Key Individuals
Individuals play a strong role in local government. Con-
sequently, one or more key individuals can make or
break a program. Often one individual causes the inter-
nal and external coordination problems noted above.
Also, in the early development stages of a program, until
dedicated personnel are identified, individuals may re-
sist the additional work load required of them to make
the program work.
Financial Resources
Without dedicated financial resources, a stormwater
management program is destined to fail. Programs that
do not start the process to secure dedicated funds early
in program development find themselves unable to com-
mit to a meaningful program. The process of estab-
lishing a stormwater utility, assessment district, or other
funding mechanisms is cumbersome and requires stra-
tegic planning.
Legal Issues
Initial review of existing local ordinances may result in
the conclusion that sufficient legal authorities already
exist. Later on in the development process, however,
when specific implementation activities are identified,
the existing authority may be found to be too vague or
unsuitable. Review of legal issues should be part of the
annual evaluation process.
Competing Mandates
Mandates by other programs within a municipality or by
external agencies may directly conflict with stormwater
program mandates. Examples include fire departments
prohibiting inside or covered storage of certain materi-
als or the obvious conflict between eradication of vege-
tation with herbicides in flood control channels and
water quality concerns.
Problem Awareness/Understanding
To solve or manage a problem, one must first under-
stand the problem. Effective pollution prevention re-
quires a new way of thinking that may be foreign to those
accustomed to more conventional engineering solu-
tions. A subset of this issue involves those who deny that
a problem exists.
Resistance to Maintenance Responsibility
Municipal programs are expected to result in installation
of some structural controls, particularly in areas of new
development or significant redevelopment. A frequently
encountered barrier is that municipalities are not willing
to take on the additional maintenance responsibility as-
sociated with new structural controls.
Problem Sources Beyond Municipal Authority
Many sources of stormwater pollution involve atmos-
pheric emissions, automobile wear (e.g., brakes, tires),
and household products over which a municipality has
no control. Transportation related issues are beyond the
control of a single municipality. State and federal coor-
dination with local programs is essential.
Lack of Tools To Evaluate Effectiveness
The effectiveness of pollution prevention measures is
difficult to quantify. Natural variability in stormwater qual-
ity may mask improvements associated with certain
control measures. Surrogate measures and analytical
tools to evaluate stormwater management program ef-
fectiveness should be better defined.
74
-------
Municipal Stormwater Program Solutions
The efforts of the Regional Board and the municipalities
in the San Francisco Bay area have overcome many of
the problems noted above. The essential ingredient of
the process that has enabled progress has been a
cooperative, proactive relationship between the Re-
gional Board and municipalities. A discussion of some
of the solutions that have evolved follows.
Carrot and Stick Approach
At the onset of each new municipal program, the Re-
gional Board has made it clear that stormwater pollution
is a serious problem that must be dealt with and that the
best solutions will only happen at the local level. The
carrot has been an offer to the municipalities to control
their own destinies rather than waiting for the powers
that be in Sacramento or Washington to determine what
they can or cannot do. This approach allows the munici-
palities to identify and select the measures that are
workable for them and, most importantly, that are most
cost-effective. On the other hand, the Regional Board
has also made it clear that participation is not voluntary
and that failure to commit to meaningful actions will
result in enforcement actions.
Round Table Forum
Contrary to the conventional regulatory approach, in
which the regulator demands and the regulatee reacts,
the Regional Board has promoted a round table forum
in which all involved parties work collectively and coop-
eratively to identify solutions that address the concerns
and means of all involved. This approach has also pro-
vided a mechanism for participation by all affected inter-
nal and external public agencies.
Regular Meetings
The Regional Board has met in the round table format
with municipalities throughout the program development
process. Meetings have been held at least monthly. This
has allowed for timely and effective decision-making.
Focused work groups to address specific problems or
program elements have also been formed.
Minimization of Bureaucracy
The stormwater pollution problem is not a conventional
problem that can be solved by conventional means. Any
program is doomed to fail if it is mired in red tape. To
promote innovative solutions, the regulators must be
willing to promote innovative regulatory mechanisms.
Flexibility
To truly present a carrot to entice municipalities and
promote innovative solutions, the regulator must be will-
ing to be flexible. No one solution exists for stormwater
pollution problems. What works in one municipality may
not work in another. Also, flexibility provides a reward
mechanism for those municipalities who are committed
and proactive.
Phased Approach
The phased approach promotes a strategy based on
goal setting, identification of actions, planning and
preparation for planned actions, small-scale implemen-
tation, and finally full-scale implementation. Evaluation
is essential to each step. It must be recognized that
some actions may be implemented immediately or in the
short term, while others may take many years to fully
implement.
Pilot Studies
Although many control measures have been demon-
strated to be effective, such measures often need test-
ing within the conditions of a specific municipality. Pilot
studies also provide an opportunity to identify factors
such as operation and maintenance parameters or non-
technical factors such as legal issues that may not be
apparent. They also provide a mechanism for demon-
strating acceptability to concerned parties and should
be considered a first step leading to successful wide-
scale implementation.
Annual Program Audit
Recurring evaluation is essential. At a minimum, pro-
gram participants and the regulator should annually
evaluate program progress. This comprehensive annual
audit should identify program successes as well as fail-
ures and should provide a mechanism to steer the pro-
gram in the most effective direction.
Conclusions
Focusing on the described municipal stormwater program
elements and taking a cooperative approach to solving
problems have led to the development of successful
stormwater management programs by municipalities in
the San Francisco Bay area. Although program implemen-
tation is in the early stages and total success cannot be
claimed, the programs are successful in that they present
a workable framework for implementation of meaningful
actions. Essential to the process is strategic planning,
accountability, and recurring evaluation of program direc-
tion, success, and failure.
The NPDES permit issued to a municipality is not going
to solve the stormwater pollution problemit can only
serve as a tool to facilitate action. The success of the
municipal stormwater permit program will be recognized
when municipalities are committed to action, and
NPDES permits merely require municipalities to do what
they have committed to do.
75
-------
RGfGTGnCGS 2. Santa Clara Valley Nonpoint Source Pollution Control Program.
1991. Stormwater management plan.
1. Mumley, I.E. 1993. Urban runoff pollution prevention and control 3. Alameda County Urban Runoff Clean Water Program. 1991.
planning, San Francisco Bay experiences. In: Proceedings of the Stormwater management plan.
U.S. EPA National Conference on Urban Runoff Management,
Chicago, IL.
76
-------
Stormwater Management Ordinance Approaches in Northeastern Illinois
Dennis W. Dreher
Northeastern Illinois Planning Commission, Chicago, Illinois
Abstract
Stormwater drainage and detention is widely regulated
by local ordinances in northeastern Illinois. Early ordi-
nances, going back to about 1970, focused exclusively
on the prevention of increased flooding and nuisance
drainage problems. Recent ordinances address the ob-
jectives of preventing flooding and channel erosion, pre-
serving predevelopment hydrology, protecting water
quality and aquatic habitat, providing recreational op-
portunities, and enhancing aesthetic conditions.
The basis for many of the newer ordinances is a model
ordinance developed by the Northeastern Illinois Plan-
ning Commission. The "Model Stormwater Drainage
and Detention Ordinance" calls for "natural" drainage
practices to minimize increases in runoff volumes and
rates and for detention basins that control the full range
of flood events and effectively remove Stormwater pol-
lutants.
The model ordinance requires detention designs that
limit the 100-year release to 0.15 ft3/sec/acre and the
2-year release to 0.04 ft3/sec/acre. These rates are
actually lower than the local predevelopment runoff
rates and are based on observed capacities of the
downstream channel system. Detention design also
must incorporate water quality mitigation features, in-
cluding permanent pools or created wetlands, stilling
basins, and the ability to avoid short-circuiting. Further,
the model ordinance strongly discourages detention in
onstream locations or in existing wetlands.
As multipurpose ordinances are implemented, several
issues remain. Some municipal officials are concerned
about the aesthetics and maintenance needs of wet-
land-type detention basins and natural drainage prac-
tices, such as vegetated swales. Technical debate
continues over the effectiveness of on-line and on-
stream detention, both from a water quality and flood
prevention perspective. Also, the appropriateness of us-
ing existing wetlands for Stormwater detention remains
to be determined.
History
Stormwater drainage and detention has been widely
regulated by local ordinances in northeastern Illinois
since the early 1970s. Early ordinances were imple-
mented because of a recognition that rapid suburban
development was causing more frequent and more
damaging flooding and drainage problems. Flooding
and drainage problems in the region are exacerbated by
the very flat landscape; typical ground slopes range
from 0.5 to 4 percent. As a result, even a slight increase
in flood volumes and rates can expose large additional
areas to flooding.
Most early ordinances required storage of the 100-year
rainfall event. These ordinances were based on require-
ments developed by the Metropolitan Water Reclama-
tion District of Greater Chicago (MWRDGC). MWRDGC
requires sewer permits for new development within
Cook County, the largest and most populous in the
six-county northeastern Illinois region. Many communi-
ties in the outer "collar" counties followed MWRDGC's
lead and developed similar ordinances.
At the same time that municipalities began to implement
Stormwater detention controls for new development,
most also required via subdivision ordinances that new
development be drained by curb and gutter and storm
sewer systems. This drainage philosophy was intended
to reduce local drainage problems but resulted in in-
creased rates and volumes of runoff.
The quality of urban runoff began to receive some atten-
tion in the late 1970s with the completion of the
Areawide Water Quality Management Plan by the North-
eastern Illinois Planning Commission (NIPC) (1). This
plan reported much higher pollutant loads for urban
land-use categories compared with rural land uses. As
a consequence, the plan recommended that Stormwater
loadings of suspended solids and biological oxygen de-
mand (BOD) be reduced by 50 percent by appropriate
best management practices (BMPs) for all new devel-
opment. Despite the recommendations of the plan, few
77
-------
changes occurred in the stormwater management strat-
egy of local governments, which addressed exclusively
the quantity of runoff but not the quality.
Assessment of Ordinance Effectiveness
In 1986 and 1987, large areas of northeastern Illinois
were besieged by major floods, with total damage esti-
mates exceeding $100 million. In some locales, flood
flows exceeded the reported 100-year frequency event.
Of particular concern was the observation that large
flood damages had occurred in watersheds that had
developed extensively since the implementation of de-
tention ordinances in the early 1970s. This lead to the
suspicion that detention was not preventing increases
in flood flows.
To address these concerns, NIPC was funded by the
Illinois Department of Transportation, Division of Water
Resources, to investigate the effectiveness of existing
stormwater detention ordinances. First, a literature re-
view was performed to assess the effectiveness of de-
tention in various locales around the country. Next, a
comprehensive watershed modeling study was per-
formed to evaluate both the effects of urbanization and
a range of existing and proposed stormwater detention
controls. The study concluded that the detention stand-
ards that most communities required were not adequate
to prevent increases in flooding due to new development
(2). Other local studies initiated by the Soil Conservation
Service reached similar conclusions (3). Several spe-
cific weaknesses were identified:
Detention volumes were inadequate to store the in-
tended 100-year design event due to outdated rainfall
statistics and/or simplistic hydrologic design tech-
niques.
Required 100-year release rates were typically based
on site predevelopment runoff rates rather than ob-
served instream flood flow rates.
Because detention outlets were designed to explicitly
control only the 100-year event, smaller flood events
(e.g., the 2-year event) typically passed through de-
tention facilities with inadequate control.
The study also noted two problems in addition to flood-
ing impacts. The first was increased stream channel
erosion, caused in part by the increased magnitude and
frequency of small floods. The second was water quality
impairment due to inadequately controlled urban runoff.
New Model Ordinance Approach
With the preceding problems in mind, NIPC was con-
tracted to develop an updated model stormwater ordi-
nance. This "Model Stormwater Drainage and Detention
Ordinance" (4) was developed with the assistance of a
regionwide, multiagency technical advisory committee.
The primary purposes of the ordinance are to minimize
the stormwater-related effects of development on down-
stream and local flooding, stream channel erosion,
water quality, and aquatic habitat.
The model ordinance is intended to apply to all devel-
opment, including redevelopment. It requires the sub-
mittal of a basic drainage plan consisting of a
topographic map, a detailed description of the existing
and proposed drainage system, and a description of
sensitive environmental features such as wetlands. An
advanced drainage plan is required for sites larger than
10 acres. The advanced plan should include flow rates,
velocities, and elevations at representative points in the
drainage system for events up to the 100-year. The
following are some important ordinance standards and
criteria:
Runoff reduction hierarchy: The ordinance requires
the evaluation of site design practices that minimize
the increase in runoff volumes and rates. A prefer-
ence is stated for, in order, minimization of hydrauli-
cally connected impervious surfaces, use of open
vegetated swales and channels and natural depres-
sions, and infiltration practices. Traditional storm
sewer approaches are discouraged unless other
measures are not practical.
100-year release rate: The peak 100-year discharge
should not exceed 0.15 ft3/sec/acre. This release rate
is related to the capacity of the downstream channel/
floodplain system for extreme flood events. The refer-
enced detention effectiveness evaluation indicated that
this release rate should prevent development-related
increases in flooding for watersheds up to at least 30
square miles in size (and probably much larger).
2-year release rate: The peak discharge for events
up to the 2-year event should not exceed 0.04
ft3/sec/acre. This release rate is designed to minimize
increases in the magnitude and frequency of the in-
stream 2-year event, which is sometimes associated
with bankfull flow conditions. This requirement is in-
tended to minimize increases in stream channel ero-
sion. This release rate also will provide extended
ponding for small storm events, which will enhance
pollutant removal.
Detention storage requirements: The design maxi-
mum storage should be based on the runoff from the
100-year, 24-hour event. Storage should be com-
puted based on hydrograph methods, such as TR-55
or TR-20. Design rainfall should be based on the
Illinois State Water Survey's Bulletin 70 (5), which
supersedes the U.S. Weather Bureau's Technical Pa-
per No. 40 (6). Bulletin 70, which is based on a
precipitation database that is more extensive and
more current, reports a 100-year, 24-hour rainfall of
7.6 in., while Technical Paper 40 recommends 5.8 in.
78
-------
Water quality design features for detention: The or-
dinance indicates a preference for wet detention
basins over dry extended detention facilities to maxi-
mize pollutant removal potential. For wet basins, the
ordinance includes design criteria for depths, shore-
line slopes, permanent pool volume, and inlet/outlet
orientation. For dry extended detention basins, the
ordinance includes design criteria for velocity dissi-
pation at inlets and inlet/outlet orientation.
Detention in floodways and stream channels: The
ordinance discourages detention in designated flood-
ways, particularly in onstream locations with upstream
drainage areas larger than about 1 to 2 square miles.
The principal concerns with onstream detention are
that it may be less effective in mitigating stormwater
pollutants and it allows stormwater pollutants to be
discharged into stream channels without adequate
pretreatment.
Detention in wetlands: Use of existing wetlands to
accommodate stormwater detention requirements is
strongly discouraged. The ordinance requires that all
stormwater be stored and routed through a 2-year
water quality detention facility (consistent with the
previous design criteria) before being discharged to
a wetland. The ordinance allows additional storage,
up to the 100-year event, to be provided in a wetland
if it can be shown that the wetland is low in quality
and that proposed detention modifications will main-
tain or improve its habitat and other beneficial func-
tions.
Overall, the new model ordinance is one of the most
stringent in the country in its storage and release rate
requirements for minimizing the effects of development
on downstream flooding. The new ordinance also in-
cludes, for the first time, some basic requirements for
BMPs to mitigate stormwater quality effects.
Recent Improvements in Local
Stormwater Regulations
As an advisory agency, NIPC has no authority to require
compliance with its model ordinances. Similarly, there is
no comprehensive state requirement for local stormwa-
ter regulations. Because of recent experience with dev-
astating floods, however, many communities were
eager to consider alternatives to stormwater standards
that were a decade or more old.
The process of evaluating new ordinances was facili-
tated by state legislation, passed after the floods of
1986 and 1987, that authorized northeastern Illinois
counties to establish stormwater management committees
(SMCs). These committees, with equal representation
from county government and municipalities, were
authorized to develop comprehensive, binding storm-
water management plans. These plans included both
watershed-based flood remediation measures as well
as uniform, countywide stormwater regulations.
So far, comprehensive countywide ordinances have
been implemented in two counties, DuPage (7) and
Lake (8). These ordinances address traditional storm-
water drainage and detention concerns as well as flood-
plain management, soil erosion and sediment control,
and stream and wetland protection. The ordinances
incorporate many standards from the NIPC models and
address multipurpose objectives of preventing flooding
and channel erosion, preserving predevelopment hy-
drology, protecting water quality and aquatic habitat,
providing recreational opportunities, and enhancing
aesthetic conditions. Probably the most remarkable ele-
ment of these new ordinances is their inclusion of some
basic stormwater BMPs that are intended to address
both stormwater quantity and quality concerns.
Countywide stormwater planning efforts also have be-
gun in Cook, Kane, and McHenry Counties. Many com-
munities in these counties have individually begun to
update their ordinances. Some of the impetus for ordi-
nance updates has come from watershed-based
groups, such as the Butterfield Creek Steering Commit-
tee. This group developed a comprehensive ordinance
for seven watershed communities all faced with similar
problems of overbank flooding, stream channel erosion,
and water quality degradation (9).
Other communities are updating ordinances based on
requirements of the Illinois Environmental Protection
Agency (IEPA) as a condition for facility planning area
amendments for expanded wastewater service. These
requirements are based on provisions of the Illinois
Water Quality Management Plan and essentially require
that development within new FPA expansions not ad-
versely affect water quality, either due to point or non-
point sources.
The IEPA also is delegated to implement the new
NPDES requirements for stormwater discharges. In par-
ticular, as part of its new general permit for construction
site activities, IEPA requires the development of a pollu-
tion prevention plan that must include provisions for soil
erosion and sediment control as well as stormwater
BMPs such as detention facilities, vegetated swales and
natural depressions, infiltration practices, and velocity
dissipation measures (10). While the construction site
general permit does not mandate the adoption of ordi-
nances, it does provide further incentive to local govern-
ments to begin to add stormwater quality control
measures to their existing ordinances.
Regionwide enthusiasm for inclusion of water quality
BMPs in stormwater ordinances is still somewhat limited
because of a lack of awareness among many stormwa-
ter engineers, local officials, and the public of the ad-
verse effects of stormwater runoff on water quality and
79
-------
aquatic life. This perception appears to be at least partly
related to the long-term degradation of urban water
bodies in the region and the lack of a prominent focal
point, such as a Chesapeake Bay or Puget Sound, for
viewing stormwater quality impacts.
Some Current Issues
As multipurpose stormwater ordinances are adopted
throughout the region, several issues remain. Some
municipal officials are concerned about the aesthetics
and maintenance needs of wetland-type detention ba-
sins and natural drainage practices, such as vegetated
swales. Technical debate continues over the effective-
ness of on-line and onstream detention, both from a
water quality and flood prevention perspective. Also, the
appropriateness of using existing wetlands for stormwa-
ter detention remains to be resolved.
Perhaps the most important consideration of local gov-
ernment officials regarding stormwater drainage is pub-
lic acceptance, which generally translates as the
avoidance of "nuisance" drainage conditions. Some
commonly cited nuisance concerns include extended
saturation or ponding on lawns or swales, "weedy" vege-
tation, mosquito breeding potential, and wet detention
areas. These concerns have driven many communities
to require highly engineered drainage systems, includ-
ing curbs and gutters, storm sewers, and concrete chan-
nels, which rapidly convey runoff from the site. Some
public works officials also argue that engineered drain-
age systems are less expensive to maintain.
There is growing support, however, in other parts of the
country and in a few northeastern Illinois communities
for "natural" drainage practices using vegetated swales,
channels, and filterstrips and created wetlands. In addition
to providing significant pollutant removal and runoff re-
duction benefits, natural practices may be much less
expensive to install and, at least to some, are preferred
aesthetically over engineered systems. Progress in gain-
ing acceptance of natural drainage systems has been
slow in northeastern Illinois. Successful ongoing de-
monstration projects, innovative new corporate campus
developments, and improved public education should
be helpful in advancing natural drainage approaches.
Onstream stormwater detention is a desirable alterna-
tive to many site design engineers in the region. In a
typical situation, such facilities generally do not provide
regional detention for the entire upstream watershed;
rather, they serve the storage requirements of a devel-
opment adjacent to the floodplain. As previously men-
tioned, however, there are significant concerns about
the effects and effectiveness of onstream facilities.
These facilities alter the free-flowing nature of streams,
creating impoundments susceptible to sedimentation
and eutrophication. Impoundments can impede the up-
stream migration of fish and the downstream drift of
benthic organisms. Onstream detention essentially uses
the stream as a treatment device. Because of typically
shorter residence times relative to offline facilities, how-
ever, onstream facilities may not be very effective in
trapping stormwater runoff pollutants and protecting
downstream water bodies. While the appropriateness of
onstream detention in northeastern Illinois merits addi-
tional debate, currently this debate is not fully consider-
ing the potential adverse water quality and habitat
impacts of onstream facilities.
Another unresolved issue is the appropriateness of us-
ing existing wetlands for stormwater detention. Section
404 permits have been issued for the incorporation of
detention into existing wetlands and mitigation wet-
lands. If a wetland is impounded without the introduction
of fill material, a Section 404 permit may not even be
required. Limited water quality protection is provided by
several new stormwater ordinances and the NIPC
model ordinance, which require pretreatment of storm-
water before it is discharged into a wetland. Even if
stormwater quality effects are reasonably mitigated,
however, detention in a wetland can radically affect its
hydrology. In particular, detention is likely to pond water
more frequently and at greater depths than in a natural
wetland. Such alterations can adversely affect sensitive
plant communities and wildlife.
Conclusions
Stormwater management ordinances have evolved dra-
matically in northeastern Illinois since their introduction
over 20 years ago. Always a leader in flood prevention,
northeastern Illinois now has some of the most stringent
standards in the nation for detention volumes and re-
lease rates.
Evolving from an early emphasis on local drainage
and flood prevention, many ordinances now recognize
the importance of water quality mitigation and habitat
protection. Some newer ordinances reflect a revised
philosophy of stormwater management that takes ad-
vantage of natural drainage and storage functions, with
the objective of limiting stormwater runoff rates, volume,
and quality to predevelopment conditions. Much re-
mains to be learned, however, about effective designs
for BMPs such as wetland detention, filter strips, and
infiltration practices.
References
1. Northeastern Illinois Planning Commission. 1979. Areawide water
quality management plan. Chicago, IL.
2. Dreher, D.W., G.C. Schaefer, and D.L. Hey. 1989. Evaluation of
stormwater detention effectiveness in northeastern Illinois. Chi-
cago, IL: Northeastern Illinois Planning Commission.
3. Bartels, R.M. 1987. Stormwater management: When onsite de-
tention reduces stream flooding. In: Proceedings of the Eleventh
Annual Conference of the Association of State Floodplain Manag-
ers, Seattle, WA (June).
80
-------
4. Northeastern Illinois Planning Commission. 1990. Model storm-
water drainage and detention ordinance. Chicago, IL.
5. Huff, R, and J. Angel. 1989. Frequency distributions and hydro-
climatic characteristics of heavy rainstorms in Illinois. Bulletin 70.
Urbana, IL: Illinois State Water Survey.
6. Hershfield, D.M. 1961. Rainfall frequency atlas of the United
States. Technical Paper 40. U.S. Department of Commerce,
Weather Bureau.
7. DuPage County Stormwater Management Committee. 1991.
Countywide stormwater and floodplain ordinance. Wheaton, IL:
DuPage County Environmental Concerns Department.
8. Lake County Stormwater Management Commission. 1992. Lake
County watershed development ordinance. Libertyville, IL.
9. Butterfield Creek Steering Committee. 1990. Model floodplain
and stormwater management code for the Butterfield Creek wa-
tershed communities. Cook and Will Counties, IL.
10. Illinois Environmental Protection Agency. 1992. NPDES Permit
No. ILR100000: Construction site activities. Springfield, IL.
81
-------
The Lower Colorado River Authority Nonpoint Source
Pollution Control Ordinance
Thomas F. Curran
Lower Colorado River Authority, Austin, Texas
Abstract
Urban development can be managed to control nonpoint
source pollution using a variety of methods. The method
selected is typically a function of the jurisdictional
agency's authority (or lack thereof), the use and desired
quality of the receiving waters, and the impact on and
acceptance by the public.
The Lower Colorado River Authority (LCRA) is a conser-
vation and reclamation district created by Texas legisla-
tion. LCRA is responsible for the conservation, control,
and preservation of the waters of the Colorado River and
its tributaries within a 10-county area. Given this respon-
sibility but not land-use control authority, LCRA has
developed a nonpoint source pollution control ordinance
with a technology-based approach.
The ordinance requires a large percentage of the pollut-
ants generated from new development to be removed
before stormwater discharge from the property. A tech-
nical manual accompanies the ordinance and explains
how to calculate the expected increase in pollution and
the various management practices a developer may
employ to achieve the required pollutant removal stand-
ards. The developer and engineer determine what com-
bination of management practices are most compatible
with their site and development plan.
This paper provides the methodology and primary features
of the ordinance and technical manual. The reasoning
behind this approach is explained, with discussion re-
garding the strengths and weaknesses of a technology-
based ordinance.
Introduction
The Lower Colorado River Authority (LCRA) is a conser-
vation and reclamation district created by the Texas legis-
lature in 1934. LCRA is also a self-sufficient public utility
company. The authority's responsibilities are many and
include energy generation, water supply, flood control,
management of certain public lands, and preservation
and conservation of the waters of the lower Colorado
River.
While given these responsibilities, LCRA has limited
authority and can only exercise powers expressly given
by the legislature. As such, LCRA cannot regulate land
use, impose zoning or site development restrictions, or
assess taxes. LCRA can, however, promulgate ordi-
nances to control water pollution within its 10-county
statutory area.
With these powers and limitations, LCRA has developed
an ordinance to control nonpoint source (NPS) pollution
from urban development. The ordinance does not im-
pose any land-use regulations other than to establish a
technology-based pollutant reduction standard for new
development.
Background
In 1988, the LCRA board of directors approved a water
quality leadership policy stating LCRA's goals regarding
water quality protection. This policy directed staff to
develop a program to control NPS pollution within the
10-county area, commencing with the area of the High-
land Lakes.
The Highland Lakes are a chain of seven lakes located
west of Austin, Texas. The lakes were created in the
1930s and 1940s for flood control, water supply, and
hydroelectric generation. In the early 1980s, the area
around the lakes experienced tremendous growth in
development activity. This growth prompted concern
about the long-term health of the lakes.
A Pollution Control Approach
From the outset, LCRA was limited in the number of
options available to manage development for control of
NPS pollution. We realized, however, that it must be
attacked in several ways. The initial effort was a public
82
-------
education program, the highlight of which was a 30-min-
ute video entitled, "Pointless Pollution: America's Water
Crisis," narrated by Walter C ran kite.
Realizing that public education alone would not protect
water quality, LCRA staff began addressing the control
of NPS pollution through a regulatory program. Lacking
land-use control or zoning power, LCRA selected a strat-
egy to reduce the quantity of pollution generated by new
development that would otherwise be received by the
lakes.
In December 1989, the LCRA board of directors adopted
the Lake Travis NPS Pollution Control Ordinance, the
first of its kind ever promulgated by a river authority in
the state of Texas. In March 1991, a similar ordinance
was passed to cover the upper Highland Lakes, which
includes Lakes Buchanan, Inks, LBJ, and Marble Falls.
A Nonpoint Source Control Ordinance
The main strategy of the Lake Travis NPS Pollution
Control Ordinance is to establish a set of pollution re-
duction performance standards. Pollution reduction
would be through three methods: 1) removal of a speci-
fied percentage of the projected increase in annual NPS
pollution load; 2) streambank erosion protection via
stormwater detention requirements; and 3) employment
of erosion controls during construction.
Pollution Reduction Standards
LCRA's primary goal was to develop a pollution preven-
tion strategy to protect the lakes. At the same time,
consideration was given to producing feasible standards
that would not prevent development activity.
The basic requirement of the ordinance is the removal
of 70 percent or more of the increased pollution generated
over background or undeveloped conditions. Higher re-
moval rates are required for steeply sloped property or land
located adjacent to the lakes. The required removal rates
were chosen first from a water quality standpoint, but also
were considered feasible. Analysis of existing develop-
ments and the anticipated performance of best manage-
ment practices (BMPs) showed possibilities of significant
land-use restriction if higher removal standards were em-
ployed. Additionally, members of LCRA's board of directors
represent their respective counties or service areas, a
majority of which are predominantly rural. While the board
adopted an environmental leadership policy, its concern
about imposing regulations that could adversely affect
local economic development was clear.
Streambank Erosion Control
Urbanization of a site or area can have a great impact
on the downstream conveyance system. As pavement
and rooftops replace the natural soil and vegetative
cover, the magnitude and frequency of runoff increases
dramatically.
Just as runoff from an undeveloped watershed has
carved out a stream channel overtime to convey typical
runoff events, the increased volume and frequency of
runoff from an urbanized area will reconfigure the
streambank to create a larger conveyance system. The
result is erosion of streambanks transporting sediment
to receiving water bodies, degrading of undercut
streams, removal of aquatic habitat, and loss of public
and private property.
The approach LCRA has taken to control streambank
erosion is to require detainment of postdeveloped runoff
to predeveloped runoff conditions for the 1-year design
storm. Stream morphology is generally dictated by the
2-year storm event.
To simplify the permitting process, the technical manual
provides the required detention volume in inches of runoff
as a function of impervious cover. These detention volume
requirements can be incorporated into the use of BMPs to
meet the pollutant removal performance standards.
Temporary Erosion Control
The ordinance requires erosion and sedimentation to
be controlled throughout the development process. For
permitted activities, an erosion control plan is required
for review and approval. Activities not requiring a per-
mit, such as the construction of a single-family home,
also require erosion controls to be in place until revege-
tation occurs.
The technical manual provides guidance for appropriate
erosion controls. These strategies include minimization
of area cleared; physical controls such as silt fences,
brush berms, and rock berms; downstream vegetative
buffers; diversion of upstream flow; flow spreading; con-
tour furrowing; loose straw or jute netting for soil protec-
tion; and use of structural BMPs as sedimentation
basins during construction.
Technical Manual
The ordinance is accompanied by a technical manual that
provides explanation and guidance for the applicant or
engineer. Included in the technical manual are permitting
procedures, pollutant loading calculations, and design
standards and efficiencies of management practices.
Types of Pollution
Urbanization causes numerous forms of pollution.
Analysis of all pollutant elements through a permitting
program would encumber both the applicant and review
body. LCRA has classified these forms of pollution into
three distinct groups important to the protection of the
lakes: sedimentation, eutrophication, and toxins. LCRA
83
-------
then selected an indicator pollutant to represent these
categories. Indicator pollutants are total suspended sol-
ids (TSS) for sedimentation, total phosphorus (TP) for
eutrophication, and oil and grease (O&G) for toxins.
TSS consist of colloidal and settleable particulate mat-
ter. In alkaline waters such as those of the Highland
Lakes, metals tend to precipitate and become particu-
late matter. In addition, some organic compounds such
as chlordane and polychlorinated biphenyls tend to be
adsorbed onto sediment particles.
TP can be indicative of other nutrients. While the
nitrogen cycle is different, plant and microbial uptake
occurs for both elements.
O&G, while encompassing both nontoxic and toxic
organic compounds, represents petroleum hydrocar-
bon pollutants, including carcinogens such as ben-
zene and toluene and chlorinated compounds such
as pesticides and herbicides.
These indicator pollutants are used to represent the
array of pollutants generated. It is reasonable to assume
that removal of these indicator pollutants will result in
removal of other pollutants not specifically analyzed.
Pollutant Loads
A mass loading equation is used to calculate the pollut-
ant load under existing and developed conditions. This
determines the increase in pollution generated over
background conditions. The equation is a product of
annual runoff volume and the average stormwater pol-
lutant concentration.
The pollutant load is calculated in pounds per year and
is represented as follows:
L = A * RF * Rv * C * K,
where L = annual pollutant load (pounds)
A = area of development (acres)
RF = average annual rainfall (inches)
Rv = average runoff-to-rainfall ratio
C = average pollutant concentration (mg/L)
K = unit conversion factor (0.2266)
The runoff-to-rainfall ratio equation used is as presented
in the Metropolitan Washington Council of Governments
document Controlling Urban Runoff: A Practical Manual
for Planning and Designing Urban BMPs. This regres-
sion equation simplifies the runoff-to-rainfall relationship
to a function of impervious cover as follows:
Rv= 0.05+ (0.009 * 1C),
where 1C is impervious cover in percent
Background and developed pollutant concentrations for
the indicator pollutants are provided. These values were
acquired primarily from screening local and national
reports. The average pollutant concentrations used for
indicator pollutants under background and developed
conditions are shown in Table 1.
Table 1. Average Pollutant Concentrations for Indicator
Pollutants
Background (mg/L) Developed (mg/L)
TSS
TP
O&G
48
0.08
0
130
0.26
15
The manner in which this information is supplied within
the technical manual results in reasonable estimates of
a development's potential pollution impact while making
calculations simple and consistent.
Selection of Management Practices
The technical manual provides design criteria and esti-
mated removal efficiencies for BMPs. The manual is
intended to provide guidance to the applicant in select-
ing BMPs. The applicant must select the BMPs that
will enable the development to meet the criteria of the
ordinance. The basic strategy for selecting BMPs is to
match the pollutant removal requirements with site and
development characteristics. Consideration must be
given to drainage area, soil type, and topography to
select BMPs effectively.
The technical manual provides the expected removal
efficiencies for BMPs with a performance history. Most
of this data is based on criteria presented in nationally
published documents. For structural BMPs, a percent
removal efficiency is provided for each indicator pollut-
ant. This is then multiplied by the percent of the total
average annual runoff volume to be captured by the
proposed BMP. The product is the expected removal
efficiency of that BMP. This is done for each indicator
pollutant. The analysis and performance standard for
O&G is applied only to developments other than single-
family residential use. The focus on O&G is on commer-
cial land and parking lots instead of single-family
residential neighborhoods. Efficiencies used for each
BMP are shown in Table 2.
Other BMPs for which removal efficiencies are provided
include vegetated filter strips, street sweeping, and pollu-
tion source removal credit for using an integrated pest
management plan.
The manual promotes the use of innovative practices as
long as the applicant can document the potential effective-
ness of the practice. LCRA may also require, by ordinance,
84
-------
Table 2. Expected Removal Efficiencies of Selected BMPs
Pollutant
Table 3. Pollutant Concentrations for Austin Example Site
Background (mg/L) Developed (mg/L)
Best Management Practice
Sedimentation basin
Sand filtration
Extended detention
Retention basin
Infiltration practices
TSS
60
70
70
80
80
TP
20
33
60
80
80
O&G
10
30
30
80
80
that innovative BMPs be monitored at a cost borne by
the applicant. Some innovative practices include water
quality catch basins (oil/grit separators), peat/sand fil-
ters, zeolite filters, and wet ponds. While wet ponds
have a proven track record in portions of the United
States, their performance, and more particularly their
maintenance requirements, in semiarid regions war-
rants further scrutiny.
BMPs in Series
Based on the removal efficiencies of known BMPs and the
removal requirements of the ordinance, development with
moderate or high impervious cover may need to provide
BMPs in series to meet the ordinance performance stand-
ards. One of the unknowns at this juncture is how BMPs
operate in series. LCRA currently assumes that the total
removal is the sum of the individual BMP removal perform-
ances. This is an assumption that warrants further analysis
from monitoring BMPs in series.
Example of Ordinance Application
A commercial establishment desires to develop 200,000
ft2 of retail space and is looking at a 23-acre undeveloped
site in the Austin, Texas, area. What would be required for
the development to meet LCRA's NPS ordinance?
The site plan layout shows parking for 1,200 vehicles.
With access drives and loading areas, the impervious
cover provided for vehicular traffic is about 400,000 ft2.
The proposed total impervious cover is 600,000 ft2, or
60 percent of the site area.
The average annual rainfall in Austin is 32.5 in. Applying
the pollutant load calculations shown in the technical
manual,
L = A * RF * Rv * 0.2266 * C,
yields the average pollutant concentrations shown in
Table 3.
With a pollutant removal standard for the site of 70
percent:
TSS
TP
O&G (calculated
for paved area
only, at 100% 1C)
407
0.68
0
12,992
26.0
963
TSS removal = (12,992 - 407) * 0.70 = 8,810 Ib
TP removal = (26.0 - 0.68) * 0.70 = 17.7 Ib
O&G removal = (963 - 0) * 0.70 = 674 Ib
The applicant proposes a weekly street sweeping pro-
gram for general maintenance of the area. The pollutant
removal efficiencies assumed for this practice with a
vacuum-type sweeper are 20 percent for TSS, 10 per-
cent for TP, and 15 percent for O&G.
The site is gently sloping and does have adequate soil
for percolation. Infiltration is desirable; however, it must
be preceded by a sediment removal practice according
to the technical manual.
To meet the streambank erosion control criteria, a site
with 60 percent 1C must provide detention for 1 in. of
runoff. Therefore, structural BMPs should be sized to
also meet this criteria.
The designerdecides to try a sedimentation basin followed
by an infiltration basin. With 60 percent 1C, a 1-in. capture
volume will collect 89.7 percent of the average annual
runoff based on historical rainfall data and runoff/rainfall
relationships. The removal efficiencies of these ponds are
the product of the BMP efficiency and percent of average
annual runoff captured, as shown in Table 4.
Table 4.
Remove Efficiencies of Sedimentation Basin and
Infiltration Basin BMPs
Sedimentation Basin
Infiltration Basin
TSS - 0.60 * 0.897 = 53.7
-------
TSS Eff. (total) = [1-((1-0.2)*(1-0.537)*(1-0.716))]*100
= 89.5 percent
TP Eff. (total) = [1-((1-0.1)*(1-0.179)*(1-0.716))]*100
= 79.0 percent
O&G Eff. (total) = [1-((1-0.15)*(1-0.09)*(1-0.716))]*100
= 78.0 percent
Therefore, the above controls would meet the perform-
ance standard requirements of the ordinance. Had infil-
tration not been a viable option, other potential solutions
include 1) a street sweeping program with a 1-in. volume
extended detention basin followed by 8.4 acres of vege-
tative filter strip (fair condition, 2- to -7 percent slope) or
2) a street sweeping program with three extended de-
tention ponds, each of 2-in. capture volume.
Administration
Maintenance Agreements
Maintenance of BMPs is critical to their long-term per-
formance. Without maintenance, the effective life of a
BMP may be limited to a couple of years. Relying on
good faith or volunteer efforts has not shown to be an
effective way to maintain these pollution controls.
The ordinance requires that a NPS Best Management
Practice Maintenance Permit be issued upon accept-
able completion of construction. Whether through a
homeowner's association or through the land owner as
an individual, a maintenance association must be
formed. The maintenance association is to post financial
security or create a fund for the purpose of maintaining
all BMPs implemented to meet the ordinance.
Enforcement
A necessary portion of any regulatory program is the
ability to impose penalties for not complying with the
regulations. The ordinance contains a violations section
that allows financial penalties to be imposed for viola-
tions of a provision of the ordinance.
Case Application
The ordinance is relatively new, and there have been
few opportunities to evaluate its effectiveness. Two
projects of note have shown the impact that the ordi-
nance has had on development.
LCRA Office Complex
The first project of note is construction of LCRA's gen-
eral office buildings. While not located in an area under
the purview of the ordinance, LCRA chose to make a
leadership statement by applying ordinance standards
to the office complex.
The offices are located on 11.7 acres of land and consist
of 250,000 ft2 of office space with close to 600 parking
spaces. Site 1C is approximately 55 percent. Due to site
constraints, innovation had to be applied to achieve the
performance standards of the ordinance.
A series of BMPs are employed on the site, including a
full integrated pest management and xeriscape plan, a
street sweeping program, five surface ponds composed
of extended detention ponds, a peat/sand filter, and an
enhanced (partial wet pond) extended detention pond.
There are also subsurface treatment devices that in-
clude off-line water quality catch basins conveying to a
sand filtration system beneath a parking lot and
peat/sand filtering system under an open-space front
yard area. Infiltration practices could not be used due to
soil conditions. LCRA has acquired grants from the U.S.
Environmental Protection Agency to monitor the effec-
tiveness of some of the innovative practices being ap-
plied on this project.
The total construction cost associated with the NPS
controls on this project was $250,000. This represents
about 1.5 percent of the total project cost.
Sun City Development
The Del Webb Corporation is in the planning stages of
developing a 2,400-acre active adult community west
of Austin, Texas. The project is within the jurisdiction of
the Lake Travis NPS Pollution Control Ordinance. Del
Webb is presently going through a master plan ap-
proval phase with LCRA.
The development is predominantly single-family resi-
dential and entails 4,200 single-family homes with rec-
reational amenities. The overall proposed 1C for the site
is slightly less than 30 percent. The project has incorpo-
rated in the preliminary design 60 to 70 structural BMPs
to meet the performance requirements of the ordinance.
Over 90 percent of the runoff from the development will
convey to a structural BMP of some form. The structural
practices proposed include extended detention ponds,
wet ponds, retention ponds, sedimentation ponds, and
infiltration practices. These structural facilities take up 5
percent of the total land area.
In addition, the development includes a roadway system
that has vegetated filter strips throughout and grass-lined
swales for stormwater conveyance. Commercial areas
include a street sweeping program, and areas left as
native open space receive credit for pollution reduction
as low-maintenance landscapes.
The cost of meeting the performance standards of the
ordinance has been estimated by the applicant to be about
$1,300 per single-family home. It is quite possible that
86
-------
an economy of scale is realized, as studies before ordinance
implementation estimated a per-unit cost of almost twice
this amount for developments of similar net density.
Pros and Cons
The quality of any development management strategy
has to be measured on the basis of what it achieves
versus the impacts it may create.
Strengths of a Technology-Based Approach
A technology-based approach to control NPS pollution
from urbanization has several strengths. The first is the
transferability of this approach to other jurisdictions.
Creating pollution reduction strategies of this kind can
be applied on a city, county, watershed, or statewide
basis. The only variables may be in the selection of
BMPs that are compatible with a region and the percent-
age of annual runoff captured based on rainfall patterns.
Implementing land-use restrictions from a density or 1C
standpoint can be difficult due to public opposition. The
technology-based approach gives the landowner the
freedom to determine the highest use of the land with
consideration given to the increasing costs of providing
and maintaining additional BMPs to compensate for
dense development. It is theoretically possible for a
landowner to use every square inch of land for develop-
ment purposes if the developer is willing to incur the
increased cost of subsurface stormwater treatment or
even mechanical treatment.
The standards for achieving compliance with a
technology-based ordinance are clear. The approach
is simple, with straightforward calculations. This cook-
book approach minimizes staffing requirements for re-
view of applications.
Density or 1C limitations are a best management prac-
tice. More pollution could be discharged, however, from
a less dense development with no other BMPs than from
a more intense development with BMPs. There is also
concern that density controls contribute to urban sprawl,
which may result in poorer water quality on a regional
basis and may adversely affect air quality through in-
creased vehicular operating time.
Finally, there is no question that implementation of this
technology-based practice mitigates some of the water
quality impacts associated with urbanization.
Weaknesses of a Technology-Based Approach
The sole use of a technology-based pollution reduction
strategy has weaknesses as well. First and foremost is
the full reliance on this new technology to maintain a
high level of pollution removal over the long term. Rec-
ognition of the requirements for maintaining these facili-
ties at their expected performance standards over the
long term has yet to occur.
Notwithstanding the urban sprawl issue, there is no ques-
tion that on a site-specific basis the reduction of 1C and
maintenance of land in a natural vegetative state are more
foolproof means of reducing pollution from that site.
The technology-based approach only considers water
quality issues. Land use is at the disposal of the land-
owner. There are locations where aesthetics, views, and
protection of existing vegetation and habitat are equally
as important as the quality of water. This ordinance does
not directly address these other considerations.
Conclusion
LCRA considers the NPS ordinance to be an excellent
beginning in protecting the quality of the waters of the
Highland Lakes and Colorado River. Close to a million
people rely on the Highland Lakes for drinking water sup-
ply and countless thousands for recreational and aesthetic
purposes.
LCRA is committed to evaluating the effectiveness of
this ordinance. Depending on the actual development
that takes place around the Highland Lakes, the actual
pollution removal achieved, and the change in water
quality evidenced, more or less restrictive standards or
alternate practices may be required. The effectiveness
of the ordinance must be analyzed as development
takes place to ensure good water quality.
There are limitations in our knowledge of BMPs and of
pollution generation from various land uses. The current
version of the technical manual is already in need of
revision to account for research performed over the last
few years. The calculations do not adequately address
certain land uses, such as golf courses, nurseries, or
parks, due to the low 1C yet high maintenance associ-
ated with these land uses, particularly as they pertain to
pesticides and nutrients.
Finally, it is LCRAs desire to ultimately connect the
pollution removal standards of the ordinance to estab-
lished water quality standards of the receiving waters.
There is much work to be performed before a full under-
standing of the dynamics of the lakes and Colorado
River permit us to achieve this goal.
87
-------
New Development Standards in the Puget Sound Basin
Peter B. Birch
Washington Department of Ecology,
Olympia, Washington
Abstract
The Puget Sound Water Quality Management Plan
(PSWQMP) calls for all counties and cities in the Puget
Sound drainage basin to adopt ordinances that require
stormwater control for new development and redevelop-
ment. Ordinances were to be adopted by July 1, 1994.
The PSWQMP also directed the Washington Depart-
ment of Ecology to prepare technical guidance and a
model ordinance to assist local governments in imple-
menting these standards.
In response, the Department of Ecology has prepared
several sets of minimum requirements that are applied
based on the type and size of proposed development.
These include:
Simplified erosion and sediment controls and a small
parcel erosion and sediment control plan for small
developments (under 5,000 ft2 impervious surface),
single-family homes, and land-disturbing activities
under 1 acre.
A set of 11 minimum requirements for proposed new
development of large parcels (5,000 ft2 impervious
surface and greater) and/or land-disturbing activities
over 1 acre. The requirements include erosion and
sediment control, and source control and treatment
best management practices designed to prevent or
minimize impacts to receiving waters. A stormwater
site plan is also required for this level of development.
The same 11 requirements apply to large parcels with
less than 1 acre of land-disturbing activities except
that the small parcel erosion and sediment require-
ments are substituted for the large parcel erosion and
sediment controls.
If redevelopment is proposed, the same minimum re-
quirements apply, subject to a set of thresholds and
criteria for applying the minimum requirements to all or
part of the site.
Introduction
Puget Sound, which is located in western Washington
State, has been the focus of a comprehensive water
quality improvement effort in recent yearsespecially
since documentation of liver tumors in English sole and
toxics in sediments and with increasing closures of
shellfish beds (1). Initial efforts culminated in 1986, with
the publication of the Puget Sound Water Quality Man-
agement Plan (PSWQMP) and subsequent amend-
ments in 1989 and 1991 (2). In 1991, Puget Sound was
listed as an Estuary of National Significance under Sec-
tion 320 of the federal Clean Water Act.
The section of the PSWQMP that covers stormwater
management calls for all counties and cities in the Puget
Sound drainage basin to adopt ordinances that require
stormwater control for new development and redevelop-
ment by July 1, 1994. The plan also requires all local
governments in the basin to adopt operation and main-
tenance programs for new and existing public and pri-
vate stormwater systems. Local governments located
within census-defined urbanized areas have additional
requirements that include:
Identification and ranking of significant pollutant
sources.
Corrective actions for problem drains.
A water quality response program.
Assurance of funding.
Local coordination.
Public education.
Compliance measures.
An implementation schedule.
As a last resort in problem areas, retrofitting of control
measures.
88
-------
The PSWQMP also directed the Washington State De-
partment of Ecology (Ecology) to prepare a best man-
agement practices (BMPs) technical manual (3) and a
program guidance manual containing model ordinances
and other supplemental guidance (4) to assist local
governments in implementing plan requirements. The
guidance prepared for new development and redevel-
opment consists of several sets of minimum require-
ments that are applied depending on the type and size
of proposed development. In summary, these include:
Simplified erosion and sediment controls (ESCs) and
a small parcel ESC plan for small developments (under
5,000 ft2 impervious surface), detached single-family
homes and duplexes, and land-disturbing activities
under 1 acre.
A set of 11 minimum requirements for proposed new
development of large parcels (5,000 ft2 impervious
surface and greater) and/or land-disturbing activities
over 1 acre. The requirements include ESC and source
control and treatment BMPs designed to prevent or
minimize impacts to receiving waters. A stormwater
site plan is also required for this level of development.
The same 11 requirements apply to large parcels with
less than 1 acre of land-disturbing activities except
that the small parcel ESC are substituted for the large
parcel ESCs.
If redevelopment is proposed, the same minimum re-
quirements apply, subject to a set of thresholds and
criteria for applying the minimum requirements to all or
part of the site.
The BMP manual that Ecology prepared contains a full
description of the minimum requirements and technical
guidance on how to meet them. In essence, develop-
ment sites are to demonstrate compliance with the re-
quirements by preparing and implementing a
stormwater site plan that includes an appropriate selec-
tion of BMPs from the manual.
Two major components of a stormwater site plan are an
ESC plan and a permanent stormwater quality control
(PSQC) plan. The ESC plan is intended to be temporary
in nature to control pollution generated during the con-
struction and landscaping phase only, primarily erosion
and sediment. The PSQC plan is intended to provide
permanent BMPs for the control of pollution and other
impacts from stormwater runoff after construction is
completed. For small sites, this is met by implementing a
small parcel erosion and sediment control (SPESC) plan.
Further details of these plans are contained in the Storm-
water Management Manual for the Puget Sound Basin (3).
The following sections describe the minimum requirements
as they apply to local governments in the Puget Sound
basin and have been adapted directly from the
technical manual (3). The description also includes sev-
eral associated requirements specific to Washington
laws; therefore, some modifications would be needed
for application of the minimum requirements to areas
outside of Washington. The model ordinance that was
prepared as guidance for enacting the minimum require-
ments is contained in the program guidance manual (4).
The full guidance package may be ordered from Ecology
by calling (206) 438-7116. The current cost of the tech-
nical manual is $24.85 plus postage, and of the program
guidance manual is $28.00 plus postage.
Definitions
The following definitions are useful to the understanding
of the minimum requirements:
Approved manual: A technical manual that is substan-
tially equivalent to the Stormwater Management Man-
ual for the Puget Sound Basin (3). (The PSWQMP
requires all counties and cities located in the Puget
Sound basin to adopt a manual that is the same or
substantially equivalent to this manual by July 1,1994.)
New development: Development consisting of land-
disturbing activities; structural development, includ-
ing construction, installation or expansion of a
building or other structure; creation of impervious sur-
faces; Class IV general forest practices that are con-
versions from timber land to other uses; and
subdivision and short subdivision of land as defined in
RCW58.17.020. All other forest practices and commer-
cial agriculture are not considered new development.
Redevelopment: On an already developed site, the
creation or addition of impervious surfaces; structural
development including construction, installation, or
expansion of a building or other structure, and/or
replacement of an impervious surface that is not part
of a routine maintenance activity; and land-disturbing
activities associated with structural or impervious re-
development.
Impervious surface: A hard surface that either pre-
vents or retards the entry of water into the soil mantle
as under natural conditions prior to development,
and/or a hard surface area that causes water to run
off the surface in greater quantities or at an increased
rate of flow from the flow present under natural con-
ditions prior to development.
Land-disturbing activity: Any activity that results in a
change in the existing soil cover (both vegetative and
nonvegetative) and/or the existing soil topography.
Land-disturbing activities include, but are not limited
to, demolition, construction, clearing, grading, filling,
and excavation.
Source control BMP: A BMP that is intended to pre-
vent pollutants from entering stormwater. Examples
include covering an activity, controlling erosion,
89
-------
directing wash water to a sanitary sewer, and altering
a practice that results in pollution prevention.
Exemptions
Commercial agriculture and forest practices regulated
under Title 222 WAC, except for Class IV general forest
practices that are conversions from timber land to other
uses, are exempt from the provisions of the minimum
requirements. All other new development is subject to
the minimum requirements.
Small Parcel Minimum Requirements
The following new development shall be required to
control erosion and sediment during construction, to
permanently stabilize soil exposed during construction,
to comply with Small Parcel Requirements 1 through 5,
and to prepare a SPESC plan:
Individual, detached single-family residences and du-
plexes.
Creation or addition of less than 5,000 ft2 of imper-
vious surface area.
Land-disturbing activities of less than 1 acre.
Supplemental Guidelines
The objective of these requirements is to address the
cumulative effect of sediment coming from a large
number of small sites. The SPESC plan is meant
to be temporary in nature to deal with erosion and
sediment generated during the construction phase
only. Local governments may choose to apply addi-
tional permanent, site-specific stormwater controls to
small parcels.
Small Parcel Requirement 1:
Construction Access Route
Construction vehicle access shall be limited to one route
whenever possible. Access points shall be stabilized
with quarry spall or crushed rock to minimize the track-
ing of sediment onto public roads.
Small Parcel Requirement 2:
Stabilization of Denuded Areas
All exposed soils shall be stabilized by suitable applica-
tion of BMPs, including but not limited to sod or other
vegetation, plastic covering, mulching, or application of
ground base on areas to be paved. All BMPs shall be
selected, designed, and maintained in accordance with
an approved manual. From October 1 through April 30,
no unworked soils shall remain exposed for more than
2 days. From May 1 through September 30, no un-
worked soils shall remain exposed for more than 7 days.
Small Parcel Requirement 3:
Protection of Adjacent Properties
Adjacent properties shall be protected from sediment
deposition by appropriate use of vegetative buffer strips,
sediment barriers or filters, dikes or mulching, or by a com-
bination of these measures and other appropriate BMPs.
Small Parcel Requirement 4: Maintenance
All ESC BMPs shall be regularly inspected and main-
tained to ensure continued performance of their in-
tended function.
Small Parcel Requirement 5: Other BMPs
As required by the local plan-approval authority, other
appropriate BMPs to mitigate the effects of increased
runoff shall be applied.
Application of Minimum Requirements for
New Development and Redevelopment
New Development
All new development that includes the creation or addi-
tion of 5,000 ft2 or greater of new impervious surface
area and/or land-disturbing activities of 1 acre or greater
shall comply with Minimum Requirements 1 through 11
below and be in agreement with a stormwater site plan.
All new development that includes the creation or addi-
tion of 5,000 ft2 or more of new impervious surface area
and land-disturbing activities of less than 1 acre shall
comply with Minimum Requirements 2 through 11 below
and the Small Parcel Minimum Requirements listed
above. This category of development requires preparation
of a stormwater site plan that includes a SPESC plan.
Redevelopment
Where redevelopment of 1 acre or greater occurs, new
development Minimum Requirements 1 through 11 ap-
ply to that portion of the site that is being redeveloped,
and source control BMPs shall be applied to the entire site,
including adjoining parcels if they are part of the project.
Where one or more of the following conditions apply, a
stormwater site plan shall be prepared that includes a
schedule for implementing Minimum Requirements 1
through 11 below to the maximum extent practicable for
the entire site, including adjoining parcels if they are part
of the project:
Existing sites greater than 1 acre in size with 50
percent or more impervious surface.
Sites that discharge to a receiving water that has a
documented water quality problem.
90
-------
Sites where the need for additional stormwater control
measures has been identified through a basin plan
or other local planning activities.
Note: An adopted and implemented basin plan (Mini-
mum Requirement 9) may be used to develop require-
ments that are tailored to a specific basin.)
Minimum Requirement 1: Erosion and
Sediment Control
All new development and redevelopment that includes
land-disturbing activities of 1 acre or more shall comply
with Large Parcel ESC Requirements 1 through 15 be-
low. Compliance shall be demonstrated through imple-
mentation of a Large Parcel ESC plan.
All proposed developments where land-disturbing ac-
tivities 5,000 ft2 and greater but less than 1 acre are
planned shall implement the Small Parcel Minimum
Requirements above, as well as Minimum Require-
ments 2 through 11 below.
Large Parcel ESC Requirement 1: Stabilization
and Sediment Trapping
All exposed soils shall be stabilized by suitable application
of BMPs. From October 1 to April 30, no unworked soils
shall remain exposed for more than 2 days. From May 1
to September 30, no unworked soils shall remain exposed
for more than 7 days. Prior to leaving the site, stormwater
runoff shall pass through a sediment pond or sediment
trap, or other appropriate BMPs shall be employed.
Supplemental Guidelines. This criterion applies both
to soils not yet at final grade and soils at final grade. The
type of stabilization BMP used may differ depending on
the length of time that the soil is to remain unworked.
Soil stabilization refers to BMPs that protect soil from the
erosive forces of raindrop impact, flowing water, and wind.
Applicable practices include vegetative establishment,
mulching, plastic covering, and the early application of
gravel base on areas to be paved. Soil stabilization meas-
ures should be appropriate for the time of year, site condi-
tions, and estimated duration of use. Soil stockpiles must
be stabilized or protected with sediment trapping meas-
ures to prevent soil loss, including loss to wind.
These requirements are especially important in areas
adjacent to streams, wetlands, or other sensitive or
critical areas.
Large Parcel ESC Requirement 2: Delineated
Clearing and Easement Limits
In the field, clearing limits and/or any easements, set-
backs, sensitive/critical areas and their buffers, trees,
and drainage courses shall be marked.
Large Parcel ESC Requirement 3: Protection of
Adjacent Properties
Properties adjacent to the project site shall be protected
from sediment deposition.
Supplemental Guidelines. This may be accomplished
by preserving a we 11-vegetated buffer strip around the
lower perimeter of the land disturbance; by installing
perimeter controls such as sediment barriers, filters or
dikes, or sediment basins; or by using a combination of
such measures.
Vegetated buffer strips may be used alone only where
runoff in sheet flow is expected. Buffer strips should be
at least 25 ft wide. If at any time the vegetated buffer
strip alone is found to be ineffective in stopping sedi-
ment movement onto adjacent property, additional peri-
meter controls must be provided.
Large Parcel ESC Requirement 4: Timing and
Stabilization of Sediment Trapping Measures
Sediment ponds and traps, perimeter dikes, sediment
barriers, and other BMPs intended to trap sediment on
site shall be constructed as a first step in grading. These
BMPs shall be functional before land-disturbing activities
take place. Earthen structures such as dams, dikes, and
diversions shall be seeded and mulched according to the
timing indicated in Large Parcel ESC Requirement 1.
Large Parcel ESC Requirement 5: Cut and Fill
Slopes
Cut and fill slopes shall be designed and constructed in
a manner that minimizes erosion. In addition, slopes
shall be stabilized in accordance with Large Parcel ESC
Requirement 1.
Supplemental Guidelines. Consideration should be
given to the length and steepness of the slope, the soil
type, upslope drainage area, ground-water conditions, and
other applicable factors. Slopes that are found to be erod-
ing excessively within 2 years of construction must be
provided with additional slope stabilizing measures until
the problem is corrected.
Large Parcel ESC Requirement 6: Controlling
Offsite Erosion
Properties and waterways downstream from develop-
ment sites shall be protected from erosion due to in-
creases in the volume, velocity, and peak flow rate of
stormwater runoff from the project site.
Large Parcel ESC Requirement 7: Stabilization of
Temporary Conveyance Channels and Outlets
All temporary onsite conveyance channels shall be
designed, constructed, and stabilized to prevent erosion
from the expected velocity of flow from a 2-year, 24-hour
91
-------
frequency storm for the developed condition. Stabiliza-
tion adequate to prevent erosion of outlets, adjacent
streambanks, slopes, and downstream reaches shall be
provided at the outlets of all conveyance systems.
Large Parcel ESC Requirement 8: Storm Drain
Inlet Protection
All storm drain inlets made operable during construction
shall be protected so that stormwater runoff shall not
enter the conveyance system without first being filtered
or otherwise treated to remove sediment.
Large Parcel ESC Requirement 9: Underground
Utility Construction
The construction of underground utility lines is subject
to the following criteria:
Where feasible, no more than 500 ft of trench shall
be opened at one time.
Where consistent with safety and space considera-
tions, excavated material shall be placed on the uphill
side of trenches.
Trench dewatering devices shall discharge into a
sediment trap or sediment pond.
Large Parcel ESC Requirement 10: Construction
Access Routes
Wherever construction vehicle access routes intersect
paved roads, provisions must be made to minimize the
transport of sediment (mud) onto the paved road. If
sediment is transported onto a road surface, the roads
shall be cleaned thoroughly at the end of each day.
Sediment shall be removed from roads by shoveling or
sweeping and shall be transported to a controlled sedi-
ment disposal area. Street washing shall be allowed
only after sediment is removed in this manner.
Large Parcel ESC Requirement 11: Removal of
Temporary BMPs
All temporary erosion and sediment control BMPs shall
be removed within 30 days after final site stabilization is
achieved or after the temporary BMPs are no longer
needed. Trapped sediment shall be removed or stabi-
lized on site. Disturbed soil areas resulting from removal
shall be permanently stabilized.
Large Parcel ESC Requirement 12: Dewatering
Construction Sites
Dewatering devices shall discharge into a sediment trap
or sediment pond.
Large Parcel ESC Requirement 13: Control of
Pollutants Other Than Sediment on Construction
Sites
All pollutants other than sediment that occur on site
during construction shall be handled and disposed
of in a manner that does not cause contamination of
stormwater.
Large Parcel ESC Requirement 14: Maintenance
All temporary and permanent erosion and sediment con-
trol BMPs shall be maintained and repaired as needed
to ensure continued performance of their intended func-
tion. All maintenance and repair shall be conducted in
accordance with an approved manual.
Large Parcel ESC Requirement 15: Financial
Liability
Performance bonding or other appropriate financial in-
struments shall be required for all projects to ensure
compliance with the approved ESC plan.
Minimum Requirement 2: Preservation of
Natural Drainage Systems
Natural drainage patterns shall be maintained and dis-
charges from the site shall occur at the natural location
to the maximum extent practicable.
Supplemental Guidelines
Natural drainage systems provide many water quality
benefits and should be preserved to the fullest extent
possible. In addition to conveying and attenuating
stormwater runoff, these systems are less erosive, pro-
vide ground-water recharge, and support important
plant and wildlife resources. Effective use of the natural
system can maintain environmental and aesthetic attrib-
utes of a site as well as be a cost-effective measure to
convey stormwater runoff.
Creating new drainage patterns requires more site dis-
turbance and can upset the stream dynamics of the
drainage system, thus tending to increase erosion and
sedimentation. Creating new discharge points can cre-
ate significant streambank erosion problems because
the receiving water body typically must adjust to the new
flows. Newly created drainage patterns seldom, if ever,
provide the multiple benefits of natural drainage sys-
tems. Where no conveyance system exists at the adja-
cent downstream property line and the discharge was
previously unconcentrated flow or significantly lower
concentrated flow, then measures must be taken to
prevent downstream impacts. Necessary drainage
easements may need to be obtained from downstream
property owners.
92
-------
Minimum Requirement 3: Source Control of
Pollution
Source control BMPs shall be applied to all projects to
the maximum extent practicable. Source control BMPs
shall be selected, designed, and maintained according
to an approved manual.
An adopted and implemented basin plan (Minimum Re-
quirement 9) may be used to develop source control
requirements that are tailored to a specific basin; how-
ever, in all circumstances, source control BMPs shall be
required for all sites.
Objective
The intention of source control BMPs is to prevent
stormwater from coming in contact with pollutants. A
cost-effective means of reducing pollutants in stormwa-
ter, source control BMPs should be a first consideration
in all projects.
Minimum Requirement 4: Runoff Treatment
BMPs
All projects shall provide treatment of stormwater. Treat-
ment BMPs shall be sized to capture and treat the water
quality design storm, defined as the 6-month, 24-hour
return period storm. The first priority for treatment shall be
to infiltrate as much as possible of the waterquality design
storm, if site conditions are appropriate and ground water
quality will not be impaired. Direct discharge of untreated
stormwater to ground water can cause serious pollution
problems. All treatment BMPs shall be selected, designed,
and maintained according to an approved manual.
Stormwater treatment BMPs shall not be built within a
natural vegetated buffer, except for necessary convey-
ance as approved by the local government.
An adopted and implemented basin plan (Minimum Re-
quirement 9) may be used to develop runoff treatment
requirements that are tailored to a specific basin.
Supplemental Guidelines
The water quality design storm (the 6-month, 24-hour
design storm, in this instance) is intended to capture
more than 90 percent of annual runoff.
Infiltration can provide both treatment of stormwater,
through the ability of certain soils to remove pollutants,
and volume control of stormwater, by decreasing the
amount of water that runs off, to surface water. Infiltra-
tion can be very effective at treating stormwater runoff,
but soil conditions must be appropriate to achieve effec-
tive treatment while not affecting ground-water re-
sources. Methods currently in use, such as direct
discharge into dry wells, do not achieve adequate water
quality treatment.
Minimum Requirement 5: Streambank
Erosion Control
The requirement below applies only to situations where
stormwater runoff is discharged directly or indirectly to
a stream, and must be met in addition to the requirements
in Minimum Requirement 4, Runoff Treatment BMPs.
Stormwater discharges to streams shall control stream-
bank erosion by limiting the peak rate of runoff from
individual development sites to 50 percent of the exist-
ing condition, 2-year, 24-hour design storm while main-
taining the existing condition peak runoff rate for the
10-year, 24-hour and 100-year, 24-hour design storms.
As the first priority, streambank erosion control BMPs
shall utilize infiltration to the fullest extent practicable,
only if site conditions are appropriate and ground-water
quality is protected. Streambank erosion control BMPs
shall be selected, designed, and maintained according
to an approved manual.
Stormwater treatment BMPs shall not be built within a
natural vegetated buffer, except for necessary convey-
ance as approved by the local government.
An adopted and implemented basin plan (Minimum
Requirement 9) may be used to develop streambank
erosion control requirements that are tailored to a
specific basin.
Supplemental Guidelines
This requirement is intended to reduce the frequency
and magnitude of bankfull flow conditions, which are
highly erosive and increase dramatically as a result of
development. Conventional flood detention practices do
not adequately control streambank erosion because
only the peak rate of flow is decreased, not the fre-
quency nor duration of bankfull conditions.
Reduction of flows through infiltration decreases
streambank erosion and helps to maintain base flow
throughout the summer months. Infiltration should only
be used, however, where ground-water quality is not
threatened by such discharges. The use of an artificial
treatment system, such as an aquatard, should be con-
sidered in areas with highly permeable soils. Treatment
of the water quality design storm must be accomplished
before discharge to these soils. If highly permeable soils
are present, they should be utilized for streambank ero-
sion control by infiltrating flows greater than the water
quality design storm.
Minimum Requirement 6: Wetlands
The requirements below apply only to situations where
stormwater discharges directly or indirectly through a
conveyance system into a wetland, and must be met in
addition to the requirements in Minimum Requirement
4, Runoff Treatment BMPs:
93
-------
Stormwater discharges to wetlands must be control-
led and treated to the extent necessary to meet state
water quality standards.
Discharges to wetlands shall maintain the hydrope-
riod and flows of existing site conditions to the extent
necessary to protect the characteristic uses of the
wetland. Prior to discharging to a wetland, alternative
discharge locations shall be evaluated, and natural
water storage and infiltration opportunities outside
the wetland shall be maximized.
Created wetlands that are intended to mitigate the
loss of wetland acreage, function, and value shall not
be designed to also treat stormwater.
For constructed wetlands to be considered treatment
systems, they must be constructed on sites that are
not wetlands managed for stormwater treatment. If
these systems are not managed and maintained in
accordance with an approved manual for a period
exceeding 3 years, these systems may no longer be
considered constructed wetlands.
Stormwater treatment BMPs shall not be built within
a natural vegetated buffer, except for necessary con-
veyance as approved by the local government.
An adopted and implemented basin plan (Minimum Re-
quirement 9) may be used to develop requirements for
wetlands that are tailored to a specific basin.
Objective
This requirement seeks to ensure that wetlands receive
the same level of protection as any other state waters.
Wetlands are extremely important natural resources that
provide multiple stormwater benefits, including ground-
water recharge, flood control, and streambank erosion
protection. Development can readily affect wetlands un-
less careful planning and management are conducted.
Stormwater discharges from urban development due to
pollutants in the runoff and also due to disruption of
natural hydrologic functioning of the wetland system
severely degrade wetlands. Changes in water levels
and the duration of inundations are of particular con-
cern.
Minimum Requirement 7: Water Quality
Sensitive Areas
Where local governments determine that the minimum
requirements do not provide adequate protection of
water quality sensitive areas, either on site or within the
basin, more stringent controls shall be required to pro-
tect water quality.
Stormwater treatment BMPs shall not be built within a
natural vegetated buffer, except for necessary convey-
ance as approved by the local government.
An adopted and implemented basin plan (Minimum Re-
quirement 9) may be used to develop requirements for
water quality sensitive areas that are tailored to a spe-
cific basin.
Supplemental Guidelines
Water quality sensitive areas are areas that are sensi-
tive to a change in water quality, including but not limited
to lakes, ground-water management areas, ground-
water special protection areas, sole source aquifers,
critical aquifer recharge areas, well head protection ar-
eas, closed depressions, fish spawning and rearing
habitat, wildlife habitat, and shellfish protection areas.
Areas that can cause water quality problems, such as
steep or unstable slopes or erosive stream banks,
should also be included. Water quality sensitive areas
may be identified through jurisdiction-wide inventories,
watershed planning processes, local drainage basin
planning, and/or on a site-by-site basis.
Minimum Requirement 8: Offsite Analysis and
Mitigation
All development projects shall conduct an analysis of
offsite water quality impacts resulting from the project
and shall mitigate these impacts. The analysis shall
extend a minimum of one-fourth of a mile downstream
from the project. The existing or potential impacts to be
evaluated and mitigated shall include, but not be limited
to:
Excessive sedimentation.
Streambank erosion.
Discharges to ground-water contributing or recharge
zones.
Violations of water quality standards.
Spills and discharges of priority pollutants.
Minimum Requirement 9: Basin Planning
Adopted and implemented watershed-based basin
plans may be used to modify any or all of the Mini-
mum Requirements provided that the level of protec-
tion for surface or ground water achieved by the
basin plan will equal or exceed that which would be
achieved by the Minimum Requirements in the ab-
sence of a basin plan. Basin plans shall evaluate and
include, as necessary, retrofitting of BMPs for existing
development and/or redevelopment in order to achieve
watershed-wide pollutant reduction goals. Standards
developed from basin plans shall not modify any of
the above requirements until the basin plan is formally
adopted and fully implemented by local government.
Basin plans shall be developed according to an ap-
proved manual.
94
-------
Supplemental Guidelines
While Minimum Requirements 3 through 7 establish
protection standards for individual sites, they do not
evaluate the overall pollution impacts and protection
opportunities that could exist at the watershed level. For
a basin plan to serve as a means of modifying the
Minimum Requirements, it must be formally adopted by
all jurisdictions that have responsibilities under the basin
plan, and construction and regulations called for by the
plan must be complete; this is what is meant by an
"adopted and implemented" basin plan.
Basin planning provides a mechanism by which the
onsite standards can be evaluated and refined based on
an analysis of an entire watershed. Basin plans are
especially well suited to develop control strategies to
address impacts from future development and to correct
specific problems whose sources are known or sus-
pected. Basin plans can be effective at addressing both
long-term cumulative impacts of pollutant loads and
short-term acute impacts of pollutant concentrations, as
well as hydrologic impacts to streams and wetlands.
In general, the standards established by basin plans will
be site-specific but may be augmented with regional solu-
tions for source control (Minimum Requirement 2) and
streambank erosion control (Minimum Requirement 4).
Minimum Requirement 10: Operation and
Maintenance
An operation and maintenance schedule shall be pro-
vided for all proposed stormwater facilities and BMPs,
and the party (or parties) responsible for maintenance
and operation shall be identified.
Minimum Requirement 11: Financial Liability
Performance bonding or other appropriate financial in-
struments shall be required for all projects to ensure
compliance with these requirements.
Exceptions
Exceptions to Minimum Requirements 1 through 11 may
be granted prior to permit approval and construction. An
exception may be granted following a public hearing,
provided that a written finding of fact is prepared that
addresses the following:
The exception provides equivalent environmental pro-
tection and is in the public interest, and the objectives
of safety, function, environmental protection and fa-
cility maintenance, based upon sound engineering,
are fully met.
Special physical circumstances or conditions affect-
ing the property are such that strict application of
these provisions would deprive the applicant of all
reasonable use of the parcel of land in question, and
every effort to find creative ways to meet the intent
of the minimum standards has been made.
The granting of the exception will not be detrimental
to the public health and welfare, nor injurious to other
properties in the vicinity and/or downstream nor to
the quality of state waters.
The exception is the least possible exception that
could be granted to comply with the intent of the
Minimum Requirements.
Supplemental Guidelines
The Plan Approval Authority is encouraged to impose
additional or more stringent criteria as appropriate for
its area. Additionally, criteria that may be inappro-
priate or too restrictive for an area may be modified
through basin planning (Minimum Requirement 9).
Modification of any of the Minimum Requirements that
are deemed inappropriate for the site may be done by
granting an exception.
The exception procedure is an important element of
the plan review and enforcement programs. It is in-
tended to maintain a flexible working relationship be-
tween local officials and applicants. Plan Approval
Authorities should consider these requests judiciously,
keeping in mind both the need of the applicant to maxi-
mize cost-effectiveness and the need to protect offsite
properties and resources from damage.
References
1. PSWQA. 1988. State of the Sound report. Puget Sound Water
Quality Authority, Olympia, WA (May).
2. PSWQA. 1992. Puget Sound water quality management plan.
Puget Sound Water Quality Authority, Olympia, WA (February).
3. Washington State Department of Ecology. 1992. Stormwater man-
agement manual for the Puget Sound basin. Publication No. 91-75
(February).
4. Washington Department of Ecology. 1992. Stormwater guidance
manual for the Puget Sound basin. Publication Nos. 92-32 (Vol. I)
and 92-33 (Vol. II) (July).
95
-------
Ordinances for the Protection of Surface Water Bodies: Septic Systems, Docks
and Other Structures, Wildlife Corridors, Sensitive Aquatic Habitats, Vegetative
Buffer Zones, and Bank/Shoreline Stabilization
Martin Kelly
Southwest Florida Water Management District, Tampa, Florida
Nancy Phillips
U.S. Environmental Protection Agency, Region 5, Chicago, Illinois
Introduction
Local government can substantially protect surface
water bodies by enacting and enforcing appropriate or-
dinances. As part of its Surface Water Improvement and
Management (SWIM) Program, the Southwest Florida
Water Management District (SWFWMD) in consultation
with advisory committees developed a list of seven is-
sues that needed ordinance models. As a result, the
SWFWMD outlined and funded a project for model ordi-
nance development. The scope of the project included
preparing model ordinance language to address seven
specific issues, drafting individual papers addressing
the ecological and legal significance of each issue, and
developing a decision model for local government plan-
ners to use in determining the applicability or need for
ordinance adoption. The private consulting firm Henigar
and Ray, Inc., of Crystal River, Florida, developed under
contract the model ordinances, issue papers, and deci-
sion model.
This paper highlights the results of and recommendations
for ordinances addressing six of the seven project issues:
Placement and maintenance of individual septic systems
Regulation of docks and other appurtenance structures
Establishment of wildlife corridors
Protection of environmentally sensitive habitats
Vegetative buffer zones
Erosion control and bank stabilization
The seventh issue, "Stormwater Management and Treat-
ment," is covered in other papers in this publication.
Because any ordinance is likely to face challenges,
often from a number of opposing camps, issue papers
were drafted to support an ecologically and legally
defensible argument. While legal information contained
in the detailed issue papers focuses on the Florida
experience, the ecological arguments are valid over a
much larger geographic area.
It is not possible to consider in detail the products of
this project; however, this paper attempts to transfer the
flavor and scope of information available on each of the
issues. The paper provides an overview on the need/
justification for a particular ordinance, mentions some
of the technical issues that should be considered, and
recommends necessary components of a viable ordi-
nance. (The U.S. Environmental Protection Agency [EPA]
is currently condensing the body of this work [1].)
Project History
The State of Florida passed the SWIM Act in 1987 estab-
lishing a program similar to the Clean Lakes Program but
encompassing all surface waters (i.e., estuaries, rivers,
springs, lakes, and swamps [2]). The Act mandated that
each of the state's five water management districts de-
velop a list of priority water bodies and begin developing
management plans for each of them. Once a manage-
ment plan received approval, monies from the SWIM
Trust Fund could help implement projects outlined in the
specific management plan for each water body.
During plan development for a number of water bodies,
several advisory committees suggested that drafting
and enacting ordinances at the local government level
(municipality or county), particularly with regard to land
development issues, could do much to protect water
bodies from degradation. Such ordinances would be
proactive in that they would avoid or minimize antici-
pated deleterious impacts. SWIM staff at the SWFWMD
96
-------
in consultation with various members of advisory com-
mittees identified the seven issues that required model
ordinances.
Although passage of the SWIM Act gave the state's
water management districts no new regulatory authority,
the SWFWMD felt it was appropriate to develop model
ordinances for consideration by local governments. Be-
cause enactment of ordinances that affect development
are likely to invoke challenges, SWFWMD deemed it
necessary not only to develop model ordinance language
but also to develop "issue papers" detailing the ecological
justifications for a given ordinance. Issue papers would
also review similar ordinances already enacted in Flor-
ida and elsewhere (i.e., establish precedence) and con-
sider the legality of enacting a particular ordinance.
Henigar and Ray, Inc., employed the appropriate tech-
nical and legal authorities to draft the issue papers and
ordinance language. The project resulted in a series of
seven issue papers, five model ordinances, a decision
model (planning document), and a report summarizing
"The Law of Surface Water Management in Florida."
Placement and Maintenance of Individual
Septic Systems (3)
Almost invariably when potential sources of pollutants
to a water body are discussed, the topic of septic tanks
arises. Many people assume that their septic systems are
operating effectively simply because failure is not obvi-
ous (i.e., blocked plumbing, standing water over the drain
field). As Brown (4) has pointed out, a system's technical
failure (the inability to effectively process the waste)
goes unnoticed; as long as the homeowner is not incon-
venienced, the system usually remains unrepaired.
Septic systems can fail for two basic reasons: poor
design or poor maintenance. Design includes not only
the tank and drain field layout, but also the soils and
hydrologic character of the site. Maintenance implies a
periodic check and cleaning of the tank and possibly the
drain field, and a consideration of the substances dis-
charged to the system.
Effective treatment in the drain field requires soils of the
proper permeability. For example, soils that are too per-
meable permit the tank effluent to travel too rapidly away
from the drain field and do not allow for proper biologic
treatment in the biomat. Alternately, impermeable soils
become clogged with effluent, causing lateral or upward
seepage. In the latter case, the homeowner may be
inconvenienced, but in the former the owner may as-
sume everything is working fine.
Soil absorption fields must lie above the surficial water
table. If not, the system will cease to function effectively.
An unsaturated zone ensures a desirable effluent veloc-
ity away from the drain field and good aeration in the
zone where aerobic decomposition should occur. Atypi-
cal onsite sewage disposal system (OSDS) ordinance
might require, for example, a minimum of at least 24 in.
between the bottom of the absorption (drain) field and
the seasonal high watertable. Virtually every Health and
Rehabilitative Services (MRS) worker in Florida who is
familiar with OSDS permitting can cite at least one
example of a drain field totally submersed underwater
during Florida's summer wet season.
Design, siting, and construction of a proper OSDS do
not ensure proper long-term operation. Maintenance is
absolutely necessary. The typical OSDS owner is often
unknowledgeable regarding proper OSDS mainte-
nance. In fact, many owners are unaware that septic
tanks should be pumped out periodically to remove
accumulated septage. Ayers and Associates (5) re-
ported that it is "relatively common for homeowners to
have never serviced the septic tank during their occu-
pancy in the home."
Water conservation within the home can reduce waste
flow and attendant pollutant load. This extends the life
of the drain field, reduces system failures, and saves
money by increasing the time between needed pumpouts.
Low-flow toilets and shower heads and "graywater"
reuse are examples of water conservation measures
that can reduce potable water consumption. Siegrist (6)
reported that eliminating the use of garbage disposals
in connection with OSDSs could decrease the total sus-
pended solids load by as much as 37 percent.
A host of findings in the literature support the develop-
ment of ordinances to regulate septic systems. Interest-
ingly, Cooper and Rezek (7) found that most of the
heavy metals in the typical OSDS effluent stream origi-
nated from pigments used in cosmetics. In addition, EPA
(8) found that compounds from septic tank cleaning
solvents (i.e., methylene chloride and trichloroethane)
actually hinder septic tank operation by killing bacteria
that promote decomposition. Bicki et al. (9) concluded
that nitrate-nitrogen contamination of ground water by
OSDSs is a national problem and that high concentra-
tions in many areas pose a health risk to infants. Yates
and Yates (10) documented the extreme distances that
certain microorganisms can move and remain viable.
Certain viruses, because of their small size and long
survival times, were found as far as a mile from their
source in karst areas, an especially significant subsur-
face geologic feature in Florida.
Certain authors have also correlated septic tank density
(allowable units per acre) with ground-water contamina-
tion (10, 11). Recommended acceptable densities vary
greatly, with densities being a function of soils, depth to
watertable, and distance from surface water bodies.
Any entity considering a local ordinance to regulate
septic tanks can, based on the literature, consider sev-
eral options that might be more restrictive (protective)
97
-------
than existing regulations. These can relate to soils,
depth to ground water, densities, and distance to surface
water. These options may take the form of pumpout and
inspection requirements, alternative septic systems,
prohibitions (e.g., no garbage disposals), and even
"moratoriums" in already contaminated or totally unsuit-
able areas.
Regulation of Docks and Appurtenance
Structures (12)
Czerwinski and McPherson (12) thoroughly defined the
various classes of docks and marinas (e.g., private sin-
gle family, multislip residential, and commercial marinas).
The intended use and size of a facility are important from
both an impact and a regulatory standpoint, but space
does not allow us to consider these in detail; the inter-
ested reader should consult the original document orthe
condensation being prepared by Simpson (1). To be
effective, an ordinance must clearly define what is to be
regulated. It is advantageous to include definitions
within the body of the ordinance to avoid ambiguity that
could seriously limit ordinance effectiveness.
The potential need to adopt an ordinance on a local level
may be determined by considering projected increases
in the number of registered boats in an area. As an
example, in Florida there are approximately 48 boats per
thousand residents. This reflects a 300-percent increase
in the number of registered boats since 1964. Florida
ranks fourth nationally in the number of registered boats,
and the Florida Department of Natural Resources has
projected a 48-percent increase to 712,349 boats by the
year 2005 (13).
Environmental impacts associated with docks and ap-
purtenance structures (e.g., boathouses, gazebos, and
diving platforms) can be direct or indirect. Direct impacts
relate to areas adjacent to and covered by these struc-
tures, and would typically include the transitional zone
between the upland, wetland, and open water. The "lit-
toral zones provide many valuable ecological functions,
including flood storage, erosion and sedimentation con-
trol, filtration of surface water runoff, and essential habi-
tat for flora and fauna" (12). Indirect effects, which are
due to the attendant use of these structures, include
effects attributable to outboard exhausts, fuel spills,
sanitation facilities, and prop scour.
When regulating these structures, the actual construc-
tion materials should be considered. The list is long and
varied. Wood is probably the most widely used material,
particularly for single-family facilities. Whereas un-
treated wood is no match for the aquatic environment,
chemically treated wood may last for 15 to 20 years
without replacement. Chemicals used in treatment proc-
esses include ammoniacal copper arsenate (ACA),
chromated copper arsenate (CCA), creosote-coal tar
(CCT), acid copper chromate (ACC), chromated zinc
chloride (CZC), fluorochrome arsenate phenol (FCAP),
pentachlorophenol (which provides a clean, paintable
surface), and creosote-petroleum solutions (14). "Although
the pertinent regulatory agencies . . . test and register
these substances as generally safe for use," Czerwinski
and McPherson (12) concluded that "research con-
ducted in preparation of this paper revealed little data or
information on the biologic effects of wood preservatives
on (nontarget) aquatic and marine organisms."
Other construction materials include steel, aluminum,
reinforced concrete, fiberglass, and polyvinyl chloride
(PVC). Styrofoam (expanded bead foam polystyrene) is
still common in floating docks, although it may not be
the most suitable floatation material available today.
Unfortunately, bead foam polystyrene tends to break up
easily, has a long life, and may be ingested by and be
harmful to wildlife. In addition, chlorofluorocarbons are
used in the manufacturing process. Safer but more ex-
pensive alternatives such as petroleum-resistant poly-
styrene and sealed solid (as opposed to extruded) foam
are available.
Docks and appurtenance structures should not inter-
fere with navigation. In Florida, for example, a dock is
not considered a navigation hazard if it does not exceed
20 to 25 percent of the distance across the water body,
is limited to the minimum distance necessary to provide
reasonable access to navigable waters (which is gener-
ally defined to be approximately 4 ft below mean or
ordinary low water), and does not infringe upon the
main navigational channel or upon the riparian rights of
adjacent property owners. For safety reasons, docks
may be required to be fitted with navigational aids
(e.g., lights or reflectors).
Turbidity and sedimentation problems can result from
construction activities. Such impacts, however, are likely
to be small compared with other activities unless the
construction requires a large area and considerable
time, as might be the case with commercial marinas.
Florida water quality regulations, however, do not allow
turbidity in excess of 29 nephelometric turbidity units
above background in any case, and regulatory agencies
may require the installation of turbidity screens or other
protective barriers. Turbidity problems more likely arise
indirectly from effects such as prop scour as boats make
use of docking facilities.
Shading of the water column and the littoral shelf can
also affect the environment. Shading may not be a
problem in areas where a tree canopy already exists,
but obviously it can affect areas previously unshaded.
Czerwinski and McPherson (12), however, cite no sci-
entific studies on the direct effects of shading by docks
or appurtenance structures. Employing some simple sit-
ing and design criteria can avoid or at least lessen any
potential detrimental effects. Suggestions include:
98
-------
Siting in areas already shaded or in areas low in
light-dependent resources requiring protection.
Elevating structures in areas high in light-dependent
resources (e.g., grass beds).
Substantially elevating accessways, boardwalks, or
other appurtenance structures that are not as water
dependent.
Spacing of planking to allow sunlight to penetrate
(e.g., leaving 1-in. gaps between boards).
Another obvious effect is that installation of docks and
attendant structures directly alters the shoreline. In Flor-
ida, for example, a lakefront resident desiring access
may remove a 25-ft wide band of vegetation to open
water without a permit and without revegetating the
area. These areas frequently suffer clearing in associa-
tion with docks and similar structures. Depending on lot
size, then, it is conceivable that residents may remove
as much as half of the shoreline vegetation for access
without needing a permit.
Fortunately regulatory agencies may have the ability to
consider the cumulative impacts of projects in deciding
whether to issue a permit. Florida's Department of En-
vironmental Regulation, by virtue of its "dredge and fill"
responsibilities, requires a permit to construct a dock or
other structures that affect wetlands. "Therefore, these
agencies have the authority to review, suggest alterna-
tives ... or deny projects based upon the 'foreseeable,'
future cumulative impacts. However, the ability to deny
a project based upon future, anticipated cumulative im-
pacts can be subjective and is cautiously exercised due
to the potential for legal challenge. This is most likely to
be a supportable factor in project review when specific
endangered species concerns are at issue" (12).
Of course, not all shoreline changes are detrimental. For
example, a dock could expose previously densely vege-
tated areas, thus creating open sandy areas that can
provide valuable fish bedding areas. Docks and related
structures can also provide cover or serve as substrate
for aquatic organisms.
Most indirect environmental effects ascribable to docks
and appurtenance structures result from recreational
boating activity. These include potential effects from
outboard motor exhaust contaminants, prop dredging,
sanitation devices, fuel and oil spills, and antifouling
boat paints. Rather than consider most boating impacts
in detail here, the reader can refer to the review by
Wagner (15).
Antifouling paints, which prevent fouling of hulls by ma-
rine organisms (e.g., barnacles), pose an unusual prob-
lem. Traditional coatings contain lead, copper, and
organotin compounds. For antifouling, the organotins
are especially effective because they continuously re-
lease active ingredients into the water. One of the or-
ganotins, tributyltin (TBT), has gained recent notoriety.
EPA, due to the results of documented acute and
chronic effects, has proposed maximum concentrations
of 26 and 10 parts per trillion in fresh and marine water,
respectively, for the protection of fish and other aquatic
organisms. They have further proposed restricting sales
of TBT to certified commercial pesticide applicators for
use only on vessels greater than 65 ft in length.
The concepts of cumulative impacts and carrying capac-
ity are important considerations. They are, however,
difficult to implement with respect to docks and other
water-dependent structures. Czerwinski and McPher-
son (12) did not cite studies that defined how one might
set scientifically defensible limits. This is clearly an area
needing research. Although often discussed and de-
bated, regulation is difficult on this premise due to the
lack of quantifiable data.
Docks and water-dependent structures should be
located so as to minimize adverse environmental im-
pacts. Where possible, authorities should encourage
multislip facilities over the use of many individual docks.
Approval of docks should include criteria for preserving
a portion of the remaining unaffected shoreline, such as
conservation easements or shoreline buffers. Another
helpful measure may be to consider construction of
boat ramps in lieu of docks; a careful analysis, however,
is necessary to ensure consideration of increases in
boat traffic and of the need for appropriate provisions to
limit ramp usage.
The Need for, Rationale for, and
Implementation of Wildlife Dispersal
Corridors (16)
The SWIM Act was careful to stress the state's desire to
restore or preserve the natural systems associated with
its surface water bodies as well as its water quality.
There is a growing awareness among resource manag-
ers that preserving fauna and flora involves strategies
that stretch beyond watershed and governmental
boundaries. The need to implement a system of faunal
corridors may be the hardest issue to grasp in this paper,
and it is doubtful that the authors can do more than
introduce the topic. In fact, to a resource manager with
a background in water-related issues, the issue paper
developed by Harris (16) may appear exhaustive and
rhetorical and is almost certain to pose unfamiliar ques-
tions and problems.
Model ordinance language proposed with regard to this
topic (i.e., faunal corridors) was unlike the others devel-
oped. Accordingly, we have referred to the work as an
"article" rather than an ordinance. The proposed article
serves only to provide a means by which the
boundaries and natural amenities of a WCSD [Wild-
life Corridor Special District], as well as nonnatural
99
-------
characteristics and associated implications, can be
identified. Once the WCSD has been identified, and
a strategy for its protection and management devel-
oped, an ordinance is required to actually create the
WCSD. Due to the many site-specific charac-
teristics involved in defining the areal extent, physi-
cal characteristics and management implications of
the WCSD, such an ordinance is impossible to de-
velop in a "generic" form that would be applicable to
all jurisdictions and geographic areas in which the
ordinance potentially would be used. This Article
does, however, provide general guidelines for the
creation of a WCSD, while also providing a method
by which virtually all information needed for a
WCSD-creation ordinance can be collected.
Harris (16) suggested that it is not possible to appreciate
the need for implementing a system for faunal movement
corridors without first comprehending three major issues:
"1. Throughout most of North American history,
humans and their developments have oc-
curred as localized entities in an expansive
and interconnected matrix of undeveloped
natural ecosystems; now, it is the natural
systems that occur as localized entities in a
matrix of human development.
"2. The second issue is the current biological
diversity crisis. Without a keen awareness
of the breadth of the dimensions and rapid-
ity at which biological diversity (biodiversity)
is currently being eroded there can be no
grasp of the gravity of the remedial actions
that must be taken.
"3. The third critical issue concerns the need of
plants and animals to move; without care-
fully weighing the value of plant and animal
movement corridors against other alterna-
tive conservation actions it is not possible
to achieve balance and perspective in ap-
proaching these concerns."
Harris (16) makes a semantical distinction between the
terms "wildlife" and "faunal," with faunal relating specifi-
cally to native animal species. Although it is important to
appreciate how others may apply these terms, this pa-
per applies them more or less interchangeably.
The need for implementing a system of faunal corridors
is recent. Depending on the degree of development in
an area, the need becomes more pressing in some
areas than others. The need appears great in Florida.
Historically, human developments have occurred as is-
lands in a matrix of natural ecological communities; now,
however, the pattern has changed, with unaltered natu-
ral communities occurring as islands in a predominately
human-altered environment.
As noted by Harris (16), "a confusing paradox to many
is the fact that habitat fragmentation may enhance local
wildlife diversity while simultaneously reducing native
biotic diversity at a somewhat larger scale." Harris ex-
plains this paradox is due to the action of the following
mechanisms:
Populations lose genetic integrity due to being se-
questered within patches (i.e., islands).
"Forest-interior" and "area-sensitive" species that
cannot exist within small habitat patches are lost.
Weedy species that are characteristic of disturbed
environments increase in abundance.
Important ecological processes are disrupted.
Geographic separation of populations and gene pools
can, over geologic time, lead to new species. Spatial
separation, however, which creates small isolated popu-
lations preventing gene flow, can lead to elimination of
populations and even extinction of species. As an exam-
ple, Harris (16) cites the following statistics on the de-
gree of inbreeding depression that has already occurred
in isolated populations of the Florida panther:
Of all the Florida panthers known to exist in the wild
today, less than a dozen are reproductively unrelated.
The percentage of infertile spermatozoa in all male
Florida panthers examined in recent years exceeds
90 percent.
Of all the male Florida panthers examined, only about
50 percent have two distended testicles, and "it re-
mains a matter of speculation if or when the highly
inbred males might exhibit bilateral cryptorchidism
and be unable to reproduce at all."
Roads are a significant fragmenting force because, un-
like the passive fragmentation caused by areas such as
farm fields, roads possess an active mortality-causing
forcethe associated traffic. Lalo (17) has estimated
that nationally trucks and automobiles kill as many as
100 million vertebrates annually. Over 146,000 deer
were killed on U.S. highways in 1974 (18). Adams and
Geis (19) and Voorhees and Cassel (20) present statis-
tics showing that within the contiguous 48 states and
within individual states, the amount of land set aside in
the form of national parks, wildlife refuges, and game
management areas is smaller than the land that roads
and rail right-of-ways occupy. Vehicles, including boats,
represent one of the most significant sources of mortality
for all of Florida's large threatened, rare, and endan-
gered vertebrates. These include the panther, key deer,
black bear, eagle, crocodile, and manatee. Data cited by
Harris (16) even suggest that the number of road kills
increases in direct proportion to vehicle speed.
100
-------
Roads create barriers in several ways:
They alter light, wind, temperature, humidity, evapo-
ration rates, and noise level as they create a different
microclimate in and near the right-of-ways.
Exhaust fumes cause avoidance by some species,
and heavy metals accumulate in those that occur
adjacent to roadways (21).
Pesticides used to maintain right-of-ways affect non-
target plants and animals as well.
Right-of-ways have led to the creation of a different type
of ecological community. Harris (16) cites numerous
examples of opportunistic predators that "run roadsides"
in search of prey.
Over the last 20 years, there has been an increasing
realization that habitat fragments, even relatively large
fragments, are not adequate protection for many spe-
cies; if these species are to be protected, corridors must
connect these habitat fragments. Simple green belts are
not sufficient because corridors of non-native habitat
welcome "weedy" species. Interconnecting corridors
must be consistent with the habitats they are connecting
to avoid "edge effects"; the wrong types of corridors
could conceivably hasten the spread of exotic or weedy
species. Currently in Florida, considerable funds are
being spent to "Save Our Rivers" and protect the water
quality of streams. Careful consideration and planning
could ensure that these programs accomplish a dual
function by protecting our biological diversity as well. As
Harris (16) states, "When sufficiently wide, streamside
management zones serve as critically important habitat
for many rare and endangered native species. But un-
less the streamside zones connect larger tracts of habi-
tat or protected areas they may function simply as long
narrow fragments of habitat."
Corridors are necessary to keep small fragmented
populations from being expatriated, to preserve biodi-
versity, and ultimately to allow populations to adapt to
major climatic and geologic changes. Because of the
geographic scope involved, corridors are an issue that
will require cooperation and coordination between local,
regional, and state governments and agencies.
Protecting Environmentally Sensitive
Aquatic Habitats (22)
Aquatic habitats include lakes, rivers, streams, estuaries
and bays, springs, and wetlands. These habitat areas are
typically subject to a variety of differing agency jurisdic-
tions. Quite commonly, though, ordinances developed at
the local level protect wetlands (including marshes,
swamps, bogs, ponds, and wet prairies). Local wetland
resource areas promote the local quality of life as well
as the quality of the environment. The advantages in-
clude hydrologic functions (flood control, runoff velocity
control, ground-water and surface-water recharge), water
quality benefits (erosion and sedimentation control and
removal of pollutants such as nutrients and heavy met-
als), and wildlife habitat benefits (food source, breeding,
nesting, spawning, and wildlife protection) (23).
When wetlands are allowed to remain in their natural
state, they maximize multiple benefits and achieve eco-
logical stability. Anthropogenic changes, however, can
affect the natural function and resultant benefits of the
wetland, such as change the quality of the water enter-
ing the wetland, the hydrologic cycle of the wetland, and
the physical structure of the wetland (24). Several
sources can affect the quality of water entering the
wetland, including point and nonpoint pollution, nutrient
enrichment, and sedimentation (25). The hydrologic cy-
cle of the wetland can be disrupted by well pumping,
channelization, sedimentation, upstream diversions, in-
creased surface flows, and decreased ground-water
base flows. In addition, filling, dredging, and channeliza-
tion can affect the physical structure of the wetland (26).
By identifying the sources of impacts to these valuable
areas, one can begin to develop the necessary elements
of a local ordinance that would help to restore and
maintain ecological integrity. An ordinance should ad-
dress the wetland system from a holistic perspective,
not as isolated areas. Some recommendations for a
wetlands protection ordinance include the following:
Consider individual and cumulative impacts on
aquatic habitats from anthropogenic alterations. En-
vironmentally sensitive systems can degrade from
the accumulation effect of many individual human
activities (27).
Develop specific performance standards. Perform-
ance standards will allow local governments to use
environmentally sensitive lands in a manner that
minimizes negative impacts (28).
Develop financial incentives that encourage local
property owners to protect aquatic habitats. If envi-
ronmentally sensitive areas are to be protected
through long-term management of private lands, land
owners must be compensated accordingly (29).
Develop mechanisms by which local government fa-
cilitates the property owner's efforts to protect aquatic
habitats. If proper channels exist for conservation
easements and reduced tax assessments, voluntary
efforts to protect environmentally sensitive areas may
increase (29).
Coordinate state and federal permitting processes.
Coordination at the local level will ensure compliance
with all requirements that serve to protect, enhance,
or restore environmentally sensitive areas.
101
-------
Identify state and federally exempted activities that
contribute to the degradation of aquatic habitat, and
regulate those activities locally.
Develop an appropriate definition of aquatic habitat.
An adopted definition will define the areas of jurisdiction
for local, state, and federal regulations; few definitions,
however, adequately describe all environmentally
sensitive areas (29). Local definitions can provide
greater protection for those areas not adequately pro-
tected by state or federal regulations.
Develop a long-term plan for the protection of aquatic
resource areas, and develop management objectives
that will provide the desired level of protection.
Provide for local enforcement. Taking responsibility
for local environmentally sensitive areas ensures
maximum protection.
Along with the above requirements, additional elements
can be considered:
Create a mechanism to develop site-specific upland
buffer zones.
Create a mechanism to implement fixed-distance up-
land buffer zones.
Create a mechanism to implement no construction/no
disturbance zones.
Allow for restoration of disturbed areas at ratios
greater than 1:1.
Incorporate endangered, threatened, and special-
concern species into upland buffer zone consid-
eration.
Encourage the use of creative site planning to pre-
serve and protect sensitive aquatic habitats.
Vegetative Buffer Zones (30)
A transition zone is an area between a water body
(e.g., wetland, lake, river) and upland areas. The area
of land that a transition zone occupies varies and is
greatly influenced by topography. In areas of major topo-
graphic changes, the transition zone tends to be small
(1 to 2 ft). In areas where topographic changes are
slight, the transition zone tends to increase in size sub-
stantially (30 to 50 ft).
Vegetative transition zones provide multiple benefits to
the surrounding area. First, they are ecologically com-
plex, as the assemblage of plants and animals can be
characteristic of the nearby water body as well as the
upland area. Within these areas, substantial ecological
diversity can occur (31-33). Second, transition zones
help maintain a balanced hydrologic cycle by retarding
the flow of surface runoff volumes through absorption
and by allowing for infiltration into the ground water.
Vegetative transition zones also play a major role in the
maintenance of the quality of the nearby water resource.
Processes such as deposition, absorption, and transfor-
mation help remove pollutants such as sediment, phos-
phorus, nitrogen, and heavy metals from overland flows.
Also, when vegetation is present, it tends to reduce the
temperature of storm flows, thereby maintaining water
body temperatures (34, 35).
When activities related to urbanization disturb vegetative
transition zones, the benefits realized can be diminished
or even lost. With the removal of the complex ecological
area, habitat values decrease, resulting in a loss of
species diversity and richness (36, 37). Urbanization
activities can also disrupt the hydrologic balance of the
nearby water bodies. Typically, surface water hydrology
changes to reflect the increase in the volume and rate
of surface flows. This causes increased streambank
erosion adjacent to the disturbed area as well as down-
stream. Streambank erosion reduces water clarity, de-
stroys benthic habitat, interferes with aquatic plant
transpiration processes, and reduces stream storage
capacity. Removal of vegetative transition zones affects
ground-water flow by reducing the overall infiltration rate
of surface water to ground water. The decrease in sur-
face water recharge can affect the hydroperiod of
nearby wetlands, which are heavily dependant on
ground-water discharge, and nearby stream base flows.
Removing transition zones also affects water quality by
allowing pollutants to enter the watercourse untreated.
One of the most obvious water quality impacts is the
increase in sedimentation to the receiving waters (30).
Because vegetative transition areas provide such valu-
able ecological benefits, protection measures need to
be implemented to ensure their preservation. The size
of these areas, however, tends to be site specific and
requires individualized management approaches.
Therefore, local ordinances are the most effective and
adaptive tool to facilitate preservation.
In developing an ordinance for vegetative transition zones,
efforts should maximize the benefits to wildlife, habitat,
hydrology, and water quality. Methodologies have been
developed to "engineer" vegetative transition areas in a
supportable, defensible manner. In general, the recom-
mendations for vegetative transition areas are:
Minimize disturbances of vegetative transition zone
when possible through the use of site fingerprinting.
Limiting the extent of disturbance will greatly reduce
the potential of negative water quality impacts.
Develop local requirements for "no-build" and "no-
disturbance" zones. Protective buffer zones can be
implemented in such a way to allow for construction
while minimizing the impact of development.
Encourage alternative land use planning that can pro-
tect vegetative transition areas. Planning techniques
102
-------
are valuable tools that can afford long-term protection
and management of vegetative areas.
Develop criteria for vegetative transition areas based
on defensible procedures. This is an important step
that will implement vegetation protection measures in
a nonbiased manner. Based on identifiable and sci-
entific procedures, arguments can be made for suc-
cessful long-term implementation.
Examples of recommendations for vegetative transition
follow:
Area size of 30 to 550 ft may be necessary when
ground-water drawdown is an issue (using surficial
aquifer data and structure drawdown calculations).
Area size of 75 ft for coarse sand, 200 ft for fine sand,
and 450 ft for silty soils should be considered to
protect water quality (utilizing Technical Release [TR]
55, local soils data, and soil deposition formula).
Area size of 322 ft for fresh and saltwater marshes,
550 ft for hardwood swamps, and 732 ft for border-
ing sandhill communities to protect wildlife habitat
(based on indicator species and 50 percent other
present species).
Providing for Erosion Control and
Bank/Shoreline Stabilization (38)
Banks and shorelines are those areas that occur along
streams, lakes, ponds, rivers, wetlands, and estuaries
where water meets land. The topography of banks and
shorelines can range from very steep to very gradual.
These areas can be considered a subset of the vegeta-
tive transition areas.
Banks and shorelines provide many benefits to the
environment, including prevention of erosion, storage
and attenuation of runoff, and provision of valuable
habitat for fish and wildlife (39). Stabilization, which
prevents erosion, occurs below the water line via root
systems, as well as above the water line through ab-
sorption of raindrop energy and overland flow velocity.
Both physical characteristics and stability of the bank
and/or shoreline accomplish the storage and attenuation
of runoff. The provision of habitat is also accomplished
through physical stability and the unique physical char-
acteristics of the bank and/or shoreline. Often, ecologi-
cal zones will be apparent and consistent with the
shoreline, and provide special habitat for various plant
and animal species (29).
As water bodies continue to support human activities
both on and near the water, impacts will occur to the
bank and shoreline area. Flows of increased water
movement from activities such as boating can cause
erosion, damage to vegetation, and increased turbidity
in aquatic habitat areas (40). Urbanization commonly
results in a change in the surface water hydrology, in-
creasing storm volumes and rates of discharge. This
movement of storm flows through water channels tends
to erode and undercut banks and shorelines overtime.
The resultant erosion reduces water quality through in-
creased turbidity as well as destruction of existing bank
and shoreline habitat and smothering of downstream
habitat areas (29, 41).
Bank and shoreline stabilization is an important element
necessary to protect multiple ecological benefits. Ordi-
nances that recognize this can be developed to address
local management needs. Bank and shoreline stabiliza-
tion typically should include an array of approaches as
outlined below:
Promote nonstructural methods such as revegetation
and preservation of vegetation because they are an
inexpensive and beneficial approach. Studies have
shown that nonstructural practices can provide mul-
tiple benefits to bank and shoreline areas where im-
plemented. Also, construction costs are substantially
lower than traditional structural methods (41, 42).
Limit use of structural methods to when erosive
forces are significant. Public perceptions and aesthet-
ics have led to the construction of structural methods
in areas where nonstructural methods could have
worked. Structural methods should be the last option
when addressing bank or shoreline erosion.
Develop an appropriate definition for banks and
shorelines. Good definitions provide jurisdictional
boundaries to those attempting to implement protec-
tion measures.
Develop a long-term comprehensive plan for the pro-
tection of banks and shorelines. Comprehensive
planning will ensure that bank areas and shorelines
remain in their natural state.
Additional recommendations for the protection and pres-
ervation of banks and shorelines can include:
Meet environmental goals through shoreline stabili-
zation regulations that are performance based (not
numerical).
Allow for flexibility to integrate structural and non-
structural methods.
Address instability caused by water-based and land-
based activities.
Develop financial incentives that encourage the local
property owner to employ nonstructural techniques.
Prohibit the use of noxious plants while encouraging
the use of native plant species.
Provide design standards.
103
-------
Conclusion
While much of the information considered in this paper
was gathered with a focus on Florida, we feel it can
be extrapolated to other states. Although ordinances
can be enacted to address singular issues, it is better
to develop a more comprehensive approach to devel-
opment review. This kind of approach can eliminate
potential duplicity while maximizing environmental benefit.
The issues addressed above range widely, but environ-
mental integrity and preservation are common themes.
Enactment of an ordinance rarely occurs without chal-
lenge, but its chance of passage can only be increased by
a scientifically justifiable and legally defensible argument.
References
1. Simpson, J. 1994. Developing local ordinances: Seven issues
related to nonpoint source management. Herndon, VA: Olem
Associates. In press.
2. Kelly, M.H. 1990. Florida's SWIM program for lake, wetland, and
estuarine management. In: Enhancing states' lake/wetland pro-
grams. Chicago, IL: Northeastern Illinois Planning Commission.
pp. 159-166.
3. Martin, R., and J.K. McPherson. 1991. Placement and mainte-
nance of individual septic systems (onsite disposal systems).
Crystal Lake, FL: Henigar and Ray, Inc.
4. Brown, R.B. 1990. Soils and onsite wastewater disposal. Soil Sci.
41:1-11.
5. Ayers and Associates. 1987. Onsite sewage disposal system
research in Florida: Impact of Florida's growth on the use of
onsite sewage disposal systems. Report to the Florida Depart-
ment of Health and Rehabilitative Services, Tallahassee, FL.
6. Siegrist, R.L. 1977. Waste segregation to facilitation onsite waste-
water disposal alternative. In: Home sewage treatment. Pro-
ceedings of the Second National Home Sewage Treatment
Symposium. Publication No. 5-77. St. Joseph, Ml: American
Society of Agricultural Engineering.
7. Cooper, I.A., and J.W Rezek. 1977. Septage disposal in waste-
water treatment plants. In: McClelland, N.I., ed. Individual onsite
wastewater systems. Ann Arbor, Ml: Ann Arbor Science Publica-
tions.
8. U.S. EPA. 1986. Septic systems and ground-water protection: A
program manager's guide and reference book. EPA/440/6-86/006.
9. Bicki, T.J., R.B. Brown, E.E. Collins, R.S. Mansell, and D.F. Roth-
well. 1984. Impact of onsite sewage disposal systems on surface
and groundwater quality. Report to the Florida Department of
Health and Rehabilitative Services, Tallahassee, FL.
10. Yates, M.V., and S.R. Yates. 1989. Septic tank setback distances:
Away to minimize virus contamination of drinking water. Ground
Water 27:202-208.
11. Brown, R.B., and T.J. Bicki. 1987. Onsite sewage disposal: Influ-
ence of system densities on water quality. Soil Sci. 31:1-10.
12. Czerwinski, M.G., and J.K. McPherson. 1991. Regulation of docks
and appurtenance structures. Crystal River, FL: Henigar and Ray, Inc.
13. Florida Statistical Abstract. 1989. University of Florida, Bureau of
Economic and Business Statistics.
14. Webb, D.A., and L.R. Gjovik. 1988. Treated wood products:
Their effect on the environment. American Wood Preservers
Association.
15. Wagner, J.J. 1991. Assessing impacts of motorized watercraft on
lakes: Issues and perceptions. Proceedings of a National Confer-
ence on Enhancing States' Lake Management Programs, May
17-18, 1990. Chicago, IL: Northeastern Illinois Planning Com-
mission, pp. 77-93.
16. Harris, L.D. 1991. The need, rationale, and implementation of
wildlife dispersal corridors. Report submitted to the Southwest
Florida Water Management District, Brooksville, FL.
17. Lalo, J. 1987. The problem of road kill. Am. Forests 72:50-52.
18. Feldhamer, G.A., J.E. Gates, D.M. Harman, A.J. Loranger, and
K.R. Dixon. 1986. Effects of interstate highway fencing on white-
tailed deer activity. J. Wildlife Mgmt. 50:497-503.
19. Adams, L.A., and A.D. Geis. 1981. Effects of highways on wildlife.
Report No. FHWA/RD-81/067. Washington, DC: U.S. Depart-
ment of Transportation.
20. Voorhees, L.D., and J.F. Cassell. 1980. Highway right-of-way
mowing versus succession as related to duck nesting. J. Wildlife
Mgmt. 44:155-163.
21. Scanlon, P.F. 1987. Heavy metals in small mammals in roadside
environments: Implications for food chains. Science and Total
Environment 59:317-323.
22. Cook, C., J.K. McPherson, and R. McCormick. 1991. Protecting
environmentally sensitive aquatic habitats. Crystal River, FL:
Henigar and Ray, Inc.
23. Zedler, J.B., R. Lnagis, J. Cantilli, M. Zalejko, K. Swift, and S.
Rutherford. 1988. Assessing the functions of mitigation marshes
in Southern California. In: Urban wetlands. Proceedings of the
National Wetland Symposium, Oakland, CA. pp. 323-330.
24. Lewis, R.R. 1988. Restoring altered urban wetlands: Manage-
ment problems and recommended solutions. In: Urban wetlands.
Proceedings of the National Wetland Symposium, Oakland, CA.
25. Huber, W.C., PL. Brezonik, J.P. Heaney, R.E. Dickinson, S.D.
Preston, D.S. Dwornik, and M.A. DeMaio. 1982. A classification
of Florida lakes. Publication No. 72. Tallahassee, FL: University
of Florida, Water Resources Research Center.
26. Association of Wetlands Managers. 1988. Urban wetlands. Pro-
ceedings of the National Wetland Symposium, Oakland, CA.
27. Estevez, E.D., J. Miller, J. Morris, and R. Hamman. 1986. Execu-
tive summary of the Conference on Managing Cumulative Effects
in Florida Wetlands, Sarasota, FL, October 1985. New College
Environmental Studies Program Publication No. 37. Madison, Wl:
Omnipress.
28. Kusler, J.A. 1983. Our national wetland heritage: A protection
guidebook. Environmental Law Publication.
29. Thurow, C., W Toner, and D. Erley. 1977. Performance controls
for sensitive lands: A practical guide for local administrators.
American Society of Planning Officials.
30. Brown, M.T., J.K. McPherson, and R. McCormick. 1991. Vegeta-
tive buffer zones. Crystal River, FL: Henigar and Ray, Inc.
31. Allen, D.L. 1962. Our wildlife legacy. New York, NY: Funk and
Wagnalls.
32. MacArthur, R.H., and E.R. Planka. 1966. An optimal use of a
patchy environment. Am. Natl. IOU 916:603-609.
33. Ranney, J.D. 1977. Forest island edges: Their structure, devel-
opment, and importance to regional forest ecosystem dynamics.
Publication No. 1069. Oak Ridge, TN: Oak Ridge National Labo-
ratory, Environmental Sciences Division.
34. Hewlett, J.D. 1982. Principles of forest hydrology. Athens, GA:
University of Georgia Press.
35. Borman, F.G. et al. 1968. Nutrient loss acceleration by clearcut-
ting of a forest ecosystem. Science 159:882-884.
104
-------
36. Faaborg, J. 1980. Potential uses and abuses of diversity concepts
in wildlife management. Trans. Missouri Acad. Sci. 14:41-49.
37. Noss, R.F. 1983. A regional landscape approach to maintain
diversity. BioScience 33(11):299-309.
38. Shaw, S.K., J.K. McPherson, and F. Flannery. 1991. Providing for
erosion control and bank/shoreline stabilization while maintaining
the natural shoreline and its associated biota. Crystal River, FL:
Henigar and Ray, Inc.
39. Keown, M.P. 1983. Streambank protection guidelines for land-
owners and local governments. Vicksburg, MS: U.S. Army Engi-
neer Waterways Experiment Station.
40. Liddle, M.J., and H.R.A. Scorgie. 1980. The effects of recreation
on freshwater plants and animals: A review. Biol. Conserv. 17:183.
41. Allen, H.H. 1990. Biotechnical reservoir shoreline stabilization.
U.S. Army Corps of Engineers, Environmental Impact Research
Program, Wildlife Resource Note 8(1):1.
42. McComas, S., D. Jansen, J. Marter, and D. Roseboom. 1986.
Shoreline protection. In: Lake and reservoir management, Vol. II.
Proceedings of the Fifth Annual Conference and International Sym-
posium on Applied Lake and Watershed Management, pp. 421-4.
105
-------
Urban Runoff Pollution Prevention and Control Planning:
San Francisco Bay Experiences
Thomas E. Mumley
California Regional Water Quality Control Board, San Francisco Bay Region
Oakland, California
Abstract
The California Regional Water Quality Control Board,
San Francisco Bay Region, began a program for con-
trol of urban runoff pollution in 1987. The initial focus of
the program has been on the municipalities in Santa
Clara and Alameda counties. Both county programs
followed a similar methodology consisting of the follow-
ing steps: establish program goals and framework; com-
pile existing information; assess water quality problems
through collection and analysis of data and modeling of
pollutant loads; identify, screen, and select appropriate
control measures; and establish a plan for implementa-
tion. The Alameda program had the benefit of lagging
behind the Santa Clara program by about 1 year. This
provided the Alameda program with the advantage of
streamlining efforts based on the successes of the
Santa Clara program.
The experiences of these programs provide even further
insight into streamlining and optimizing the planning
process. Understanding the benefits of each step of the
planning process enables a municipality to focus limited
resources on the more critical factors affecting develop-
ment of an implementation plan. For example, a munici-
pality may weigh the cost of obtaining new data to make
more informed decisions with the risk associated with
making assumptions in the selection and implementa-
tion of control measures in lieu of data acquisition. Les-
sons learned to date are now being utilized by other
municipalities in the San Francisco Bay area, leading
towards timely and cost-effective development of urban
runoff management programs.
Introduction
The California Regional Water Quality Control Board,
San Francisco Bay Region (Regional Board), is the
state water pollution control agency responsible for pro-
tecting the beneficial uses of San Francisco Bay and its
tributaries. San Francisco Bay is a highly urbanized
estuary and, as such, receives significant loads of pol-
lutants through discharges of urban runoff. The Re-
gional Board began a program for control of urban runoff
on a watershed basis in 1987. The goals of the Regional
Board's program are to protect beneficial uses through
attainment of water quality standards in waters of the
region and to reduce pollutants in urban runoff to the
maximum extent practicable. These two goals reflect a
dual water quality and technology based approach and
serve to integrate specific regulatory programs such as
the stormwater National Pollutant Discharge Elimination
System (NPDES) permit program. The Regional Board
has promoted an areawide approach, with the initial
focus of the program on the municipalities in Santa Clara
and Alameda counties. This has led to the development of
a pseudowatershed-based program in each county.
The Regional Board program goals also serve as the
primary goals of the specific municipal urban runoff
programs. We recognize, however, that attainment of
such broadly defined goals can only be achieved
through a carefully planned strategy. Both county pro-
grams followed a similar strategy consisting of the fol-
lowing steps: establish program goals and framework;
compile existing information; assess water quality prob-
lems through collection and analysis of data and mod-
eling of pollutant loads; identify, screen, and select
appropriate control measures; and establish a plan for
implementation. Normally, such steps would proceed in
sequence. Wth an understanding of the purpose of
each step and its relation to the others, however, one
may consider a nonsequential or parallel process. The
Alameda program commenced approximately 1 year
after the Santa Clara program and had the advantage
of being able to streamline efforts based on the suc-
cesses of the Santa Clara program. The lessons learned
by the Santa Clara and Alameda programs provide valu-
able insight for optimizing the planning process.
106
-------
The Regional Board served as a facilitator in the devel-
opment of both the programs, but it has been the coop-
erative, proactive approach of the municipalities that has
resulted in the development of a technically sound and
cost-effective urban runoff program. The following dis-
cussion reflects the experiences and accomplishments
of the Regional Board and the Santa Clara and Alameda
programs.
Planning Strategy Steps
Program Framework
Development of an effective urban runoff management
program first requires an effective framework that in-
volves participation by all pertinent municipal agencies.
Initiation of both county programs began with creation
of a task force with participants from city and county
public works, city and county planning, sewage treat-
ment works, and flood control. The task force served as a
forum for communication among the involved agencies, as
well as an oversight body to track all the steps of the
planning process. Specific activities included estab-
lishment of program goals, development of a memoran-
dum of agreement among the participating agencies,
designation of a lead agency for anticipated contracts, and
development of a work plan for the planning strategy. The
work plan identified the specific tasks and timelines of the
planning strategy, identified responsible parties and con-
sultant needs, and identified the financial resources nec-
essary for completion of the planning process.
Both programs relied on extensive consulting services
for preparation of the planning process work plan and
implementation of the planning tasks. Although the pro-
grams benefited from this approach, an overreliance on
outside help may result in insufficient awareness and
expertise within the ultimate implementation agencies of
the urban runoff management program. An effective
approach should use new or existing municipal person-
nel as much as possible throughout the planning proc-
ess. Outside services may play a valuable role, but they
will be most effective when specific technical or other
needs have been identified and communication and
cross training with municipal staff are provided.
Compilation of Existing Information
Identification and compilation of existing information are
essential early steps in the process. The Alameda and
Santa Clara programs benefited from these steps for
several reasons, including that they provided a learning
experience on the importance of the relationship of land-
use information to water quality. Much pertinent informa-
tion already existed, and many existing municipal
activities were involved in the management of urban
runoff and pollutant sources. This information was criti-
cal to the identification of monitoring, modeling, and
mapping needs, and to the selection of appropriate
control measures.
Neither of the programs chose to focus resources on
detailed mapping efforts. Rather, available maps were
used to compile information. Development of more de-
tailed maps, specifically geographical information sys-
tems, was deferred to the implementation phase of the
program when funding mechanisms would be in place
and the cost could be better justified.
Monitoring and Modeling
Both the Santa Clara and Alameda programs conducted
comprehensive monitoring and modeling programs (1,
2). The objectives of these programs were to charac-
terize existing water quality conditions within storm
drains and urban creeks and to estimate urban runoff
pollutant loading. The programs included hydrologic
monitoring, wet and dry weather water quality monitor-
ing, sediment monitoring, and toxicity monitoring using
acute and chronic bioassays. Data were compiled and
used to calibrate and verify the Storm Water Manage-
ment Model for estimating pollutant loads. The load
estimates were also used to compare the relative con-
tributions of treated wastewater and urban runoff dis-
charges to the bay.
Results of both monitoring programs were similar.
Heavy metal concentrations in receiving waters in-
creased during wet weather. The metals primarily de-
tected were cadmium, copper, lead, nickel, and zinc.
Pesticides and petroleum hydrocarbons were prevalent
in sediments. Metal concentrations were distinctly differ-
ent for discharges from open space, commercial/resi-
dential, and industrial areas. It was also determined that
annual urban runoff pollutant loads were equal to or
greater than treated wastewater discharges, depending
on the amount of precipitation.
Each of the monitoring and modeling programs cost
from $1 to $2 million. Much valuable information was
gained, and there were strong driving forces for obtain-
ing the pollutant load information. Future programs may
not have this level of available resources during the
planning process, however. Municipalities must weigh
the cost of obtaining new data to make more informed
decisions with the risk associated with making assump-
tions in the selection and implementation of control
measures in lieu of data acquisition. Newly developing
programs in the San Francisco Bay Area are taking this
latter approach, in part benefiting from the information
developed by the Santa Clara and Alameda programs.
Selection of Control Measures
The process of selecting appropriate urban runoff pollu-
tion control measures involves three steps: 1) compila-
tion of candidate control measures, 2) consideration of
107
-------
the candidate measures based on screening criteria,
and 3) selection of control measures (3). The key to the
success of the process was establishing meaningful
selection criteria. The selection criteria addressed pol-
lutant control effectiveness, reliability, and sustainability;
capital, operation, and maintenance costs; public and
agency acceptability; consistency with regulatory re-
quirements; and legal and environmental liability.
An inventory of candidate control measures was devel-
oped through a review of technical literature and other
urban runoff control programs. In addition, technical and
managerial personnel from other state, county, and city
agencies were interviewed. This initial screening pro-
duced a list of 92 separate candidate control measures.
Upon application of the established screening criteria,
the list was reduced to 59 control measures. The final
step involved consideration of the overall costs of imple-
menting all the control measures, with priority given to
pollution prevention and source control measures over
structural or treatment based controls. This final step
ultimately lead to the selection of 41 separate control
measures for implementation.
The Alameda program had the advantage of following
the Santa Clara program. Consequently, the Alameda
program streamlined the process by capitalizing on the
efforts and progress of the Santa Clara program. The
Alameda program also factored in the requirements of
the storm water NPDES regulations. As more programs
are developed, we expect the selection process to be-
come even more streamlined, particularly in areas of
similar land use and climatic conditions such as the San
Francisco Bay area.
Implementation Plan
The final stage of the planning process is to develop a
plan for implementation of control measures. The imple-
mentation plan should provide a clear framework of
stated goals, tasks to achieve them, an evaluation proc-
ess, and a mechanism for modification of the plan based
on program successes and failures. The task forces of
the Santa Clara and Alameda programs played a critical
role in the development of their implementation plans.
The multiagency involvement on the tasks forces al-
lowed for a consensus-building process that resulted in
establishing responsible agencies and institutional ar-
rangements for implementation.
The Regional Board did not intend to require immediate
implementation of all control measures. Through in-
volvement with the respective task forces, high-priority,
early-action measures were identified, and schedules
for phased implementation of the remaining measures
were established. For example, targeted early actions
included a public information program and surveillance for
illegal discharges. Improved operation and maintenance
activities are being implemented under a phased sched-
ule where the efficiency of various inlet cleaning proce-
dures are being evaluated on a pilot scale first (4, 5).
Development of a comprehensive and effective imple-
mentation plan for an urban runoff control program is the
most critical and difficult step in the planning process.
The difficulties encountered are generally nontechnical
in nature and involve legal, financial, and institutional
limitations. The key to avoiding or overcoming such
limitations is recognizing them early in the planning
process and integrating their solution into the planning
process. For example, the planning process work plan
should include tasks to address legal authorities, funding
mechanisms, and institutional arrangements, rather than
waiting until a technical implementation plan is drafted. In
essence, development of the implementation plan should
commence with initiation of the planning process.
Conclusions
Development of an effective urban runoff control pro-
gram requires a well-defined planning strategy. The ex-
periences of the Regional Board and the Santa Clara
and Alameda programs provide insight on how to effi-
ciently proceed through the planning process. Under-
standing the benefits of each step of the planning
process enables a municipality to focus limited re-
sources on the more critical factors affecting develop-
ment of an implementation plan. These factors include
a multiagency task force; clear goals and a work plan
for the planning process; compilation of all available
information, with a strong emphasis on review of other
programs; strategic focus of monitoring, modeling, and
mapping resources; criteria for selection of control
measures; and the foresight to commence development
of the implementation plan at the beginning of the plan-
ning process. Lessons learned to date are now being
used by other municipalities in the San Francisco Bay
area, leading to timely and cost-effective development
of urban runoff management programs.
References
1. Woodward-Clyde Consultants. 1991. Santa Clara Valley nonpoint
source study, Vol. I: Loads assessment report. Santa Clara Valley
Nonpoint Source Pollution Control Program.
2. Woodward-Clyde Consultants. 1991. Loads assessment report.
Alameda County Urban Runoff Clean Water Program.
3. Woodward-Clyde Consultants. 1989. Santa Clara Valley nonpoint
source study, Vol. II: Control measure report. Santa Clara Valley
Nonpoint Source Pollution Control Program.
4. SCVNSPCP. 1991. Storm water management plan. Santa Clara
Valley Nonpoint Source Pollution Control Program.
5. ACURCWP. 1991. Storm water management plan. Alameda
County Urban Runoff Clean Water Program.
108
-------
Whole Basin Planning: Practical Lessons Learned From
North Carolina, Delaware, and Washington
Michael L. Bowman
Tetra Tech, Inc., Owings Mills, Maryland
Clayton S. Creager
The Cadmus Group, Inc., Petaluma, California
Abstract
Governments at all levels are broadening their view of
water quality protection and are developing and imple-
menting innovative strategies to achieve greater water
resources protection. Many of these efforts center on
"whole basin planning," which encourages active coordi-
nation across the full range of resource management pro-
grams to maximize the efficiency of program planning and
administration, data collection and analysis, pollution
prevention and control implementation, habitat protection
and restoration, permitting, and enforcement.
Basin planning consists of two phases. The first develops
the design of the state- or multistate-specific framework
under which basin planning will be performed. The
second phase implements the basin planning process.
North Carolina, Delaware, and Washington have each
employed a consensus-building, workshop-based
process to develop planning frameworks. Delaware
and Washington are currently in the framework design
phase. North Carolina implemented basinwide plan-
ning in 1991. Preliminary results are encouraging, with
improvements to the state's monitoring program, data
management, analysis and assessment, and water quality
program administrative functions being demonstrated.
Several aspects of the framework development process
as employed in these three states stand out as practical
suggestions for other states and federal and local agen-
cies considering basin planning:
Clearly define the state-specific objectives to be
achieved.
Encourage stakeholder involvement at the agency
staff level.
Allow time for discussion of ideas and iterations dur-
ing framework development.
Build in flexibility to the process development and
basin planning processes.
Define issues to address in order to translate objec-
tives for basin planning into specific tasks.
Implementing basin planning, the states found, does not
necessarily lead to disruption of existing programs.
What Is Whole Basin Planning?
There is a growing awareness in the United States
that point source water pollution control programs have
been successful, but that nonpoint sources, ground-
water contamination (1, 2), and habitat degradation (3)
continue to diminish the quality of the nation's water
supply. Point source chemical controls, while largely
effective, have not led to the achievement, mainte-
nance, nor protection of the three supporting compo-
nents of clean water provided in Section 101 (a) of the
Clean Water Act (CWA): chemical, physical, and biologi-
cal integrity. Nonchemical stressors resulting from
nonpoint source pollution (e.g., "clean sediment," in-
creased stream temperature, highly modified flow re-
gimes) can lead to direct and indirect impacts on
physical and biological integrity. A broad perspective on
water resources management is required to reduce and
eliminate such stresses. Government agencies at fed-
eral, state, and local levels are widening their views of
water quality protection and are developing and imple-
menting innovative strategies to achieve greater water
resources protection. Many of these efforts center on the
concept of a "whole basin planning" (WBP) approach,
which realigns water pollution control programs to operate
in a more comprehensive and coordinated fashion.
The underpinnings of basin planning can be found in
federal legislation, notably numerous sections of the
CWA (Table 1). Section 303(e) explicitly requires each
109
-------
Table 1. Sections of the CWA That Support Basin Planning
(Adapted From Craeger et al. [4])
Section Applicable Content
201 (c) To the extent practicable, waste treatment
management shall be on an areawide basis.
208 Several clauses of this section call for areawide
planning, reporting, and pollutant control.
303(d) Subsection 1A. Each state shall identify waters within
its boundaries which are water quality limited. The
state shall establish a priority ranking for such waters.
303(d) Subsection 1C. States shall establish TMDLs for the
identified water quality limited waters.
303(e) Establishes a continuing planning process that
includes effluent limits and compliance schedules,
applicable areawide waste management plans (§208)
and basin plans (§209), TMDLs per §303(d), revision
procedures, authority for intergovernmental
cooperation, implementation including compliance
schedules, residual waste disposal controls, and a
prioritized inventory and ranking of waste treatment
construction needs.
319(a) Nonpoint source management program, state
assessment reports.
319(b) Nonpoint source management program, state
management plans.
319(b) Section 4. States shall develop and implement
management programs on a watershed basis.
320 Comprehensive management plans to be developed
over large geographic area for estuaries in National
Estuary Program.
state to develop an areawide planning process for all
navigable waters in the state to address a broad range
of water quality issues. Sections 303(d) and 319 implicitly
require or support basin planning. Section 303(d) requires
states to define total maximum daily loads (TMDLs), as
well as associated wasteload allocations for point sources
and load allocations for nonpoint sources, to ensure the
attainment of water quality standards within all surface
waters. Section 319 requires watershed-based nonpoint
source management programs. Section 320 establishes
the National Estuary Program and requires the develop-
ment of management plans for estuaries included in the
program. The estuarine zone is broadly defined as ex-
tending to the upstream limit of historic anadromous fish
migration or head of tide. Thus, the management plans
must be prepared for broad geographic areas. In addi-
tion to the CWA, the Coastal Zone Act Reauthorization
Amendments of 1990 (CZARA) included Section 6217,
which requires coastal states with approved coastal
management programs to develop Coastal Nonpoint
Source Control Programs. During a review of state
coastal zone boundaries required by Section 6217, Na-
tional Oceanic and Atmospheric Administration will use
U.S. Geological Survey (USGS) mapping units as the
basis for examining state delineations of coastal water-
sheds (5). Section 6217 requirements provide implicit
support for whole basin planning.
In a recent paper discussing integrated basin manage-
ment, Downs et al. (6) identify five main facets that should
be included when addressing the physical and biological
attributes of river basins (Figure 1). Explicit incorporation
of water, channels, land, ecology, and human activities
management into the planning, design and implementa-
tion phases of aquatic resources management increases
the likelihood that cumulative, incremental losses to
resource quality and quantity will be identified and ad-
dressed. Whole basin planning encourages active coor-
dination across the full range of resource management
programs to maximize efficiency of program planning,
data collection and analysis, pollution prevention and
control implementation, habitat protection and restora-
tion, permitting, and enforcement. Mitchell (7) recom-
mends a two-stage strategy to achieve truly coordinated
management of resources in river basins. The first,
conceptual stage is an identification of the widest pos-
sible range of issues and variables. The second, opera-
tional stage involves an integrated, focused approach that
concentrates on the issues identified as most significant.
The U.S. Environmental Protection Agency (EPA) has
recognized the value of taking a wider view of water
quality protection. Through the Office of Water, EPA
encourages states to implement watershed protection
and basin planning and has formulated three main prin-
ciples to guide its support for state efforts in this area
(8):
Risk-based geographic targeting
Stakeholder involvement
Integrated solutions
Risk-Based Geographic Targeting
"Risk" in the context of whole basin planning refers to
indication of impairment to human health, ecological
Attributes
M
)
^
anagement
Facets
Water
Aspects of Each
Management Facet
Water quality control
Hydwlogfcaf regulation
River Channe
Channel control
Land
Land degradation control
Land ase regulation
Eco!tW ' restoration
Human
Activities
Sooioeoonomic
benefits
Figure 1. Facets of river basin management to include in basin
planning (adapted from Downs et al. [6]).
110
-------
resources, designated uses of the waterbodies, or a
combination of these, resulting from manmade pollution
and natural processes, based on a review of environ-
mental data. A probabilistic approach, as is used in
ecological risk assessment (9), has not been applied in
basin planning. Phillips (10), however, argues for a prob-
abilistic approach to targeting nonpoint source pollution
control in a watershed context. Basin planning estab-
lishes a framework within which a more probabilistic risk
assessment can be performed.
Problems that may pose risks in a watershed include:
Industrial wastewater discharges.
Municipal wastewater, stormwater, or combined
sewer overflows.
Waste dumping and injection.
Nonpoint source runoff or seepage.
Accidental toxics releases.
Atmospheric deposition.
Habitat alteration, including wetlands loss.
Flow alterations.
Specific stressors within watersheds are targeted based
on their potential to produce impairment to human
health, ecological resources, or designated uses. Un-
der a whole basin planning framework, the highest risk
stressors within watersheds are identified using, for ex-
ample, waterquality and biological monitoring data, land
use information, information on location of critical re-
sources, and tools such as water quality models and
geographic information systems (CIS). The stressors
with the greatest potential to yield impairments are tar-
geted for integrated assessment and corrective action
involving cooperative efforts between multiple jurisdic-
tions and interest groups. The targeting process may
range from qualitative ranking to computerized tech-
niques that incorporate various numeric criteria and
weighting factors (11). Difficult management problems
may not be completely addressed over the course of
one basin planning cycle (5 years is being used in North
Carolina). This can be used to advantage, however, by
breaking the identified problems into components that
can be solved, or for which measurable progress toward
a solution can be made during a cycle.
The basin planning process itself can be broken into
phases with near- and long-term goals. For example,
near-term goals could include coordinating the permit-
ting and monitoring schedules by basin, promoting pub-
lic participation in basin planning, and expanding and
improving wasteload allocation analyses and evaluation
of nonpoint sources. Long-term goals could include op-
timizing the distribution of assimilative capacity within
basins and developing and implementing basinwide
management strategies.
Stakeholder Involvement
All parties with a stake in the specific local situation
should participate in problem analysis and creation of
solutions. The involvement of potentially affected parties
("stakeholders") during the development of basin plans
is crucial to the success of those plans. The manner in
which stakeholders are involved may vary from state to
state, but a key activity for them, regardless of location,
is to reach consensus on goals and approaches for
correcting a watershed's problems, specific actions nec-
essary to achieve those goals, and processes for coor-
dinating implementation activities and evaluating the
efficacy of problem solutions. The potential pool of
stakeholders can be very broad and should be tailored
to individual basins. Potential basin plan participants
include members of:
State environmental, public health, agricultural, and
natural resources agencies.
Local/regional boards, commissions, and agencies.
EPA water and other programs.
Other federal agencies (e.g., U.S. Department of Ag-
ricultureSoil Conservation Service, U.S. Depart-
ment of the Interior, U.S. Army Corps of Engineers).
Indian tribes.
The public.
Private wildlife and conservation organizations.
Industry.
The academic community.
The farming community.
Integrated Solutions
The basin approach provides a framework to design the
optimal mix of water quality management strategies by
integrating and coordinating across program and
agency boundaries. Integrated solutions implemented
by basin management teams use limited resources to
address the most significant water quality problems
without losing sight of and planning for other factors
contributing to the degradation of the resource. Integra-
tion through the basin approach provides a means to
achieve the short- and long-term goals for the basin by
allowing the application of resources both in a timely and
geographically targeted manner. Integrated solutions
are possible because of a framework that encourages
an interdisciplinary and interagency team to develop the
most appropriate plan rather than impose predeter-
mined solutions.
111
-------
Whole Basin Planning in Three States
Before basin planning (the second, operational stage in
Mitchell's construct [7]) per se is implemented, it must
be preceded by a process to design the framework
within which it will operate (Mitchell's first, conceptual
stage [7]). This design process will be specific to each
state that implements whole basin planning due to dif-
ferences in target resources (e.g., a large number of
rivers and streams versus lakes), the objectives of im-
plementing basin plans (e.g., a water quality permitting
focus versus an aquatic resources management focus),
and differing organizational structure and implementa-
tion constraints. We draw on experiences in North Caro-
lina, Delaware, and Washington during the framework
design stage of basin planning and identify several prac-
tical lessons that can be applied by other states, EPA
regions, or other government units.
North Carolina
The Framework
North Carolina Division of Environmental Management
(NCDEM) Water Quality Section considered an Na-
tional Pollutant Discharge Elimination System (NPDES)
basin permitting strategy as early as 1989. However,
due to resource limitations, NCDEM was unable to de-
velop a framework document describing the strategy to
submit to the North Carolina Environmental Manage-
ment Commission for approval. NCDEM submitted a
request for funding to the EPA Office of Policy, Planning
and Evaluation, Water Policy Branch, for a facilitator to
assist with the development of a basin approach for
North Carolina. This consensus-building process was
initiated in 1990.
The Process
The first step in the process involved a series of individ-
ual interviews with several members of the NCDEM
Water Quality Section staff, including all branch chiefs.
The benefits of expanding the focus from solely a
NPDES permitting strategy to more comprehensive in-
volvement of the water quality program soon became
apparent. It was also clear that there was broad-based
support for the basin approach but that individual views
of that approach varied in several critical areas. The goal
of the consensus process was to successfully synthe-
size those individual views.
The next step involved a series of small group meetings
to begin outlining a framework for the basin approach.
The results of these group meetings formed the basis
for a "straw outline" compiled by the facilitator. The straw
outline was used to provide structure for a "develop-
ment" workshop attended by a large portion of the Water
Quality Section staff. The purpose of the workshop was
to finalize the outline and identify consensus positions.
Workshop results were used to produce a draft internal
document describing the North Carolina Whole Basin
Water Quality Management Framework.
The draft framework document was distributed within
the Water Quality Section for review and comment. The
revised document was circulated to a broader audience,
including other state and federal agencies and selected
academics. The draft framework document was pre-
sented at an implementation workshop, which included
broader agency and public participation than previous
meetings. The document was revised once again based
on comments received at the implementation workshop
and submitted to the North Carolina Environment Man-
agement Commission (EMC) for approval. The EMC
approved the basin approach in 1991.
The framework document has been revised twice since
its approval by the EMC. These changes reflect needed
refinements recognized during the implementation and
development of specific basin plans. These revisions
have expanded the focus of basin plans and incorporate
broader elements of the water resources program in
North Carolina to ensure that the state's basin planning
objectives are being appropriately addressed.
The final consensus basin approach established a
rotating basin schedule for NPDES permitting, monitor-
ing, and nonpoint source program implementation. These
activities are performed for each basin on a 5-year
cycle, with several basins moving through the planning
cycle together. A general sequence of tasks over the
5-year planning cycle is illustrated in Figure 2. North
Carolina basin plans are viewed as reports to the public,
policymakers, and the regulated community. Revisions
to the framework are addressing an insufficient public
outreach program for the development of specific basin
plans. Basin plans report on the current status of
surface waters in the basin, identify major water quality
concerns and issues, summarize projected trends in
development and water quality, identify long-range man-
agement goals for the basin, present recommended
management options, and discuss implementation
plans (12). The plan also presents potential changes in
Activity
Public Outreach/Involvement
Canvas for Information
Analyze Information
Determine WQ Status
Identify Problems/Loadings
Define Management Goals
Prioritize Problems
Evaluate/Describe Mgt. Options
Select Mgt. Approach
Prepare Draft Basin Plan
Implement
01234
Year of Planning Cycle
Figure 2. General sequence of planning tasks.
112
-------
discharger waste limits and recommendations for reduc-
tions in nonpoint source loadings. North Carolina Bas-
inwide Water Quality Management Plans do not,
however, currently target specific physical habitat resto-
ration issues or projects.
Barriers to Implementation
A major impediment to the development and implemen-
tation of the North Carolina basinwide approach has
also been the greatest source of strength: the CWA. The
strength comes from the merger of traditionally regula-
tory programs, having strong legal precedence for en-
forcement, with voluntary compliance programs, which
have a strong public involvement component. Each ap-
proach has enhanced the application of the other.
The barriers result from the manner in which the CWA
has been implemented, using a programmatic approach
with specific grant and entitlement programs. This has
led to a lack of coordination and integration in address-
ing water quality issues that require comprehensive
strategies. The program funding requirements reduce
the flexibility of the state to commit funds to targeted
water quality issues.
Next Steps
A useful reform of the grants process would give states
with defined basin frameworks authority to establish
water quality priorities within basins. This approach
would also reduce redundant application and reporting
requirements that are fulfilled with the basin plans. Flexi-
bility in this regard would enhance the North Carolina
approach. EPA is currently using a trial block grant
funding program with North Carolina.
Delaware
The Framework
The Delaware Department of Natural Resources and
Environmental Control (DNREC) identified a need to
focus existing water resources programs on priority
watersheds. Basin planning will provide DNREC with
the ability to assess pollution, living resources, and habi-
tat problems, and manage Delaware's resources in a
comprehensive manner (13). The department's perspec-
tive on basin planning, explicitly incorporating living
resources and habitat degradation, from the outset of
the process is significant from several standpoints. By
including a wide range of basin management facets
(Figure 1), DNREC will be more likely to proactively
identify potential cost savings (e.g., combining aspects
of current water quality and fisheries monitoring activi-
ties), watershed stressors with multiple impacts (e.g.,
loss of vegetated riparian buffer zones, which increases
nonpoint source delivery to waterbodies and degrades
aquatic and terrestrial habitat), and solutions with
benefits to multiple resource categories (e.g., riparian
zone revegetation, which reduces nonpoint source load-
ings and improves habitat). It is less likely that DNREC
will need to "retrofit" the basin planning process at a later
stage.
The Process
DNREC's framework design process began with a
series of interviews of department staff by a facilitator
to gain a better understanding of their goals for basin
planning in Delaware. Following completion of these
interviews in late summer 1992, a workshop was held
for DNREC staff in September 1992 to provide detailed
background information on whole basin planning and
to begin to identify existing roles and responsibilities
of the various functional units within the department.
The workshop provided an opportunity for department
staff to identify perceived needs for basin planning in
Delaware and to begin an initial formulation of goals and
objectives (14).
A second workshop was held in January 1993 with
DNREC staff and representatives from other state, local,
and federal agencies. The goal of this session was to
establish commitment and direction for basin planning
in the state. The 3 months between the first and second
workshops proved to be a very fertile incubation period
for agency staff to consider the design of a planning
approach. Key outcomes of the discussions were:
Identification of a strategy of sequential involvement
of a larger group of participants as the framework
planning effort proceeds.
Firm commitment by agency staff to build the plan-
ning process from the bottom up, together with the
stakeholders who will actually implement it, rather
than imposing the plan without their input.
A clear statement that an expanded definition of
"clean water" (i.e., inclusive of biological resources,
physical habitat, and watershed linkages) would en-
sure that Delaware's basin approach is consistent
with the goals and objectives of programs and agen-
cies other than DNREC water programs. Maintaining
the focus on "clean water" will allow the regulatory
components of the basin approach to remain firmly
grounded in legal and policy precedents provided by
the CWA.
Detailed discussion of whether to 1) proceed with
immediate implementation of WBP in all basins at
once, or 2) proceed incrementally, implementing the
strategy in a single basin and then assessing the
results and modifying the framework as appropriate.
Tentative delineation of basin management units
that combine groups of Delaware's 35 watersheds
(Figure 3).
113
-------
Figure 3.
Tentative delineation of basin management units in
Delaware.
Workshop participants identified a wide range of issues
to address during the formulation of the basin planning
framework. Review groups were established to explore
these issues in greater detail and prepare specific com-
ponents of a planning framework document. Topical
areas being examined by these groups are:
Implementation, coordination, and institutional barriers
Management units, data management, and monitoring
Public outreach and education
Next Steps
The review groups will be the focus of planning activities
for several months. Following completion of their delib-
erations, a framework design workshop will be con-
vened to review the components of the planning process
proposed by the groups, to make appropriate modifica-
tions, and to establish a draft basin planning framework
for subsequent review by stakeholders.
Washington
The Framework
The Washington State Department of Ecology (DOE)
Water Quality Program (WQP), Environmental Investi-
gations and Laboratory Services (EILS), and Central
Programs are currently developing the water quality
component of a broader DOE basin approach to natural
resource management. The process is the culmination
of a long-term planning program that satisfies a state-
sponsored Efficiency Commission requirement and
also fulfills the requirements of a Memorandum of Un-
derstanding between EPA Region 10 and DOE. The
development of the basinwide water quality manage-
ment program framework document is not yet final.
Therefore, the summary description offered here is
subject to change. The development of the basinwide
approach in Washington was also assisted by an inde-
pendent facilitator.
The Process
The Washington basin approach for water quality man-
agement involves coordinating issuance of wastewater
discharge permits and nonpoint source planning con-
ducted by the WQP and Central Program's Industrial
Section (to the extent practicable). It also involves water
quality monitoring, intensive field investigations, and
TMDL development conducted by DOE's Environmental
Investigations and Laboratory Services Program. Other
programs within DOE also have developed or are devel-
oping basin approaches for their areas of responsibility
(e.g., Coastal Zone Management, wetlands). All of the
basin approaches within DOE will be merged into one
resource management program at a later date.
Beginning in mid-1993, each of the WQP's four regions
committed one basin per year to this geographically
targeted, risk-based approach. The 64 Water Resource
Inventory Areas (river basins) will be lumped into 20
basin management units. Each of the four regions will
complete a basin water quality management plan each
year. All of the basin management units across the
state will be completed in a 5-year cycle. Each
basin will be revisited every 5 years to restart the
cycle of data collection, assessment, public outreach,
planning, and implementation. Basin management
teams are active in each basin management unit every
114
-------
year of the 5-year basin management cycle. Basins are
simply staggered at different steps in the cycle. The
Washington approach is viewed as a long-term commit-
ment to a stable management structure that allows DOE
to build on previous efforts.
Integration of the DOE program with local planning
agencies is a key issue in Washington. DOE is placing
a strong emphasis on stakeholder involvement through
a public outreach program that is active at each step of
the basin cycle. The roles and responsibilities of all of
the participants on the basin planning team have not
been finalized. DOE, however, is looking for a mecha-
nism that promotes public and other agency involve-
ment in all phases of the basin planning process. The
exception would be when the regulatory activities of the
basin planning team might directly affect a participant.
Next Steps
EPA flexibility is needed in numerous program compo-
nents to facilitate DOE's transition to the basin ap-
proach, including:
Using extended/expired permits to achieve synchro-
nization of permits within basins, and because certain
permits will receive a low priority ranking for risk of
waterbody impairment.
Allowing basin plans to fulfill various CWA reporting
requirements (e.g., 305(b), 319).
Using basin plans as both numeric and qualitative
TMDLs.
Administering staff/financial resources among vari-
ous program components (e.g., number of inspec-
tions and audits).
Focusing on the results of the water quality program
rather than specific intermediate evaluation criteria.
Recognizing that certain state discharge permits
(e.g., ground water) may take precedence for man-
agement over certain NPDES permits.
EPA Region 10 and DOE are working together to
resolve these issues to the extent possible within the
current configuration of the CWA. The elimination of all
institutional barriers between EPA regional offices and
states may require some amendment of the CWA as
part of its reauthorization.
Washington is continuing to resolve internal implemen-
tation barriers by establishing a cross-program workgroup
to address issues that were identified at the develop-
ment workshop. DOE also considers the basinwide
water quality management framework document that is
developed through this current consensus process the
first phase of DOE's transition to basin resource man-
agement.
How Is Whole Basin Planning Working?
North Carolina
Although only one state is actually performing basin
planning, the results so far are encouraging. EPAs Of-
fice of Water, Watershed Branch, sponsored a survey of
the staff of the NCDEM Water Quality Section after
basin planning was initiated there. Potential improve-
ments and increased efficiency in North Carolina's water
quality program were suggested in several areas.
Monitoring Program
Following implementation of basin planning, NCDEM
was able to increase the number of water quality sam-
pling stations and parameters measured. The respon-
dents attributed this increase to the ability to optimize
sampling strategies under a basin approach. The ambi-
ent water quality monitoring network has been main-
tained. NCDEM staff anticipate further improvements to
the monitoring network as a result of increased coordi-
nation with other resource agencies and the larger role
of the regulated community in the monitoring program.
Data Management, Analysis, and Assessment
During development of a basin planning approach,
North Carolina identified major improvements to data
management and analysis (both hardware and soft-
ware) as being crucial to the success of the approach.
Improved capabilities in this area are expected to re-
duce the Water Quality Section's reliance on North
Carolina's central computing services and significantly
reduce the Section's computing costs. Cost savings will
be used to upgrade in-house hardware and software,
which will in turn allow ready access to monitoring and
geographic data needed to support basin planning.
Of particular note to municipalities is the ability to fund
a staff position with the Water Quality Section to assist
in the development of basin plans from the perspective
of fulfilling municipal stormwater planning and control
requirements. North Carolina cities will benefit from this
arrangement by being able to reduce or eliminate redun-
dant monitoring and modeling.
Significant improvements have been made in assessing
water quality issues. The development of a framework
for basin planning included integration of analysis time
requirements with monitoring schedules, thus monitor-
ing now more directly supports water quality modeling.
By shifting to a basin focus, modeling is performed for
a greater length of stream segments in the state. This
expansion allows consideration of more innovative so-
lutions to water resources management issues, such as
pollutant trading, and enhances the state's ability to
prepare TMDLs.
115
-------
Administration
North Carolina's basin approach was designed to avoid
agency reorganization. The approach has led to
changes in roles and responsibilities for staff and
branches within NCDEM. Staff resources have been
shifted to place a greater emphasis on data acquisition
and assessment. Information flow and coordination of
activities between branches has significantly increased.
A basin coordinator position was created to ensure the
timely flow of information throughout the preparation of
basin plans. In addition to improved communication and
coordination within the NCDEM, there is increased co-
operation with other local, state, and federal agencies.
Potential Benefits to the Regulated Community
Basin planning has not been in place long enough to
have provided directly measurable benefits to the regu-
lated community. However, the Water Quality Section
identifies several anticipated benefits. Consolidation of
dischargers into consortia along stream reaches will
provide an economy of scale with respect to permit
monitoring requirements. Dischargers in management
units are expected to be able to combine permit moni-
toring activities and cooperate in the preparation of as-
sessments. NCDEM also expects permits to be more
stable because of the expanded spatial and temporal
scope of assessments performed during the basin plan-
ning cycle. Basin planning allows more comprehensive
assessment of existing and proposed pollution sources,
and is more effective in accounting for future impacts.
Thus, permit conditions would need to be updated less
frequently, potentially reducing costs to both NCDEM
and permittees. Increased accuracy in the assessment
of a basin's assimilative capacity will allow better iden-
tification of the level and types of controls necessary to
achieve and maintain desired aquatic resources quality.
Basin planning will help lead to the selection of an
optimal set of pollution control methods, potentially re-
ducing costs.
Neuse River Basinwide Plan
North Carolina has implemented basinwide planning
beginning with the Neuse River basin (Figure 4). Basin-
wide plans will be prepared for the remaining 16 basins
Figure 4. North Carolina basins (Neuse River highlighted).
in the state over the next 5 years and will be updated at
5-year intervals.
North Carolina's basinwide planning process has as
primary goals "to identify and restore full use of pres-
ently impaired waters, to identify and protect highly val-
ued resource waters, and to manage problem pollutants
throughout the basin so as to maintain full use of unim-
paired waters while accommodating population in-
creases and economic growth" (12). NCDEM identified
near- and long-term objectives for its basinwide plan-
ning process that apply to the preparation of basin plans
(illustrated conceptually in figure 2). Near-term objec-
tives are defined as those fully or partially achievable
during the initial 5-year planning cycle. They include
implementing management strategies to significantly re-
duce point and nonpoint source pollution and making
measurable improvements toward addressing major is-
sues identified in each of the basin plans. Longer-term
objectives include refining the recommended basinwide
management strategies during subsequent planning cy-
cles based on the results of monitoring and implemen-
tation activities from the initial round of planning (12).
The Neuse River basinwide plan is a comprehensive
document that can serve as a model for other states
considering basin planning. An outline of the contents of
the document is provided in Table 2.
Practical Lessons From Framework
Development
As noted earlier, several states are in the process of
developing a whole basin planning framework, or have
completed the framework and implemented basin plan-
ning. Several aspects of the framework development
process in these states stand out as practical sugges-
tions for other state, federal, and local agencies that
may be considering basin planning:
Clearly define the specific objectives to be achieved:
This will determine the scope of the programs to be
involved. The objectives are a positive statement of
the issues to be addressed and resolved through the
basin approach. This step eliminates uncertainty re-
garding the focus of the consensus process. Basin
planning entails a considerable shift in thinking and
practice regarding the manner in which resources will
be managed. It moves agencies (and other stake-
holders) from programmatic-based management to
resource-based management. This shift does not
necessarily require agency reorganization, but it does
require emphasis on and sustained commitment to
extensive communication and information sharing
across programmatic lines.
Encourage stakeholder involvement at agency staff
/eve/: The basin approach allows redefinition of func-
tional relationships without formal reorganization.
116
-------
Table 2. Neuse River Basinwide Plan
Introduction
General Basin Description
Sources and Causes of
Water Pollution in the
Neuse Basin
Water Quality Status in the
Neuse Basin
Existing Point and Nonpoint
Source Control Programs
Basinwide Goals, Major
Water Quality Concerns
and Recommended
Management Strategies
for the Neuse Basin
Basinwide Plan Summary
and Future Initiatives
Purpose of the Neuse Basin
Management Plan
Guide to Use of Document
Introduction to the Basinwide
Management Approach
Basinwide Responsibilities Within
NCDEM Water Quality Section
Physical and Geographic Features
Land Use, Population, and
Growth Trends
Major Surface Water Uses and
Classifications
Introduction
Defining Causes of Pollution
Point Sources of Pollution
Nonpoint Sources of Pollution
Sources and Types of Water
Quality and Biological Data
Narrative Water Quality Subbasin
Summaries
Neuse River Mainstem
Methods for Determining Water
Quality "Use Support" Ratings
Introduction
Integrating Point and Nonpoint
Source Pollution Control
Strategies
Point Source Pollution Control
Through North Carolina's NPDES
Permitting Program
Nonpoint Source Control
Programs
Major Water Quality Concerns
and Priority Issues
Recommended Management
Strategies for Oxygen Demanding
Wastes
Management Strategies for
Nutrients
Toxics
Overview of Neuse Basinwide
Goals and Objectives
Neuse NPDES Permitting and
TMDL Strategies
Nonpoint Source Control
Strategies and Priorities
Future Modeling Priorities
Future Monitoring Priorities
Future Programmatic Initiatives
The more broadly based the transition effort, the less
confusion in the implementation of the approach. The
basin approach also "flattens" organizations by shift-
ing more decision-making responsibility to basin
teams. Therefore, staff involvement is critical to de-
velopment of the basin approach. Staff made many
valuable contributions to the process and more
eagerly embraced the approach in those states
where staff participation was encouraged.
Allow time for adequate, thorough discussion of ideas
and iterations during the development of the process
framework. Development of a basin planning process
is complex and, as noted above, requires a shift in
agency thinking and practice. Although no hard-and-
fast guidance can be given on the specific lengths of
time that are needed for each of the phases of the
framework development process, experience in three
states suggests that a minimum of 12 to 18 months
should be allowed. By allowing adequate time for
agency staff to thoroughly explore potential require-
ments of basin planning and issues identified during
the preparation of a planning framework, a much
stronger process will result.
Build in flexibility to the development process as well
as the whole basin planning process itself: The three
states discussed in this paper have all employed a
consensus-building, workshop-based process to de-
velop planning frameworks. On occasion, workshops
have been rescheduled at the last minute when it
became clear that adequate numbers of participants
would not be available because of scheduling con-
flicts. Also, workshop agendas underwent substantial
modification at the session when it became clear that
participants needed more in-depth discussion of ba-
sin planning concepts or particular issues they had
identified. These conditions should not be viewed in
a negative lightthey are almost certain to occur in
a consensus process, and the ability to respond with
flexibility is essential to maintaining the momentum
generated earlier in the process.
Define issues to address in order to translate objectives
for basin planning into specific tasks: Identification of
certain core issues is essential for translating state-
specific basin planning objectives to specific tasks
that will be accomplished in the development of basin
plans. Some issues that have been commonly iden-
tified across several states thus far include cross-pro-
gram coordination, roles and responsibilities in the
existing resource management scheme versus modi-
fications necessary to implement basin planning, pol-
icy and regulatory implications at the state and federal
level, and human and capital resources needs.
As noted above, basin planning emphasizes cross-
program communication and coordination. Institutional
and regulatory constraints, which vary from state to
state, may lead to some disruption of existing pro-
grams during the transition period. Such disruptions
can be minimized by carefully considering the steps
needed to move from programmatic to resource-
based management during the framework develop-
ment process.
117
-------
Acknowledgments
EPA and state funding have supported the work
described in this paper. EPA Contract 68-C9-0013,
Work Assignment WA 3-151, to Tetra Tech, Inc., has
supported the first author. We wish to thank the fol-
lowing individuals for their support and insights on
basin planning: Don Brady, EPA Office of Wetlands,
Oceans, and Watersheds, Watershed Branch; Bill
Painter, EPA Office of Policy, Planning, and Evaluation,
Water Policy Branch; Trevor Clements, North Carolina
Department of Natural Resources and Community
Development, Water Quality Section, Technical Serv-
ices Branch; Bob Zimmerman, Delaware Department of
Natural Resources and Environmental Control, Division
of Water Resources, Surfacewater Management Sec-
tion; and Dan Wrye, Washington Department of Ecol-
ogy, Water Quality Program, Alternative Strategies Unit.
References
1. U.S. EPA. 1992. The quality of our nation's water: 1990.
EPA/841 /K-92/001. Washington, DC.
2. U.S. EPA. 1992. National water quality inventory: 1990 report
to Congress. EPA/503/9-92/006. Washington, DC.
3. Judy, R.D., P.M. Seeley, T.M. Murray, S.C. Svirsky, M.R. Whit-
worth, and L.S. Ischinger. 1984. 1982 National fisheries survey,
Vol. I. Technical report: Initial findings. U.S. Fish and Wildlife
Service. FWS/OBS-84/06.
4. Creager, C.S., J.P. Baker, and North Carolina Division of Environ-
mental Management, Water Quality Section. 1991. North Caro-
lina's basinwide approach to water quality management:
Program description. Report No. 91-08. Prepared for the North
Carolina Department of Environment, Health, and Natural Re-
sources, Division of Environmental Management, Water Quality
Section, and U.S. EPA Office of Policy, Planning, and Evaluation.
Raleigh, NC.
5. NOAA and U.S. EPA. 1991. Coastal nonpoint pollution control
program: proposed program development and approval guid-
ance. National Oceanic and Atmospheric Administration and U.S.
Environmental Protection Agency. Washington, DC.
6. Downs, P.W, K.J. Gregory, and A. Brookes. 1991. How integrated
is river basin management? Environ. Mgmt. 15(3):299-309.
7. Mitchell, B. 1987. A comprehensive-integrated approach for land
and water management, occasional paper 1. Centre for Water
Policy Research. University of New England, Armidale, New
South Wales, Australia.
8. U.S. EPA. 1991. The watershed protection approach: An overview.
EPA/503/9-92/002. Washington, DC.
9. U.S. EPA. 1991. Summary report on issues in ecological risk
assessment. Prepared for U.S. EPA Risk Assessment Forum.
EPA/625/3-91/018. Washington, DC.
10. Phillips, J.D. 1989. Nonpoint source pollution risk assessment in
a watershed context. Environ. Mgmt. 13(4):493-502.
11. U.S. EPA. 1989. Selecting priority nonpoint source projects: You
better shop around. EPA 506/2-89/ 003. Washington, DC.
12. NCDEM. 1992. Neuse River basinwide water quality management
plan (draft). North Carolina Division of Environmental Manage-
ment. Raleigh, NC.
13. Clark, T. 1992. The watershed approach to resource manage-
ment. Memorandum from the Department of Natural Resources
and Environmental Control. Dover, DE.
14. Creager, C.S., and M.L. Bowman. 1992. Summary of whole basin
planning approach scoping meetings with Delaware DNREC
staff, September 8-9, 1992. Prepared for the Department of
Natural Resources and Environmental Control, Division of Water
Resources, Dover, DE.
118
-------
Application of Urban Targeting and Prioritization Methodology to Butterfield
Creek, Cook and Will Counties, Illinois
Dennis Dreher and Thomas Price
Northeastern Illinois Planning Commission,
Chicago, Illinois
Abstract
This paper describes the applicability of a methodology,
developed by a consultant for the U.S. Environmental
Protection Agency, to select, target, and prioritize best
management practices (BMPs) in an urban watershed.
The methodology was demonstrated in the Butterfield
Creek watershed in South Cook County, Illinois. This
watershed was selected because there are no major
point sources of discharge to the creek, thus the impacts
due to nonpoint sources alone could be addressed.
The methodology considered watershed land use, con-
tributing nonpoint sources, and stream use attainment
to identify priority areas for BMPs and then to prioritize
those areas. The primary focus of the methodology, as
originally developed, was to reduce problematic pollut-
ant loads via appropriate BMPs. One shortcoming of the
procedure was that it was limited to pollutant loads and,
therefore, was not readily able to address other factors,
such as the physical habitat impairments that affect
many urban streams. Several enhancements were
added to the methodology to address this situation. Also,
the watershed configuration made interpretation of the
prioritization results less straightforward.
The targeting methodology was enhanced in this appli-
cation by presenting stormwater runoff rate as an addi-
tional targeted factor. Similarly, BMP selection and
quantification were enhanced by representing the con-
trol of stormwater runoff rate by detention retrofitting.
Introduction
Purpose
The purpose of this paper is to report on a demonstra-
tion of a methodology developed by Woodward-Clyde
Consultants for the U.S. Environmental Protection
Agency (EPA) to select, target, and prioritize best man-
agement practices (BMPs) in an urban watershed (1).
This methodology considers watershed land use, con-
tributing nonpoint sources, and stream use attainment
to identify priority areas for BMPs. The primary focus of
the methodology, as developed, is to reduce problematic
pollutant loads via appropriate BMPs. The methodology
does not, however, address other constraints to stream
use attainment, such as hydrologic destabilization and
loss of physical habitat.
Butterfield Creek was selected for this demonstration for
several reasons. First, watershed impacts are primarily
due to nonpoint sources; there are no major point
sources of discharge to the creek. Second, a preliminary
nonpoint source management plan was being devel-
oped under a Section 319 grant, and this methodology
could be used to assist in development of that plan. As
a result, this paper presents analyses and results from both
the preliminary nonpoint source plan (2) and the target-
ing methodology application (3). These two projects
were originally documented separately, as referenced.
Assessment of Butterfield Creek problems has benefitted
from the presence of a group known as the Butterfield
Creek Steering Committee. The committee includes repre-
sentatives from seven local governments in the watershed,
and its mission is to address comprehensive stormwater
management issues. While the primary focus of the com-
mittee has been the reduction of existing flooding prob-
lems, it also has identified the protection and improvement
of water quality as major objectives. While committee
members are concerned about water quality, they are
also concerned about the potential expense of retrofit-
ting urban BMPs in already developed areas. Therefore,
a goal is to target BMPs to priority areas, where their
effectiveness is maximized.
Background
Butterfield Creek drains a 26-square-mile watershed
in Cook and Will Counties in northeastern, suburban
119
-------
Illinois. Its land use is largely residential and commercial
in downstream areas. Much of the upstream watershed
is presently undeveloped, although urbanization is an-
ticipated. Existing water quality and stream use data
indicate degraded conditions. There are no major per-
mitted point source discharges to the stream, leading to
the conclusion that nonpoint source impacts are the
likely causative factors for the observed conditions.
Targeting and Prioritization Procedure
The elements of the targeting and prioritization proce-
dure are as follows:
Characterization of the watershed, including:
- Subwatershed identification
- Land-use identification
- Nonpoint source impacts
Incorporation of additional relevant factors, based on
watershed conditions, into the documented targeting
procedure.
Calculation of pollutant loads and completion of tar-
geting table.
Prioritization of drainage areas for nonpoint control.
Characterization of Butterfield Creek
Subwatershed Identification
Butterfield Creek is composed of three primary sub-
watersheds; the mainstem, the east branch, and the
west branch. The two branches are parallel systems that
are tributary to the mainstem. Approximately 25 percent
of the watershed drains to the east branch, and approxi-
mately 36 percent of the watershed drains to the west
branch. The remaining 39 percent of the watershed
drains directly to the mainstem, which is entirely down-
stream of the two branches.
Land-Use Identification
Land use in the Butterfield Creek watershed was inter-
preted from 1990 aerial photographs (1 in. equals 400
ft). This information was then digitized and entered into
an ARC/INFO geographic information system. Sub-
watershed boundaries also were entered into the sys-
tem, and land-use totals were cumulated for both the
total watershed and the three distinct subwatersheds
(west branch, east branch, and mainstem). This infor-
mation is presented in Table 1.
About 55 percent of the watershed has been developed
into the following urban land-use categories: industrial,
commercial/institutional, residential, highway/arterial
roadway, railroad, and urban park and golf course. The
remainder, including woodland/wetland areas, agricultural
land, and vacant land, remains undeveloped. Most of
the undeveloped land lies in upstream parts of the wa-
tershed, particularly the west branch.
Stream Conditions
Stream conditions were assessed based on review of
existing aquatic life, water quality, and sediment quality
data as described in the preliminary nonpoint source
plan (2). Physical habitat data were collected during
development of the preliminary nonpoint source man-
agement plan.
Aquatic Life, Water Quality, and Sediment Quality
The existing data indicated degraded fish community
conditions throughout the watershed. As is typical with
many urban streams, species diversity and number are
quite low relative to less urbanized streams in Illinois.
Water quality conditions were also generally degraded,
particularly in the more urban reaches. Sediment quality
data paralleled the water quality data, with more ele-
vated levels recorded in urban reaches.
Physical Habitat
Physical habitat conditions in Butterfield Creek were
assessed during field visits to the creek. Data were
collected on stream condition reporting forms created
for the nonpoint source management planning effort.
Conditions such as degree of channelization, stream
and riparian vegetation, substrate material, erosion
and sedimentation, and observations of benthics and
macroinvertebrates and fish species were recorded.
The site visits indicated highly variable conditions. The
west and east branches tended to be highly chan-
nelized as a result of agricultural and urban drainage
activities. Mainstem reaches tended to be less altered
but appeared to suffer from the effects of flow desta-
bilization due to urban stormwater runoff. Channel
erosion and widening was prevalent in many down-
stream reaches.
Assessment of Nonpoint Source Impacts
Considering all available information from Butterfield
Creek and comparing its characteristics to other streams
in Illinois, the following conclusions were made regard-
ing nonpoint source impairment in Butterfield Creek.
Stream Uses
Many potential stream uses identified by the Illinois
Environmental Protection Agency (IEPA) are inherently
constrained by the size and flow of Butterfield Creek.
Uses that Butterfield Creek can be expected to support
and that were evaluated are fish and aquatic wildlife
(including warm water fishery), body contact recreation,
and noncontact recreation. IEPA assessments indicate
that present stream uses are moderately impaired.
120
-------
Table 1. Watershed Land Use (square miles)
Land Use Category
Industrial
Commercial/Institutional
Low-density residential
High-density residential
Vacant
Open land/urban park
Highway/arterial road
Agriculture
Woodland/wetland
Railroad
Watershed total
Watershed rank value
West Branch
0.037
0.196
1.342
0.230
0.980
0.171
0.541
3.954
1.828
0.019
9.30
3.63
Subwatershed
East Branch
0.079
1.027
1.369
0.188
1.236
0.152
0.265
1.816
0.274
0.082
6.49
2.53
Total Watershed
Mainstem
0.022
0.669
4.035
1.655
0.657
1.552
0.296
0.233
0.568
0.143
9.83
3.84
Square Miles
0.14
1.89
6.75
2.07
2.87
1.87
1.10
6.00
2.67
0.24
25.62
10.00
Percent of
Total
0.54
7.38
26.33
8.09
11.22
7.32
4.30
23.43
10.43
0.95
100.00
While Butterfield Creek is not presently used to a great
degree for water-based recreation, it is a potentially
valuable unit of the downstream Thorn Creek and Little
Calumet River systems. Also, Butterfield Creek is a
valuable indicator of the nonpoint source effects of ur-
banization on receiving stream quality in northeastern
Illinois. Improvement of uses in the larger streams will
require the successful restoration of streams such as
Butterfield Creek.
lization, channel erosion, bacterial contamination, nutri-
ent enrichment, and noxious aquatic plants/algae.
Other suspected causes of use impairment include
heavy metals, pesticides, oil and grease, unknown tox-
icity, organic enrichment, and suspended solids. Again,
relying on the existing database, determining the degree
to which these latter causes adversely affect stream use
attainment is difficult.
Stream Use Impacts
Based on existing data, the most readily identified im-
pacts to uses in Butterfield Creek are related to de-
graded physical conditions. These conditions include
degraded physical habitat, as evidenced by artificially
modified or eroded channels, and impaired aesthetics,
due in part to debris and trash. Low dissolved oxygen
also appears to be a limiting constraint to improved
aquatic life uses, particularly in the east branch and
several reaches of the mainstem.
Several other water quality factors, including toxicity to
aquatic life, turbidity, and siltation, were identified as
contributing constraints to improved stream uses. Based
on existing data from Butterfield Creek and other urban
streams, however, whether these water quality factors
by themselves limit the potential stream uses in much
of Butterfield Creek is unclear.
Causes of Stream Use Impacts
The primary causes of stream use impacts in Butterfield
Creek include physical habitat alterations, flow destabi-
Contributing Nonpoint Sources
The most prevalent nonpoint source responsible for use
impairment in Butterfield Creek is urban runoff, which
causes both physical and chemical degradation of the
creek. Other significant nonpoint sources include stream-
bank modifications, channelization, and removal of ri-
parian vegetation.
Several other sources have been identified as contrib-
uting to stream use impairment, although their relative
effects are much less certain. These include onsite waste-
watersystems, illicit sewer connections, golf course runoff,
draining/filling of wetlands, construction site runoff, debris
jams/beaver dams, carp/nuisance fish, and nonirrigated
crop production.
Finally, potential point-source-related impacts were noted
but could not be quantified. These included the treated
wet-weather discharge from the former Homewood waste-
water treatment plant, wastewater discharges from
Ely's Mobile Home Park and Idlewild Country Club, and
sanitary sewer overflows.
121
-------
Application of Urban Targeting
Methodology
Overview of Procedure
Objectives of Butterfield Creek Application
This section describes the application of the targeting
methodology to Butterfield Creek. The major purpose of
this effort is to assess the applicability of the methodol-
ogy for nonpoint source watershed planning in north-
eastern Illinois streams.
Comparison of Butterfield Creek Application to
Example Watershed
The assessment of nonpoint source impacts has led to
some very important conclusions that drive the applica-
tion of the targeting methodology for Butterfield Creek.
Perhaps unlike many other urban watersheds, the non-
point source assessment of Butterfield Creek did not
identify pollutants delivered by urban runoff (e.g., heavy
metals, toxic organics) as the primary cause of use
impairment. Instead, physical disturbances, including
stream channelization and flow destabilization, appear
to be among the most significant causes of impairment.
(Considering both physical and chemical effects, urban
runoff is the most important nonpoint source requiring
remediation in the mainstem of the creek.) This conclu-
sion causes the BMP selection procedure to emphasize
measures that control runoff rate as well as runoff qual-
ity. Because there is not a wide range of potential BMPs
addressing this problem, BMP selection becomes more
straightforward. As a result, this paper places more em-
phasis on the targeting aspect of the methodology.
Another difference between Butterfield Creek and the
example watershed presented in the methodology re-
port is that stream use attainment in Butterfield Creek
does not vary dramatically among subwatersheds. All
three subwatersheds of Butterfield Creek are signifi-
cantly impaired, although the causes of impairment vary
substantially among the subwatersheds.
Still another difference between Butterfield Creek and the
example watershed is the orientation of the subwater-
sheds. In the example, there were three parallel stream
segments. In Butterfield Creek, there are two parallel
stream segments that are tributary to the third. Therefore,
BMPs implemented in the two upstream watersheds
affect both the local watershed and the downstream
watershed. Similarly, adequately addressing problems
in the downstream subwatershed without applying some
BMPs in upstream areas may be impossible.
Further, the three watersheds differ significantly in the
levels of potential use attainability. Both the west and
east branches are headwater streams with low dry-
weather flows. Mainstem flows are more substantial,
and its larger channel dimensions allow greater potential
for full stream use.
Computation of Pollutant Loadings
The methodology report describes a procedure for esti-
mating pollutant loadings by land-use category. The
procedure involves the assignment of runoff coefficients
and pollutant concentrations to watershed land uses.
Runoff Coefficients
The first step is to assign a dimensionless runoff
coefficient to each land use. The runoff coefficient is a
measure of the watershed response to rainfall events
and is intended to be equivalent to the total storm runoff
divided by the total rainfall volume for runoff-producing
rain events. The runoff coefficient (Rv) is estimated from
the percent imperviousness of individual land uses by
the following equation (4):
Rv= 0.05 + (0.009 * percent impervious). (Eq.1)
While this methodology is quite simplistic with respect
to true watershed hydrologic response, it is an appropri-
ate way to represent the relative runoff responses of
different land uses to pollutant-generating rainfall/runoff
events. As such, it represents only the short-term sur-
face component of runoff and is not intended to repre-
sent the complete storm hydrograph.
Pollutant Concentrations
The methodology report also includes suggested pollut-
ant concentrations for different land uses. These con-
centrations can be used in conjunction with the runoff
coefficients to estimate differences in expected pollutant
loads for different land uses. The methodology report
makes it clear, however, that these concentrations are
not intended to be used in the estimation of actual
pollutant loads for the area. Also, the methodology re-
port provides concentrations for just six land-use types.
Four additional land uses were used to represent But-
terfield Creek, and pollutant concentrations for these
were derived from both local sources (5) and the meth-
odology report.
Table 2 summarizes the runoff coefficient and pollutant
concentration assumptions forthe Butterfield Creek land
uses. These estimates are used to reflect relative differ-
ences in runoff rates and pollutant loads and are not
intended to estimate actual loads.
Pollutant Loadings
Pollutant loads from runoff and concentration are com-
puted as follows:
122
-------
Table 2. Runoff Coefficients and Pollutant Concentrations by Land Use
Pollutant Concentrations (mg/L)
Land-Use Category
Industrial
Commercial/institutional
Low density residential
High density residential
Vacant
Open land/urban park
Highway/arterial road
Agriculture
Woodland/wetland
Railroad
Runoff
Coefficient
0.60
0.80
0.20
0.40
0.10
0.10
0.60
0.10
0.05
0.20
TSS
120
80
100
90
60
50
80
150
50
80
O&G
20
20
5
10
0
0
15
0
0
15
TP
0.20
0.20
0.60
0.40
0.20
0.60
0.20
0.80
0.20
0.20
Copper
0.05
0.05
0.03
0.04
0.01
0.01
0.05
0.01
0.01
0.05
Mass load (pounds) =
Rv * area (acres) * concentration (mg/L) * 0.227.
(Eq. 2)
This computation provides an estimate of the relative
pollutant load per inch of runoff-producing rain.
Runoff Rates
As previously indicated in the summary of nonpoint
source impacts to the watershed, pollutant loadings in
stormwater runoff do not appear to be the limiting cause
of stream use attainment. The quantity or rate of runoff
from urban land uses, however, does appear to be a
limiting constraint to improved stream uses, especially
for aquatic life. In particular, the expansion of impervious
surfaces increases the rate and volume of runoff for
storm events and reduces stream base flow. This altered
hydrology destabilizes the receiving stream channel and
adversely affects habitat. Another cause of physical
habitat impairment is channel modification (e.g., chan-
nelization, armoring).
Although runoff rate was not used as a targeting factor
during development of the methodology, it can be incor-
porated readily. The runoff coefficient provides a similar
indicator of runoff "load" as the product of runoff coeffi-
cient and concentration provides for pollutant load.
Comparison of Relative Loads: Targeting
Watershed Pollutant Loads
Using the methodology described in the previous sec-
tion, pollutant and runoff loads were estimated by land-
use category for each subwatershed and the overall
watershed. Tables 3 through 6 summarize pollutant
loadings for total suspended solids (TSS), oil and
grease (O&G), total phosphorus (TP), and copper, and
Table 7 summarizes storm runoff.
Total Suspended Solids. Evaluation of Table 3 indi-
cates that TSS loads vary by subwatershed, but not to
a great degree. There is, however, a great deal of vari-
ability in loadings between land-use categories. This
variability is based on differences in runoff coefficients
and pollutant concentrations (summarized in Table 2).
Figure 1 presents TSS loadings in a different fashion.
This map visually represents loading intensity. It sug-
gests, for example, that TSS loads could be reduced
significantly by targeting just those areas of the water-
shed that contribute at high rates (e.g., greater than
4,000 Ib/mi2). The nonpoint source assessment of But-
terfield Creek identified TSS as a contributing cause of
use impairment, particularly for aquatic life and recrea-
tional uses. While TSS does not appear to be as impor-
tant as some other identified causes of use impairment
(such as flow destabilization, physical habitat alteration,
and channel erosion), it still should be addressed in the
final watershed management plan. The targeting infor-
mation presented in this section will be useful in deter-
mining a comprehensive control strategy.
Oil and Grease. O&G loadings as presented in Table 4
vary dramatically by both subwatershed and land use.
The reason for this greater variability is the fact that oil
and grease is assumed to originate completely from
developed urban areas. Therefore, there is a relatively
small loading in the mostly nonurbanized west branch
subwatershed.
As with TSS, if O&G control was a high priority for
stream use remediation, it would be relatively easy to
identify areas for BMP targeting by using a map similar
to Figure 1 for O&G. As indicated in the nonpoint source
123
-------
Table 3. Total Suspended Solids Loading
Land-Use Category
Industrial
Commercial/institutional
Low-density residential
High-density residential
Vacant
Open land/urban park
Highway/arterial road
Agriculture
Woodland/wetland
Railroad
Watershed total
Watershed rank value
Table 4. Oil and Grease
Land-Use Category
Industrial
Commercial/institutional
Low-density residential
High-density residential
Vacant
Open land/urban park
Highway/arterial road
Agriculture
Woodland/wetland
Railroad
Watershed total
Watershed rank value
West Branch
389.0
1,817.5
3,892.9
1 ,200.3
852.5
124.1
3,765.5
8,601.9
663.0
44.1
2,1351
2.8
Loading (pounds
(pounds per inch of rain)
Subwatershed
East Branch
827.1
9,533.2
3,970.6
983.5
1 ,075.7
110.0
1,846.2
3,951.9
99.4
189.3
22,587
3.0
per inch of rain)
Subwatershed
West Branch East Branch
64.9
455.1
195.0
133.6
0.0
0.0
707.2
0.0
0.0
8.3
1,564
1.9
138.1
2,387.1
198.8
109.4
0.0
0.0
346.7
0.0
0.0
35.5
3,216
3.8
Mainstem
233.8
6,207.7
11,700.0
8,641.3
572.2
1,125.6
2,061.7
505.8
206.1
332.4
31 ,587
4.2
Mainstem
39.0
1 ,554.4
586.3
961.7
0.0
0.0
387.2
0.0
0.0
62.4
3,591
4.3
Total
Pounds
1,450
17,558
19,564
10,825
2,500
1,360
7,673
13,060
968
566
75,524
Total
Pounds
242
4,397
980
1,205
0
0
1,441
0
0
106
8,371
Watershed
Pounds/
Sq. Mile
10,357
9,290
2,898
5,229
871
727
6,976
2,177
363
2,357
2,948
10.0
Watershed
Pounds/
Sq. Mile
1,739
2,319
145
580
0
0
1,304
0
0
435
327
10.0
assessment, O&G is identified as a potential, but not
major, contributor to use impairment.
Total Phosphorus. Total phosphorus loadings as pre-
sented in Table 5 vary the least among the land-use
categories. This is explained by the fact that relatively
high concentrations are assumed for low-density resi-
dential and agricultural land uses, and these concentra-
tions counterbalance the relatively low runoff
coefficients for these uses.
Copper. The last pollutant to be presented is copper.
Copper loadings are presented in Table 6 and Figure 2.
Relative differences in copper loadings are similar to
those observed for O&G in that the heaviest loadings
come exclusively from intensely developed urban land
uses. Figure 2 makes clear that effective reduction of
total copper loadings could be achieved by targeting a
relatively small fraction of the total watershed for BMPs.
Available data, however, suggest that copper is not a
major cause of stream use impairment in Butterfield
Creek. While violations of the copper water quality
standard occur with some frequency, acute toxicity to
fish due to copper concentrations in stormwater does
not appear to be problematic. Nonetheless, copper may
be used as an effective surrogate for other urban runoff
toxicants, particularly other heavy metals, which are
believed to play a role in limiting aquatic life in the creek.
124
-------
West Branch
Mainstem
East Branch
TSS Loading
[^ <2,000 Ib/mi2
g] 2,001-4,000 Ib/mi2
H 4,001-6,000 Ib/mi2
>6,000 Ib/mi2
0_ 1/2 1 2
Miles
Figure 1. TSS loading per inch of rain, Butterfield Creek.
Table 5. Total Phosphorus Loading (pounds per inch of rain)
Subwatershed
Total Watershed
Land-Use Category
Industrial
Commercial/institutional
Low-density residential
High-density residential
Vacant
Open land/urban park
Highway/arterial road
Agriculture
Woodland/wetland
Railroad
Watershed total
Watershed rank value
West Branch
0.648
4.544
23.358
5.335
2.842
1.489
9.414
45.877
2.652
0.110
96.3
2.9
East Branch
1.379
23.833
23.824
4.371
3.586
1.320
4.615
21.077
0.397
0.473
84.9
2.6
Mainstem
0.390
15.519
70.238
38.406
1.907
13.507
5.154
2.698
0.824
0.831
149.5
4.5
Pounds
2.4
43.9
117.4
48.1
8.3
16.3
19.2
69.7
3.9
1.4
330.6
10.0
Pounds/
Sq. Mile
17.4
23.2
17.4
23.2
2.9
8.7
17.4
11.6
1.5
5.8
12.9
125
-------
West Branch
Mainstem
East Branch
Copper Loading
j 0-0.50 Ib/mi2
H 0.51-2.00 Ib/mi2
U 2.01-3.50 Ib/mi2
11 >3.5 Ib/mi2
Fjd
Figure 2. Copper loading per inch of rain, Butterfield Creek.
Table 6. Copper Loading (pounds per inch of rain)
Subwatershed
Total Watershed
Land-Use Category
Industrial
Commercial/institutional
Low-density residential
High-density residential
Vacant
Open land/urban park
Highway/arterial road
Agriculture
Woodland/wetland
Railroad
Watershed total
Watershed rank value
West Branch
0.16
1.14
1.17
0.53
0.14
0.03
2.36
0.57
0.13
0.03
6.3
2.1
East Branch
0.35
5.97
1.19
0.44
0.18
0.02
1.16
0.26
0.02
0.12
9.7
3.3
Mainstem
0.10
3.89
3.52
3.85
0.10
0.23
1.29
0.03
0.04
0.21
13.2
4.5
Pounds
0.6
11.0
5.9
4.8
0.4
0.3
4.8
0.9
0.2
0.4
29.2
Pounds/
Sq. Mile
4.3
5.8
0.9
2.3
0.1
0.1
4.3
0.1
0.1
1.4
1.1
10.0
126
-------
Storm Runoff. Although runoff is not a pollutant, it has
been shown to be nearly as important as pollutant load-
ing for causing degradation of stream uses. Storm runoff
"loadings" in units of acre-inch/inch of rain are presented
in Table 7 and Figure 3. Relative differences in storm
runoff loadings are similar to those observed for O&G
and copper, and high rates of runoff are from intensely
developed urban land uses. Table 7 suggests that, as
with the urban pollutants, targeting a relatively small
area could reduce the overall loading by a substantial
proportion. Figure 3 indicates that the same areas con-
tributing high copper loads are contributing high storm
runoff rates.
Evaluation of BMP Alternatives for Butterfield
Creek
The methodology report describes several BMP types,
including detention, retention, vegetative controls, and
source controls. Each of these were discussed briefly in
the Butterfield Creek targeting report (3), and that dis-
cussion will not be repeated here. The important conclu-
sions from that discussion follow.
The feasibility of implementing certain BMPs differs dra-
matically between remedial applications (i.e., existing
development) and preventative applications (i.e., new
development or redevelopment). Most of the municipali-
ties in the Butterfield Creek watershed have recently
adopted comprehensive stormwater management ordi-
nances that require implementation of effective deten-
tion designs for development activities and require
site-by-site evaluation of other BMPs, such as infiltration
trenches, filter strips, and vegetated buffers. The ordi-
nance discussed here was developed by the Butterfield
Creek Steering Committee.
The limiting cause of stream use impairment in Butter-
field Creek is hydrologic destabilization and streambank
modification/channelization. After addressing these
problems, however, full uses still may not be supported
without addressing contributing water quality factors.
Thus, BMPs for Butterfield Creek must control both
runoff rates or volumes and pollutant loadings.
Stormwater detention is a widely accepted practice in
the watershed, and recent experience indicates that the
stringent designs that accommodate pollutant removal
functions are implementable. The generally accepted
detention design for new development among water-
shed communities calls for limiting the runoff rate for the
2-year storm to 0.04 ft3/sec/acre. This should provide
effective pollutant removal as well as control of rates for
most storm events. Virtually the only other management
practice capable of controlling runoff volumes (and
rates) is infiltration (retention devices). This practice,
however, has not been widely applied in the watershed
or throughout the northeastern Illinois region. The pri-
mary constraint to using infiltration practices is the rela-
tively impervious soils of the region.
Most existing detention facilities in the watershed were
built without consideration of pollutant removal functions
or rate control of more routine events. Investigation of
typical facilities, however, suggests that most could be
readily retrofitted by installing new outlet controls and
performing minor regrading to achieve substantial water
quality and rate control benefits. Similarly, there are
open areas (e.g., school yards, parks, vacant land) in
the watershed where detention could be constructed
adjacent to existing uncontrolled developments.
Detention retrofitting has the benefit of controlling both
water quality and runoff rate to address stream use
impairments as well as flood control benefits, which are
often perceived as greater needs. Thus, detention
retrofitting has the greatest potential for reducing con-
straints to stream uses as well as the greatest imple-
mentability. Targeting of detention retrofitting is
discussed in the following section.
Reduction of Pollutant and Storm Runoff Loads
via Detention Retrofitting
To demonstrate how targeting of BMPs can remediate
high pollutant loadings in Butterfield Creek, it was as-
sumed that detention basin retrofitting would be applied
to land uses contributing high copper loads. These in-
cluded industrial, commercial/institutional, and high-
density residential uses, representing 16 percent of the
total watershed area. For purposes of this evaluation, it
is assumed that under existing conditions there is no
effective detention-based control of copper runoff from
these land uses. This is generally true in that much of
the historical development in the watershed occurred
without detention requirements. Further, most detention
facilities built subsequent to the promulgation of ordi-
nance requirements did not include pollutant removal
features. Another significant contributor of copper loads,
highways/arterial roads, was not considered for this
BMP because of the general unavailability of land within
right-of-ways to implement detention.
Targeting is also demonstrated for remediating high
storm-runoff rates. Because the same land uses that
contribute high copper loadings also contribute the high-
est runoff rates, the same 16 percent of the area will be
targeted for runoff rate control. As with copper, it is
assumed that under existing conditions there is no ef-
fective control of the 2-year and smaller storm events
most affected by urbanization.
Effective detention retrofitting designs, based on fully
detaining runoff from the 2-year storm (as now required
by most Butterfield Creek communities), was assumed
to remove 60 percent of the copper load. Table 8 and
Figure 4 show the effects of this action. By controlling
127
-------
West Branch
Mainstem
East Branch
Storm Runoff
rj~ 0-0.10 in.
[H 0.11-0.30 in.
H 0.31-0.50 in.
>0.50in.
3 1/2 1
!^S
Miles
Figure 3. Storm runoff per inch of rain, Butterfield Creek.
Table 7. Storm Runoff (inch-acres per inch of rain)
Land-Use Category
Industrial
Commercial/institutional
Low-density residential
High-density residential
Vacant
Open land/urban park
Highway/arterial road
Agriculture
Woodland/wetland
Railroad
Watershed total
Watershed rank value
West Branch
14.3
100.2
171.8
58.8
62.7
10.9
207.7
253.0
58.5
2.4
940
2.6
Subwatershed
East Branch
30.4
525.8
175.2
48.2
79.1
9.7
101.8
116.2
8.8
10.4
1,106
3.0
Total Watershed
Mainstem
8.6
342.4
516.5
423.7
42.1
99.3
113.7
14.9
18.2
18.3
1,598
4.4
Inch-Acres
53
968
863
531
184
120
423
384
85
31
3,644
10.0
Inches
0.60
0.80
0.20
0.40
0.10
0.10
0.60
0.10
0.05
0.20
0.22
128
-------
just 16 percent of the watershed via detention retrofit-
ting, the total watershed copper load is reduced from
29.2 Ib/in. of rain to 19.3 Ib/in. of rain, a 34-percent
reduction. This example demonstrates quite clearly the
value of being able to target BMPs within a watershed.
It is assumed that effective detention retrofitting, which
includes control of runoff from the 2-year storm to 0.04
ft3/sec/acre, can limit the storm runoff rates (not vol-
umes) for high-intensity land uses to the runoff rate from
nonurbanized land. Table 9 and Figure 5 illustrate the
effects of this control being applied to industrial, com-
mercial/institutional, and high-density residential land
uses. Comparing Table 7 to Table 9 indicates that the
short-term, storm runoff rate is reduced by 35 percent
for the entire watershed, from 0.22 in. per in. of rain to
0.14 in. per in. of rain. The reduction in storm runoff rate
is even more dramatic for the mainstem (39 percent). In
other words, if detention retrofitting can be implemented
for just 16 percent of the creek watershed, short-term
storm runoff can be reduced dramatically, thereby re-
ducing downstream bank erosion and habitat destabili-
zation effects. While detention retrofitting will have
relatively little effect on total runoff volumes, it will damp-
en stormwater runoff peaks substantially and also pro-
duce significant pollutant removal benefits.
Application of Watershed Prioritization
Analysis
The methodology report briefly describes a procedure
for prioritizing subwatersheds for BMP targeting. This
procedure relies on a number of factors (including water
body importance; type, status, and level of use; pollutant
loads, and implementability of controls) to rank sub-
watersheds. The relative importance of these factors is
indicated by assigning weights. As discussed previously,
the Butterfield Creek watershed orientation is different
from the example presented in the methodology report
and, as a result, may not be as appropriate for this type
of prioritization as the example. Nonetheless, the sug-
gested prioritization methodology is illustrated in the
following example.
Assignment of Prioritization Factors
The methodology report recommends the assignment of
factors based on relative rankings. For purposes of this
evaluation, the ranking scale ranges from 0 to 10.
Water Body Importance/Stream Size
Stream size factors are assigned in proportion to the
total drainage area providing flow to the stream. Subwa-
tershed drainage area rank values were previously com-
puted and are presented in Table 1.
Beneficial Use Type
Use-type ranks are based on the nature of potential use
of the stream reach. The mainstem is assigned a rela-
tively high rank because of the presence of riparian
public open space and because its size and physical
characteristics offer the most potential for aquatic life
and recreational uses. The west and east branches are
assigned relatively lower ranks because of their more
limited potential and because of the perception, particu-
larly for sections of the east branch, that the stream's
primary function is drainage.
Table 8. Copper Loading3 With Detention Basin Retrofitting for Industrial, Commercial/Institutional, and High-Density Residential
Areas (pounds per inch of rain)
Subwatershed
Total Watershed
Land-Use Category
Industrial
Commercial/institutional
Low-density residential
High-density residential
Vacant
Open land/urban park
Highway/arterial road
Agriculture
Woodland/wetland
Railroad
Watershed total
Watershed rank value
West Branch
0.07
0.45
1.17
0.21
0.14
0.03
2.35
0.57
0.13
0.03
5.2
2.7
East Branch
0.14
2.38
1.19
0.18
0.18
0.02
1.15
0.26
0.02
0.12
5.6
2.9
Mainstem
0.04
1.55
3.51
1.54
0.10
0.23
1.29
0.03
0.04
0.21
8.5
4.4
Pounds
0.2
4.4
5.9
1.9
0.4
0.3
4.8
0.9
0.2
0.4
19.3
Pounds/Sq.
Mile
1.7
2.3
0.9
0.9
0.1
0.1
4.3
0.1
0.1
1.4
0.8
10.0
a60 percent loads reduction assumed for targeted areas
129
-------
West Branch
Mainstem
East Branch
Copper Loading
Q| 0-0.50 Ib/mi2
Ql 0.51-2.00 Ib/mi2
g 2.01-3.50 Ib/mi2
>3.5 Ib/mi2
Figure 4. Copper loading per inch of rain, Butterfield Creek (with detention basin retrofitting for industrial, commercial/institutional,
and high-density residential areas).
Table 9. Storm Runoff3 With Detention Basin Retrofitting for Industrial, Commercial/Institutional, and High-Density Residential
Areas (inch-acres per inch of rain)
Land-Use Category
Industrial
Commercial/institutional
Low-density residential
High-density residential
Vacant
Open land/urban park
Highway/arterial road
Agriculture
Woodland/wetland
Railroad
Watershed total
Watershed rank value
West Branch
2.4
12.5
171.8
14.7
62.7
10.9
207.7
253.0
58.5
2.4
797
3.4
Subwatershed
East Branch
5.1
65.7
175.2
12.1
79.1
9.7
101.8
116.2
8.8
10.4
584
2.5
Mainstem
1.4
42.8
516.5
105.9
42.1
99.3
113.7
14.9
18.2
18.3
973
4.1
9
121
863
133
184
120
423
384
85
31
2,354
0.10
0.10
0.20
0.10
0.10
0.10
0.60
0.10
0.05
0.20
0.14
10.0
Deduction of runoff coefficient to 0.1 for targeted areas
130
-------
West Branch
Mainstem
East Branch
Storm Runoff
Q 0-0.10 in.
US 0.11-0.30 in.
M 0.31-0.50 in.
>0.50 in.
D 1/2 1
~^^
Miles
Figure 5. Storm runoff per inch of rain, Butterfield Creek (with detention basin retrofitting for industrial, commercial/institutional,
and high-density residential areas).
Beneficial Use Status
The methodology report is somewhat unclear regarding
the determination of this factor. It is assumed in this
example that use status reflects the degree of restora-
tion and protection needed to achieve desired beneficial
uses. Because each of the branches is similar in its
relative degree of aquatic life use impairment, similar
factors are assigned. The mainstem's ranking is slightly
lower, however, because of the greater level of stream-
side activities presently supported.
Beneficial Use Level
This factor reflects the level of stream use relative to
other water bodies in the target watershed. For Butter-
field Creek subwatersheds, use level considers acces-
sible riparian and accessible open space (e.g., parks
and golf courses) and the presence of residential land
use adjacent to the stream corridor. With these factors
considered, the mainstem is assigned the highest rank-
ing, followed by the east branch and the west branch.
Pollutant Loads
This factor represents the degree of pollutant loading or
some other cause that is impairing water body use. In
this example, runoff rate (rather than quality) is used to
reflect this factor. Storm runoff rate factors are derived
from Table 7.
Implementability of Controls
This factor is assumed to represent the relative degree
of implementability of control measures. In this example,
the recommended control measure to reduce storm run-
off rates is detention basin retrofitting. As was discussed
previously, retrofitting of existing highway/arterial roads
probably will not be feasible in most areas. Beyond that,
distinguishing the relative implementability of retrofitting
based on institutional or technical factors is not easy. For
this reason, ranks are assigned on the basis of water-
shed size and the relative degree of high-density urban
development. Another factor that could have been con-
sidered is the relative proximity of targeted land uses.
Large concentrations of targeted land uses could more
readily be addressed through more cost-effective re-
gional controls.
Table 10 presents ranks for each of these factors by
subwatershed. It includes an assignment of factors for
the total watershed as well. The recommended basis for
assignment of total watershed factors is not described
in the methodology report. In the Butterfield Creek ex-
ample, totals of the subwatershed ranks are used for
both stream size and stormwater rate. For the remaining
131
-------
Table 10. Butterfield Creek Prioritization Analysis
Beneficial Use
Watershed
Weights
West branch
East branch
Mainstem
Total watershed
Stream
Size
25
3.63
2.53
3.84
10.00
Type
10
4
3
6
5
Status
10
7
8
6
7
Level
5
3
4
7
5
Stormwater
Rate
25
2.6
3.0
4.4
10.0
Ability To
Implement
25
8
4
4
5
Target
Score
100
4.81
3.68
4.61
7.70
Target score = weighted average of rank points = sum (rank score * weight) / sum (weights)
factors, approximate averages of the subwatershed
ranks are used.
Assignment of Relative Weights
The methodology report recognizes that some factors
may be more important than others and suggests that
these differences be accounted for by assigning differ-
ent weights to each factor. The report also recognizes
that considerable subjectivity is involved in the selection
of factors and the assignment of ranks and relative
weights.
Discussions with representatives from the watershed,
primarily the Butterfield Creek Steering Committee, were
considered in assigning relative weights for Butterfield
Creek. The actual assignment, however, becomes some-
what challenging for several reasons. First, as indicated,
evaluation of the different factors is quite subjective, and
quantification, even in relative terms, is difficult. Second,
while the listed evaluation factors are clearly important
to the efficient remediation of use constraints in Butterfield
Creek, they are difficult to compare and weight relative
to each other. Third, as discussed previously, because
two of the stream branches flow into the third, the reme-
diation of problems in the third branch (the mainstem) is
clearly not independent of remedial activities in the other
branches. The example from the methodology report
does not directly reflect this interdependence.
Bearing in mind these qualifications, weights were
assigned to the identified factors by following the proce-
dure described in the methodology report. As seen in
Table 10, equal weights of 25 are assigned to the four
factors. For the beneficial-use category, weights are as-
signed to the three subcategories so that they total 25.
Results of Watershed Prioritization
On the basis of the assignment of weights and factors
as described above, stormwater rate controls should
be applied first to the west branch, followed closely by
the mainstem, and then the east branch. Just as in the
example in the methodology report, however, the "total
watershed" receives the highest target score, implying
priority control of the entire watershed.
In evaluating the results of this prioritization to Butter-
field Creek, the west branch apparently receives the
highest subwatershed priority primarily because it
scores quite well in the ability-to-implement category. In
reality, its high score in this category is due to the
relatively little high-density urbanization within its water-
shed and, therefore, its relative ease of control. The east
branch receives the lowest targeting score because it is
smallest in watershed size and because it scores poorly
relative to potential beneficial uses.
The interpretation of the total watershed score of 7.7,
higher than each of the subwatershed scores, is some-
what perplexing. The procedure applied to Butterfield
Creek, which establishes total watershed ranks as av-
erages or sums of the subwatershed ranks, always
results in the total watershed receiving the highest
score. This implies that problem remediation (or preven-
tion) always should be addressed watershedwide, de-
spite the results of subwatershed prioritization. It also
may suggest that the assumptions used in arriving at
total watershed ranks are not appropriate and, there-
fore, the total watershed score should not be compared
with the subwatershed scores.
Overall, the results of this simple analysis are quite
interesting. Intuitively, if limited funds are available for
remedial measures, it makes sense to spend them in
subwatersheds in which stream use has the most po-
tential for improvement and in which remedial activities
are most implementable. The results for Butterfield
Creek, in which the mainstem and west branch receive
similarly high targeting scores, are generally consistent
with this logic. Because conditions in the mainstem also
are dependent on nonpoint contributions from the east
branch, however, it may not be possible to eliminate
critical use constraints and to fully restore mainstem
stream uses without applying effective BMPs water-
shedwide.
132
-------
Other Prioritization Applications
Application of the prioritization in this watershed was not
straightforward due to the configuration of the water-
shed. Based on the experience gained in this applica-
tion, however, it is apparent that there are two cases in
which the prioritization methodology would be more use-
ful and straightforward. The first case would be in priori-
tizing restoration efforts between separate watersheds
under a single management agency or funding source.
The second case would be in prioritizing efforts within a
single watershed tributary to a critical resource (e.g.,
recreational lake, high-quality stream segment, water
supply reservoir).
Prioritizing Between Distinct Watersheds
During development of a statewide or regionwide non-
point source control program, limited funds often must
be prioritized between distinct watersheds within the
region. This methodology provides a relatively objective
method for assigning priorities to watersheds competing
for funds. To ensure acceptance of the results of the
prioritization and to avoid conflicts between competing
watershed officials, involving the officials and interested
parties from all of the watersheds in the assignment of
ranking and weighting factors is very important. Be-
cause they all have participated in that process and
agreed on the ranks and weights, it will be difficult for
them to dispute the outcome of the prioritization results.
Therefore, a rational schedule can be developed for
expenditures and efforts in the various watersheds.
Prioritizing Within a Watershed
During development of a watershed nonpoint source
management plan, a particular resource within the wa-
tershed often motivates development of the plan. The
methodology could be used readily to prioritize targeted
land uses within that watershed. In this case, however,
the beneficial use and probably even the stream size
factors would be meaningless because all subwater-
sheds would be tributary to the same resource whose
uses are being protected. The only two factors that
would be used would be the pollutant load (or stormwa-
ter rate) and the ability to implement.
Summary and Conclusions
This report has discussed some of the strengths and
weaknesses of the urban targeting and prioritization meth-
odology as applied to Butterfield Creek in northeastern
Illinois. Highlights of this evaluation are discussed below.
Technical Representation
The methodology recommends a relatively simple meth-
odology for generating pollutant loads and assessing
BMP effectiveness. For purposes of this type of applica-
tion, which emphasizes relative loadings among land-
use types and subwatersheds, this simplicity is appro-
priate and appears to produce reasonable results for
Butterfield Creek. One shortcoming is that the technical
procedure is limited to pollutant loads. Inclusion of runoff
rates was readily incorporated into the methodology,
however, making it more useful for urban streams such
as Butterfield Creek.
Urban Targeting
The urban targeting component of the methodology
worked quite well, especially when combined with map-
ping, which highlighted relative pollutant contributions
by land use. Targeting also provided a fairly clear indi-
cation of the relative pollutant (and high runoff rate)
contributions by subwatershed.
BMP Selection
Effective BMP selection must take into account the
causes of stream use impairment as well as the physical
characteristics of the watershed and the drainage sys-
tem. In the application of the recommended BMP selec-
tion methodology to Butterfield Creek, it was clear that
BMPs that control both pollutant loads and runoff rates
would be required. As a result, detention facility retrofit-
ting became, somewhat by default, the selected BMP
for evaluation. The quantification procedure recom-
mended in the methodology report worked quite well
and was enhanced by the mapping of pollutant loadings.
Watershed Prioritization
The application of watershed prioritization to Butterfield
Creek, based on assigning ranks and weights to priori-
tization factors among subwatersheds, was accom-
plished with some difficulty. Part of this difficulty was
related to the subwatershed orientation in Butterfield
Creek, in which two stream segments were tributary to
a third. The existing methodology is not structured to
address this situation. A related difficulty was the sub-
jectivity involved in assigning relative ranks and weights
to unrelated prioritization factors. The methodology
would be more useful for prioritizing between distinct
watersheds or prioritizing within a watershed all tributary
to a single critical resource.
Remedial Versus Preventative Applications
The Butterfield Creek application of the targeting and BMP
selection methodology focused on BMPs to remediate
existing stream use impairments. This methodology could
potentially be applied to assess preventative BMPs as
well. In this context, pollutant loads could be assessed
for a nonurbanized watershed, for a fully urbanized water-
shed without BMPs, and for a fully urbanized watershed
with BMPs. For a nonurbanized watershed, however,
133
-------
some of the stream-use prioritization factors become
irrelevant, assuming that stream use is relatively unim-
paired before urbanization. In the Butterfield Creek wa-
tershed, several preventative BMPs have already been
chosen for newly urbanizing areas. These include soil
erosion and sediment control measures, effective storm-
water drainage and detention controls, and stream and
wetland protection requirements. These preventative
BMPs have been endorsed by most watershed commu-
nities because of their multipurpose benefits (i.e., non-
point control, flood prevention, channel erosion control,
and aesthetic enhancement) and implementability.
Partly for reasons of equity, local officials have no strong
desire to target or prioritize these BMPs to particular
land uses or subwatersheds.
Conclusions
One of the major benefits of this approach is that the
user can document the decision-making process in a
systemized fashion. The methodology also forces con-
sideration of the interdependence of various technical
and institutional factors in the decision-making process.
In addition, the methodology enables the presentation
of complex decision-making factors in a visual format.
As a result, this methodology could be very useful in
targeting BMPs in stream watersheds throughout north-
eastern Illinois. For successful application of the meth-
odology, however, existing stream use impairments,
causes, and nonpoint sources must be clearly under-
stood. In most watersheds, this will require the collection
and assessment of additional stream use and water
quality data.
The primary limitations of the methodology may be its
subjectivity and the fact that it attempts to represent
complex watershed interrelationships in a relatively sim-
ple fashion. These shortcomings can be addressed by
properly qualifying assumptions and providing thorough
documentation of results, as well as by involving all of
the interested parties in the ranking and weighting proc-
ess. Without the proper awareness of critical assump-
tions, however, the methodology is capable of producing
misleading or counterintuitive results. Another potential
shortcoming of the methodology, revealed in its applica-
tion to Butterfield Creek, is the difficulty in representing
interdependent (i.e., upstream-downstream) subwater-
sheds and stream reaches.
References
1. Woodward-Clyde Consultants. 1989. Urban targeting and BMP
selection: An information and guidance manual for state NPS
program staff engineers and managers. Oakland, CA.
2. Dreher, D., T. Gray, and H. Hudson. 1992. Demonstration of an
urban nonpoint source planning methodology for Butterfield Creek.
Northeastern Illinois Planning Commission, Chicago, IL.
3. Dreher, D., and J. Clark. 1992. Application of urban targeting and
BMP selection methodology to Butterfield Creek, Cook and Will
Counties, Illinois. Northeastern Illinois Planning Commission, Chi-
cago, IL.
4. Schueler, T.R. 1987. Controlling urban runoff: A practical manual
for planning and designing urban BMPs. Washington, DC: Metro-
politan Washington Council of Governments.
5. Northeastern Illinois Planning Commission. 1979. Areawide water
quality management plan. Chicago, IL.
134
-------
Development of a Comprehensive Urban Nonpoint Pollution Control Program
Jennifer M. Smith and Larry S. Coffman
Prince George's County Government, Department of Environmental Resources,
Landover, Maryland
Abstract
Comprehensive urban nonpoint pollution control is a new,
rapidly developing multidisciplinary field. Significant water
quality improvements will be achieved when all state and
local governments have the necessary resources, knowl-
edge, skills, and vision to implement effective programs.
Urban nonpoint pollution has traditionally been addressed
by relying heavily on structural stormwater control devices
to treat contaminated runoff. Yet, this "band-aid" approach
has proven relatively ineffective for controlling such a ubiq-
uitous and poorly defined problem.
The objective of this paper is to illustrate some of the
many problems, issues, and obstacles that federal,
state, and local government agencies must address to
facilitate further advancements in urban water quality
control. A more comprehensive, watershed approach
must be developed, specifically focusing on source pre-
vention programs, improved technology, and intra-
agency coordination. Measuring the effectiveness of
innovative source control programs, such as public edu-
cation, will become essential for targeting problems,
focusing goals, and allocating resources to areas need-
ing improvement.
Guidance for implementing these nonpoint pollution con-
trol strategies is needed to assist state and local govern-
ments. The nature, magnitude, and scope of urban
nonpoint source pollution, one of the most fundamental
and universal problems facing local governments, are is-
sues that have yet to be adequately resolved. Without
program guidance and leadership, the urban nonpoint
pollution problem will persist and the quality of our nation's
waters will further deteriorate.
Introduction
To address the complex nature of the national water
pollution problem and the comprehensiveness of non-
point pollution control, all states and municipalities must
have access to the understanding, expertise, knowledge,
resources, and insight needed to respond to difficult
challenges and provide the most appropriate services
and solutions. Effective water quality improvement will
depend on the ability of municipalities to appropriately
implement an array of preventative measures, manage-
ment strategies, and treatment technologies for dealing
with all aspects of water pollution.
Traditional offsite structural treatment is only one of the
tools available for addressing this national problem. At the
local level a variety of other innovative tools must be
tailored to the unique problems and characteristics of a
particular site, land use, community, or watershed. Non-
point source pollution will be fully and effectively controlled
only when municipalities understand how to identify prob-
lems, evaluate alternatives, and implement solutions.
Discussion
The magnitude and scope of critical issues associated
with current urban nonpoint source control programs, such
as the National Pollutant Discharge Elimination System
(NPDES) program, must be appreciated to ensure suc-
cess. To effectively implement the NPDES regulations,
municipalities must address the following questions:
How Will the NPDES Goals Be Met?
The success of the municipal NPDES program in achiev-
ing the water improvement objectives of the Clean Water
Act will depend heavily on the ability and commitment
of each municipality to develop focused and effective
comprehensive pollution control programs. To reduce
nonpoint pollution to the maximum extent possible,
local governments must be prepared to support and
effectively implement the full range of necessary pro-
gram components and to shift their programs to a more
balanced approach between prevention and treatment.
Municipal governments need active leadership that em-
powers each jurisdiction with the necessary knowledge,
135
-------
tools, skills, and resources to implement effective pro-
grams. Ultimately, each municipality's success will be
judged based on the ability to effectively implement
program constituents related to planning, coordination,
integration, education, prevention, management, mainte-
nance, inspection, enforcement, funding, and appropriate
use of technology. Many roadblocks, however, will in-
hibit the ability to accomplish these objectives. Funding
and competition with other local programs are obvious
barriers, while misunderstanding the nature of the prob-
lem, setting incorrect priorities, and focusing programs on
nontraditional prevention strategies are less obvious pit-
falls.
What Does Each Jurisdiction Need?
The successful integration of effective nonpoint source
pollution reduction programs into traditional local storm-
water programs is more easily accomplished if imple-
mentation problems are identified and thoroughly
addressed. These problems can concern:
Legal, financial, and political liabilities and issues.
Public awareness, acceptance, and education.
Development and implementation of adequate in-
spection programs for construction and maintenance.
Development and implementation of effective en-
forcement programs.
Funding options for various programs.
Integration, coordination, and enhancement of exist-
ing programs.
Allocation and sharing of private, public, and corpo-
rate resources.
Understanding the techniques, approaches, strate-
gies, and philosophies of comprehensive water qual-
ity planning.
Development of mechanisms for technology transfer
and implementation of innovative practices.
The need for practical guidance on program devel-
opment.
Local governments will be looking for guidance on how
to overcome these obstacles. Thus guidance on effec-
tive model programs must take into account the effect
policy decisions have at the local level.
Can We Depend on Treatment Technology?
Historically, stormwater programs have addressed water
pollution from a treatment standpoint, making them rather
symptomatic and ineffective. Typical programs rely heavily
on structural treatment devices to control contaminated
runoff from new development. As a result, current water
pollution control programs address problems through a
"band-aid" approach instead of a more comprehensive
approach in which both preventative and treatment
measures are employed within a watershed.
With the many years of experience that some munici-
palities now have using treatment devices, it is becom-
ing clear that many current treatment practices are
riddled with inherent problems that may be difficult, if not
impossible to overcome. Problems such as burdensome
maintenance, improper construction, inadequate design,
ineffective site management, and the latest obstacles
posed by federal and state wetland permitting require-
ments have left many local governments frustrated. Thus
the proper role, long-term impacts, and effectiveness of
current treatment practices in urban nonpoint source pol-
lution control need to be carefully evaluated.
Reliance on treatment technology as the primary ap-
proach to pollution control can result in failure of a
program. Many current treatment practices cause prob-
lems that limit, restrict, or prohibit their use. Thus, in a
more recent study, Prince George's County, Maryland,
found that of 151 urban nonpoint source treatment de-
vices constructed or put into operation within the past 5
years, only 60 percent were functioning as designed.
Given such limitations, it would be inappropriate to guide
other local jurisdictions to heavily rely on treatment tech-
nology in the hope of greatly improving water quality.
Do We Effectively Control New Development?
One problem that has yet to be adequately addressed
is an effective and comprehensive approach to environ-
mentally safe development. Current programs primarily
focus on treatment controls for new development and
generally do not consider or incorporate other important
pollution reduction and prevention strategies.
New development must be designed in such a manner
that onsite treatment of stormwater runoff can be effec-
tive. In addition, prevention must become an integrated
part of site development through public education, im-
plementation of site maintenance and management
plans, and industrial process changes.
The goal of an effective stormwater management site
plan should be the integration of preventive, manage-
ment, and treatment devices that can effectively mitigate
all adverse water quality impacts associated with the
development. New development can be easily regulated
and pollution abatement requirements selected from a
broad range of options can be imposed, including:
Greater use of open and surface drainage systems.
Limited and creative grading to encourage onsite re-
tention and to enhance ground-water recharge.
Treatment of surface water by maximizing biological,
chemical, and physical treatment devices.
136
-------
Requiring grounds maintenance plans.
Education programs for developers and the public.
Use of effective construction and maintenance in-
spection and enforcement programs.
Greater preservation of existing natural water quality
and habitat features.
What Do We Do About Existing Development?
Controlling nonpoint source pollution from existing devel-
opment represents the greatest challenge but offers the
most potential for attainment of overall pollution reduc-
tion goals. Water pollution problems associated with
existing development are the most difficult to control and
require the most complicated mix of approaches. Typical
issues include a lack of regulations requiring retrofitting
of facilities, a lack of available space to construct onsite
controls, limited incentives, difficulty in identifying prob-
lems and solutions, a lack of public awareness, a lack
of funding, and limited experience with source control
and prevention programs. To address these issues, mu-
nicipalities should consider the following:
A community and/or watershed-based approach.
Baseline data collection needs.
A comprehensive nonpoint source reconnaissance study.
Investigative approaches and tools.
Water quality data collection and use.
Public outreach programs.
Regulatory actions.
Inspection.
Enforcement.
Comprehensive maintenance and management plans.
Retrofit opportunities.
Innovative control technology.
Lake, stream, and wetland restoration and enhance-
ment.
How Comprehensive Is Comprehensive?
A comprehensive program not only uses dedicated local
government personnel, but also integrates existing pro-
grams and personnel at the state and federal level.
Coordination, cooperation, communication, and partici-
pation among all agencies involved with programs re-
lated to water quality improvements are essential for
efficient use of available resources.
Many important water quality-related programs have
been independently developed overtime that achieve a
variety of environmental objectives. Identifying all such
programs and directing and focusing them on a common
goal would be extremely valuable and useful. Although
many water quality-related pollution control programs
exist, few coordinate oversight in order to pool re-
sources and combine efforts.
Existing water quality protection and community out-
reach programs can be easily enhanced or expanded to
incorporate additional water quality education and en-
forcement programs. For example, in Prince George's
County, the police community relations program is work-
ing with the state's Department of Environmental Re-
sources, the U.S. Attorney's Office, and local citizens
groups to incorporate water pollution control educational
information into the program. In conjunction with a state,
federal, and local enforcement training program, this effort
focuses on the enforcement of water quality regulations.
The final aspect of a comprehensive program is to con-
sider all possible sources of water pollution, point and
nonpoint source alike. Combining the investigation and
enforcement efforts of both programs could help elimi-
nate loopholes in the system and facilitate effective use
of existing resources. Investigators and enforcement
agents at all levels of government must pool their re-
sources and continuously exchange information regarding
known sources of water pollution. Leadership will be criti-
cal for facilitating such communication and coordination.
How Will We Measure the Effectiveness of
NPDES Programs?
Municipal governments, scientists, environmentalists,
and the public will continue to ask, How effective are
source controls? Various plans have been discussed as
a result of the NPDES stormwater permit application
requirements to quantify the effectiveness of municipal
programs. Among these is the water quality standards
approach that is currently used in the NPDES industrial
point source discharge program.
The water quality standards approach to measuring the
effectiveness of urban nonpoint source control/prevention
programs will require extensive water quality base-flow
and storm-event monitoring. In the past, however, water
quality monitoring programs, either with automated equip-
ment or manual sampling, have proven to be difficult and
costly to implement. Problems with drought conditions,
weather predictions, equipment errors, and the physical
constraints associated with manual sampling present par-
ticular challenges. Ultimately, municipalities, which will be
responsible for implementing source control programs and
measuring their effectiveness, will need to rely on the
availability of low-cost, flexible alternatives.
The success of source control programs will rest on
the ability of small and medium-size municipalities to
implement comprehensive and effective water quality
control programs. How these programs are structured
137
-------
and the number of programs implemented will ultimately
determine the effectiveness of urban nonpoint source
pollution control efforts. The focus of efforts should not
be on the development of water quality standards but
on the development and implementation of a wide range
of prevention, management, and treatment programs.
Summary
Significant reductions in urban nonpoint pollution will
be achieved only when effective treatment, prevention,
management tools, strategies, and programs have
been fully developed and implemented. Given the
clearer picture of the nature and scope of the problem,
how the pieces will fit together is better understood. None-
theless, effective efforts will require time, patience, and
cooperation. All governments, agencies, and organizations
dealing with these issues must work together to develop
the technology necessary for a nationally comprehen-
sive urban nonpoint source control program. Momentum
for change must be sustained by continued strong lead-
ership, and expertise in this ever-growing and compli-
cated field must be appropriately channeled to develop
state-of-the-art technology, and not just to restate it.
138
-------
Site Planning From a Watershed Perspective
Nancy J. Phillips
U.S. Environmental Protection Agency, Region 5, Chicago, Illinois
Elizabeth T. Lewis
U.S. Department of Agriculture, Soil Conservation Service, Grayslake, Illinois
Abstract
The site planning review process involves consideration
of the impacts on water resources that can result from
the proposed activity, including changes in water quality
and quantity. These changes can affect areas immedi-
ately adjacent to the site, as well as distant areas of the
watershed. Therefore, site-specific and watershed is-
sues must be considered when developing solutions for
proper management.
An important first step in the process involves locating
the project site within the watershed and becoming fa-
miliar with the watershed characteristics. Secondly,
analysis of the impact of site development on the re-
source areas within the watershed should be conducted
so that management objectives can be identified. This
aids in the identification of best management practices
that can meet management objectives for the site and
the watershed.
Introduction
Site planning tends to occur on a limited scale, usually
when developing individual sites, such as subdivisions,
commercial developments, industrial parks, residential
areas, and schools, as well as infrastructure such as
roadways and bridges. Together, these sites compose
an urban area.
As sites within the urbanizing area develop, water re-
sources such as streams, lakes, wetlands, and ground
water degrade. Because of the incremental nature of
development and the cumulative effect that develop-
ment can have on resources, the site planning process
must involve consideration of the watershed within
which the development is occurring. The watershed
approach, which allows for a comprehensive evaluation
of the development process, contains several elements
that together form a review process: 1) delineation of
the watershed and subbasins, 2) inventory of soils,
3) inventory of natural systems, 4) identification of im-
pacts from development, 5) development of manage-
ment goals and objectives, and 6) development of
recommendations for mitigation.
Delineation of the Watershed
A watershed is an area of land that drains to a water
resource such as a wetland, river, or lake. Depending
on the size and topography, watersheds can contain
numerous tributaries, such as streams and ditches, and
ponding areas such as detention structures, natural
ponds, and wetlands.
Rainwater and snowmelt that do not evaporate or infil-
trate into the soil run off into a nearby tributary or
ponding area, then flow to the main wetland, river, or
lake within that watershed. Through this linkage, the
upper portions of a watershed can affect downstream
areas. Thus, the quality of a wetland, stream, or lake
often reflects the land use and other activities being
conducted in upstream areas. Because the relationship
of cause and effect can extend for large distances
throughout the entire watershed, it is important to ad-
dress environmental management issues from a water-
shed perspective.
Use of topographic maps is a common method of locat-
ing and delineating the boundaries of watersheds. To
locate a site on a topographic map, the site plan should
be closely examined. A topographic map represents the
physical features of the land such as hills, valleys, ba-
sins, ridges, and channels. The mapping technique
used is based on elevation data (usually mean sea
level) and contour intervals (commonly of 10 ft). Distinc-
tive features such as road intersections and curves,
towns, agricultural field boundaries, streams, and lakes
make acceptable landmarks. These landmarks can be
used to locate the approximate site on a topographic
map. The next step is to delineate the watershed that
139
-------
contains the site. Below is an outline of steps necessary
to delineate a watershed:
1. Use a topographic map(s) to locate the river, lake,
stream, wetland, or other water bodies of interest
(see Figure 1).
2. Trace the watercourse from its source to its mouth,
including the tributaries. This step determines the
general beginning and ending boundaries (see
Figure 2).
3. Examine the lines on the topographic map that are
near the watercourse; these are referred to as con-
tour lines (see Figure 3). Contour lines connect all
points of equal elevation above or below a known
reference elevation. The thick contour lines have a
number associated with them, indicating the eleva-
tion. The thin contour lines are usually mapped at
10-ft intervals, and the thick lines are usually
4.
mapped at 50-ft intervals. Contour lines spaced far
apart indicate that the landscape is more level and
gently sloping. Contour lines spaced very close to-
gether indicate dramatic changes (rise or fall) in
elevation over a short distance (see Figure 4). To
determine the final elevation of a location, simply
add or subtract the appropriate contour interval for
every thin line or the appropriate interval for every
thick line.
Check the slope of the landscape by locating two
adjacent contour lines and determine their respective
elevations. The slope is calculated as the change in
elevation divided by the distance. A depressed area
(valley, ravine, swale) is represented by a series of
contour lines "pointing" towards the highest eleva-
tion (see Figure 5). A higher area (ridge, hill) is repre-
sented by a series of contour lines "pointing" towards
the lowest elevation (see Figure 6).
40
Figure 3. Contour lines.
Figure 1. Big River watershed.
Mouth
Sources
(Wetlands, Ponds,
Lakes, Depressions)
Tributaries
(Streams, Drains,
Swales, Channels)
Floodplain
Figure 2. West Branch subwatershed.
Figure 4. Floodplains and ridges.
140
-------
5. Determine the direction of drainage in the area of
the water body by drawing arrows perpendicular to
a series of contour lines that decreases in elevation.
Stormwater runoff seeks the path of least resistance
as it travels downslope. The "path" is the shortest
distance between contours, hence a perpendicular
route (see Figure 7).
6. Mark the break points surrounding the water body.
The "break points" are the highest elevations where
half of the runoff would drain towards one body of
water and the other half would drain towards an-
other body of water (see Figure 8).
100
Contour Lines
Figure 5. Valley.
Contour Lines
Figure 6. Ridge.
150
Figure 7. Direction of drainage.
7. Connect the break points with a line following the high-
est elevations in the area. The completed line rep-
resents the boundary of the watershed (see Figure 9).
Inventory of Soils
Locating the site on the soils map requires a U.S.
Department of Agriculture (USDA) Soil Conservation
Service (SCS) soil survey of the county. Select the
appropriate soil sheet for the site by examining the Index
to Map Sheets. Each numbered section corresponds to
a soil sheet. After obtaining the necessary soil sheet,
locate the site by using distinguishing landmarks, such
as road intersections, field outlines, creeks, and rivers.
Note the map unit symbols that are in that area. Map
unit symbols in a soil survey may consist of numbers or
letters, or a combination of numbers and letters. Soil
surveys differ from state to state and county to county.
Some soils are symbolized by letters and others by
numbers. Figure 10 depicts a typical soils map found in
an SCS soil survey.
A variety of information that can be used to evaluate
sites is contained within the soil survey and maps. The
different types of information contained in the soil survey
include land capability classification, suitability tables,
slopes, erosiveness, wetness, permeability, and drain-
age patterns.
Land Capability Classification
The land capability classification shows the suitability of
the soils for various types of activities, from farming to
engineering. The capability classification, denoted by
roman numerals, suggests ways to manage and use the
soils and highlights any potential hazards. Included in
the capability classification are subclasses of erosion,
wetness, shallowness, and climate limitations, indicated
by small letters after the roman numerals. These sub-
classes signal a soil's tendency, for example, towards
erosiveness or wetness.
Suitability Tables
Suitability tables are found in the section located after
the soil descriptions and management capability group-
ings. They designate the soil's suitability for various cate-
gories of uses, including wildlife plantings, septic fields,
building foundations, and road subgrades. This table
can highlight some potential hazards for sites planned
on questionable soils. For example, soils that are appro-
priate for a road subgrade may not always do as well for
septic fields.
Slopes
Steepness of slopes can be easily determined by look-
ing for the capital letter posted behind the first series of
numbers or letters. The "A" slopes are usually very
141
-------
Break Points
Figure 8. Identify break points.
Figure 9. Watershed boundary.
Watershed
Boundary
v^-it'-.';, ',, i *
u.V" ';><' '*',.**** ^
^f|$^?5
Figure 10. Soils map.
142
-------
gentle, with B, C, and D slopes progressively steeper.
Knowing the slopes on the site helps determine the
amount of grading required and the amount of earth to
be moved. Slope steepness also indicates the potential
for problems with erosion and stabilization of the site.
Erosiveness
The soil survey sections entitled "Detailed Soil Map
Units" and "Classification of the Soils" provide more
specific information regarding the soils and their forma-
tions and uses. It is important to scan these sections for
any potential erosion problems. Knowing a soil is ero-
sive in nature is useful when analyzing how construc-
tion, mass grading, and clearing could affect the site.
This can help predict how much soil loss could occur
and pinpoint the best erosion and sediment controls to
be used on the site. An erosion problem already present
on the site may be indicated by the use of a number after
the symbol depicting the soil type and slope on the map
(e.g., 104B2).
Wetness
To determine if the soil present on the site is hydric or
"wet," the soil description section and land capability
classification indicate whether or not that soil has a
water table at or near the surface. Most of the wet soils
occur in valley bottoms or depressional areas. On the
soil map itself, the wetness may be designated with a
"W" preceding or following the soil symbol. Knowing if a
soil has a tendency towards wetness can signal poten-
tial hazards. A site originally planned for septic systems
may have to turn to sewer and water, or a site could
contain wetlands that require protection.
Permeability
Soil permeability is important to a variety of people when
looking at a potential construction site. The permeability
of the soil can determine if the site is appropriate for a
detention pond, a septic field, or an infiltration trench. In
addition, knowing if the soil has a slow or fast perme-
ability can alert the planner to the potential for ponding
or ground-water vulnerability.
Drainage Patterns
Soil surveys typically have a smaller scale than a topo-
graphic map; therefore, more detail pertaining to the
landscape can be shown. Drainage patterns are impor-
tant to identify. Drainage patterns highlight how the land
slopes and drains and in what direction. This is impor-
tant when considering a site for development, as it is
advisable to keep the natural drainage pattern intact
whenever possible. Utilizing natural drainage can elimi-
nate the need for regrading and rerouting of runoff from
the site.
Other Information
Other symbols used on a soil survey may denote a
wetland or marsh, or the presence of heavy clays, de-
pressional areas, intermittent streams, springs, and ero-
sion spots. These features are not always found on a
topographic map. This information is particularly impor-
tant when doing cursory site evaluations.
The most important point to remember when using the
information in a soil survey is to recognize that it has
inherent limitations. Due to the scale in the field versus
that of an aerial photograph, the soil survey can only
point towards a situation that may need further investi-
gation. Any questions raised by the soil survey should
be followed by an onsite soil determination by a qualified
soil scientist.
Inventory of Natural Systems
Most areas have National Wetland Inventory (NWI)
maps produced by the U.S Fish and Wildlife Service. On
the NWI maps, the wetlands are defined as "lands tran-
sitional between aquatic and terrestrial systems where
the water table is usually at or near the surface, or the
land is covered by shallow water." In addition, the defi-
nition requires that one or more of the following three
attributes be present: "1) at least periodically the land
supports predominantly hydrophytes, 2) the substrate is
predominantly undrained hydric soil, or 3) the substrate
is nonsoil and is saturated with or covered by shallow
water at some time during the growing season of each
year." Therefore, these maps contain information on
sites that have lakes, rivers, and streams, as well as
such areas as marshes, bogs, and swamps.
Some counties have advanced wetland mapping that
delineates critical areas in need of protection from con-
struction disturbances using the NWI maps as one of
their criteria. Recently, SCS has inventoried wetlands in
agricultural fields and adjacent areas. In addition, SCS
has also identified highly erodible cropland fields. These
areas, if developed, will have special needs for soil
erosion and sediment control measures.
Other natural systems that need to be included in the
watershed review process are ground-water resources,
such as aquifers, and recharge areas to public and
private wells. Many states have mapped their ground-
water resource areas, and local municipalities should
have maps showing the location of and contribution
zones to public wells.
It is important to examine several additional maps to
gain a proper perspective on other developments in the
watershed. Comprehensive zoning and plan maps re-
veal current land use and plans for the future of the area.
These maps are invaluable when determining what
stormwater best management practices (BMPs) should
be applied to the site. If development currently exists
143
-------
upstream or more development is planned, caution may
need to be taken when situating homes or businesses
near a stream. Conversely, if the proposed development
will be upstream of existing developments, detention
measures may be needed to prevent downstream flood-
ing. Whatever the situation, knowing where develop-
ments are and where they will be helps determine what
means and methods of prevention and protection need
to be taken.
Identification of Impacts From
Development
Once the locational information for the project has been
gathered and the contributing watershed identified, it is
necessary to consider the impacts the development will
have on the watershed. In general, the major impacts
will be alterations in water quality, water hydrology, and
terrestrial and/or aquatic habitat. Some simple methods
allow initial judgments to be made as to the extent of the
impact and the level of mitigation required to protect the
surrounding ecosystem (1).
Changes in Water Quality
As people inhabit and use the lands around them, they
deposit various pollutants on the land. When rainfall and
runoff occur, these pollutants are washed into receiving
waters. As urban development occurs within the water-
shed and the land use changes, pollutants, loading
rates, and the concentration of pollutants discharged to
the receiving waters also change. Many studies have
been conducted during the past 20 years to characterize
the types and amounts of pollutants associated with
various land uses, including urban land uses. A review
of the results indicates that different types of land use
generate "typical" pollutants, at amounts within a range
of values (2). (These values have been consolidated
into a single value based on statistical analysis of all
data.)
Pollutant Concentration
Some pollutants are more likely to have short-term
(acute) effects on environmental systems because of
the pollutant concentration. Typically, the pollutants con-
sidered to have an acute impact on water quality are
oxygen-demanding substances and bacteria. Using an
equation that considers normal probability, median pol-
lutant concentrations, and variability, estimates can be
made of the probability that pollutant concentrations will
exceed acceptable water quality standards. The equa-
tions used for estimating concentrations and probability
of exceedances are found in Equations 1 and 2 (2).
where (for log-transformed data)
Cx = expected concentration of pollutant x
Z = standard normal probability (for specified
probability of occurrence)
Cm = median pollutant concentration
COV = coefficient of variation
To estimate probability, use the equation
Z = (ln[Cx/Cm])/[ln(1+COV2)1/2]
Pollutant Loads
(Eq. 2)
Some pollutants are likely to have long-term (chronic)
effects on environmental systems because of pollutant
loading rates. Typically, the pollutants considered to
have a chronic impact on water quality are nutrients,
sediments, toxic metals, organics, and some oxygen-
demanding substances. One approach relies on the
development of unit area loading rates for various pol-
lutants for different land uses. The unit area loading
values are generally a numerical value based on the
area of land use (1).
Many methods have been developed to estimate the
pollutant load that would be expected from a proposed
development. The anticipated value can be compared
with the existing pollutant loads to determine the in-
crease in pollutant loading. One of the easiest methods
to use is the Simple Method (3). This method uses
readily available information but is limited to sites less
that 1 square mile in area. Loading information gathered
can be used to judge whether some type of runoff
treatment will be needed before discharging to the re-
ceiving waters. The equation for estimating pollutant
loads is found in Equation 3.
When concentration is in mg/L,
L = (P) (Pi) (Rv) (C) (A) (0.227)
(Eq. 3)
where
L = annual mass of pollutant export (Ib/yr)
P = annual precipitation (in.)
PJ = correction factor for smaller storms that do not
produce runoff (dimensionless)
Rv = runoff coefficient (dimensionless)
C = average concentration of pollutant
A = site area (acres)
To estimate expected concentrations, use the equation When concentration is in u,g/L,
Cx = Cm (exp [Z (1 n {1+COV}2)1'2]) (Eq. 1) L = (P) (Pj) (Rv) (C) (A) (0.000227)
(Eq. 4)
144
-------
Changes in Water Hydrology
As development occurs within the watershed, the degree
of imperviousness within the watershed often increases.
Impervious surfaces do not allow rainfall to infiltrate as
would occur in an undeveloped setting; as a result, more
rainfall becomes runoff. As the amount of impervious-
ness increases, so does the amount of runoff from the
site. Taken individually and cumulatively, the increase in
runoff will change the hydrology of the watershed. De-
pending on the location of the site within the watershed
and on development conditions in other areas of the
watershed, changes in watershed hydrology can nega-
tively affect downstream properties, causing flooding
and property destruction, and also lead to downstream
bank destabilization, erosion, and scouring. In some
areas of the country, land subsidence becomes an issue
if the water table is lowered because of the lack of
ground-water recharge. This problem can be addressed
through ordinances that stipulate all pre- and postde-
velopment runoff rates for the entire watershed be con-
sidered when a single site is being developed.
A commonly used method for determining the pre- and
postdevelopment runoff rates for a site and watershed
is SCS Technical Release 55, "Urban Hydrology for
Small Watersheds." TR55 can serve as an initial screen-
ing procedure for estimating runoff values. An advan-
tage of the procedure is its ease of use through charts
and availability on computer disk (4).
Alterations in Terrestrial and Aquatic Habitat
As more undisturbed lands near shore areas are con-
verted into urban and suburban land uses, areas once
inhabited by terrestrial and aquatic animal and plant
species are minimized or destroyed. As native habitats
have continued to decrease over the years, more atten-
tion has been given to the need to protect and preserve
them. In many areas, endangered species laws serve
to protect habitat areas for those plants and animals
appearing on state and federal endangered species list.
Although this is helpful, it does little to protect more
prolific and less sensitive plant and animal species that
are burdened by urban development. Consideration of
and accommodations for plant and animal species
should and can be incorporated into the individual site
planning process as well as the watershed management
strategy.
Development of Management Goals and
Objectives
An effective method to review site development is to first
consider what the overall watershed management ob-
jectives are. One place to start looking for this type of
information is within the existing state water quality
standards. Water quality standards give numerical val-
ues and narrative descriptions for various pollutants, at
levels that are protective to human and biological health,
and assign designated uses forthe resource. A manage-
ment approach can consist of a review of the existing
and potential designated uses for the resources within
the watershed, and can attain or preserve these uses.
In addition, local agencies may have developed man-
agement objectives through such mechanisms as wa-
tershed protection districts.
A simple hierarchy of management objectives has been
presented by Schueler et al. (5), which consists of
the following:
Reducing increases in pollutant loading and
concentration.
Reducing the severity of impacts of pollutant loading
and urbanization.
Addressing specific pollutants.
Protecting sensitive areas.
Controlling floods.
Restoring the area.
Whipple (6) also uses a hierarchical method of desig-
nated uses as management objectives:
Habitat of threatened or endangered species and out-
standing natural resource waters.
Water supply from both surface and ground.
Other areas to be protected.
Those not needing protection.
Figure 11 presents a resource area hierarchy con-
sisting of:
Baseline urban nonpoint source pollutant control
Baseline urban resource protection
Control of specific pollutants
Protection of sensitive resource areas
Flood control
Management Objectives
"Baseline" Urban Nonpoint Source Pollutant Control
"Baseline" Urban Resource Protection
Control of Specific Pollutants
Protection of Sensitive Resource Areas
Flood Control
Figure 11. Resource area hierarchy.
145
-------
Development of Recommendations for
Mitigation
After consideration has been given to the degree to
which changes in water quality, hydrology, and habitat
alterations potentially affect the watershed and the site
and after management goals and objectives have been
identified, it is necessary to develop management strate-
gies that mitigate impacts to the level desired. This is
accomplished through the use of mitigation techniques,
commonly referred to as BMPs. These practices can take
the form of engineered practices, called structural BMPs,
or nonengineered practices, called nonstructural BMPs.
BMPs can be implemented on a site-specific basis and
on a regional orwatershed basis. The overall management
objectives and the severity of impacts from development
may dictate the degree of mitigation required (7).
In selecting BMPs for a site, it is important to consider
1) how the BMPs will function as a system; 2) how the
practice will meet watershed- and site-specific manage-
ment objectives, such as pollutant load and concentra-
tion reduction, control of storm volumes, and provision
of habitat; and 3) what some of the limitations and uses
of the practices are.
Best Management Practice Systems
Structural and nonstructural BMPs differ in their design,
limitations, and optimal applicability (i.e., addressing pol-
lutant loads, habitat, or hydrology). While some BMPs are
implemented to provide a primary objective, secondary
mitigation and benefits also are commonly provided. For
example, a wet detention pond optimally functions to im-
prove water quality through pollutant load reduction but
can also function to balance water hydrology and pro-
vide habitat. BMPs can be grouped into discrete func-
tional units that address different aspects of stormwater
management. These units are pollution prevention, habitat
protection, runoff attenuation, runoff conveyance, runoff
pretreatment, and runoff treatment. The units, taken
together, form the BMP system. The BMPs selected to
meet watershed- and site-specific objectives generally
will be from all of these functional units. Figure 12 de-
picts a BMP systems approach, described below:
Pollution prevention: An effective approach to man-
aging pollutants in urban settings is to prevent or
reduce the potential for pollutant loading. Many of the
pollution prevention practices are referred to as non-
structural BMPs. These practices can include such
activities as public education, zoning ordinances, site
planning procedures, restricted use policies, and
overlay districts.
Habitat protection: An effective tool for the restoration
and management of habitat areas is the implemen-
tation of measures to ensure long-term protection.
Habitat protection is usually accomplished through non-
structural BMPs, such as river corridor programs,
wetland protection programs, critical habitat protection
programs, and zoning tools such as open space require-
ments and creative land-use planning techniques
(cluster development).
Runoff attenuation: One of the most effective ways
to manage stormwater flows is to prevent and reduce
them. Much of this can be accomplished through a
reduction in site impervious cover. Reduction in im-
pervious cover allows for increased infiltration. Other
practices that attenuate runoff are drywells, depres-
sion storage, and appropriately placed infiltration
trenches. Implementing these practices reduces the
other impacts of development by reducing runoff vol-
ume, flood occurrence, pollutant loads and concen-
trations, and stream degradation.
Runoff conveyance: Runoff conveyance systems
serve to transport the storm flows from the point of
origin to the runoff pretreatment and treatment sys-
tem. Runoff conveyance systems can allow for lim-
ited treatment levels, as in the case of grassed
swales with check dams and exfiltration devices.
Other conveyance systems for stormwater include
structural elements, such as pipes with flow splitters.
Runoff pretreatment: Runoff pretreatment is the proc-
ess whereby runoff is diverted through pretreatment
practices. These practices usually prolong and im-
prove the efficiency of the treatment device. Pretreat-
ment practices include vegetated filter strips, riparian
systems, settling basins, and water quality inlets.
Runoff treatment: Runoff treatment practices are de-
vices designed to treat stormwater runoff and remove
pollutants through a number of processes, including
adsorption, transformation, and settling before entry to
Water Quality
Pollution Prevention
Runoff Attenuation
Habitat
Protection
Maintenance
Hydrology
Runoff Conveyance Runoff Pretreatment Runoff Treatment
Secondary Impacts
Figure 12. BMP systems approach.
146
-------
the resource area. Treatment devices are considered
the final component of the BMP system. Some famil-
iar treatment devices include detention, retention, and
infiltration.
Several additional issues need to be considered when
developing recommendations for practices. Among these
are acceptance of practices by landowners and the
aesthetic quality of the practices. Although these issues
seem minor, disgruntled landowners can inhibit implemen-
tation of effective long-term management programs.
A frequently overlooked but critical consideration for storm-
water management is the development of long-term main-
tenance and financing programs. BMPs, once installed,
require upkeep and periodic repairs. Long-term urban
runoff management programs require a commitment to
maintain technical and program support staff.
Determine Reduction or Protection Measures
Necessary To Achieve Objectives and Meet
Watershed and Site-Specific Needs
To develop a management strategy, it is important to
integrate watershed needs with site-specific needs. The
simplest approach is to first consider the broader water-
shed needs and then "work in" site-specific needs around
them. Examples of broad watershed management needs
are protecting public water supplies, river corridors and
riparian areas, wetlands and wildlife habitat; preserv-
ing/expanding open space; or meeting a watershedwide
pollutant reduction goal. To address these needs, man-
agement practices such as no construction/no distur-
bance buffer zones, creative site layout practices,
impervious cover limitations, tree disturbance restric-
tions, total site disturbance limitations, and riparian en-
hancement zones may be utilized. These management
practices tend to define or refine areas for the actual site
development and site-specific practices.
On the site level, with broader watershed management
practices incorporated, more specific needs can be ad-
dressed. Examples of site-specific management needs
are preventing or managing soils loss, lowering the
postdevelopment discharge rate and volume, enhanc-
ing riparian areas, and reducing pollutant loads from the
site. To address these needs, management practices
such as developing and implementing a preventive soil
erosion control plan, and installing such items as tem-
porary sediment basins, siltation fencing, dry wells, in-
filtration trenches, wet ponds, and native plant species
planting may be utilized.
It is important to remember that a combination of BMPs
is often necessary to achieve desired objectives. No one
single practice will provide all necessary mitigation or
benefits. Table 1 provides an example of how watershed
objectives can direct selection of various practices.
Best Management Practice Limitations
To provide information on the limitations and uses of
BMPs, several charts have been developed. The most
recently completed of these is found in Schueler et al.
(5). Summary information can also be found in Schueler
(3) and U.S. EPA (8). Information contained in the charts
includes advantages, disadvantages, cost efficiency,
limitations for ground-water depth, and soils. Schueler
and colleagues consolidated information on reported
BMP efficiency in a similar chart form (5). All of this
information can help the decision-maker determine the
most effective mix of practices to meet stated objectives.
Figures 13 and 14 provide an example of the BMP
limitation charts available.
Benefits of Watershed Planning
The most obvious benefit realized from a watershed
planning approach is the installation of BMPs to mitigate
water management issues before serious problems re-
sult. Advance planning saves valuable resources at the
state and local level, which could be used in other areas.
Economies of scale can also be realized as a result of
the watershed approach. When installing regional prac-
tices, larger areas within the watershed can be treated
on a per unit area cost basis. This will be beneficial to
the development community and the local jurisdictions.
Restoration is always more expensive than prevention.
Most restoration costs are associated with damage off
site and downstream by runoff and sedimentation. As
emphasized earlier, the amount and velocity of runoff
flowing off site can cause severe erosion of stream-
banks and watercourses. Watershed planning can elimi-
nate restoration costs by examining the surrounding
area proposed for development. With preliminary runoff
control measures, much downstream and offsite dam-
age can be prevented and controlled.
Another hazard of poor planning involves dredging of
sediment-laden streams, channels, and lakes. Dredging
is a very expensive solution to a problem that could have
been prevented for a fraction of the cost. Again, proper
examination of an area on a watershed basis can target
erosive soils and extensive urbanization with BMPs to
keep offsite erosion and sedimentation from occurring.
Mitigation involves creating sensitive habitat areas, usu-
ally wetlands, after they have been replaced by filling or
construction. Mitigation can often be avoided if some
advanced watershed planning is undertaken. By deline-
ating sensitive areas early, alterations in construction
plans can be worked around the sites. In planning large
areas, sensitive areas can be designated and protected
through land acquisitions and greenbelt planning.
Finally, by doing advanced watershed planning the po-
tential for court actions in the case of flooding, erosion
147
-------
Table 1. Tools To Achieve Watershed Objective
Watershed Objective
BMP System Component
Tools
Baseline nonpoint source pollutant
control
Pollution prevention
Runoff conveyance
Runoff pretreatment
Erosion control
Buffer requirements
Pesticide/Fertilizer reduction
Grassed swales with check dams
Vegetated buffer strips
Baseline urban resource protection
Pollution prevention
Runoff attenuation
Runoff pretreatment
Runoff treatment
Steep slope restriction
Site fingerprinting
Minimum site disturbance
Cell closure/opening
Construction phasing
Erosion control
Buffer requirements
Infiltration trenches
Drywells
Reduced directly connected impervious areas
Stream buffers
Wetlands buffers
Infiltration basins
Specific pollutants
Pollution prevention
Runoff conveyance
Runoff pretreatment
Runoff treatment
Septic system density
Restricted use areas
Nitrogen overlay district
Grassed swales with check dams
Vegetated buffer strips
Riparian buffers
Water quality inlets
Wet extended detention ponds
Sensitive areas
Pollution prevention
Habitat protection
Hazardous waste recycling
Stenciling storm drains
Industrial cross connections
Underground storage tank regulations
Protection districts
Restricted uses
Decreased DCIA
Nitrogen overlay zones
Septic density requirements
Extensive erosion/sediment control
Wellhead protection program
River corridor program
Open space requirements
Cluster development
Wetlands protection program
Critical habitat program
Riparian zone requirements
Resource area buffer requirements
Flood control
Runoff attenuation
Runoff conveyance
Infiltration trench
Drywells
Riprap swales
Detention ponds
Retention ponds
148
-------
J? *f?
,*P
BMP
Extended Detention Pond
Wet Pond
Infiltration Trench
Infiltration Basin
Porous Pavement
Water Quality Inlet
Grassed Swale
Filter Strip
x
O O
3^00
O O O O O
3 O O 3 3 O
3
3 O
O
O
oooooooo
00*000
OO3300OO
3333
3
O
O May Preclude the Use of a BMP
(P Can Be Overcome With Careful Site Design
W Generally Not a Restriction
Figure 13. Other common restrictions on BMPs (3).
damages, sedimentation removal, dredging, and sensi-
tive habitat areas may be lessened. By looking at the
watershed area in total and addressing probable haz-
ards both upstream and downstream, the chances of
causing damage downstream will be minimized.
References
1. Marsalek, J. 1991. Pollutant loads in urban stormwater: Review of
methods for planning level estimates. Water Res. Bull. (April).
2. U.S. EPA. 1983. Results of the Nationwide Urban Runoff Program,
Vol. I. Final report. NTIS PB84185552.
3. Schueler, T.R. 1987. Controlling urban runoff: A practical manual
for planning and designing urban BMPs. Washington, DC: Metro-
politan Washington Council of Governments.
4. U.S. Department of Agriculture, Soil Conservation Service. 1986.
Urban hydrology for small watersheds. Technical Release 55.
5. Schueler, T.R., PA. Kumble, and M.A. Heraty. 1992. A current
assessment of urban best management practices: Techniques for
reducing nonpoint source pollution in the coastal zone. Washing-
ton, DC: Metropolitan Washington Council of Governments.
6. Whipple, W, Jr. 1991. Best management practices for storm water
and infiltration control. Am. Water Res. Bull. (December).
7. Phillips, N. 1992. Decisionmaker's stormwater handbook: A
primer. Washington, DC: Terrene Institute.
8. U.S. EPA. 1993. Guidance specifying management measures for
sources of nonpoint pollution in coastal waters. EPA/840/B-92/002.
Washington, DC.
Additional Reading
1. Bannerman, R.T., R. Dodds, D. Owens, and P. Hughes. 1992.
Sources of pollutants in Wisconsin stormwater. Wisconsin Depart-
ment of Natural Resources grant report.
2. Dennis, J., J. Noel, D. Miller, and C. Eliot. 1989. Phosphorus
control in lake watersheds: A technical guide to evaluating new
development. Maine Department of Environmental Protection.
3. Schueler, T.R. 1991. Mitigating the adverse impacts of urbaniza-
tion on streams. In: Watershed restoration sourcebook. Washing-
ton, DC: Metropolitan Washington Council of Governments.
4. Schueler, T.R., J. Galli, L. Herson, P. Kumble, and D. Shepp. 1991.
Developing effective BMP systems for urban watersheds. In: Wa-
tershed restoration sourcebook. Washington, DC: Metropolitan
Washington Council of Governments.
5. Shelly, P. 1988. Technical memorandum to SAIC. Novembers.
6. U.S. Department of Agriculture, Soil Conservation Service. 1986.
Soil survey of Ford County, Illinois.
7. U.S. EPA. 1990. Urban targeting and BMP selection. Washington,
DC: Terrene Institute.
8. U.S. EPA. 1985. Water quality assessment: A screening proce-
dure for toxic pollutants. EPA/600/6-85/002A.
149
-------
BMP/Design
Extended Detention Pond
Design 1
Design 2
Design 3
Wet Pond
Design 4
Design 5
Design 6
Infiltration Trench
Design 7
Design 8
Design 9
Infiltration Basin
Design 7
Design 8
Design 9
Porous Pavement
Design 7
Design 8
Design 9
Water Quality Inlet
Design 10
Filter Strip
Design 11
Design 12
Grassed Swale
Design 13
Design 14
3
*
O
(5
O
O
(3 (3 3
3 (5 3
33
3
3 (5
3 (3 (3
33
339
339
3 3
3 3
3
(5 (3
3
3
O O O
339
O O O
3 (3 (3
(3
O
O
Moderate
Moderate
High
Moderate
Moderate
High
Moderate
High
High
Moderate
High
High
Moderate
High
High
Low
Low
Moderate
Low
Low
Key:
O 0 to 20% Removal
(5 20 to 40% Removal
(J 40 to 60% Removal
^ 60 to 80% Removal
9 80 to 100% Removal
Q£) Insufficient Knowledge
Design 1: First-flush runoff volume detained for 6 to 12 hr.
Design 2: Runoff volume produced by 1.0 in., detained for 24 hr.
Design 3: As in Design 2 but with shallow marsh in bottom stage.
Design 4: Permanent pool equal to 0.5 in. of storage per impervious acre.
Design 5: Permanent pool equal to 2.5 (Vr), where Vr = mean storm runoff.
Design 6: Permanent pool equal to 4.0 (Vr); approx. 2 weeks of retention.
Design 7: Facility exfiltrates first-flush; 0.5 in. of runoff/impervious acre.
Design 8: Facility exfiltrates 1-in. runoff volume per impervious acre.
Design 9: Facility exfiltrates all runoff up to the 2-year design storm.
Design 10: 400 ft3 of wet storage per impervious acre.
Design 11: 20-ft-wide turf strip.
Design 12: 100-ft-wide forested strip with level spreader.
Design 13: High-slope swales with no check dams.
Design 14: Low-gradient swales with check dams.
Figure 14. Comparative pollutant removal of urban BMP designs (3).
150
-------
The Soil Conservation Districts' Role in Site Plan Review
Glenn Bowen
Kent Conservation District, Dover, Delaware
Eric H. Buehl and John M. Garcia, Jr.
Sussex Conservation District, Georgetown, Delaware
Abstract
Officially organized nearly 50 years ago, both the Kent
and Sussex Conservation Districts have been at the
forefront of soil and water conservation. The more spe-
cific role of the conservation districts in sediment control
and stormwater management is tied to two legislative
initiatives. In 1978, the Delaware State Legislature
passed an Erosion and Sediment Control Law (Chapter
40, Title 7, Delaware Code). In 1991, this law was
amended to include stormwater management.
Because certain types of construction can increase
sediment yields by 2,000 times, sediment control is a
necessary first step on any construction site. The con-
servation districts' role in reviewing site plans is based
on the importance of sediment control for limiting the
degradation of surface water.
The conservation districts review site plans for stormwa-
ter management quantity control to ensure that the risk
of downstream flooding is reduced and stream channel
erosion is controlled. This is achieved by sustaining
predevelopment runoff rates for the 2-, 10-, and 100-
year storm events at the postdevelopment state and
maintaining similar hydrograph timing for peak flows
before and after development.
When reviewing site plans, the conservation districts also
consider the quality of stormwater runoff. The order of
preference for practices to improve water quality, accord-
ing to Delaware law, is as follows: ponds with a perma-
nent pool, extended detention ponds without a permanent
pool, and infiltration systems. The acceptability of other
practices that can remove up to 80 percent of the sus-
pended solids in runoff is determined on a case-by-case
basis. The Kent and Sussex Conservation Districts have
promoted sand filtration systems and biofiltration swales
for water quality treatment where applicable.
Background
Delaware, the first state to ratify the Constitution, in
1787, has a rich history dating back to pre-colonial
times. Delaware is 1,978 square miles; only Rhode
Island has less land mass. Located entirely on the Del-
MarVa (Delaware, Maryland, and Virginia) Peninsula,
Delaware is a 2- to 3-hour drive from Baltimore, Mary-
land; Washington, DC; Philadelphia, Pennsylvania; and
Norfolk, Virginia.
Location between the Chesapeake and Delaware Bays
and the Atlantic Ocean provides for a moderate climate.
Delaware receives 45 in. of rainfall annually, and Kent
and Sussex Counties experience an average of 187
frost-free days a year. New Castle County, the north-
ernmost of the three Delaware counties, is partially lo-
cated in the Piedmont region, while the rest of the state
is in the Atlantic coastal plain. Delaware's gently rolling
topography starts at sea level and peaks at 368 ft in the
northern part of the state.
With a statewide population of just over 666,000, Dela-
ware has unique demographics. Currently, two-thirds of
the population is located on less than one-third of the
land in the state. Northern New Castle County, in which
the city of Wilmington lies, is within easy commuting
distance of Philadelphia and northeastern Maryland.
The city of Dover, located in Kent County in the central
portion of the state, is not only the state capital but in
1992 was officially designated a metropolitan area. Kent
County, which has considerable land in agricultural pro-
duction, is also the home of Dover Air Force Base, a
central military airlift command facility. Both of these
factors have combined to produce considerable growth
around the capital city.
Sussex County, the southernmost of the three counties,
has two areas of interest that have brought considerable
development to a primarily rural area. One is a 25-mile
151
-------
stretch of Atlantic Ocean shoreline. The other area is
commonly referred to as the "Inland Bays" region, which
has 80 miles of shoreline located directly behind the
coastal barrier dune system. Although the resident
population of Sussex County is just over 113,000, during
the peak of the tourist season (July 4th weekend) the
population balloons to an estimated 300,000 people.
In 1969, Governor Russell Peterson assigned a task
force to study the steady decline of shellfish and finfish
populations as well as related environmental issues of
concern for the Inland Bays region. Reports and studies
over the subsequent two decades pointed to the neces-
sity of encouraging land-use planning and establishing
various water quality initiatives regarding agricultural
land and land that could be developed.
Steady growth in the state's metropolitan areas was not
surprising. The increasing development in the two more
rural counties of Kent and Sussex, however, brought the
conservation districts to the forefront of soil and water
conservation efforts at land development projects.
The Role of the Conservation Districts
In their first 50 years, the conservation districts were
primarily involved in agricultural issues affecting local
landowners. Historically, each district has been run by a
board of seven elected supervisors, most of whom are
local farmers, and has functioned as a clearinghouse for
current information about the construction and mainte-
nance of drainageways, wildlife ponds, and water con-
trol structures; updates on the availability of technical
and financial assistance for farmers and other residents;
and education activities related to resource manage-
ment and protection.
In 1978, Delaware passed an Erosion and Sediment
Control Law covering most types of residential, commer-
cial, industrial, and institutional construction. In 1980,
the conservation districts were enlisted to implement the
law by the Delaware Department of Natural Resources
and Environmental Control (DNREC). DNREC turned to
the conservation districts because of their intimate know-
ledge of the counties in terms of constituents, soils, topo-
graphy, and local and county governmental structure.
Moreover, the conservation districts had a proven ability
to run cost-effective programs with a minimum of "red tape."
From 1980 to 1987, development authorities were pri-
marily concerned with erosion and sediment control in
regard to all types of new construction. Stormwater
management was handled by various state and munici-
pal agencies on an "as needed" basis to control flooding.
Then, in 1989, DNREC began the long process of es-
tablishing a statewide stormwater management law to
address both runoff quantity control and water quality
concerns. Using an approach that involved not only the
regulators but also the regulated community, DNREC
encountered a minimal amount of public opposition and
gained the full support of the state legislature.
Thus, on July 1, 1991, the Erosion and Sediment Con-
trol Law was amended to include stormwater manage-
ment. The conservation districts are now the lead
agencies implementing this law. The program is consid-
ered by many to be a model of efficiency, not only from
a cost perspective but also in terms of the rapid turn-
around time for plan reviews, which is extremely impor-
tant for interested parties in this age of fax machines,
electronic mail, and cellular phones.
Scope of Site Plan Review
Review of site plans for construction projects has evolved
from mere suggestions provided by a district employee
concerning what might work best at a particular location
to an engineered topographic plan showing the project's
location, the site's details, and specifications for all prac-
tices to be used. To illustrate the plan review process,
we occasionally refer in this paper to a project for "Run-
ning Brook Estates and Business Park" (Figure 1).
Plan review goes beyond looking at blueprints to see
that specifications meet minimum standards set forth in
state laws and regulations. Material that district inspec-
tors frequently use to assess a project include:
The state erosion and sediment control handbook.
The district sediment and stormwater manual.
County soil surveys.
U.S. Geological Survey topographic maps.
Federal Emergency Management Agency floodzone
maps.
State/Federal wetland inventories.
The Delaware Department of Transportation (DelDot)
specification book.
Equipment manufacturer specifications and literature.
The most important tool for ensuring a thorough design
as well as a consistent and efficient review is the man-
agement plan checklist. Figure 2 presents the checklist
used by the Kent Conservation District.
Sediment Control
A plan for sediment control and stormwater manage-
ment usually evolves from the site or grading plan but
includes the location, dimensions, and details for the
required erosion and sediment controls.
In some cases, designers or developers choose to use
the stormwater facility as a sediment trap or basin. This
is easily accomplished by modifying the facility's outlet
control structure to include the necessary filtration
devices (Figure 3). Although use of an infiltration basin
152
-------
Figure 1. Site map for Running Brook Estates and Business Park.
as a sediment trap is generally discouraged, on occasion
it may be necessary. For such cases, several approaches
are recommended. One is to direct any sediment-laden
runoff to a trap (Figure 4, northeast corner). Another is
to leave the basin 12 to 24 in. above finished grade until
the site is stabilized; excess material is then removed and
the basin graded according to the plan's specifications.
The management plan must describe the construction
sequence and establish the points at which various
control installations must be added, removed, com-
pleted, or activated. For certain features, such as em-
bankment ponds, the contractor may be required to
notify the district inspector when construction is about
to commence. This gives the inspector the opportunity
to reemphasize the importance of such aspects of the
installations as a cutoff trench and the emergency spill-
way's dimensions and to visually inspect riser struc-
tures, antiseep collars, and the foundation preparation.
Additional sediment control features commonly presented
in the plan include the following (see also Figure 4):
Rock-check dams: Used for velocity and erosion
control in ditches and swales.
Perimeter dikes/swales, earth dikes, temporary
swales: Used to convey runoff to a trap or as a
clean-water diversion.
A stabilized construction entrance: Stone structure
used to minimize sediment tracking onto roadways.
Vegetative requirements list (permanent and tempo-
rary): Used to specify amounts and types of seed,
mulch, and soil amendments needed.
A silt fence: Commonly used downstream of dis-
turbed soils as a perimeter filtration device.
Often the review process reveals unique or unexpected
site features requiring that the district inspector make
additional site visits, hold meetings with designers, and
seek technical guidance from the Soil Conservation
Service or the DNREC Division of Soil and Water Con-
servation. For example, because of the unique soils on
the DelMarVa Peninsula, erosion problems necessi-
tated that a list of soil erodibility (K) values (Figure 5),
as determined by the Universal Soil Loss Equation, be
compiled for the predominant soil types shown on the
sediment and stormwater plan for Running Brook Es-
tates and Business Park (Figure 6). Such lists not only
expedite the review process but also help designers
better prepare for the review comment period.
Stormwater Management for Quantity
Control
The adverse impacts of stormwater runoff have been
well documented. Damage caused by flooded streams
and rivers has cost millions of dollars in property losses
and has degraded the quality of the nation's waters.
Reducing the risk of downstream flooding and stream-
channel erosion after land development is the primary
153
-------
KENT CONSERVATION DISTRICT
SEDIMENT AND STORMWATER MANAGEMENT PLAN
SUBMISSION REQUIREMENTS
1 Review is predicated upon receipt of one set of plans and applicable review
and inspection fee.
2 Upon notification of approval, one additional set of plans must be
submitted to be stamped and kept available on the construction site a£ all
times.
REQUIRED STATEMENTS
1 Provide the name, mailing address, and phone number of the owner of the
property, the land developer, the engineer or consultant and the applicant.
Provide names of adjacent property owners on the plan.
2 Include the following notes:
A The Kent Conservation District must be notified in writing five (5)
days prior to commencing with construction. Failure to do so
constitutes a violation of the approved Sediment and Stormwater
Management Plan.
B Review and or approval of the Sediment and Stormwater Management Plan
shall not relieve the contractor from his or her responsibilities for
compliance with the requirements of the Sediment and Stormwater
Regulations, nor shall it relieve the contractor from errors or
omissions in the approved plan.
C If the approved plan needs to be modified, additional sediment and
Stormwater control measures may be required as deemed necessary by
the Kent Conservation District.
D The Kent Conservation District reserves the right to enter private
property for purposes of periodic site inspection.
E Following soil disturbance or redisturbance, permanent or temporary
stabilization shall be completed within 14 calender days as to the
surface of all perimeter sediment controls, topsoil stockpiles, and
all other disturbed or graded areas on the project site.
3 Include signed Owner's Certification of the following statements (these
must be signed in ink on each plan submitted):
A I, the undersigned, certify that all land clearing, construction and
development shall be done pursuant to the approved plan.
B I, the undersigned, certify that responsible personnel certified by
DNREC will be in charge of on-site clearing and land disturbing
activities.
GENERAL REQUIREMENTS
1 Provide a legend on the Sediment and Stormwater Management Plan.
2Provide a "limit of disturbance" line and the disturbed area in acres.
3 Provide a vicinity map with a scale of 1" = 1 mile.
4 Provide a north arrow on the plan.
5 Maximum plan scale of 1" = 100'
6 Plans must be submitted on 24"x36" sheets.
7 When two or more sheets are used to illustrate the plan view, an index
sheet is required, illustrating the entire project on one 24"x36" sheet.
8 Provide existing and proposed contours based on mean sea level datum
provided at one foot intervals. Total contributing drainage area must be
shown regardless of being located on or off-site.
9 For small projects, provide existing and proposed spot elevations on a 50
foot grid system, based on mean sea level datum, with high and low points.
10 State and Federal wetlands must be accurately delineated.
11Delineate the National Flood Insurance Program 100 Year Flood Zone.
12Provide soils mapping on plan with a general description of each soil.
13 Streams must be delineated.
Figure 2. Sample sediment and Stormwater management plan checklist.
154
-------
EROSION AND SEDIMENT CONTROL
1 All erosion and sediment control practices shall comply with the Delaware
Erosion and Sediment Control Handbook 1989.
2 Projects must be phased so that no more than 20 acres is cleared at any one
time. Once grading is initiated in one 20 acre section, a second 20 acre
section may have stumps, roots, brush, and organic material removed.
Grading of the second 20 acre section may not proceed until temporary or
permanent stabilization of the first 20 acre section is accomplished.
3 Stone check dams are required in all swales, ditches and channels. Provide
details, cross sections and specifications, including check dam station
locations. Check dam depth must be such that a maximum stone depth is
achieved while ensuring that flow will continue over the center of the dam.
A minimum 6" depth from the weir to the top of the structure is required.
4 All stone, with the exception of check dams, must be underlined by a filter
fabric. Filter fabric specifications must be provided for various
applications.
5 Outlet protection is required at all points of discharge from pipes,
"channels, and spillways. Provide details, cross-sections and
specifications, including d50 stone size, stone depth, outlet dimensions
and type of filter fabric.
6 Provide inlet and outlet invert elevations for all drainage structures and
facilities.
7 Provide profiles for all outfall pipes and channels.
8 Erosion control matting is required on slopes of 3:1 or greater.
9 Provide corner and lowest floor elevations for all buildings.
10 Specify what stabilization measures will be used if dust control becomes
a problem.
11 Sediment traps and basins shall be utilized and sized to accomodate 3600
cubic feet of storage per acre of contributing drainage area until project
stabilization is complete. These structures must be located at the base
of the drainage area. The following information is required: top of slope,
bottom, and outlet elevations, dimensions, proposed and required volumes,
type of trap or basin, and contributing drainage area. Include details,
cross sections and specifications; a minimum 2:1 length to width ratio is
required.
12 Diversions must be used to direct runoff into traps. When sediment-laden
stormwater is directed to traps or basins by closed pipe systems, temporary
diversions roust be used to direct stormwater to traps and basins until
closed pipe systems are operational.
13 Provide a detailed sequence of construction, at a minimum, include the
following activities: clearing and grubbing those areas necessary for the
installation of perimeter controls, construction of perimeter controls,
remaining clearing and grubbing, road grading, grading for remainder of
site, utility installation and whether storm drains will be used or blocked
until after completion of construction, final grading, landscaping or
stabilization, and removal of sediment control practices.
14 Soil stockpile areas must be delineated, locate stockpiles on areas with
little or no slope. Stockpiles must be surrounded with silt fence or a
stabilized earthen berm.
Figure 2. Sample sediment and stormwater management plan checklist (continued).
155
-------
STORMWATER MANAGEMENT
1 Show drainage calculations considering off-site contributing drainage.
Provide pre and post-development velocities, peak rates of discharge, and
inflow and outflow hydrographs of stonnwater runoff at all existing and
proposed points of discharge from the site for the 2 year and 10 year
frequency storms. Show site conditions around points of all surface water
discharge including vegetation and method of flow conveyance from the land
disturbing activity and design details for structural controls.
2 All hydrologic computations shall be accomplished using the most recent
version of USDA Soil Conservation Service TR-20 or TR-55, with the Delmarva
Unit Hydrograph. The storm duration for computational purposes shall be
the 24 hour rainfall event. The pre-development peak discharge rate shall
be computed assuming that all land uses in the site to be developed are in
good hydrologic condition.
3 Sub-watershed areas must be delineated on the plan for both the pre and
post-development conditions. Provide the area in acres of each sub
'watershed.
4 Provide directional stonnwater flow arrows for all existing and proposed
channels, pipes, etc.
5 QUANTITY: Post-development peak rates of discharge for the 2 and 10 year
frequency storm events shall not exceed the pre-development peak rates of
discharge for the 2 and 10 year frequency storm events.
6 QUALITY: Water quality structures having a permanent pool shall be designed
to release the first 1/2 inch of runoff from the site over a 24 hour
period. Practices not having a permanent pool shall be designed to release
the first inch of runoff from the site over a 24 hour period.
INFILTRATION
1 Infiltration practices shall be used only when the following criteria can
be met or exceeded:
A Systems shall be designed to accept, at least, the first inch of
runoff from all streets, roadways and parking lots. (Including all
contributing drainage areas.)
B Areas draining to these practices must be stabilized and vegetative
filters established prior to runoff entering the system.
C - A suspended solids filter accompanies the practice, when vegetation
is used there shall be at least a 20 foot length of vegetative
filter.
D The bottom of the infiltration practice is at least 3 feet above the
seasonal high water table.
E The system shall be designed to drain completely in 48 hours.
F Infiltration practices are limited to soils having an infiltration
rate of at least 1.02 inches per hour. On site soil borings and
textural classifications must be done to verify site conditions and
seasonal high water table. This information must be submitted with
the plan.
G Infiltration practices greater than 3 feet deep shall be located at
least 20 feet from basement walls.
H Infiltration practices designed to handle runoff from impervious
parking areas shall be a minimum of 150 feet from any public or
private water supply well.
I Infiltration practices shall have overflow systems with measures to
provide a non-erosive velocity of flow along its length and at the
outfall.
J The slope of the bottom of the infiltration practice shall not exceed
5 percent.
K Infiltration practices shall not be installed on or atop a slope
whose natural angle of incline exceeds 20 percent,
L Infiltration practices shall not be installed in fill material.
M Unless allowed on a specific project, infiltration practices will
only be permitted for the primary purpose of water quality
enhancement.
Figure 2. Sample sediment and stormwater management plan checklist (continued).
156
-------
PONDS
1 All ponds constructed for stormwater management shall be designed and
constructed in accordance with USOft Soil Conservation Service Small Pond
Code 378, dated September 1990, as approved for use in Delaware.
2 All ponds shall have a forebay or other design feature to act as a sediment
trap, a 10 foot reverse slope bench must be provided 1 foot above the
normal pool elevation for safety purposes, a 10 foot level bench 1 foot
below the normal pool elevation, and all embankment ponds having a
permanent pool shall have a drain installed.
DETAILS
1 Provide details and specifications for all erosion and sediment control and
stormwater management practices used.
2 Provide details of temporary and permanent stabilization measures.
3 Provide details, cross-sections and specifications (including
stabilization) for diversions, ditches, ponds, swales, infiltration
structures, etc.
4 Specify details of any unusual practices used.
MAINTENANCE
1 Specify whose responsibility it will be to maintain and repair all erosion
and sediment control and stormwater management practices during utility
installation.
2 Maintenance set aside areas for disposal of sediments removed from
stormwater management facilities must be provided. Set aside areas shall
accomodate at least 2 percent of the stormwater management facility volume
to the elevation of the 2 year storm volume elevation, maximum depth of set
aside volume shall be 1 foot, and the slope of the set aside area shall not
exceed 5 percent.
3 A clear statement of defined maintenance responsibility shall be
established during the plan review and approval process.
Figure 2. Sample sediment and stormwater management plan checklist (continued).
157
-------
(a) Outlet control
structure for
stormwater
control only
Dewatering Device Wrapped in Filter
Cloth and Encased in a Gravel Jacke
(b) Outlet control
structure with filters
for both
stormwater and
sediment
control
Elbow Supported
With Stone Pile
Figure 3. Outlet control structures for sediment and stormwa-
ter control.
reason for establishing a program that encourages
stormwater management quantity control. Indeed, it has
also been shown that flood peaks after development can
increase by more than 500-fold.
The conservation districts' role in stormwater manage-
ment quantity control is to ensure that discharge rates
for the 2- and 10-year, 24-hour duration storm events do
not increase following development. The districts also
review management plan data on hydrograph timing
and runoff volumes to ensure that areas downstream of
development sites are not adversely affected. The dis-
tricts prefer multiple-storm control because it is gener-
ally accepted as the most appropriate management
approach for a wide range of storm discharges.
To compute stormwater discharges, procedures described
in the Soil Conservation Service's Technical Release
(TR) 20 and TR55 are used. Along with being generally
user friendly, TR20 and TR55 procedures facilitate the
production of required hydrographs and the computing
of runoff storage requirements. Sussex and Kent Coun-
tiesand the DelMarVa Peninsula generallyfall under
the TR20 and TR55 Type II rainfall distribution.
Early in the model's development, concerns were ex-
pressed that this rainfall distribution did not accurately
represent the DelMarVa Peninsula, with its generally
gently rolling topography, sandy soils, and limited out-
falls. As a result, studies were performed and a new
-SCE
SCt
-24-
"" "" ~* = Temporary Dike/Swale
\f = Rock-Check Dam
y = Outlet Structure
SCE = Stabilized Construction Entrance
Figure 4. Sediment control features at Running Brook Estates and Business Park.
= Sediment Trap
158
-------
Sussex Conservation District
P.O. Box 8 : Georgetown, Delaware 19947 - Phone (302) 856-2105 or 7219
LIST OF HIGHLY ERODIBLE SOILS*
*S.C.S. FIELD OFFICE TECNNICAL GUIDE ("F VALUE OF 0.20 OR GREATER)
SOIL NAME SOIL SYMBOL "K" VALUE
ELKTON SANDY LOAM El 0.43
ELKTON LOAM Em 0.43
Ju V .Li G Ovx Jt\ v/ Ool'l JJ ********«»*** JuQJDJL/ ftM***t***»*»***»«»**»***»*U»*IO
(0-15%)
FALLSINGTON SANDY LOAM Fa 0.28
FALLSINGTON LOAM..'. Fs 0.28
KALMIA SANDY LOAM Ka 0.28
KENANSVILLE LOAMY SAND KbA/B 0.24
(0-5%)
KEYPORT FINE SANDY LOAM KfA/B2# 0.43 fERODED
(0-5%)
MATAWAN LOAMY SAND Mm 0.28
MATAWAN SANDY LOAM Mn 0.32
POCOMOKE SANDY LOAM Pm 0.28
PORTSMOUTH LOAM Pt 0.28
RUMFORD LOAMY SAND RuA-C 0.20
(0-10%)
SASSAFRAS SANDY LOAM SaA/B/C2#/D 0.28 IERODED
SASSAFRAS LOAM SfA/B 0.28
(0-5%)
WOODSTOWN SANDY LOAM Wo 0.28
WOODSTOWN LOAM Ws 0.28
Figure 5. Erodibility values for predominant soils on the DelMarVa Peninsula.
159
-------
Figure 6. Predominant soils at Running Brook Estates and Business Park.
dimensionless, synthetic unit hydrograph was devel-
oped to be used with the Type II rainfall distribution. This
hydrograph, named the DelMarVa Hydrograph, is used
in Kent and Sussex Counties. The DelMarVa Hy-
drograph can develop peak flow rates up to 60 percent
of those using just the given dimensionless synthetic
hydrograph with the Type II rainfall distribution.
Stormwater is primarily managed for quantity control
with ponds. In the Running Brook Estates and Business
Park example, three stormwater management ponds
are used (see Figure 7). The two ponds at the south side
of the site were sized in accordance with standard cri-
teria (i.e., using the 2- and 10-year, 24-hour duration
storm events for discharge rates). The third pond is
sized for a watershed with no positive outfall, a unique
situation that often exists on the DelMarVa Peninsula. In
such situations, when all possibilities to achieve an out-
fall have been exhausted, the facility is sized for the
10-year storm event runoff volume. A modified 100-year
flood zone is then determined to establish finished floor
elevations for any properties that could be affected by
storms larger than the 10-year event. Infiltration can be
factored in to reduce the size of such structures.
When development is proposed in urban areas and site
space is limited, the district inspector has the flexibility
to reduce the stormwater management quantity require-
ments to those related to quality, as discussed in the
next section.
Stormwater Management for Quality
Control
The preferred method for water quality treatment is use
of a retention, or "wet," pond. Such a pond has a perma-
nent pool capable of holding up to 1/2 in. of runoff over
the drainage area. The elevation of the pool is deter-
mined by the low flow orifice of the outlet structure
(Figure 3), from which the first 1/2 in. of runoff flows.
Thus, above this elevation, 24-hour extended detention
is provided for the 1/2 in. of runoff. Another feature
required in the construction of a wet pond is the level
bench. The bench is a 10-ft wide ledge around the
perimeter of the pond, approximately 1 ft below the
design elevation of the permanent pool, on which vege-
tation may be planted or allowed to grow naturally. The
establishment of a thick mat of vegetation offers water
quality improvements through sedimentation, filtration,
and nutrient uptake. In addition, once this marshy area
is established, it may help deter public access to the
permanent pool area. Conservation districts often en-
courage addition of a wet pond as a water quality meas-
ure when soil and ground-water conditions are
appropriate.
Figure 7 shows a wet pond in the southwest corner of
Running Brook Estates and Business Park that was
installed to capture and provide water quality treatment
for a majority of the site's runoff. The pond's irregular
shoreline and its proximity to wetlands (south of the site)
make the pond aesthetically appealing and provide an
160
-------
Analysis
= Postdevelpment Drainage Boundary
Figure 7. Stormwater management ponds at Running Brook Estates and Business Park.
extension of the natural area. Picnic tables were placed
in the area for tenants' use.
More common for new construction projects is the de-
tention, or "dry," pond, which detains runoff during a
storm but then drains completely to a dry state. To meet
regulations, a dry pond must be designed with a low flow
orifice that provides extended detention of the first inch
of runoff for a 24-hour period. While this appears to be
an increase from the 1/2 in. required for wet ponds,
actually the reverse is true. The first flush is generally
accepted to be the first inch of runoff, but because wet
ponds have been shown to provide better sedimentation
and nutrient uptake, a volume credit is given for the use
of a wet pond. This reduces the extended detention
requirements by 50 percent.
Figure 7 shows a pond at the southern edge of Running
Brook Estates and Business Park that provides extended
detention for runoff from a large portion of the residential
development. Discharge is to the wetland areas south
of the site. Based on studies by the Mercer County
Conservation District in New Jersey, the bottom and
sides of this pond need to be planted with a wildflower
mix. This type of vegetation will reduce the necessity of
mowing to once a year, in the fall, greatly reducing
maintenance expenses and increasing visual appeal.
While state law requires a 3:1 side slope ratio for ponds
in residential areas, the conservation districts encour-
age owners and consultants to design milder slopes.
If the use of ponds is not feasible on a site, an infiltration
system should be considered. Infiltration trenches, in
which perforated pipe is placed on a stone bed sur-
rounded by filter fabric, are often preferred for urban
sites, where higher land values make such systems
particularly cost efficient. Infiltration trenches are gener-
ally considered less cost-effective for larger sites.
Another type of infiltration system is the basin. The infiltra-
tion basin depicted in the northeast corner of Running
Brook Estates and Business Park in Figure 7 is used for
the no-positive-outfall situation described above. The infil-
tration method of runoff management is encouraged for
water quality enhancement but is discouraged for water
quantity control due to the high potential for failure.
State law also allows the use of any practice that can
achieve 80-percent removal of suspended solids in storm-
water runoff. One such practice, the use of sand filters,
has been effective in Delaware. Sand filtration can also
be effective for capturing hydrocarbons, which can escape
from ponds. Such systems function much like a septic
system, with a sediment chamber leading to a filtration
chamber (Figure 8); however, the majority of runoff is
stored ahead of the structure in two grassed swales.
Because this design is new, a strict maintenance sched-
ule has been developed that must be followed until
161
-------
Catch Basin Grate
Filtration
Chamber
Sedimentation
Chamber
Inlet
Weir Flow
10 Ft
Figure 8. Septic tank modified for sand filtration.
performance can be verified. The system must be in-
spected every 3 months and any large debris removed.
Once a year, the sedimentation chamber must be evalu-
ated and the polluted top layer of sand removed and
replaced. Every 5 years, the entire volume of sand must
be replaced.
Another acceptable method of infiltration is the use of
vegetated swales, an approach referred to as biofiltra-
tion. Given their linear configuration, vegetated swale
systems may be especially appropriate for space-limited
urban sights where a water quality pond might otherwise
be used.
Runoff from the northwest corner of Running Brook
Estates and Business Park is treated in two biofiltration
swales before it enters the tax ditch that separates the
residential subdivision from the business park. The
swales are located on either side of the forestry lane
leading to the tax-ditch crossing. The forestry lane,
which was installed because fire laws require two ac-
cess points for developments of this size, is demarcated
with a combination of fescue and a wildflower mix, which
the conservation district mandates for the quality and
aesthetic aspects of swales.
Because these swales at Running Brook Estates and
Business Park only receive water quality treatment, a
TR20 analysis was performed on the entire site to as-
sess flows at the analysis point shown in Figure 7. Other
factors were also considered in finalizing review of the
site plan (see Figures 9, 10, and 11).
Site Inspection
Plan review is not the only element of sediment control
and stormwater management delegated to the conser-
vation districts. To keep day-to-day operation of the
program within one agency, the conservation districts
also conduct site visits periodically during construction
and then on an annual basis to perform maintenance
inspections of all completed facilities. A long-term main-
tenance plan for each facility, identifying the responsible
parties, must be established during the plan review
stage.
Conclusion
The most important role the conservation districts have
in site plan review is providing technical assistance to
landowners, designers, and contractors with respect to
sediment control and stormwater management. The dis-
tricts' staff pride themselves on their working relation-
ships and knowledge of the evolving situations in the
state's counties.
162
-------
28
IPO I
20) 30
= Existing contours
Figure 9. Existing contours at Running Brook Estates and Business Park.
= Existing contours
= Proposed contours
Figure 10. Existing and proposed contours at Running Brook Estates and Business Park.
163
-------
Lead
"TTTTTT = Predevelopment Drainage Boundary
Figure 11. Predevelopment drainage boundary at Running Brook Estates and Business Park.
164
-------
The Role of Landscapes in Stormwater Management
Steven I. Apfelbaum
Applied Ecological Services, Inc.,
Brodhead, Wisconsin
Abstract
This paper presents evidence that many existing
streams did not have conspicuous channels and were
not identified during presettlement times (prior to 1830s
in the midwestern United States). Many currently iden-
tified first-, second-, and third-order streams were iden-
tified as vegetated swales, wetlands, wet prairies, and
swamps in the original land survey records of the U.S.
General Land Office.
The data presented show that significant increases in
discharge for low, median, and high flows have occurred
since settlement. Stream channels have formed inad-
vertently or were created to drain land for development
and agricultural land uses. Currently, discharges may be
200 to 400 times greater than historical levels, based on
data from 1886 to the present for the Des Plaines River
in Illinois, a 620-square-mile watershed. Historic data
document how this river had no measurable discharge
or very low flow conditions for over 60 percent of each
year during the period from 1886 to 1904.
This study suggests that land-use changes in the pre-
vious upland/prairie watershed have resulted in a
change from a diffuse and slow overland flow to in-
creased runoff, concentrated flows, and significantly re-
duced lag time. Preliminary modeling suggests the
following results: reduced infiltration, reduced evapora-
tion and evapotranspiration, greatly increased runoff
and hydraulic volatility, and increased sediment yields
and instream water quality problems caused by desta-
bilization of streambanks.
The opportunity to emulate historical stormwater be-
havior by integrating upland landscape features in
urban developments and agricultural lands offers
stormwater management options that are easier to
maintain, less expensive over time, attractive, and
possibly more efficient compared with many conven-
tional stormwater management solutions and the use
of biofiltration wetlands.
Introduction
Diverse and productive prairies, wetlands, savannas,
and other ecological systems occupied hundreds of mil-
lions of acres in presettlement North America. These
ecological systems have been replaced by a vast acre-
age of tilled and developed lands. Land-use changes
have modified the capability of the upland systems and
small depressional wetlands in the uplands to retain
water and assimilate nutrients and other materials that
now flow from the land into aquatic systems, streams,
and wetlands. The historical plant communities that
were dominated by deep-rooted, long-lived, and produc-
tive species have been primarily replaced by annual
species (corn, soybeans, wheat) or shallow rooted non-
native species (bluegrass lawns, brome grass fields).
The native vegetation was efficient at using water and
nutrients, and consequently maintained very high levels
of carbon fixation and primary productivity. Modern com-
munities, in turn, are productive but primarily above-
ground, in contrast to the prairie ecosystem where
perhaps 70 percent of the biomass was actually created
belowground in highly developed root systems. These
changes in the landscape and vegetation coupled with
intentional stormwater management have changed the
lag time forwaterto remain in uplands and consequently
the rate and volume of water leaving the landscape.
The Des Plaines River
Changes that have occurred on the uplands and how
these changes have affected the hydrology of wetlands
and aquatic systems can be illustrated using historical
and more recent data to illustrate trends in discharge of
major river systems. The Des Plaines River was chosen
as a study watershed because of available historical
data and trackable changes in watershed land uses.
The Des Plaines River originates southeast of Burling-
ton in southeastern Wisconsin, flows for over 90 river
miles through agricultural, urban, and suburban land-
scape through northeastern Illinois and the Chicago
165
-------
region, then flows west and south, meeting with another
river and becoming the Illinois River. The historical data
presented are from a case before the Illinois Supreme
Court and a circuit court (U.S. Department of War vs.
Economy Power and Light, 1904) that dealt with the
navigability of the Des Plaines River. The data were
derived from a gauge station installed and operated at
present-day Riverside, Illinois, from 1886 to 1904. The
U.S. Geological Survey has maintained this same sta-
tion since 1943. Historical data from 1886 to 1904 in-
clude a single-stage measurement per day and weekly
discharge measurements (rating curves). For our stud-
ies, duration flow curves were created for the years
1886 to 1904 and 1943 to 1990. The data were com-
pared using median values of discharge (50 percent)
and also using low and high levels of discharge as
indicated by the 75 percent and 10 percent values de-
rived from the annual duration flow curves 1886 to 1904
and 1943 to 1990. The watershed area gauged at Riv-
erside is approximately 620 square miles (400,000
acres).
In the late 1800s, about 40 percent of the watershed had
been tilled and/or was developed. In contrast, approxi-
mately 70 to 80 percent of the watershed is now devel-
oped or under annually tilled agriculture land uses.
Annual duration flow curve values based on linear re-
gression analysis suggested very significant increases
in discharge since 1886; perhaps 250 to 400 times
(Figure 1). In 1886, the median discharge was 4 ft3/sec.
In contrast, in recent years the median discharge has
been 700 to 800 ft3/sec. Trends in low, medium, and high
flow values for the Des Plaines River have undergone
very significant increases.
Preliminary watershed hydrologic modeling suggests
that the watershed and discharge data for 1886 to 1904
had already been modified by development and agricul-
tural land uses; the Des Plaines River watershed was
settled in the late 1830s, and thus 50 years of land use
and development had passed before the 1886 data
were collected. Other data resulting from the litigation
suggested very clearly that the discharge of water from
the Des Plaines River was significantly less between
1886 to 1904 compared with present day discharge.
Because the litigation contested navigability, evidence
was presented using daily stage, discharge, and water
depth data on the opportunity for commercial navigation
on the river. The data suggested that between 1886 and
1904, for an average 92 days per year, the river had no
measurable discharge. An additional 117 days per year,
the river had 60 ft3/sec or less discharge, which was
equal to a depth of less than 3 in. at Riverside. Based
on these statistics, over 60 percent of the year the
400,000 acre watershed yielded no water or such low
flows that navigation was not possible or reliable. An-
other 10 to 25 percent of the year the river was covered
with ice.
CFS 3,000
2,000
1,000
High Flows
High flows have doubled
From 1886-1904, 1.5 high flows/year
Presently, 3 to 5 high flows/year
-1,000
CFS 800
600
400
200
0
-200
0 10 20 30 40 50 60 70
VO
Median Flows
Median flows are 400 times higher
o
in
0 10 20 30 40 50 60 70
VO
Low Flows
Low flows are 250 times higher
-100
0 10 20 30 40 50 60 70
VO
Figure 1. Linear regression analysis and raw data plots of Des
Plaines River discharge at Riverside, Illinois, 1886 to
1988. Low, median, and high flow data were derived
from duration-flow curves for 75, 50, and 10 percen-
tile annual flow levels (1).
Additional supporting evidence of the significance of
changes in the watershed and river is available. The
original land survey records for parts of the Des Plaines
River where section lines were surveyed identified that
reaches of the river had no discernable channels.
Where channels now occur, in the 1830s surveyors
found wet prairies, swamps, and swales but usually no
conspicuous or measurable channel widths. Channels
166
-------
and "pools" were identified in some locations and with
greater frequency downstream in the watershed. The
original land surveyors were under contract by the
U.S. Government Land Office to document the vege-
tation types covering the land and to identify, where
possible, the widths and depths of streams when they
were encountered during the process of laying out the
section lines.
Conclusions and Applications of the
Findings
These data suggest very clearly that highly significant
changes in the hydrology, hydraulics, and water yield
from the Des Plaines River watershed have occurred
since settlement. Other major river and watershed
systems have yielded similar results, suggesting the
transferability of the concepts and general conclu-
sions reached from the studies of the Des Plaines
River. These findings and their applications are dis-
cussed below.
Natural Ecological System Functions and
Processes Should Be Emulated
Water Yield
The historical landscapes "managed" stormwater very
differently than it is managed by present-day strategies.
Historical data clearly indicate that a relatively small
percentage of the precipitation in a watershed actually
resulted in measurable runoff and water leaving the
watershed. In fact, preliminary analysis suggests very
strongly that an average 60 to 70 percent of the precipi-
tation in the watershed did not leave the watershed from
the Des Plaines River; this water was lost through
evaporation and evapotranspiration. Analysis predicts
that approximately 20 to 30 percent infiltrated and may
have contributed indirectly to base flow in the streams
and directly to base flow in wetlands in the watershed.
During a full year, the balance of the water directly
contributed to flow in the "river," where an identifiable
river channel now occurs.
Present-day water management strategies involve col-
lection, concentration, and managed release of water.
These activities are generally performed in developed
parcels in the lower topographic positions. Historically,
a greater percentage of water was lost through evapo-
ration and evapotranspiration from upland systems. In
these situations, microdepressional storage and dis-
persed rather than concentrated storage occurred.
Weaver (1) documented the ability of the foliage of
native perennial grassland vegetation to intercept over
an inch of rain with no runoff generated.
Sediment and Pollutant Management
Because many pollutants in stormwater require water to
dislodge and translocate the suspended solids to which
they are adsorbed, there is a great opportunity to emu-
late historical functions by using upland systems to per-
form biofiltration functions, increase lag time, and
reduce total volume and rate of runoff.
Increased discharge and velocity of water moving
through channels has been documented to greatly affect
instream water quality. Perhaps as much as 70 percent
of instream sediment loads come from channel and
bank destabilization associated with the higher velocity
waters and with solufluction and mass wasting of banks
after flood waters recede (2). Stabilizing (or at least
reducing) hydraulic pulsing in streams can best be ac-
complished by desynchronization and reduction of tribu-
tary stormwater volumes and runoff rates from uplands.
This can be accomplished by integrating substantial
upland perennial vegetated buffers throughout develop-
ments and agricultural land uses. Buffers are designed
not only to convey water and minimize erosion (i.e.,
grassy waterways) but also to attenuate hydraulic puls-
ing, settle solids and adsorbed nutrients, and reduce
and diffuse the velocity, energy, and quantity of water
entering rivers, wetlands, and other lowland habitats.
Using upland microdepressional storage, perhaps in the
form of ephemeral wetland systems and swales in the
uplands, also would emulate the historical landscape
conditions and functions.
Applications
Several example projects of "conservation develop-
ments" are now being completed, which integrate up to
50 to 60 percent of the urban development as open
space planted to perennial native prairie, wet swales,
and other upland communities (as site amenities). Hy-
bernia is a 132-acre residential development in Highland
Park, Illinois, designed and constructed by Red Seal
Development Corporation, Northbrook, Illinois. Empiri-
cal data from Hybernia suggest that the use of upland
vegetation systems in combination with ponded areas
has resulted in the rate and volume of discharge being
essentially unchanged before and after development.
Another project, Prairie Crossing, is a 677-acre residen-
tial project designed to offer comprehensive onsite
stormwater management in uplands and created lake
systems. Extensive upland prairie and wet swale sys-
tems biofilter runoff and enhance the quality and reduce
the quantity of water reaching wetlands and lakes in the
development.
In these types of projects, upland vegetation takes sev-
eral years to fully offer stormwater management bene-
fits. In planted prairies, surface soil structure develops
a three-dimensional aspect in 3 to 5 years. The devel-
opment of this structure seems to have an important role
167
-------
both in offering microdepressional storage and increas-
ing the lag time for retaining water in upland systems.
Restoration and native species plantings also have pro-
vided benefits where ecological system degradation has
led to increased water and sediment yields. Where eco-
logical degradation is occurring indirectly because hu-
man activities on the landscape have reduced or
eliminated major processes (such as natural wildfires),
restoration can provide vegetation and stormwater man-
agement benefits. Wildfires have been all but eliminated
since human settlement has occurred, especially in ar-
eas that contain forests, savanna, or oak woods. In the
absence of fires in many oak woods and savannas, a
dense shading develops caused by increased tree can-
opy and dense shrub development. Where this has
occurred, a reduced ground cover and soil stabilizing
vegetation grows under the low-light conditions. Conse-
quently, highly erodible topsoils containing the seeds,
roots, and tubers of the soil stabilizing vegetation and
higher volumes and rates ofwatercan run off from these
degraded savanna sites. The process of savanna dete-
rioration has been documented; restoration has used
prescribed burning and other strategies (3-5). Reestab-
lishment of ground cover vegetation is key to reducing
runoff, improving water quality, and reestablishing an
infiltration component in degraded, timbered systems.
Should Wetlands Be Used for Sediment
Management, or Should This Occur on the
Uplands?
Because wetlands often provide what little wildlife habi-
tat remains in developed landscapes, and because they
are attractive to wildlife, their use for stormwater man-
agement must be carefully considered. Currently, a na-
tional movement is afoot to use created (and often
natural) wetlands for stormwater management and
biofiltration. Many studies of existing high-quality wet-
lands, however, provide little or no evidence that they
historically served important biological filtration and
sediment management functions. Sediment deposition
was generally episodic (e.g., after wildfires), was of
short duration, and yielded small sediment loads com-
pared with loads from present-day agricultural and de-
veloped lands.
Use of wetlands for biofiltration can actually aggravate
existing problems for many wetland wildlife species. For
example, in the Chicago region it is not unusual to find
100 to 200 parts per million lead (and other contami-
nants) in tadpoles (especially in frog species with a
2-year tadpole stage, such as leopard frogs, bullfrogs,
and green frogs) found in wetlands receiving highway
stormwater. It is imperative to understand the potential
long-term toxic effects on biological systems associated
with stormwater management in wetlands and contami-
nant mobility.
Proposals have been made to allow the materials con-
centrated in biofiltration wetlands to simply be buried by
each additional sediment load or to be intentionally bur-
ied by adding additional soil. Contaminant mobility
through biological pathways still occurs, however, from
beneath considerable sediment burial. In fact, in the
Great Lakes, contamination from PCBs that are often
several feet below the surface of the sediments have
contributed to major increased mortality rates and major
morphological problems in predacious birds such as
cormorants, terns, and gulls (6, 7). The literature on
wetland biofiltration inadequately addresses contami-
nant mobility routes through biological systems and the
potential threat to the viability of biological systems.
Because wetlands are so attractive to biological organ-
isms (and, in fact, the biological organisms are often key
to the successful functions of the biofiltration wetlands),
it is necessary to rethink and carefully design biofiltration
wetland systems in the future.
Far too often, people view the lowland environments
(i.e., rivers, wetlands) as the locations for treating or
physically removing problems created in the upland en-
vironments. The studies briefly described in the previous
section, however, suggest that stormwater, sediment
loads, and the varied contaminants may be best man-
aged on upland systems. Although the land cost for
using upland rather than lowland environments for
stormwater management may be higher, the efficiency
and reduction in potential contaminant problems may be
greater. A landscape with many upland microdepres-
sional storage opportunities and a large buffering capac-
ity might offer more efficient processing than would a
single biofiltration wetland at the downstream end. Each
buffer or depressional wetland would need to treat a
smaller volume of water and contaminants. Also, upland
or dispersed stormwater treatment facilities would have
significantly reduced long-term maintenance costs and
represent a more sustainable approach to management
of stormwater. Centralized biofiltration wetlands, on the
other hand, have high maintenance requirements and
potential problems that include decreases in removal
efficiency for some materials in the short and long term.
There Are No Controlled Year-Round (and
Long-Term) Studies of Removal Efficiencies
Comparing Uplands and Wetlands
The stormwater treatment literature indicates that use of
wetlands and measurements of removal efficiencies
have been based primarily on removal during storm
events passing through the biofiltration wetlands. Year-
round contaminant mass-balance data are largely un-
available. Nongrowing season studies have
documented the export of materials to be significant;
consequently, removal efficiencies for some materials
(e.g., metals, phosphorus) are not likely to be signifi-
cantly reduced from what has been documented for
168
-------
storm event sampling. Wetland efficiencies need to be
experimentally controlled and compared with upland
removal efficiencies, which also have not been studied
in detail (with the exception of removals for several key
elements such as phosphorus). The ability of upland
(soil colloids) systems to provide reliable and long-term
binding and retention for many contaminants has been
demonstrated (8).
Acknowledgments
Funding for a series of ongoing studies summarized in
this paper were provided by the Wisconsin Chapter of
the Nature Conservancy and by Cook and DuPage
County (Illinois) Forest Preserve Districts.
Assistance in these ongoing studies was provided by Dr.
Luna B. Leopold, Dr. James P. Ludwig, Dr. Alan W
Haney, Mr. Robert A. Riggins, and Mr. Brett Larson and
others at Applied Ecological Services, Inc. Mr. and Mrs.
George Ranney and Mr. David Hoffman of Prairie Hold-
ing Corporation and Red Seal Development Corporation,
respectively, allowed their conservation development
projects to be presented as examples here.
References
1. Weaver, J.E. 1968. Prairie plants and their environment, a 50-year
study in the Midwest. Lincoln, NB: University of Nebraska Press.
2. Dunne, T., and L.B. Leopold. 1978. Water in environmental plan-
ning. San Francisco, CA: WH. Freeman and Company.
3. Haney, A., and S. Apfelbaum. 1990. Structure and dynamics of
Midwest oak savannas. In: Sweeney, J.M., ed. Management of
dynamic ecosystems. West Lafayette, IN: The Wildlife Society,
North Central Section, pp. 19-30.
4. Haney, A., and S. Apfelbaum. 1993. Characterization of midwest-
ern oak savannas. In: U.S. EPA. Proceedings of the Workshop
on Oak Savannas, February. Chicago, IL. In press.
5. Apfelbaum, S.I., and A. Haney. 1991. Management of degraded
oak savanna remnants in the upper Midwest: Preliminary results
from 3 years of study. In: Ebinger, J., ed. Proceedings of the Oak
Woods Management Workshop, Peoria, IL. pp. 81-89.
6. Schneider, S., and R. Campbell. 1991. Cause-effect linkages II.
Abstract presented at the Michigan Audubon Society Symposium,
Traverse City, Ml, September 27-28.
7. Gilbertson, M., and R.S. Schneider. 1991. J. Toxicol. Environ.
Health 33(4).
8. Leeper, G.W 1978. Managing the heavy metals on the land. New
York, NY: Marcel Dekker.
169
-------
The U.S. Environmental Protection Agency's
Advanced Identification Process
Sue Elston
U.S. Environmental Protection Agency, Region 5, Chicago, Illinois
Abstract
Advanced Identification (ADID) is a planning process
designed to identify and help protect high-quality wet-
land resources. The ADID process is a joint effort be-
tween the U.S. Environmental Protection Agency (EPA)
and the U.S. Army Corps of Engineers, in which wetland
functions and values are evaluated to determine which
wetlands within an ADID study area are high quality and
should be protected from future fill activities or, in some
cases, which wetlands are of ecologically low value and
could be considered as potential future fill sites. ADID
provides the local community with information on the
value of wetland areas that may be affected by their
activities as well as a preliminary indication of factors
that are likely to be considered during permit review of
a Section 404 permit application.
Final ADID products usually consist of a technical report
that includes the data gathered during the ADID study,
a description of how the wetland evaluation was done,
and a set of maps that identify the sites determined to
be either unsuitable or suitable for filling activities. EPA
works closely with other federal, state, and local agen-
cies as well as the public throughout the ADID process.
Each ADID process is designed a little differently to
meet the specific wetland planning needs of the local
area.
Introduction
In an effort to provide protection to remaining wetlands, the
U.S. Environmental Protection Agency (EPA), in coopera-
tion with the U.S. Army Corps of Engineers (COE) and
other federal, state, and local agencies, may identify wet-
lands and other waters of the United States as generally
unsuitable or suitable for the discharge of dredged or fill
material before receiving a Section 404 permit application.
This Advanced Identification (ADID) process is authorized
by the regulations pertaining to Section 404 of the Clean
Water Act. During the ADID process, EPA, COE, and
other federal, state, and local agencies collect informa-
tion on the values of wetlands and other waters of the
United States to determine which wetlands in the ADID
study area are of high functional value and should be
protected from future fill activities and, in some cases,
which wetlands are of low functional value and could
be considered as potential fill sites.
What Is an ADID?
ADID is an advanced planning process designed to
provide an additional level of protection to wetlands and
other waters of the United States. The ADID process is
one of the few tools currently available to EPA and other
regulatory agencies that can help address resource-
specific issues from a broader perspective. Typically,
Section 404 permitting actions are considered on a
case-by-case basis. ADID provides the opportunity to
evaluate permit requests against wetland resource con-
cerns from a watershed or regional perspective. There-
fore, ADID can be used to address large geographic
issues such as regional wetland loss, to provide the
information needed to better evaluate cumulative loss
impacts, and to provide more detailed ecological infor-
mation than is typically available to regulatory decision-
makers.
A planning tool, ADID is advisory not regulatory in na-
ture. ADID provides landowners and developers with
advance information, allowing them to plan with more
predictability regarding the Section 404 permitting pro-
gram. ADID can provide environmental groups, re-
source agencies, or other groups with information that
can be used to guide protection or restoration efforts.
ADID also can give information on local wetland loss
trends. Most importantly, ADID can provide local com-
munities with information on specific values of local
wetlands that can be used to help develop local ordi-
nances or other planning efforts designed to protect
wetlands with values important to the community.
170
-------
ADID projects vary in size and scope. Study areas range
in size from 100 acres to 4,000 square miles and have
been initiated throughout the country. Nationally, 35
ADID projects have been completed, and 36 are ongo-
ing. The ADID process can be very resource intensive,
depending on the scope of the project. From start to
finish, the time to complete the ADID process can range
from 6 months to several years.
Final ADID products vary from project to project. Typi-
cally, a completed ADID includes a map that identifies
areas that are either unsuitable or suitable for fill, a
database that contains the information used to make the
ADID determination, and a technical summary docu-
ment that explains how the wetland evaluations were
done and what criteria were used to make the unsuit-
able/suitable determinations. Before ADID is completed,
a joint public notice is issued by EPA and COE and a
public meeting is held to solicit public comment on the
products. Public comments are considered before the
final ADID determinations are made. The final maps,
supporting data, and technical summary document are
all available to the public upon request.
In Region 5's experience, ADID is most effective where
there is strong local support for such a project. ADID
projects that involve local agencies can be tailored to
address local needs or problems, such as flood control,
water quality problems, or habitat loss. Participation of
local agencies in the ADID process not only provides
valuable local perspective and expertise but also the
opportunity for ADID determinations to be included in
local comprehensive planning efforts and wetland pro-
tection ordinances.
Lake County ADID
EPA Region 5, in cooperation with COE and several
other federal, state, and local agencies, completed an
ADID project in Lake County, Illinois, in January 1993.
The following is a brief overview of how the ADID proc-
ess worked in Lake County.
Lake County is 460 square miles and is located in
northeastern Illinois. This county has been under inten-
sive development pressure for the last 5 to 10 years.
Lake County also contains a significant proportion of the
wetlands and lakes within Illinois. The majority of wet-
lands within Lake County are isolated or above the
headwaters; therefore, many small wetland fills (less
than 10 acres) were authorized under Nationwide Per-
mit 26. EPA and COE were concerned that, cumula-
tively, these fills could have a significant negative effect
on aquatic resources in Lake County.
Lake County was interested in supporting an ADID study
because local citizens were raising many wetland devel-
opment issues. The county hoped that the ADID process
would provide an additional level of protection for the
high-quality wetlands, as well as an opportunity for the
county to work with federal agencies to resolve local
wetland issues. In addition, the county was beginning to
work on a stormwater and wetland protection ordinance.
The county viewed the ADID process as an opportunity
to work with federal and state agencies to develop an
evaluation methodology for local wetlands that could be
used to guide implementation of the proposed ordi-
nance.
The Lake County ADID process was started in the fall
of 1989. The first meeting included representatives from
federal, state, and local agencies and public interest
groups. The goals of the ADID process were explained,
and the wetland functions and values to be evaluated
were selected based on local needs. A technical advi-
sory committee was formed consisting of repre-
sentatives from EPA, COE, the U.S. Fish and Wildlife
Service, the Soil Conservation Service, the Illinois De-
partment of Conservation, the Lake County Forest Pre-
serve District, the Lake County Department of
Management Services, the Lake County Department of
Planning, the Lake County Soil and Water Conservation
District, the Lake County Stormwater Management
Commission, and the Northeastern Illinois Planning
Commission. The committee's task was to develop the
methodologies to evaluate the selected wetland func-
tions and values. Due to resource constraints, the com-
mittee decided to focus on identifying high-quality
wetland sites only. Sites identified as being of high
functional value would be considered unsuitable for fill-
ing activities.
Lake County, Illinois, contains many lakes and wetlands
and is undergoing rapid urban development. Issues
such as degradation of water quality, flooding problems,
and habitat loss are of local concern. Based on these
concerns, the committee selected the following five wet-
land functions to evaluate for the ADID study:
Biological community value
Stormwater storage value
Shoreline/bank stabilization value
Sediment/toxicant retention value
Nutrient removal/transformation value
In considering evaluation methodologies, the committee
immediately determined that the selected approach
must be capable of dealing with a very large number of
wetlands. The final evaluation methodologies devel-
oped for use in the Lake County ADID process were
combinations of portions of the Wetland Evaluation
Technique (WET) developed for COE (1) and the Min-
nesota Wetland Evaluation Methodology (2) developed
by the St. Paul District of COE. Portions of these meth-
odologies were adapted to meet the needs of the Lake
County ADID process. The evaluation methodologies
171
-------
and the criteria used to determine which wetlands and
streams were of high functional value are described in
detail in the Lake County ADID final report (3).
The wetlands identified as being of high functional value
were considered generally unsuitable for filling activi-
ties. A wetland was determined to be of high functional
value, or unsuitable, if the site included high-quality
biotic communities or if the site provided three of the four
stormwater storage or water quality functions. This ADID
study also identified high-quality stream corridors that
are designated as being unsuitable.
The preliminary Lake County ADID designations were
published in a joint public notice issued by COE and
EPA. Also available for public review and comment were
the evaluation methodologies used, scale maps (1 in. =
1,000 ft) showing the location of all sites of high func-
tional value, and data sheets corresponding to each site
identified as being of high functional value. A public
meeting also was held to gather further public comment.
After considering all the public comments, five sites
were added to the list of areas of high functional value.
Approximately 24,000 acres of wetlands, lakes, and
streams were identified as high functional value sites.
These sites include both public and privately owned
property and represent about 39 percent of the wetlands
and lakes remaining in the county. The Record of Deci-
sion, final public notice, report, and finalized maps were
published in January 1993.
Results and Effectiveness
It is difficult to accurately assess how effective the Lake
County ADID study was in providing an additional level
of protection for wetlands. The ADID maps have been
used by both developers and public entities such as the
Illinois Department of Transportation during site plan-
ning. In addition, COE relies heavily on the information
provided by the ADID study to guide permit decisions for
ADID sites. The county, however, has not yet imple-
mented its wetland protection ordinance. Once the
county's wetland protection ordinance is in place, not
only will the county provide protection for ADID sites but
the ordinance will also require that a buffer area be
maintained around all ADID sites.
While ADID or similar advanced planning processes are
resource intensive, these types of studies can be well
worth the effort if the projects are well designed and the
resulting information is incorporated into local compre-
hensive planning efforts that will guide local land-use
decisions. In addition to focusing on Section 404 issues,
ADID can be tailored to provide information needed for
a variety of other wetland related issues. For example,
ADID can be designed to provide information that assists
in the selection of wetland restoration sites. Advanced
wetland planning studies also can be components of larger
planning efforts (e.g., watershed protection strategies) or
parts of geographic initiatives (e.g., remedial action plans
and lakewide management plans).
Summary
ADID is one of the few tools available to EPA and other
regulatory agencies that can substantially address re-
source-specific issues from a broader ecological per-
spective. ADID can be used in an innovative manner to
address large, geographically based issues. Within an
urban setting, ADID can provide information to commu-
nities regarding the functions and values of local wet-
lands and can guide local protection and restoration
efforts while focusing on local problems or concerns.
References
1. Adamus, P.R., E.J. Clairain, R.D. Smith, and R.E. Young. 1987.
Wetland evaluation technique (WET). Vicksburg, MS: Depart-
ment of the Army, Waterways Experiment Station, Corps of Engi-
neers.
2. U.S. Army Corps of Engineers, St. Paul District. 1988. The Min-
nesota wetland evaluation methodology for the north central
United States.
3. Dreher, D.W, S. Elston, and C. Schaal. 1992. Advanced Identifi-
cation (ADID) study, Lake County, Illinois. Final report (Novem-
ber).
172
-------
Wisconsin Smart Program: Starkweather Creek
William P. Fitzpatrick
Wisconsin Department of Natural Resources, Madison, Wisconsin
Abstract
Starkweather Creek drains a 23-square-mile urban wa-
tershed in the city of Madison, Wisconsin. Urban runoff
had resulted in elevated levels of biochemical oxygen
demand, mercury, lead, zinc, cadmium, and oil and
grease in the sediments and a severely degraded fish
and macroinvertebrate habitat. Historically, the creek
had received significant amounts of stormwater and
industrial waste discharges. Industrial activities in the
watershed had included metal fabrication, battery
manufacturing, meat packing, and food processing.
Starkweather is the second largest tributary and the
largest source of mercury to Lake Monona, a principal
recreation lake for the Madison area. Downstream
transport of sediments and associated pollutants from
the Starkweather watershed effects the quality of this
important lake, which is under a fish advisory to anglers
to restrict consumption of larger walleyes due to ele-
vated mercury levels.
To address contamination in the creek and Lake
Monona and to implement the recommendation of the
local priority watershed plan, Wisconsin's Sediment
Management and Remediation Techniques program se-
lected Starkweather as a sediment remediation demon-
stration project. A joint U.S. Environmental Protection
Agency, Wisconsin Department of Natural Resources,
county, and city project was developed to 1) reduce
nonpoint loading, 2) control the impacts of in-place con-
taminants, and 3) restore the recreational value and
aquatic habitat of the creek. This $1 million program
included the dredging of 17,000 yd3 of contaminated
sediments, construction of stormwater detention ponds,
development of streambank erosion controls, and
aquatic habitat restoration.
Introduction
Starkweather Creek, located on the northeast side of
Madison, Wisconsin, is the city's largest urban water-
shed, draining 23 square miles (Figure 1). The creek
discharges to Lake Monona, a principal recreation lake
located in the city of Madison. The creek and its water-
shed have been extensively altered as a result of urbani-
zation. Extensive ditching, channelization, wetland
draining and filling, and impervious structure develop-
ment have shaped the hydrology and water quality of
the creek.
Starkweather Creek has been affected by both point and
nonpoint pollution overtime. The creek drains a heavily
industrialized portion of the city where metal fabrication,
battery manufacturing, meat packing, and food processing
occurred. Urban nonpoint runoff is believed to have con-
tributed significant levels of pollutants in recent years.
Recent monitoring indicated that the creek had elevated
levels of sediment oxygen demand, biochemical oxygen
demand (BOD), mercury, lead, zinc, cadmium, and oil
and grease in the sediments and a severely degraded
fish and macroinvertebrate habitat. Concern for the lev-
els of contaminants in the sediments of the creek ex-
tended beyond the stream channel and its habitat and
also encompassed the downstream impacts of the sedi-
ments on Lake Monona.
Lake Monona has a mercury advisory on large walleye
due to excessive levels of the metal in the tissues of this
fish. Starkweather Creek, identified as the largest
source of mercury to the lake, was targeted for remedia-
tion to restore the aquatic habitat of the creek and to
protect Lake Monona.
Wisconsin Sediment Management and
Remediation Techniques Program
In response to the growing awareness of natural re-
sources managers of the continuing impacts of in-place
pollutants associated with sediment deposits in the
state's waterways, the Wisconsin Department of Natural
Resources (DNR) established an interdisciplinary team
to develop necessary assessment and remediation
tools to restore affected waters of the state. The Wis-
consin Sediment Management and Remediation Tech-
niques (SMART) Program has brought together
173
-------
East Branch
Starkweather Cr.
2 3
Scale in Miles
Figure 1. Location map of Starkweather Creek and the restoration project area.
expertise in environmental toxicology, aquatic habitat
assessment, hydrographic surveying, sediment map-
ping, sediment engineering, and remedial technology.
The SMART Program has two basic responsibilities: 1)
define the extent of sediment contamination and im-
pacts on the waters of the state and 2) guide the reme-
diation of contaminated sediments.
The SMART Program coordinates the state's contami-
nated sediment activities with various universities and
federal programs, such as the U.S. Environmental Pro-
tection Agency's Superfund and Great Lakes National
Program Office Assessment and Remediation of Con-
taminated Sediment (ARCS) programs.
Monitoring Data
Starkweather Creek, the first sediment cleanup demon-
stration of the Wisconsin SMART Program, provided an
opportunity to use advance monitoring of the many com-
ponents of an aquatic system affected by contamination
in sediments. Several assessment techniques were used
to define the degree of sediment contamination, stream
water quality, and aquatic habitat (Table 1). Later sec-
tions of this paper address monitoring performed during
dredging to assess on- and offsite impacts of the cleanup.
Postremediation monitoring will continue for 2 years to
document the changes and response of the creek.
Remediation Planning
Starkweather Creek was selected as the first sediment
remediation demonstration for the SMART Program
based on recommendations from the state's DNR man-
agement districts, on the relative small scale of the site,
and on ranking of the site with the SMART selection
criteria. This criteria included:
Impaired uses of the water body
Adequate data for feasibility analysis
Upstream pollution source controls
Local support
174
-------
Adequate access
Integration with other state and local programs
The specific project goals and objectives were developed
by a project implementation team assembled from repre-
sentatives of relevant state and local agencies and bu-
reaus who guided the development of the project work
plan, schedule, and budget. Individual members were
responsible for ensuring that their program's relevant
Table 1. A Summary of Starkweather Creek Preremediation
Monitoring Data
Range
Average
Total
Weight
Sediment Chemistry
Mercury <0.1-3.5 1.1 ppm 40 Ib
Lead 33-320 130 ppm 2.4 tons
Chromium 9-31 19 ppm 0.35 tons
Oil and grease 1,500-3,600 2,800 ppm 51 tons
PCBs <0.14ppm <0.14ppm
Bulk density 65-106 80 Ib/ft3 18,400 tons
Water Column
Mercury (total) 1.69-1.70 ng/L
Mercury (methyl) 0.033-0.050 ng/L
Lead <3-10 |ig/L
Chromium <3-18|ig/L
Phosphorus-P 0.03-0.37 mg/L
DO
COD
Ammonia-N
Fish Tissue
3.3-14.6 mg/L
(37.5-120%
saturation)
10-38 mg/L
0.04-1.8 mg/L
Freshwater drum 0.16-0.48 ppm
(three samples, mercury
10-19 in.)
Carp (three 0.09-0.11 ppm
samples, 18-26 in.) mercury
Caged Fish Bioaccumulation
Minnows, 2-wk 0.012-0.018 ppm
exposure mercury
Minnows, 4-wk 0.012-0.016 ppm
exposure mercury
Toxicity Characteristic Leaching Procedure (TCLP)
Sediment leaching <1 mg/L lead
test (three samples)
Sediment Mapping
Surveyed cross sections at 100-ft intervals
17,000 yd3 of soft sediment measured
regulations were followed and the work plan was con-
sistent with program policies and goals.
Following the development of the initial work plan, public
informational meetings were held to solicit comments
and suggestions. Presentations were also given to
neighborhood associations and local environmental
groups. Fact sheets outlining the proposed scope of
work were distributed at these meetings. These meet-
ings provided the implementation team with feedback
on the scope and schedule of the work plan and a sense
of the public's priorities regarding the restoration. Most
of the public responses were requests for further clarifi-
cation of the monitoring data, the permitting process,
environmental safeguards during remediation, and po-
tential exposure of local residents to contaminants in the
sediments. One of the most frequent concerns for local
residents was the removal of trees along the creek. The
comments provided by the public and interested organi-
zations were, where practical, incorporated into the work
plan. For example, the replanting and vegetative resto-
ration aspects of the project were developed in greater
detail and the scope of the replanting was increased to
address the concerns expressed at the public meetings.
Press releases and direct mailing to interested citizens
and residents were used to keep the public involved and
informed on the progress of the project.
Work Plan
The Starkweather Implementation Team developed the
remediation work plan to achieve the goals of reducing
pollutant loading to Lake Monona, restoring the aquatic
habitat and fishery, and improving recreational use and
access to the creek. The work plan included the follow-
ing tasks to achieve these goals:
Dredge 17,000 yd3 of contaminated sediments.
Improve the habit for fish and aquatic life through
riprapping.
Regrade and stabilize the eroding creek banks.
Establish shoreline buffer zones.
Use vegetative management to improve terrestrial
habitat.
Create public access paths and fishing platforms.
Enhance public awareness and stewardship.
Dredging was selected as the means to remove the
contaminated sediments, eliminate downstream loading
of these contaminants, and restore the depth and diver-
sity of the aquatic habitat. Survey cross sections of the
creek were established at 100-ft intervals through the
project site and were measured for water depth and
sediment thickness. These data were used to model the
volume and mass of contaminated sediments to be
175
-------
removed from the channel. In addition to removing con-
taminants from the creek, the enlarged cross-sectional
area of the channel would maintain a greater depth of
water capable of holding more dissolved oxygen and
would provide more cover and structure for aquatic life.
The dredging of the creek channel increased the aver-
age depth from 1.5 to 4 ft. The average maximum depth
of the channel thalweg was increased from 2 to 7 ft.
Figure 2 is a typical cross section of the creek showing
the pre- and postproject channel geometry and changes
in water depth and streambanks.
Hydraulic studies of the creek channel and Lake
Monona were performed to assess the local and re-
gional impacts of dredging Starkweather Creek. This
work was performed to assess issues related to
changes in water surface elevations, channel stability,
base level lowering, and potential upstream bed ero-
sion. Starkweather Creek throughout the project area is
in the backwater of Lake Monona. The water surface
elevation of the creek is the same as the downstream
lake. Therefore, the deepening of the creek by dredging
would not decrease the water surface elevation or pro-
mote upstream bed or bank erosion.
Riprapping was selected for shoreline protection to pro-
tect the bank soils from waves and currents and to provide
structure for fish and aquatic life. Sheet pile was used
in selected areas where the steepness of the shoreline
required vertical protection and regrading was not fea-
sible (e.g., near buildings, roadways, and bridges). Vertical
shore protection (sheet pile) was avoided in most areas
because it presents a less than natural appearance and
forms a barrier to aquatic life migration from water to land.
The banks of Starkweather Creek exhibited significant
undercutting and failure and were a significant source
of sediment to the creek. The failure of the creek banks
undermined shoreline trees and vegetation and pro-
duced a perpetuating process of landward erosion of
Figure 2. Starkweather Creek example cross section showing
the channel profile before and after dredging.
increasingly steep banks. Eventually, the creek would
have reached a hydraulic equilibrium by reshaping the
channel geometry to a much wider and shallower channel.
This process would have eliminated the fishery and small
boating uses, however, and would have undermined
local structures such as roadways, bridges, and buildings.
The banks of the creek were stabilized by regrading the
abovewater slopes from vertical to 3:1 (horizontal:verti-
cal), covering with protective riprap, and finally topping
with a 6-in. seed bed planted to native grasses, shrubs,
and trees. The near shore areas of the creek banks were
planted to provide a vegetative buffer zone to filter pol-
lutants carried by overland flows to the creek.
The terrestrial habitat along Starkweather Creek, although
degraded, did provide important food and cover to insects,
birds, and animals. Principal goals of the remediation
project were to carefully manage all construction activi-
ties to minimize disturbances to the existing vegetation,
to restore quality terrestrial areas disturbed by the creek
restoration construction activities, and to improve the
habitat where possible. A vegetation management and
restoration plan was developed by the city's landscape
architects to identify existing important tree and shrub
specimens along the creek that were to be protected
during construction work. The management and resto-
ration plan was integrated with the construction plans,
and close cooperation between the landscape archi-
tects, contractors, the DNR, and city engineering staff
was used to resolve conflicting needs for access and
mobility of the heavy equipment and the need to pre-
serve desirable species. Trees and shrubs were initially
either classified for saving or removal before construc-
tion. To reduce disturbance to the site and the costs of
revegetation, the landscape architects and construction
supervisors performed a final walking tour of the site to
identify additional trees and shrubs, initially classified for
removal, that could be saved if practical. This process
provided the supervising field engineer with the discre-
tion to either modify the construction plans and activities
in the field to try to preserve existing vegetation or to
permit the construction contractors to remove the speci-
mens to facilitate access and work activities.
The project area was scheduled for replanting in the
early spring of 1993. In addition to native and park
grasses, 1,400 trees and shrubs were to be planted,
including white ash, basswood, oak species, and ma-
ples. Planting would be located and spaced to provide
optimal habitat areas along the shore of varying species,
heights, and distribution.
Public access was provided to allow pedestrians to
walk the site without disturbing the wildlife areas or
trampling the banks of the creek. Landscape architects
designed walkways to connect the project site with
existing city parks and natural areas. Access to the creek
was provided by low-lying shore areas and fishing/canoe
176
-------
access platforms constructed into higher creek banks
near the water line.
Public awareness and stewardship was encouraged
from the start to involve local groups throughout the
project from early project design through final restora-
tion. Regular press releases, media interviews, talks to
neighborhood groups, direct mailings on project activi-
ties, aquatic education tours, fishing-for-kids clinic, and
a volunteer planting event were used to keep people
involved in and informed about the restoration.
Permits and Regulatory Review
The environmental restoration of Starkweather Creek
included construction activities that were under the ad-
ministrative and regulatory jurisdiction of several pro-
grams and agencies. Guidance from members of the
implementation team representing the state's Water
Regulation and Zoning, Solid Waste, and Environmental
Assessment Bureaus were incorporated in the develop-
ment of the project work plan and construction plans.
City personnel guided the planning for compliance with
local ordinances and coordination with local utilities.
Permits were necessary for dredging and shoreline ex-
cavation and filling. In addition, regulatory review and
approval was requested for the management of sedi-
ments dredged from the creek. Related regulations re-
quiring compliance were historical and archeological
site assessment, floodplain zoning regulations, and
state environmental assessment guidelines. The city of
Madison was the applicant for the construction work.
Because many portions of the creek shoreline in the
work area are privately owned, the permit required that
either all riparian landowners individually apply for per-
mits or that they assign the city to act as their agent for
the permit application. A form letter was sent to the
riparian landowners requesting their approval forthe city
to apply for the permit in their behalf. All riparian land-
owners in the project area approved, and copies of the
signoff letters were then submitted to the U.S. Army
Corps of Engineers and DNR.
Construction
Following completion of the construction plans, sealed
bids were requested from qualified, interested contrac-
tors. The lowest of five bids was accepted. Speedway
Sand and Gravel, Inc., of Madison, Wisconsin, was
awarded the contract with a bid 17 percent lower than
the highest bid.
Retention Site
The sediment retention and dewatering facility, 6 miles
southeast of the project area, was built in January 1992.
The site covered 2.8 acres and was built on county-
owned land at the local municipal landfill. The sediment
retention facility was designed to dewaterthe sediments
and contain the sediment and carriage water. The facility
is square in plan view with 7-ft berms built of local clay
soils. The bottom was unlined but consisted of several
feet of clay. Local monitoring wells provide data on
potential leachate from the facility. A concrete drop inlet
spillway was built into the facility to allow excess water
to be pumped to a sanitary sewer if necessary.
The retention site was built to contain 17,000 yd of
sediment with a 25-percent bulking factor and to provide
a minimum of 1.5 ft of freeboard to contain direct pre-
cipitation and provide a margin of safety.
Dewatered sediments from the facility are available for
use as cover on the landfill.
Site Preparation
A double silt curtain of geotextile fabric was placed
across the creek at the downstream end of the project
in mid-November 1992. The silt curtains were intended
to trap debris in the streamflow generated by construc-
tion activities. In addition, the porous fabric was in-
tended to trap sediments resuspended by the dredging.
The curtains were held in place at the top by a half-inch
steel cable tied to trees on the bank and weighted at the
bottom by a heavy logging chain.
Utility representatives identified and marked all pipe-
lines, cables, and utility facilities along the creek in the
project area.
Site clearing and grading for heavy equipment access
followed the installation of the silt curtains. Access roads
and trees to be left undisturbed were clearly identified
to minimize site disturbances and the cost of restoring
vegetation.
Dredging
Dredging began on the upstream end of the west branch
of Starkweather Creek on November 19, 1992. Dredg-
ing was performed with a backhoe. Construction activi-
ties were staged through the project area such that
approximately 100 yd of streambed was dredged, the
banks were shaped to a stable slope, and then the site
was riprapped. The goal of this sequence was to mini-
mize the size of the project area opened by construction.
In addition, because the project is in a residential neigh-
borhood, keeping the principal work confined to a limited
area at one time minimized noise and dust in the area.
Dredging, bank shaping, and stabilization proceeded in
a downstream direction on the west branch to the con-
fluence with the east branch. When the west branch was
finished, work moved to the upstream end of the east
branch. Approximately 12 dump trucks were used to
haul the dredged sediments to the retention facility.
Trucks were loaded on average every 5 minutes. To
prevent leakage from the trucks, the tailgates were fitted
177
-------
with neoprene seals, and chain binders were used to
provide a backup to the tailgate lock. No sediment spills
occurred during hauling. Dredging was completed on
January 27, 1993. Bank shaping and stabilization work
finished 2 weeks later.
Nearly 14,000 tons of riprap and 3,400 tons of crushed
stone were used on the project. Bank shaping involved
3,200yd3 of soil.
Dredge Monitoring
Monitoring during dredging and other construction work
was performed to track the impact of these activities on
the creek and Lake Monona. Visual observations were
made daily of the degree of turbidity changes caused by
construction. Best management practices related to the
work on site were used to minimize the instream and
offsite impacts. Water sampling for chemical analyses
was performed on a weekly basis at upstream reference
sites, downstream of the dredging, and above and be-
low the silt curtains. Creek water samples were ana-
lyzed for metals (arsenic, cadmium, calcium, copper,
chromium, iron, lead, magnesium, nickel, zinc), nutrients
(ammonia, nitrate and nitrite, total Kjeldahl nitrogen,
total phosphorus), and general water quality parameters
(suspended solids, chemical oxygen demand, BOD, con-
ductivity, pH, alkalinity, hardness, temperature, dis-
solved oxygen).
Monitoring results indicate that there was no significant
difference between the water quality parameters at the
upstream reference sites and at the downstream end of
the project on the dates of sampling. Figure 3 is a plot
of selected water quality parameters measured on De-
cember 3, 1992, during the dredging activities. On this
date, dredging was performed approximately 300 yd
downstream of the upstream reference sampling site on
the west branch. Sampling was also performed at the
first bridge downstream of the dredging site. Other data
shown in Figure 3 were obtained on the same date at a
reference site on the east branch above the project and
at two locations on the downstream end at the silt cur-
tains. In can be seen in this figure that data from the
dredging site show significantly higher values than at
other sampling sites. The concentrations from the down-
stream end of the project (at the silt curtains), however,
are equivalent to the undisturbed reference sites for
most parameters, indicating that the resuspension of
sediment and pollutants from the dredging had minimum
offsite impacts. Lead and zinc values did exhibit an
increase at the downstream site samples (Figure 3)
compared with the upstream reference sites; however,
the values at the downstream sites were within the
1,000 _
0)
I 100
c
ro
B Upstream Reference West Br.
[] Upstream Reference East Br.
B 100 Yd Downstream of Dredge
0 Above Curtain at Downstream End of Project
Below Curtain at Downstream End of Project
~
-------
range of values measured overtime at the undisturbed
reference sites. Lead and zinc concentrations in water
at the downstream end of the project were well below
State Water Quality Criteria NR105 for acute and
chronic toxicity in all water samples.
The silt curtains had little effect on the water quality of
the streamnearly all parameters were at the same
levels above and below the curtains. Sediments and
associated contaminants resuspended by the dredging
work settled fairly quickly in the creek channel, and
downstream loading to Lake Monona remained at back-
ground levels during the construction work. This project
deployed the silt curtains normal to the streamflow (i.e.,
across the width of the channel) in an attempt to trap
debris generated by the construction activity and to
control resuspended sediments. The curtains were ef-
fective in trapping floating debris; however, they were
not always effective in filtering solids from the stream-
flow. Figure 3 shows a slight drop in solids concentration
across the silt curtain; however, the difference in con-
centration is fairly low and was not seen in most water
sampling days. Field observations of the performance
of the curtains showed that during all but the lowest
base flow, the curtains would "billow out" to the down-
stream, allowing the streamflow to pass beneath the
curtains.
Postremediation Monitoring
Routine water quality sampling will continue on a
monthly basis for a least a year following the completion
of construction work. Additional monitoring intended to
document the restored conditions of the creek include
fish shocking surveys, caged fish bioaccumulation, sedi-
ment bioassay, sediment chemistry, qualitative habitat
assessment, and macroinvertebrate sampling (sedi-
ment and artificial substrate). These additional activities
will be performed over the next 2 years to assess the
success or failure of the restoration work, help to refine
further work at other aquatic restoration projects, and
guide the development of standard procedures for sedi-
ment assessment work.
Summary and Conclusion
Contaminated sediments can be managed to restore
lost beneficial uses of a degraded waterway. The envi-
ronmental restoration of Starkweather Creek has dem-
onstrated that the knowledge and skills of various
environmental programs can be successfully coordi-
nated to accurately assess the degree of contamination,
identify necessary sediment removal and disposal tech-
niques, develop and implement a cross-program work
plan, and carefully monitor the site disturbance and final
restoration.
Some important aspects of this project that were critical
to its successful implementation were cross-program
coordination and communication, public communica-
tions and feedback, construction field supervision, and
a significant investment in environmental monitoring to
guide the development of the work plan and document
the results of the restoration.
179
-------
Wolf Lake Erosion Prevention
Roger D. Nanney
Soil Conservation Service, Crown Point, Indiana
Abstract
The U.S. Environmental Protection Agency (EPA), Re-
gion 5, in cooperation with the Lake County, Indiana,
Soil and Water Conservation District, the City of Ham-
mond, Board of Park Commissioners, and the U.S. De-
partment of Agriculture, Soil Conservation Service,
prevented bank erosion on over 300 m of the east shore
of Wolf Lake. This project was funded through a $70,000
grant from EPA under Section 319(h) of the Clean Water
Act. EPA had identified Wolf Lake as part of the Internal
Joint Commission's Great Lakes Area of Concern, along
with the Grand Calumet River Basin in northwest Indi-
ana. Various sources of sediment were contaminating
the lake, but the Park Board determined that the shore-
line erosion was the highest priority. The bank was also
one of the few remaining habitats of silverweed (Poten-
tilla anserina), a plant on the Indiana endangered spe-
cies list. A member of the rose family (Rosaceae),
silverweed grows on wet, sandy shores in Canada south
to Iowa, the Great Lakes, and coastal New England.
When the Indiana Department of Natural Resources
identified the plant at the site, the project was in jeop-
ardy until a compromise was reached. Limestone riprap
was chosen as the nonpoint source pollution/best man-
agement practice material to stabilize the 0.3- to 1.0-m
bank. Wave action induced by wind was the cause of
the bank erosion problem. Average fetch exposure,
shore geometry, and shore orientation proved to be the
significant factors in designing a successful shoreline
protection system.
Introduction
The southern shoreline of Lake Michigan, in northwest-
ern Indiana, is one of the major urban and industrial
centers in the Great Lakes region and includes the cities
of East Chicago, Gary, Hammond, and Whiting in Lake
County, Indiana (1). The heavy industry in this area
contains approximately 40 percent of the steel making
capacity of the United States, and one of the largest
petrochemical complexes in this country. This combina-
tion has created one of the most environmentally de-
graded areas within the entire Great Lakes basin.
Wolf Lake is located in the northwest corner of the
region and is an important remnant of what once was a
large Lake Michigan bay. As the Great Lakes' levels
dropped from the Nipissing through the Algona to the
present-day Lake Michigan, several coastal area lakes
developed (2). Among these lakes were Calumet, Hyde,
Wolf, Berry, and George. Today, only Calumet, Wolf, and
small remnants of Lake George remain; the others were
drained and filled to allow for development (3).
The present surface area of the lake is 156 ha in Indiana
and 170 ha in Illinois. As would be expected because it
was once a shallow bay, Wolf Lake is shallow, with a
mean depth of only 1.5 m. The maximum depth is listed
as 5.5 m in areas influenced by past sand mining (1).
Wolf Lake is not protected by natural features such as
hills or stands of trees. Therefore, strong winds fre-
quently cause wave action to pound the eastern shore-
line and create erosion and sediment.
Shoreline Erosion and Protection
Few things are a bigger eyesore and problem for lake-
shore users than an eroding shoreline. A variety of lake
shoreline protection practices are designed to stabilize
and protect these areas against the forces of erosion,
such as scour and erosion from wave action, ice action,
seepage, and runoff from upland areas. These practices
are both nonstructural (vegetation or beach sloping) and
structural (flexible structures such as riprap and rigid
structures such as seawalls).
Shoreline erosion is a significant problem in several
areas along Wolf Lake's shoreline. The problem has
been documented by historical photographs and per-
sonal accounts, but estimating the volume of shoreline
eroded is difficult. Photographs indicate that the eastern
shore has receded 15 m. Photographs from 1938, when
compared with recent photographs, show that the area
has receded at a rate of about 0.3 m/yr.
180
-------
The lake's shallow water depth, long wind fetch, and
motor boat use all contribute to the waves eroding the
shoreline. The scarcity of rooted littoral vegetation and
the sand, slag, and gravel texture of the scoured littoral
sediment are further evidence of wave action. Fetch is
defined as the distance a wind blows unobstructed over
water, especially as a factor affecting the buildup of
waves. The average fetch exposure, shore geometry,
and shore orientation are significant factors in success-
ful shoreline stabilization (4).
Vegetation effectively controls runoff erosion on slopes
or banks leading down to the water's edge; however,
vegetation is ineffective against direct wave action or
seepage-caused bank slumping (5). Diverse, moder-
ately dense stands of aquatic plants are desirable in a
lake's littoral zone. Emergent aquatic plant communities
protect the shoreline from erosion by damping the force
of waves and stabilizing shoreline soils (6).
Riprap armoring is a flexible structure constructed of
stone and gravel that is designed to protect steep shore-
lines from wave action, ice action, and slumping due to
seepage. The riprap is flexible in that it will move slightly
under certain conditions. This improves its ability to
dissipate wave energy.
Seawalls, bulkheads, and retaining walls are rigid struc-
tures used where steep banks prohibit the sloping forms
of protection. Seawalls do not primarily dissipate wave
energy but rather redirect the wave energy away from
the shore (7).
Site Evaluation
The Hammond Park Board had been in contact with the
U.S. Environmental Protection Agency's (EPA's) Region
5 office in Chicago, Illinois, about an ongoing erosion
problem at Wolf Lake in Hammond, Indiana. The site
was actively eroding and endangering the east shoreline
for 300 m. This was part of the Internal Joint Commis-
sion's Area of Concern and was identified in the area
Remedial Action Plan (RAP) by the Indiana Department
of Environmental Management (IDEM). The Park Board
called on EPAfortechnical and financial assistance, and
project development began.
In the fall of 1990, the eastern shoreline of Wolf Lake
was surveyed by the Soil Conservation Service (SCS).
The survey revealed a water depth ranging from 0.3 to
1.0 m, with a vertical dropoff. This area had been erod-
ing for an undetermined amount of time and had
reached a point where it would soon undercut a pedes-
trian trail connecting a picnic area with the beach. Over
the years, the Park Board had allowed large pieces of
broken concrete to be dumped along the shoreline to try
to control the erosion. This had slowed the erosion
process in some areas but accelerated it in others.
Where the wave action could get between the concrete,
the erosion continued to advance.
The undercutting of a fishing pier at the south end of the
area demonstrated the strength of the wave action on
the site. Although the average fetch at the site is about
1,000 m, the wave energy is funneled to the northeast
and southeast shoreline by a manmade island located
200 m offshore. The maximum depth of the bay area
created by this erosion is only 3 m, with the majority at
no more than 1.5 m.
SCS recommended that the 300-m shoreline be stabi-
lized with riprap. In the winter of 1990, the Lake County
Soil and Water Conservation District applied to EPA for
a Section 319 grant of $70,000 to stabilize the shoreline.
SCS completed the designs, and the Park Board sought
permit applications from IDEM, the Indiana Department
of Natural Resources (IDNR), and the Army Corps of
Engineers (COE). Several coordination meetings were
held with the Park Board to keep them informed of the
progress of the various activities. The Park Board ap-
proved the final plans in the spring of 1991, and permits
were approved that summer.
During the permit review process, an IDNR biologist iden-
tified the presence of silverweed (Potentilla anserina) at
the site. Silverweed, which is on the IDNR endangered
species list, was growing in patches along the eastern
shoreline. Silverweed is a prostrate species that sends up
yellow flowers with leaves on a separate stalk. The leaves
are strikingly silver beneath, divided into 7 to 25 paired,
sharp-toothed leaflets that increase in size upward. The
total plant length ranges from 0.3 to 1.0 m, and it flowers
in June through August (8). This plant was also in danger
of losing its habitat as the shoreline eroded back. The
IDNR approved of the riprap project with the stipulation
that care be taken to avoid main clusters of the plant.
Riprap Size and Placement
A stone revetment, riprap involves more than simply
dumping rocks on the shoreline. The SCS area-office
engineer developed a design, which was reviewed by
the SCS state engineer. This design included the inves-
tigation of the average depth of the bay water, wave height,
depth of dropoff, and the orientation of critical winds.
The largest wave that can reach shore is 0.8 times the
depth of the water (9). This would generate a wave
height of 1.2 m where the water depth is 1.5 m. A
maximum wave height of 0.5 m would be reached for a
1,000-m fetch over 6-m deep water with a 16 m/sec wind
speed (9). Therefore, NAS No. R-5 (46 cm maximum,
D50 23 cm, minimum 13 cm) graded riprap was chosen
for the armor stone (9). For the bedding or filter stone,
NAS No. FS-2 (5 cm maximum, average No. 4, No. 100
minimum) would be used.
181
-------
With the existing concrete in place, it was difficult to
determine the amount of riprap needed. An estimate
was made based on an average riprap thickness of 0.6
m and enough bedding stone to fill in the voids on a
typical cross section 300 m long. The plans called for a
2:1 slope for the finished riprap, which meant 800 metric
tons of bedding stone and 615 metric tons of riprap was
needed.
The Park Board received bids for the work and awarded
the contract in the late summer of 1992. The cost for
actual purchase and placement of material was $133.00
per linear meter. Additional costs associated with the
project were for design, administration, and construction
supervision. The construction of the 300-m barrier took
7 working days, including hauling the stone from a
quarry within 16 km. Stone was placed using a large
hydraulic backhoe and a front-end loader.
Chapters 16 and 17 of the SCS National Engineering
Field Manual (10) contain detailed discussions on the
selection and placement of riprap for erosion control.
Discussion and Conclusion
The nonpoint source/best management practice (BMP)
of limestone riprap was selected for the Wolf Lake pro-
ject. Selection was based on the need for the practice
to withstand wave energy, be cost effective, and be
compatible with the endangered species plant found at
the site. Revegetation was not selected as the BMP
because the site was unstable and few plants could
stand up to the wave action. The erosive force of wave
action limits plants survival in open lakes. Aquatic
macrophytes may not grow in areas where wind fetch
exceeds 850 m (11). A seawall or other rigid BMPswere
not selected because of their higher cost and the distur-
bance to the site that would be required for their instal-
lation. Another alternative not discussed here, because
of the major site disturbance it would require, is regrad-
ing of the bank to a stable slope.
The design characteristics of the site taken into consid-
eration were fetch exposure, shore geometry, and shore
orientation. In addition, the resistance of dumped stone
to displacement by waves depends on:
Weight, size, shape, and composition of the stone.
Gradation of the stone.
Height of the wave.
Steepness and stability of the protected area.
Stability and effectiveness of the filter or bedding
material (12).
References
1. Watson, L, R. Shedlock, K. Banaszak, L. Arihood, and P. Doss.
1989. Preliminary analysis of the shallow ground-water system
in the vicinity of the Grand Calumet River/Indiana Harbor Canal,
northwestern Indiana. U.S. Geological Survey Open-File Report
88-492. Indianapolis, IN.
2. Bell, J., and R. Johnson. 1990. Environmental site assessment
of Wolf Lake. TAP Report No. TAP901126. Hammond, IN: Ham-
mond Chamber of Commerce.
3. Holowath, M., M. Reshkin, M. Mukluk, and R. Tolpa. 1990. Work-
ing towards a remedial action plan for the Grand Calumet River
and Indiana Harbor Ship Canal. Unpublished paper. U.S. EPA,
Region 5, Chicago, IL; and Indiana University Northwest, Gary,
IN.
4. Berc, J., and S. Ailstook. 1989. Shoreline stabilization on Navy
property. J. Soil Water Conserv. 44(6):560-561.
5. McComas, S. 1986. Shoreline protection. Lake Reservoir Mgmt.
2:421-425.
6. Nichols, S.A. 1986. Innovative approaches for macrophyte man-
agement. Lake Reservoir Mgmt. 2:245-251.
7. Jones, W.W., and J. Marnatti. 1990. Cedar Lake enhancement
study. Bloomington, IN: Indiana University.
8. Peterson, R. 1968. Yellow flowers: A field guide to wildflowers,
Northeastern and Northcentral North America. Boston, MA:
Houghton Mifflin Company.
9. National Crushed Stone Association. 1978. Quarried stone for
erosion and sediment control.
10. Soil Conservation Service. 1989. National engineering field man-
ual. Washington, DC: U.S. Department of Agriculture.
11. Harvey, R.M., J.R. Pickett, and R.D. Bates. 1987. Environmental
factors controlling the growth and distribution of submerged
aquatic macrophytes in two South Carolina reservoirs. Lake Res-
ervoir Mgmt. 3:243-255.
12. Searcy, J.K. 1970. Use of riprap for bank protection. Hydraulic
Engineering Circular No. 11. Washington, DC: U.S. Department
of Transportation. Available as a reprint from U.S. GPO, Wash-
ington, DC.
182
-------
Incorporating Ecological Concepts and Biological Criteria in the Assessment and
Management of Urban Nonpoint Source Pollution
Chris O. Yoder
Ohio Environmental Protection Agency, Division of Surface Water,
Ecological Assessment Section, Columbus, Ohio
Abstract
The health and well-being of the aquatic biota in surface
waters are important barometers 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.
Yet, these tangible products of the CWA regulatory and
water quality planning and management efforts are fre-
quently not linked nor equated with the more popular-
ized notion of chemical-physical water quality criteria
and other surrogate indicators and endpoints. Simply
stated, biological integrity is the combined result of
chemical, physical, and biological processes. Nowhere
in water quality management and assessment is the
interaction of these three factors more apparent than
with nonpoint sources. Management efforts that rely
solely on comparatively simple chemical-physical water
quality criteria surrogates frequently do not result in the
full restoration of ecological integrity. Therefore, ecologi-
cal concepts, criteria, and assessment tools must be
incorporated into the prioritization and evaluation of non-
point source pollution abatement efforts.
Introduction
The monitoring of surface waters and evaluation of the
biological integrity goal of the Clean Water Act (CWA)
have historically been predominated by nonbiological
measures such as chemical-physical water quality (1).
While this approach may have fostered an impression
of empirical validity and legal defensibility, it has not
sufficiently measured the ecological health and well-be-
ing of aquatic resources. An illustration of this point was
demonstrated in a comparison of the abilities of chemi-
cal water quality criteria and biological criteria to detect
aquatic life impairment based on ambient monitoring in
Ohio. Out of 645 water-body segments analyzed, bio-
logical impairment was evident in 49.8 percent of the
cases where no impairments of chemical water quality
criteria were observed (2). While this discrepancy may
at first seem remarkable, the reasons for it are many and
complex. Biological communities respond to and inte-
grate a wide variety of chemical, physical, and biological
factors in the environment whether they are of natural
or anthropogenic origin. Simply stated, controlling
chemical water quality criteria alone does not ensure the
ecological integrity of water resources (1).
The health and well-being of surface water resources
are the combined result of chemical, physical, and bio-
logical processes (Figure 1). To be truly successful in
meeting these goals, monitoring and assessment tools
are needed that measure both the interacting processes
and the integrated result of these processes (3). This is
especially true for nonpoint sources because many of
the effects involve the interactions of these factors. Bio-
logical criteria offer a way to measure the end result of
nonpoint source management efforts and successfully
accomplish the protection of surface water resources.
Biological communities respond to environmental im-
pacts that chemical-physical water quality criteria alone
cannot adequately discriminate or even detect. Habitat
degradation and sedimentation are two prevalent im-
pacts of nonpoint source origin that simply cannot be
measured by chemical-physical criteria alone. As illus-
trated by Figure 1, the combination of chemical and
physical factors results in surface water use impair-
ments from nonpoint sources.
The Ohio Environmental Protection Agency (EPA) re-
cently adopted biological criteria in its water quality
standards (WQS) regulations. These criteria are based
on measurable endpoints regarding the health and well-
being of aquatic communities. They are further struc-
tured into the state's WQS regulations within a system
of tiered aquatic life uses from which numerical biologi-
cal criteria are derived using a regional reference site
approach (4-7). These numerical expressions of biologi-
cal goal attainment criteria are essentially the end
183
-------
Velocity.
. Solubility
Feeding
Competition
Predation
Nutrients
Water
Resource
Integrity
Sunlight
Energy
Source
Seasonal
Cycles
s~~\
Riparian J
Vegetation /
Organic
Matter Inputs
1°and2°
Production
Width/Depth
Bank
Stability
Channel
Morphology
Gradient
Substrate
Instream
Cover
Figure 1. The five principal factors, with some of their important chemical, physical, and biological components, that influence and
determine the integrity of surface water resources (modified from Karr et al. [1]).
product of an ecologically complex but structured deri-
vation process. While numerical biological indices have
been criticized for potentially oversimplifying complex
ecological processes (8), distillation of such information
to readily comprehendible expressions is both practical
and necessary. The advent of new-generation evalu-
ation mechanisms, such as the Index of Biotic Integrity
(IBI) (1, 9, 10), the Index of Well-Being (Iwb) (11, 12),
the Invertebrate Community Index (ICI) (5), and similar
efforts (13-16), has filled important practical and theo-
retical gaps not always fulfilled by previously available
single-dimension indices. Multimetric evaluation mecha-
nisms, such as the IBI, extract ecologically relevant
information from complex biological community data
while preserving the opportunity to analyze such data on
a multivariate basis. The problem of biological data vari-
ability is also addressed within this system. Variability is
controlled by specifying standardized methods and
184
-------
procedures (17) that are then compressed through the
application of multimetric evaluation mechanisms (e.g.,
IBI, ICI) and stratified by accounting for regional and
physical variability and potential (e.g., ecoregions, tiered
aquatic life uses). The results are evaluation mecha-
nisms, such as the IBI and ICI, that have acceptably low
replicate variability (18-20).
Ecoregional Biocriteria and Determination
of Use Attainment
Biological criteria can play an especially important role
in nonpoint source assessment and management be-
cause they directly represent an important environ-
mental goal and regulatory endpoint (i.e., the biological
integrity goal of the CWA). Numerous studies have
documented this capability. Gammon et al. (21) docu-
mented a "gradient" of compositional and functional
shifts in the fish and macroinvertebrate communities of
small agricultural watersheds in central Indiana. Com-
munity responses ranged from an increase in biomass
with mild enrichment to complete shifts in community
function. Impacts from animal feedlots had the most
pronounced effects. In the latter case, the condition of
the immediate riparian zone was correlated with the
degree of impairment.
Later work by Gammon et al. (22) suggests that non-
point sources are impeding any further biological im-
provements observed in larger rivers due primarily to
reduced point source impacts. This is similar to obser-
vations that Ohio EPA has made in the Scioto River
downstream from Columbus. Urban nonpoint source
impacts are well known and have also been docu-
mented by numerous investigators. Klein (23) docu-
mented a relationship between increasing urbanization
and biological impairment, noting that the latter does not
become severe until urbanization reaches 30 percent of
the watershed area. Steedman (24) used a modification
of the IBI to demonstrate the influence of urban land use
and riparian zone integrity in Lake Ontario tributaries.
Steedman developed a model relationship between the
IBI and these two environmental factors.
Biological monitoring of nonpoint source impacts and
pollution abatement efforts conducted in concert with the
use of more traditional assessment tools (e.g., chemi-
cal-physical) can produce the type of evaluation needed
to determine where nonpoint source management ef-
forts should be focused, what some of the management
goals should be, and what determines the eventual
success (i.e., end result) of such efforts. At the same
time, a well-conceived monitoring program can yield
multipurpose information that can be applied to similar
situations without the need to perform site-specific moni-
toring everywhere. This is best accomplished when a
landscape-partitioning framework, such as ecoregions
(25) and the subcomponents, is used as an initial step
in accounting for natural landscape variability. Because
of landscape variability, uniform and overly simplified
approaches to nonpoint source management often fail
to produce the desired results (26).
Biological criteria in Ohio are based on two principal
organism groups: fish and macroinvertebrates. Numeri-
cal biological criteria for rivers and streams were derived
from the results of sampling conducted at more than 350
reference sites that typify the "least impacted" condition
within each ecoregion (5, 6). This information was used
within the existing framework of tiered aquatic life uses
in the Ohio WQS regulations to establish attainable,
baseline biological community performance expecta-
tions on a regional basis. Biological criteria vary by
ecoregion, aquatic life-use designation, site type, and
biological index. The resulting criteria for two of the
"fishable, swimmable" uses, Warmwater Habitat (WWH)
and Exceptional Warmwater Habitat (EWH), are shown
in Figure 2.
Procedures for determining the use attainment status of
Ohio's lotic surface waters were also developed (5, 27).
Using the numerical biocriteria as defined by the Ohio
WQS regulations, use attainment status is determined
as follows:
Full: Use attainment is considered full if all of the
applicable numeric indices exhibit attainment of the
respective biological criteria; this means that the
aquatic-life goals of the Ohio WQS regulations are
being attained.
Partial: At least one organism group exhibits nonat-
tainment of the numeric biocriteria, but no lower than
a narrative rating of "fair," and the other group exhibits
attainment.
A/on: Neither organism group exhibits attainment of
the ecoregional biocriteria, or one organism group
reflects a narrative rating of "poor" or "very poor,"
even if the other group exhibits attainment.
Following these rules, a use attainment table is con-
structed on a longitudinal mainstem or watershed basis.
Information included in the table includes sampling lo-
cation (river mile index), biological index scores, the
Qualitative Habitat Evaluation Index (QHEI) score, at-
tainment status, and comments about important site-
specific factors such as proximity to pollution sources.
An example of how to construct a use attainment table
is provided in Table 1.
Aquatic Ecosystems at Risk
Ecosystems that possess or reflect integrity (as envi-
sioned by the biological integrity goal of the CWA) are
characterized by the following attributes (1):
The inherent potential of the system is realized.
185
-------
Fish Boat Sites
Fish Wading Sites
EWH
EWH
Fish Headwater Sites
Macroinvertebrates
EWH
EWH
Huron Erie Lake Plateau - HELP |^| Eastern-Ontario Lake Plain - EOLP
*"-] Interior Plateau-IP ES3 Western Allegheny Plateau-WAP
Eastern Corn Belt Plains - ECB
Figure 2. Biological criteria in the Ohio WQS for the Warmwater Habitat (WWH) and Exceptional Warmwater Habitat (EWH) use
designations arranged by biological index, site type for fish, and ecoregion. The EWH criteria for each index and site type
is located in the boxes located outside of each map.
The system and its components are stable.
The system retains a capacity for self-repair when
perturbed or injured.
Minimal or no external support for community main-
tenance is required.
Thus, ecosystems that are impaired and therefore lack
integrity have had their capacity to withstand and rapidly
recover from perturbations exceeded. Impaired ecosys-
tems are likely to become even further degraded due to
incremental increases in stress.
186
-------
Many rivers and streams nationwide fail to exhibit the
characteristics of healthy ecosystems. Recent esti-
mates indicate that as many as 98 percent of lotic
ecosystems are degraded to a detectable degree (29).
Karr et al. (30) illustrated the extent to which the Illinois
and Maumee River basin fish communities have de-
clined during the past 50 years: two-thirds of the original
fauna were lost from the former and more than 40
percent from the latter. Losses of naiad mollusks and
crayfish have been even greater. In Ohio, long-term
declines in fish communities have been extensively
documented by Trautman (31). More recent information
indicates that the fraction of the fish fauna that is imper-
iled or declining has increased from 30 to 40 percent
since 1980 (32). This information indicates that lotic
ecosystems are threatened in both Ohio and nation-
wide, an indication that existing frameworks for water
resource protection and management have been essen-
tially ineffective in preventing large-scale losses of eco-
logical integrity. This is particularly true for ecosystems
affected by habitat degradation, riparian encroachment,
excess sedimentation, organic enrichment, and nutrient
enrichment. All or most of these forms of degradation
are evident in areas affected by urban nonpoint sources.
Urban Nonpoint Source Pollution in Ohio
Urban watersheds in Ohio have exhibited a familiar and
well-known legacy of aquatic resource degradation.
Few, if any, functionally healthy watersheds exist in the
older, heavily urbanized parts of the Midwest. Good
quantitative estimates of the proportion of surface wa-
ters that are degraded by urbanization are lacking, how-
ever, particularly for headwater streams. It is also widely
perceived that the restoration of beneficial aquatic life
uses in most heavily urbanized areas is not practically
attainable. This in itself presents a barrier to any notion
of attaining existing use designations or upgrading use
designations for waters classified for less than fishable
and swimmable uses. The assignment of appropriate
aquatic life and recreational uses is a challenge that
Ohio EPA has dealt with over the past 15 years.
Urban and suburban development activities that have
the greatest impacts on aquatic life in Ohio include the
wholesale modification of watershed hydrology, riparian
vegetation degradation and removal, direct instream
habitat degradation via channelization, construction and
other drainage enhancement activities, sedimentation
and siltation caused by stream-bank erosion (which is
strongly linked to riparian encroachment), and contribu-
tions of chemical pollutants. Statewide, urban and sub-
urban sources are responsible for impairment (major
and moderate magnitude sources) in more than 927
miles of streams and rivers and more than 23,000 acres
of lakes, ponds, and reservoirs (32). These activities
also threaten existing use attainment in nearly 160 miles
of streams and rivers and may be a potential problem in
more than 4,380 miles of streams and rivers that have
not yet been fully monitored and evaluated (33).
While much attention is generally given to toxic sub-
stances in urban nonpoint source runoff, evidence sug-
gests that nontoxic effects are more widespread, at least
in Ohio and the Midwest. The second leading cause of
impairment identified by the 1992 Ohio Water Resource
Inventory, sedimentation (or siltation) resulting from ur-
ban and other land-use activities is the most pervasive
single cause of impairment from nonpoint sources in
Ohio. Sedimentation is responsible for more impairment
(over 1,400 miles of stream and rivers and 23,000 acres
of lakes, ponds, and reservoirs) than any other cause
except organic enrichment/dissolved oxygen, with
which it is closely allied in urban and agricultural areas.
Since Ohio conducted the Ohio Water Resource Inven-
tory in 1988 (34), this cause category has surpassed
ammonia and heavy metals in rank. If the statewide
monitoring database were distributed more equally
across the state, sedimentation would likely be found to
be the leading cause of impairment.
Although sediment deposition in both lotic and lentic
environments is a natural process, it becomes a problem
when the capability of the ecosystem to "assimilate" any
excess delivery is exceeded. Sediment deposited in
streams and rivers comes primarily from stream bank
erosion and in runoff from upland erosion. The effects
are much more severe in streams and rivers with de-
graded riparian zones and low gradient. Given similar
rates of erosion, the effects of sedimentation are much
worse in channel-modified and riparian zone-degraded
streams than in more natural, intact habitats. In chan-
nel-modified streams, incoming silt and sediment re-
main within and continue to degrade the stream
channel, instead of being deposited in the immediate
riparian "floodplain" during high flow periods (35). This
also adds to and increases the sediment bedload that
continues to affect the substrates long after the runoff
events have ceased.
One of the more prevalent results is substrate em-
beddedness, which occurs when an excess of fine ma-
terials, particularly clayey silts and fine sand, fills the
otherwise open interstitial spaces between larger sub-
strates (Figure 3). In extreme cases, the coarser sub-
strates may be "smothered"; in other cases, the
substrate can be cemented together, or "armor plated."
In either event, the principal ecological consequence is
the loss of available benthic surface area for aquatic
organisms (particularly macroinvertebrates) and as a
location for the development of fish eggs and larvae.
The soft substrates afforded by the increased accumu-
lation of fine materials also provide an excellent habitat
for the growth of undesirable algae. Thus, to success-
fully abate the adverse impacts of sediment, we need to
be as concerned with what each event leaves behind as
187
-------
Interstitial Spaces NOT
Filled With Fines
Aquatic Vegetation and Algae
DOES NOT Trap Excessive Clayey Silts
Some Gravel and Sand Expected
Aquatic Vegetation/Algae DO
Trap Excessive Clayey Silts
iimfNonembedded
substrates
provide living space
for a diversity of
aquatic organisms
due to increased
benthic surface area
Substrate not "Armour-Plated"
to Stream Bottom
Cobble Difficult To
Remove From Bottom
Interstitial Spaces ARE Filled With
Sand, Fine Gravel, and Clayey Silts
nmHf the substrate particle in contact with cobble/boulder is larger than
"fine sediment," then it is not considered embedded (unless armour-plated)
nmfWhen substrates are embedded by fines, the embedded
surfaces of rocks often are black or dark green as a result of
anaerobic respiration (organic fines consume oxygen in sediments)
1114-Embedded substrates are difficult to dislodge from the bottom
(cannot easily dislodge with foot)
Figure 3. Characterization of substrate embeddedness with some of the key structural signatures and a summary of some of the
ecological impacts of this form of stream substrate degradation.
much as with what takes place in the water column
during each event.
The effects of sedimentation on aquatic life are the most
severe in the ecoregions of Ohio where:
Erosion and runoff are moderate to high.
Clayey silts that attach to and fill the interstices be-
tween coarse substrates are predominant.
Streams and rivers lack the ability to expel sediments
from the low-flow channel, which results in a longer
retention time and greater deposition of silt in the
most critical habitats.
Estimates of gross erosion alone do not always corre-
late with adverse impacts to aquatic communities, al-
though this is a frequently cited criterion for prioritizing
nonpoint source management efforts. Some of the areas
of Ohio that have the highest rates of gross erosion
(e.g., East Corn Belt Plain, Interior Plateau, and Western
Allegheny Plateau ecoregions) also have some of the
most diverse and functionally healthy assemblages of
aquatic life at the least affected reference and other sites
(32). Many of the streams in these ecoregions have
relatively intact riparian and instream habitat and thus
are "buffered" against the naturally erosive conditions.
The detrimental effects of sedimentation seem to be the
worst in areas of the state where the proportion of clayey
silts are highest, stream gradient is the lowest, and
188
-------
riparian encroachment and modification are extensive
(i.e., Huron/Erie Lake Plain and portions of the East
Corn Belt Plain and Erie/Ontario Lake Plain ecoregions).
The interaction between nonpoint source runoff and
riparian and instream habitat must be appreciated and
understood if impacts such as sedimentation are to be
effectively dealt with. Figure 4 illustrates the interdepen-
dency of the rate of runoff, increased sediment delivery,
in-channel habitat degradation, riparian zone condition,
and substrate condition. An effect involving any one
Increased
Rate of Runoff
Riparian
Zone
Degradation
Increased
Sediment Delivery
Increased
Substrate
Embeddedness
In-Channel
Habitat
Degradation
Figure 4. Illustration of the complex interaction of nonpoint
source caused changes in hydrology and sediment
delivery and how each singly and in combination can
degrade instream and riparian habitat.
factor can set off a chain of events that results in cumu-
lative changes reflected by most or even all of the
interdependent factors. Two factors that are influenced
in the conversion of watersheds by urban development
are an increased rate of runoff and increased sediment
delivery. These two factors then combine to influence
other important aspects of stream habitat, such as ripar-
ian zone integrity and increased substrate embedded-
ness. In effect, a change in one of these factors can
result in a cascading chain of events that eventually
cause aquatic life use impairment or inhibit the ability of
a degraded stream to be successfully rehabilitated.
Thus, considerations of previously ignored aspects such
as riparian and instream habitat and watershed dynam-
ics must be included in urban nonpoint source assess-
ment and abatement strategies.
The direct and indirect effects of sedimentation and the
associated nutrient enrichment are becoming especially
apparent in the larger mainstem rivers. Both sediment
and nutrient enrichment impacts have largely been over-
looked and will not only require a change in the status
quo of water quality management but also in the inter-
disciplinary solutions and information gathering that
demonstrates the character and magnitude of these
impacts (36).
Bioassessment of Urban Watersheds
Biological criteria and bioassessment methods can and
do play a key role in several areas of nonpoint source
management. As a basis for determining use impair-
ments, biocriteria have played a central role in the Ohio
Nonpoint Source Assessments (33, 37), the biennial
Ohio Water Resource Inventory (305b report) (32), and
watershed-specific assessments of which Ohio EPA
completes from 6 to 12 each year. Biological criteria
represent a measurable and tangible goal against which
the effectiveness of nonpoint source pollution abate-
ment programs and individual projects can be judged.
Biological assessments, however, must be accompa-
nied by appropriate chemical-physical measures, land-
use considerations, and source information necessary
to establish linkages between the land-use activities and
the instream responses.
A great deal of uncertainty exists about the link between
steady-state water quality criteria and ecological indica-
tors. While we have observed biocriteria attainment with
chemical water quality criteria exceedences in only a
fraction of the comparisons, the chemical data are
largely from grab samples collected during summer-fall
low flow situations. In many cases, we have failed to
detect chemical criteria exceedences during low flows,
yet biocriteria impairment is apparent. The correspon-
dence of biocriteria attainment with water quality criteria
exceedences measured under elevated flows has not
been observed with any regularity. Nonetheless, we
have surmised that much of the biocriteria nonattain-
ment observed in affected urban watersheds is due to
water quality criteria exceedences that have occurred
during elevated flow events that preceded the biological
sampling. Reaching such a conclusion, however, is
made possible only by examining other evidence be-
yond water column data.
In many urban settings, sediment chemical concentra-
tions frequently are highly or extremely elevated com-
pared with concentrations measured at least-affected
reference sites. Contaminated sediments enter the
aquatic environment during episodic releases from point
sources and during runoff events from nonpoint sources.
The correspondence between increasingly elevated
sediment concentrations and declining aquatic commu-
nity performance is demonstrated by Figure 5. A sedi-
ment classification scheme derived by Kelly and Hite
(38) for Illinois streams was used to classify results for
sediment chemical analyses at sites with corresponding
biological data. Sediment chemical concentrations are
classified as nonelevated, slightly elevated, elevated,
highly elevated, and extremely elevated as the concen-
trations increase beyond the mean concentration at
background sites. The results for four heavy metal pa-
rameters (arsenic, cadmium, lead, and zinc) commonly
encountered in urban settings show that the frequency
189
-------
100
80
60
40
20
I
IBI>36
ICI > 34
Arsenic
D)
w
T3
0)
1
JD
LU
_
D)
±
T3
0)
1
0)
LU
70
60
50
^ 40
o
c
0)
§ 30
0)
LL
20
10
0
I
H IBI > 36
0 ICI > 34
Lead
0)
c
o
(0
>
0)
D)
W
(0
>
0)
T3
0)
JD
LU
D)
±
T3
0)
JD
LLJ
_>%
0)
0)
Figure 5. The frequency of occurrence of IBI and ICI scores which attain the warmwater habitat biocriteria under increasingly
contaminated levels of four heavy metals in bottom sediments. Based on data collected by Ohio EPA throughout Ohio
between 1981 and 1989.
of sites attaining the WWH use designation criteria for
the IBI and ICI sharply decline as the sediment concen-
trations of these metals increase. For arsenic, no sites
with highly or extremely elevated concentrations attain
the biocriteria. For the remaining three parameters, in a
few instances in each case, biocriteria attainment exists
with highly elevated or extremely elevated sediment
concentrations, but these are exceptions to the overall
pattern.
For bioassessments to achieve their maximum effective
use in the assessment of urban nonpoint sources, sam-
pling and analysis should be based on a watershed
design. An example of the use of biological criteria to
evaluate aquatic life-use atta i n me nt/n on attainment in an
urban watershed involves the Nimishillen Creek basin in
northeastern Ohio (Table 1). This watershed is subject
to a variety of point and nonpoint source impacts and is
extensively affected by intensive urbanization in several
190
-------
Table 1. Aquatic Life-Use Attainment Status for the Existing and Recommended Aquatic Life-Use Designations in the Nimishillen
Creek and Selected Tributaries Based on Data Collected From June to September, 1985
RIVER MILE
Use
Designation
Nimishillen Creek
VWVH
VWVH
Sherrie (Sherrick)
LRW
VWVH
Osnaburg Ditch
MWH
Hurford Run
LRW
MWH
WWH
Domer Ditch
WWH
Fish/
Invertebrate
14.2/14.2
12.7/12.7
11.7/11.7
11.2/11.1
10.2/10.3
8.8/8.8
6.7/6.7
3.2/3.2
0.6/0.6
Run
5.3/5.3
4.1/4.1
0.1/
0.7/0.7
0.1/0.1
2.0/
1.8/
1.2/
0.3/
0.1/
0.5/0.4
0.1/0.1
IBI
30d
22d
2Qd
17d
19d
19d
16
24d
2Qd
12d
17d
22
15d
12d
12d
12d
12d
^d
18d
23d
18d
Mlwb
6.7d
6.0d
48d
i3_d
Md
Z3d
3J5
42d
33a
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
ICIa
22d
22d
12d
8d
10d
8d
2
6d
Qd
pd
Ed
pd
Pd
pd
MG
pd
QHEIb
60
71.5
81
81.5
72.5
85
80.5
91
92
33.5
70 T
52
42 T
39
34.5
27
52.5
66
50.5
60
54.5
Attainment
Status0
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Non
Comment
Dst. East and Middle
Branches
Cherry Ave.
Dst. West Branch (Gregory
Galvanizing)
Dst. Hurford Run (Ashland
Oil)
Ust. Canton WWTP
Baum Rd.
Howenstine Rd.
Main St.
Ust. at mouth
Dst. Osnaburg Ditch
Ust. East Canton WWTP
Dst. East Canton WWTP
Ust. Ashland Oil
Dst. Ashland Oil
Dst. Domer Ditch
Ust. Timken
Dst. Timken
West Branch Nimishillen Creek
WWH
McDowell Ditch
MWH
Zimber Ditch
WWH
MWH
5.9/5.9
3.2/3.2
1 .6/1 .6
0.8/
0.1/0.1
1 .8/1 .8
0.1/0.1
3.8/3.8
1 .8/2.4
0.9/1.1
0.6/0.6
2Z
17d
22d
24d
21d
^d
21d
40ns
29d
23d
23d
N/A
4.8d
£5d
6.2d
Md
N/A
N/A
N/A
N/A
N/A
N/A
18d
20d
20d
12d
F
F
G
F
F
F
53
59.5
43.5
34.5
65
34
41
57
42
31
31.5
Non
Non
Non
(Non)
Non
Partial
Partial
Full
Non
Partial
Partial
At cemetery
Dst. McDowell Ditch
Ust. Tuscarawas St.
Ust. Gregory Galvanizing
Dst. Gregory Galvanizing
Ust. Everhard Rd.
At mouth
Regional reference site
Dst. Hoover Industrial Park
Ust. North Canton Ditch
Dst. North Canton Ditch
Rettig Ditch
Undesignated 0.9/0.9
North Canton Ditch
LRW 0.1/0.1
29d
32
N/A
N/A
39
46
Non
Full
Channel modified
Partially culverted (80-m
zone)
191
-------
Table 1. Aquatic Life-Use Attainment Status for the Existing and Recommended Aquatic Life-Use Designations in the Nimishillen
Creek and Selected Tributaries Based on Data Collected From June to September, 1985 (Continued)
RIVER MILE
Use
Fish/
Designation Invertebrate
IBI
Mlwb
ICIa
QHEIb
Attainment
Status0
Comment
Middle Branch Nimishillen Creek
VWVH
Swartz Ditch
MWH
Guiley (Hartville) Ditch
MWH
11.4/11.4
10.4/10.4
8.0/8.0
6.8/6.8
5.0/
2.5/2.5
1.6/
/0.8
0.2/0.1
2.6/2.6
1 .2/1 .2
0.2/0.3
/4.1
3.47
2.3/2.3
0.4/0.4
45
27d
34ns
35ns
37ns
38
43
28d
26
33
34
26
33
36
N/A
5.8d
7.7ns
8.0
76ns
8.3
8.5
7.2d
N/A
N/A
N/A
N/A
N/A
N/A
30ns
22d
30ns
40
28d
10d
14d
F
pd
F
pd
pd
F
50
38
74
47
60
34
31
45.5
27
32
44
Full
Non
Full
Full
(Full)
Partial
(Full)
(Non)
Non
Full
Non
Full
(Non)
(Full)
Partial
Full
Ust. State St.
Dst. Werner-Church Rd.
Regional reference site
Ust. 55th St.
Ust. Martindale Rd.
Dst. State Route 62
Cookes Park
Ust. Smith-Kramer Rd.
Ust. Church Rd.
Dst. Hartville Ditch
Ust. Teledyne
Ust. Hartville WWTP
Dst. Smith-Kramer Rd.
Cans Rd.-Dst. Culvert
East Branch Nimishillen Creek
VWVH
VWVH
Ecoregion Biocriteria:
INDEX - Site Type
IBI - Headwaters
IBI -Wading
Mlwb - Wading
ICI
8.6/8.6
6.4/6.3
4.7/4.7
4.2/4.2
3.4/2.8
1 .9/1 .9
0.1/0.1
Erie/Ontario
WWH
40
38
7.9
34
39ns
33d
29d
23d
24d
24d
31 d
Lake Plain
EWH
50
50
9.4
46
N/A
6.8d
6.4d
3J3d
45d
£1
8.2d
MWHe
24
24
5.8
22
40
26d
4d
f4d
20d
20d
14d
64.5
51
80
66
66
67.5
60.5
Full
Non
Non
Non
Non
Non
Partial
Regional reference site
Ust. J&L Steel
Dst. J&L Steel
Dst. Louisville South WWTP
Dst. Louisville North WWTP
Ust. LTV Steel
At mouth
a Narrative criteria used in lieu of ICI: E = exceptional, G = good, MG = marginally good, F = fair, P = poor.
b All QHEI values are based on the most recent version of the index (28).
c Use attainment is parenthetically expressed when based on one organism group.
d Significant departure from ecoregion biocriteria; poor and very poor results are underlined.
e For channel modified areas.
Dst. = downstream
LRW = Limited Resource Waters
Mlwb = modified Iwb
MWH = Modified Warmwater Habitat
ns = nonsignificant departure from WWH and EWH biocriteria (4 IBI or ICI units; 0.5 Mlwb units).
Ust. = upstream
WWTP = wastewater treatment plant
areas. As with many of the Ohio watersheds that are
more heavily affected by point and nonpoint sources, the
majority of sampling sites either fail to attain the appli-
cable biological criteria or are only in partial attainment.
Out of 57 sampling sites in the entire watershed, only 11
(19 percent) fully attained the applicable biological crite-
ria. These results demonstrate the degree of degrada-
tion that exists in most urban watersheds and the
multiple source causes.
192
-------
Another issue of critical importance to the management
of urban watersheds is also apparent in Table 1, use
attainability. Many of the use designations listed for the
various streams of the Nimishillen Creek basin are rec-
ommended uses, meaning that a different aquatic life
use applied at the time of the sampling. An important
objective of the biological sampling conducted by Ohio
EPA is to determine the appropriate aquatic life-use des-
ignation. If the results of the sampling and data analysis
suggest that the existing use designation is inappropriate
(or the stream is presently unclassified), the appropriate
use is recommended. These recommendations are then
proposed in a WQS rulemaking procedure and adopted
after consideration of public input.
Figure 6 illustrates the relative distribution of IBI scores
based on biological monitoring conducted by Ohio EPA
in several urban and suburban watersheds throughout
Ohio. These range in size from relatively small headwa-
ter streams (less than a 20-square-mile watershed area)
to increasingly larger streams and rivers. For the smaller
watersheds, there is a pattern of lower IBI scores and a
subsequent loss of biological integrity with an increasing
degree of urbanization. The baseline biological criterion
for the WWH use designation is not attained by any (or
only a few) sampling sites in the older urban water-
sheds, such as the Cuyahoga River and Little Cuyahoga
River of northeastern Ohio and Mill Creek in Cincinnati.
The IBI scores in these watersheds are indicative of
poor and very poor water resource quality. The Rocky
River basin is largely a suburban area of Cleveland upon
which municipal wastewater discharges have had an
extensive impact, but despite this the basin exhibits
higher IBI scores. The highest IBI scores were observed
in Rocky Fork (Columbus area), Taylor Creek (Cincinnati
area), and Little Miami River (southwest Ohio) tributar-
ies, which have only recently begun to be suburbanized.
These three watersheds also lack some of the compan-
ion impacts of the older urban areas, namely, combined
sewer overflows and industrial discharges.
For the larger streams and rivers, the pattern was simi-
lar, with the older urban areas exhibiting the lowest IBI
scores and the less urbanized and suburban water-
sheds exhibiting higher scores, some of which attain the
WWH criteria. The major exceptions, however, involve
the two large mainstem rivers (Great Miami River and
Scioto River) which exhibit higher IBI scores despite
flowing within urban settings. This illustrates the influ-
ence of river and upstream watershed size on the ability
of a river or stream to withstand increased urbanization.
Both the Great Miami River and Scioto River mainstems
originate in rural areas and are quite large when they
enter the Dayton and Columbus urban areas. Thus,
stream size relative to the watershed and the influence
of land-use patterns are important to understanding and
managing local nonpoint source impacts.
Applications to Nonpoint Source
Management
Steedman (24) observed the IBI to be negatively corre-
lated with urban land use. The land use within the 10 to
100 km2 area upstream from a site was the most impor-
tant in predicting the IBI, which suggests that "extrane-
ous" information was likely included if whole watershed
land-use area was used. Steedman (24) also deter-
mined that the condition of the riparian zone was an
important covariate (a measure of independent vari-
ation) with urban land use in addition to other factors,
such as sedimentation and nutrient enrichment. A model
relationship between these factors and the IBI was de-
veloped and provided the basis to predict when the IBI
would decline below a certain threshold level with cer-
tain combinations of riparian zone width and percent of
urbanization. In the Steedman (24) study, the domain of
degradation for Toronto area streams ranged from
75-percent riparian removal at 0-percent urbanization to
0-percent riparian removal at 55-percent urbanization.
These results indicate that it is possible to establish the
bounds within which the combination of watershed land
use and riparian zone condition must be maintained for
a target level of biological community performance to
persist. It seems plausible that such relationships could
be established for many other watersheds, provided the
database is sufficiently developed not only for biological
communities but also for land-use composition and ri-
parian corridor condition. Additionally including the con-
cept of ecoregions and subecoregions should lead to the
development of criteria for land use and riparian zones
that would ensure the maintenance of biocriteria per-
formance levels in streams and rivers over fairly broad
areas without the need to develop a site-specific data-
base everywhere.
Well-designed biological surveys can fit well into the
watershed approach to nonpoint source management.
Because the biota respond to and integrate all of the
various factors that affect a particular water body, they
are essentially the end product of what happens within
watersheds. The important issue is that ambient moni-
toring be conducted as part of the nonpoint source
assessment and management process, and that it be
performed correctly in terms of timing, methods, and
design. Monitoring alone is not enough, however.
Federal, state, local, and private efforts to remediate
nonpoint source impairments must include an interdis-
ciplinary approach that goes beyond water column
chemistry impacts to include the cumulative range of
factors responsible for ecosystem degradation that has
been documented over the past century. Existing regu-
lations and standards have only been locally successful
in reducing water resource declines attributable to wa-
tershed and riparian zone degradation. Effective protec-
tion and rehabilitation strategies require the targeting of
large areas and individual sites (39) as well as the
193
-------
60 -
50
I
i
EWH Criterion
(IBI=50)
40
WWH Criterion
(IBI=40)
m
30
20
12
r T^11!
O
a:
1
ro
a:
0
o
'0
w
D)
in
o
OL
I
8
a
o
,
ra
o
ce
E
ro
08
ce
.
o
o
ce
ce
ro
D)
o
O
08
o
c
ce
ro
D)
o
O
Figure 6. IBI values observed in selected Ohio headwaters streams (drainage area <20 mi.2; upper) and larger Ohio streams and
rivers between 1981 and 1992. Box and whisker plots include all values recorded in each stream or stream/river assem-
blage.
194
-------
incorporation of ecological concepts in the status quo of
land-use management practices and policies.
Ohio EPA has initiated the development of policies that
will ensure a holistic approach to nonpoint source man-
agement. For example, we have specified a minimum
width of two to three times the bank full channel width
as necessary to protect riparian zones and ensure the
integrity of instream habitat. This also ensures that the
ability of the stream to assimilate nonpoint source runoff
will be maintained. To be completely successful, how-
ever, this measure must be accompanied by the appli-
cation of best management practices in the uplands.
Such an approach goes well beyond a singular concern
for the concentration of pollutants in the water column
and must be incorporated into the total maximum daily
load approach envisioned by the U.S. Environmental
Protection Agency as an integral part of urban nonpoint
source runoff management.
Thus, it seems that we have a choice in the manage-
ment of urban nonpoint sources, as portrayed by Figure
7. Extending the traditional process by which we have
managed chemical pollutants discharged by point
sources during the past 15 to 20 years to nonpoint
sources is exemplified by treating streams as once-
through flow conduits that are essentially isolated from
interactions with the landscape. This is commonly ex-
emplified by simplified mass-balance approaches to es-
tablishing water quality-based effluent limitations for
point sources using steady-state assumptions. While
this approach has been successful in reducing point
source loadings of commonly discharged substances, it
holds much less promise for highly dynamic inputs from
diffuse sources. For nonpoint source management to
truly result in the restoration and preservation of biologi-
cal integrity, we must regard streams as an interactive
component of the landscape where multiple inputs and
influences act together to determine the health of the
aquatic resource.
Urban watershed management and protection issues
will continue to develop as new information is revealed
and relationships between instream biological commu-
nity performance and watershed factors are better de-
veloped. Nonetheless, some of what we know now
should be included in current management strategies.
Urban and suburban development must become proac-
tive; that is, developments must be designed to accom-
modate the features of the natural landscape and
include common sense features such as setbacks from
riparian zones. Regulatory agencies also share respon-
sibility, particularly in resolving use attainability issues.
Watersheds that exhibit the attainment of aquatic life-
use biocriteria should be protected to maintain the cur-
rent conditions. Frequently our attention seems to
emphasize high quality or unique habitats; however,
Static
Source
Inputs
Multiple Source,
Dynamic Inputs
'i'i'i'i^l"i'i'i1l'n'1i'rS
Mass Balance Output
A. Stream as an isolated,
once-through flow conduit
(steady-state, mechanical system)
Assimilated Output
B. Stream as an interactive
component of the landscape
(dynamic, living system)
Figure 7. Two views of a stream ecosystem: A. The stream is viewed as an isolated conveyance for static source wastes and
runoff with the net water column output as a mass balance function of flow and concentration. B. The stream as an
interactive component of the landscape with dynamic and multiple source inputs and assimilated output as affected by
the surrounding land use, habitat, geology, soils, and other biotic and abiotic factors.
195
-------
water quality standards must be maintained where they
are presently attained, if even minimally so. Strategies
should also include the restoration of degraded water-
sheds where that potential exists. In systems where the
degree of degradation is so severe that the damage is
essentially irreparable, minimal enhancement measures
should still be required, even though full use attainment
is not expected. Biocriteria and bioassessments have an
important and central role to play in this process.
References
1. Karr, J.R., K.D. Fausch, P.L. Angermeier. P.R. Yant, and I.J.
Schlosser. 1986. Assessing biological integrity in running waters:
A method and its rationale. Special Publication No. 5. Cham-
paign, IL: Illinois Natural History Survey.
2. Ohio Environmental Protection Agency. 1990. Ohio's nonpoint
source pollution assessment. Columbus, OH: Ohio EPA, Division
of Water Quality Planning and Assessment.
3. Karr, J.R. 1991. Biological integrity: A long-neglected aspect of
water resource management. Ecol. Applic. 1(1):66-84.
4. Ohio Environmental Protection Agency. 1987. Biological criteria
for the protection of aquatic life, Vol. I. The role of biological data
in water quality assessment. Columbus, OH: Ohio EPA, Division
of Water Quality Planning and Assessment.
5. Ohio Environmental Protection Agency. 1987. Biological criteria
for the protection of aquatic life, Vol. II. User's manual for biologi-
cal field assessment of Ohio surface waters. Columbus, OH: Ohio
EPA, Division of Water Quality Planning and Assessment.
6. Ohio Environmental Protection Agency. 1989. Biological criteria
for the protection of aquatic life, Vol. III. Standardized biological
field sampling and laboratory methods for assessing fish and
macroinvertebrate communities. Columbus, OH: Ohio EPA, Divi-
sion of Water Quality Planning and Assessment.
7. Yoder, C.O. 1989. The development and use of biocriteria for
Ohio surface waters. In: Flock, G.H., ed. Water quality standards
for the 21 st century. Proceedings of a U.S. EPA National Confer-
ence, Washington, DC.
8. Suter, II, G.W. 1993. A critique of ecosystem health concepts and
indexes. Environ. Toxicol. Chem. 12:1,521-1,531.
9. Karr, J.R. 1981. Assessment of biotic integrity using fish commu-
nities. Fisheries 6(6):21-27.
10. Fausch, K.D., J.R. Karr, and P.R. Yant. 1984. Regional application
of an index of biotic integrity based on stream fish communities.
Trans. Am. Fishery Soc. 113:39-55.
11. Gammon, J.R. 1976. The fish populations of the middle 340 km
of the Wabash River. Technical Report 86. Purdue University
Water Resources Research Center.
12. Gammon, J.R., A. Spacie, J.L. Hamelink, and R.L. Kaesler. 1981.
Role of electrofishing in assessing environmental quality of the
Wabash River. In: Bates, J.M., and C. Weber, eds. Ecological
assessments of effluent impacts on communities of indigenous
aquatic organisms. ASTM STP 730. Philadelphia, PA: American
Society for Testing and Materials.
13. U.S. EPA. 1989. Rapid bioassessment protocols for use in rivers
and streams: Benthic macroinvertebrates and fish. EPA/444/4-
89-001 . Washington, DC.
14. Lyons, J. 1992. Using the Index of Biotic Integrity (IBI) to measure
environmental quality in warmwater streams of Wisconsin. Gen.
Tech. Rep. NC-149. St. Paul, MN: U.S. Department of Agriculture.
15. U.S. EPA. 1991. Development of index of biotic integrity expec-
tations for the ecoregions of Indiana, Vol. I. Central corn belt plain.
EPA/905/9-91/025.
16. Kerans, B.L., and J.R. Karr. 1992. An evaluation of invertebrate
attributes and a benthic index of biotic integrity for Tennessee
Valley rivers. In: U.S. EPA. 1992. Proceedings of the 1991 Mid-
west Pollution Control Biologists' Conference. EPA/905/R-
92/003.
17. Ohio Environmental Protection Agency. 1989. Addendum to bio-
logical criteria for the protection of aquatic life, Vol. II. User's
manual for biological field assessment of Ohio surface waters.
Columbus, OH: Ohio EPA, Division of Water Quality Planning and
Assessment.
18. Davis, W.S., and A. Lubin. 1989. Statistical validation of Ohio
EPA's invertebrate community index. In: Davis, W.S., and T.P.
Simon, eds. Proceedings of the 1989 Midwest Pollution Control
Biologists' Meeting, Chicago, IL. EPA/905/9-89/007, pp. 23-32.
19. Rankin, E.T., and C.O. Yoder. 1990. The nature of sampling
variability in the Index of Biotic Integrity (IBI) in Ohio streams. In:
Davis, W.S., ed. Proceedings of the 1990 Midwest Pollution Con-
trol Biologists' Conference, Chicago, IL. EPA/905/9-90/005, pp.
9-18.
20. Stevens, J.C., and S.W Szczytko. 1990. The use and variability
of the biotic index to monitor changes in an effluent stream
following wastewater treatment plant upgrades. In: U.S. EPA.
1991. Proceedings of the 1990 Midwest Pollution Control Biolo-
gists' Meeting, Chicago, IL. EPA/905/9-90/005, pp. 33-46.
21. Gammon, J.R., M.D. Johnson, C.E. Mays, D.A. Schiappa, WL.
Fisher, and B.L. Pearman. 1983. Effects of agriculture on stream
fauna in central Indiana. EPA/600/S3-83/020.
22. Gammon, J.R., C.W Gammon, and M.K. Schmid. 1990. Land
use influence on fish communities in central Indiana streams. In:
U.S. EPA. 1990. Proceedings of the 1990 Midwest Pollution Con-
trol Biologists' Conference. EPA/905/R-92/003. pp. 111-120.
23. Klein, R.D. 1979. Urbanization and stream quality impairment.
Water Res. Bull. 15(4):948-963.
24. Steedman, R.J. 1988. Modification and assessment of an index
of biotic integrity to quantify stream quality in southern Ontario.
Can. J. Fish. Aquatic Sci. 45:492-501.
25. Omernik, J.M. 1987. Ecoregions of the conterminous United
States. Ann. Assoc. Am. Geogr. 77(1 ):118-125.
26. Omernik, J.M., and G.E. Griffith. 1991. Ecological regions versus
hydrologic units: Frameworks for managing water quality. J. Soil
Water Conserv. 46:334.
27. Yoder, C.O. 1991. The integrated biosurvey as a tool for evalu-
ation of aquatic life use attainment and impairment in Ohio sur-
face waters. In: U.S. EPA. 1991. Biological criteria: Research and
regulation. Proceedings of a U.S. EPA National Conference,
Washington, DC.
28. Rankin, E.T 1989. The Qualitative Habitat Evaluation Index
(QHEI): Rationale, methods, and applications. Columbus, OH:
Ohio EPA, Division of Water Quality Planning and Assessment.
29. Benke, A.C. 1990. A perspective on America's vanishing streams.
J. N. Am. Benthic Soc. 9(1):77-88.
30. Karr, J.R., L.A. Toth, and D.R. Dudley. 1985. Fish communities of
midwest rivers: A history of degradation. BioScience 35(2):90-95.
31. Trautman, M.B. 1981. The fishes of Ohio, 2nd ed. Columbus, OH:
Ohio State University Press.
32. Rankin, E.T, C.O. Yoder, and D. Mishne, eds. 1992. Ohio water
resource inventory, Vol. I. Summary, status, and trends, 1992.
Columbus, OH: Ohio EPA, Division of Water Quality Planning and
Assessment.
196
-------
33. Rankin, E.T., C.O. Yoder, and D. Mishne, eds. 1990. Ohio water
resource inventory, Vol. I. Summary, status, and trends, 1990.
Columbus, OH: Ohio EPA, Division of Water Quality Planning and
Assessment.
34. Rankin, E.T., ed. 1988. Water Quality Inventory1988 305(b)
report, Vol. I. Columbus, OH: Ohio Environmental Protection
Agency.
35. Hill, M.T., W.S. Platts, and R.L. Beschta. 1991. Ecological and
geomorphological concepts for instream and out-of-channel flow
requirements. Rivers 2:198-210.
36. Dickson, K.L. 1986. Neglected and forgotten contaminants affect-
ing aquatic life. Env. Tox. Chem. 5:939-940.
37. Ohio Environmental Protection Agency. 1991. Ohio nonpoint
source assessment. Columbus, OH: Ohio EPA, Division of Water
Quality Planning and Assessment.
38. Kelly, M.H., and R.L. Hite. 1984. Evaluation of Illinois stream
sediment data: 1974-1980. Springfield, IL: Illinois Environmental
Protection Agency.
39. Schaefer, J.M., and M.T. Brown. 1992. Designing and protecting
river corridors for wildlife. Rivers 3(1):14-26.
197
-------
Overview of Contaminated Sediment Assessment Methods
Diane Dennis-Flagler
U.S. Environmental Protection Agency, Great Lakes National Program Office,
Chicago, Illinois
Urban runoff has significantly contributed to the con-
tamination of lakes, rivers, and streams. After years of
accumulation in the water, toxic chemicals have found
their way to the bottom sediments. These contaminants
can be directly toxic to fish and other aquatic organisms
as well as significant sources of contaminants to wildlife.
Human health effect concerns arise primarily from con-
sumption of contaminated fish and water fowl. Assess-
ing contaminated sediments is a difficult task due to the
complex nature of the sediment matrix, contaminant
mixtures, and the physical dynamics of the waterways.
To determine the scope and extent of the sediment
contamination at a particular site, a comprehensive
sediment assessment program must be developed.
In recognition of the significance of the problem, the
Assessment and Remediation of Contaminated Sedi-
ments (ARCS) program was authorized for 6 yearsby
Congress under Section 118(c)(3) of the Water Quality
Act of 1987 and the Great Lakes Critical Program Act of
1990to develop and demonstrate new and innovative
methods both to assess and to treat contaminated sedi-
ments. The ARCS program developed an "Integrated
Contaminated Sediments Assessment Approach" for
use in the Great Lakes Areas of Concern (1). This
approach includes:
Sampling design and quality assurance
Sample collection
Chemical analysis
Toxicity testing
Benthic community structure survey
Tumors and abnormalities
These six topics are the focus of this paper.
Assessment Components
Sample Design and Collection
The ultimate goal of assessment is to determine the
scope and extent of contamination, including the mag-
nitude and spatial bounds of the problem. Assessment
needs direct sample design. Sediment sampling pro-
grams are most often undertaken to achieve one or
more of the following objectives: to fulfill a regulatory
testing requirement, to determine characteristic ambient
levels, to monitor trends in contamination levels, to iden-
tify hot spots of contamination, and to screen for poten-
tial problems. These different objectives lead to different
sampling designs. For example, a study for a dredging
project may have a specific set of guidelines on sam-
pling frequency, sample site selection methodology, and
other parameters already determined by existing, spe-
cific guidance. The design for a study to track sediment
contamination trends would expend its resources to
sample fewer sites more frequently. A study to identify
hot spots would concentrate efforts on fewer sites within
zones known to be mostly contaminated, while an initial
screening study might take few randomly distributed
samples for analysis together with some "observation"
samples to supplement the analytical results.
The most appropriate sample collection device for a
specific study depends on the study objectives, sam-
pling conditions, parameters to be analyzed, and cost.
Three general types of devices are used to collect sedi-
ment samples: dredges, grab samplers, and corers.
Core samples give byfarthe most complete information;
thus, corers should be the sampler of choice whenever
possible. Deep core sampling gives a three-dimensional
picture of the situation. This allows characterization of
the depth of contamination. Before a river or lake bottom
is dredged in an effort to remove contamination, know-
ing whether more serious contamination will be uncov-
ered is vital. All of this information guides remediation
decisions.
198
-------
The ARCS program concentrated on three levels of
sampling data:
Historical data can give some preliminary clues to
what may be present at a site. Consideration of his-
torical data can help to move the sample design proc-
ess in the proper direction. Historical data have some
limitations, however, that bear consideration. Often
data are only available for surface sediments, and
quality assurance may be in question.
Reconnaissance sampling data involve charac-
terizing a large area with "quick and dirty" screening
tests on fewer samples. This data can help eliminate
some of the parameters of concern, thus allowing
more extensive testing of toxic substances present
at the site.
Detailed assessment data involve the more extensive
chemistry and biological testing to fully characterize
a hot spot.
Chemical and Physical Analysis
Sampling efforts are performed with a variety of objec-
tives in mind. Therefore, minimal chemical and physical
parameter testing requirements vary between studies or
programs. Some chemical and physical parameters,
however, should be common to most programs unless
evidence precludes their consideration:
Particle or grain size is a physical parameter that
determines the distribution of particles. Size is impor-
tant because finer grained sediments tend to bind
contaminants more than coarse sediments do.
Total organic carbon (TOO) is an important indicator
of bioavailability for nonionic hydrophobic organic
pollutants.
Acid volatile sulfides (AVS) have been found to be
closely related to the toxicity of sediment-related as-
sociated metals.
Polyaromatic hydrocarbons (PAHs) are semivolatile
organic pollutants, several of which are potential car-
cinogens and are linked to tumors in fish.
Polychlorinated biphenyls (PCBs) are chlorinated or-
ganic compounds once used for numerous purposes,
including as a dielectric fluid in electrical transform-
ers.
Pesticides are synthetic compounds predominantly
used in agriculture to control crop-damaging insects.
Other semivolatiles include acid/base neutral com-
pounds (ABNs) such as phenols, naphthenes, and
toluenes.
Heavy metals are naturally occurring in the environ-
ment, but an excess of metals can be an indication
of anthropogenic contamination; heavy metals can be
toxic to benthic organisms.
For a typical Great Lakes site, grain size, TOC, and
AVS analyses should be done; the other five analyses
should be performed accordingly. For example, if heavy
metals in a particular area are not a problem, they could
be omitted from the scheme. Also, if certain other con-
taminants are suspected in an area, they should be
included as test parameters (e.g., tributyl tin and methyl
mercury).
Toxicity Testing
Although chemical analysis is an illuminating part of the
assessment process, chemical analysis alone does not
determine impacts. Bioavailability is key to determining
whether or not toxic contaminants will cause effects. For
example, it is possible to find a situation where high
concentrations of contaminants are present but no toxic
effects are manifested in the benthic community; in such
a situation, the contaminants may not be bioavailable to
the benthic community. In any case, further toxicity test-
ing would be required. One way to evaluate bioavailabil-
ity is by performing toxicity tests. Toxicity tests measure
the effects of sediment contamination test organisms.
Test organisms can be exposed directly to sediments
(solid phase) or to sediment slurries called elutriates.
The ARCS program evaluated over 40 toxicity tests
during the assessment program at three priority areas
of concern. Based on the results of the ARCS program,
a battery of tests should include Microtox and Daphnia
magna (7-day, three-brood survival reproduction solid
phase assay) because they are good screening assays,
relatively sensitive, discriminatory, and well correlated
with other assay responses. In addition, one or two of
the following tests should be included in the assay bat-
tery: Pimephales promelas (larval growth solid phase),
Hyalella azteca (7-day survival solid phase), Ceriodaph-
nia dubia (three-brood survival and reproduction, solid
or elutriate phase), and Hexagenia bilineata (10-day
survival and molting, solid or elutriate phase).
Benthic Community Survey
Benthic communities are communities of organisms
that live in or on sediment. In most benthic community
structure assessments, primary emphasis is placed on
determining the species that are present and the distri-
bution of individuals among those species. Information
on benthic community composition and abundance is
typically used in conjunction with information in the sci-
entific literature to infer the distribution of species and
individuals. Because sediment quality affects all major
structural and functional attributes of benthic communi-
ties in generally predictable ways, benthic community
structure assessment is a valuable tool for evaluating
sediment quality and its effects on a major biological
199
-------
component of freshwater ecosystems. Specific assess-
ment methods are available to complement the chemi-
cal and toxicological portions of the sediment quality
assessment.
Freshwater benthic macroinvertebrate communities are
used in the following ways to assess the quality of the
water resource:
Identification of the quality of ambient sites through
a knowledge of the pollution tolerances and life his-
tory requirements of benthic macroinvertebrates.
Establishment of standards based on community per-
formance at multiple reference sites throughout an
ecoregion or other regionalization categories.
Comparison of the quality of reference sites with test
sites.
Comparison of the quality of ambient sites with his-
torical data to identify temporal trends.
Determination of spatial gradients of contamination
for source characterization.
Tumors and Abnormalities
Tumors and other abnormalities are another useful
assessment tool. These abnormalities are believed to
be caused by contaminants present in the sediments,
specifically PAHs. A typical use of this type of study
would be to analyze for tumors and abnormalities
before and after cleanup to see if a change in the
incidence rate occurred. In the ARCS program, inves-
tigation of tumors and abnormalities helped to char-
acterize the different areas of concern. For example,
in the Ashtabula and Buffalo Rivers we found numer-
ous liver and external abnormalities in Brown Bull-
head, such as lip papillomas, preneoplastic lesions,
and neoplastic lesions.
Interpretation and Use of Data
All data are useless without an interpretation scheme.
Using or looking at data in isolation can lead to false
conclusions. Therefore, it is important to look at all
aspects of data using some type of integrated process
to aid decision-making.
Data Depiction
Data cannot be easily interpreted from tables. Data
need to be depicted in a visual manner, such that hot
spots, gradient depth information, and trends are evi-
dent. One way to accomplish this goal is to make a map
of the site and plot data results on the map. A three-di-
mensional map can be most useful in data depiction.
Sediment Quality Values
As stated before, the numbers obtained from chemical
testing are not very significant by themselves. If you
have a gray-area situation, in which the chemistry num-
bers are high but toxicity or biological alteration is not
necessarily evident, deciding whether this is or will be-
come a problem may be difficult. In such a case, com-
parison of one's particular program numbers with
existing numbers could give information on how to pro-
ceed. There are three general types of sediment quality
values (2):
Equilibrium partitioning is a theoretical approach that
focuses on predicting the chemical interactions between
sediments, interstitial water (i.e., the water between sedi-
ment particles), and contaminants. Chemically contami-
nated sediments are expected to cause adverse
biological effects if the predicted interstitial water
concentration for a given contaminant exceeds the
chronic water quality criterion for that contaminant.
The empirical effects-based approach (e.g., sediment
quality triad or apparent effects threshold) combines
measures of sediment chemistry, sediment toxicity,
and/or benthic infauna communities to determine the
overall sediment quality.
National status and trends is a statistical approach
that uses chemical data assembled from modeling
laboratory and field studies to determine the ranges
in chemical concentration that are rarely, sometimes,
and usually associated with toxicity.
Each approach has advantages and disadvantages.
The best approach is selected based on each programs'
particular needs.
Risk Assessment
After studying the data received from the chemistry,
toxicity, and environmental impact analysis, the final
assessment step is an evaluation of associated risk to
human, aquatic, and wildlife. What is the risk now, and
what is it potentially? This involves evaluating exposure
to and impacts resulting from contact with contaminated
sediments and media contaminated by sediment con-
taminants. If several sites are involved, a prioritization
system may be needed as a decision-making tool for
remedial actions.
The ARCS program used two levels of evaluation: base-
line and comprehensive hazard evaluations. Baseline
human health hazard evaluations were performed for all
five priority demonstration areas and were developed
from available site-specific information. The baseline
hazard evaluations described the hazards to receptors
under present site conditions. This baseline assessment
also examined all potential pathways for human expo-
sure to sediments for each given location. Comprehen-
200
-------
sive hazard evaluations were performed for the Buffalo
River and Saginaw Bay areas. Results from ARCS stud-
ies showed that consumption of contaminated fish pro-
vided the greatest risk to human health.
Conclusions
There are a number of approaches to the assessment
process. The main components are sample design,
chemical and physical analysis, biological testing and
data interpretation. Within that framework, choices are
made as to what course to follow. Regardless of which
assessment path one takes, each phase of the assess-
ment process should be carefully considered and tai-
lored to the needs and goals of that particular program.
All data must be integrated for decisions to be based on
a preponderance of evidence and to yield the most
definitive of results.
References
1. U.S. EPA. 1992. ARCS: Assessment and Remediation of Contami-
nated Sediments. 1992 work plan. Chicago, IL: Great Lakes Na-
tional Program Office.
2. U.S. EPA. 1992. Sediment classification methods compendium.
EPA/823/R-92/006. Washington, DC.
201
-------
Linked Watershed/Water-Body Model
Martin Kelly
Southwest Florida Water Management District, Tampa, Florida
Ronald Giovannelli and Michael Walters
Dames and Moore, Inc., Tampa, Florida
Tim Wool
AScI, Athens, Georgia
Abstract
With passage of the state's Surface Water Improvement
and Management (SWIM) Act of 1987, the Southwest
Florida Water Management District realized a need for
an integrated eutrophication model incorporating both a
watershed loading model and a water-body response
model. In addition, because many watershed models
depend on land use and soils mapping data, a modeling
system that could take advantage of data already stored
in the district's geographical information system (CIS)
would be useful.
This paper describes the desirable attributes of such a
modeling system, the means used to select the appro-
priate model components, the actual modeling system
developed, and an application of the model. The mod-
eling system is constructed around two U.S. Environ-
mental Protection Agency supported modelsStorm
Water Management Model (SWMM) and Water Quality
Analysis Simulation Program Model (WASP4)and is
linked to the ARC/INFO CIS. Rather than the details of
SWMM or WASP4, the paper focuses on the
SWMM/WASP Interactive Support Program (SWISP),
the interactive, menu-driven user environment that al-
lows for the easy execution of the linked water-
shed/water-body modeling system of programs. With
SWISP, the user can view and edit input data sets as
well as execute and graphically postprocess the results.
The modeling system is being tested and refined se-
quentially on three test sites. The paper presents the
results of testing to date on a specific case study: Lake
Thonotosassa, a hypereutrophic, 800-acre lake in
Hillsborough County, Florida. The objective of the mod-
eling is to allow for the assessment of various restora-
tion strategies for improving in-lake water quality. The
modeling system, which is PC based and in the public
domain, will be available for public release in the fall of
1993, along with a user's manual.
Introduction
With passage of the state's Surface Water Improvement
and Management (SWIM) Act of 1987, the Southwest
Florida Water Management District (SWFWMD) real-
ized a need for an integrated eutrophication model in-
corporating both a watershed pollutant loading model
and a water-body response model. In addition, because
many watershed models depend on land use and soils
mapping data, a modeling system that could take ad-
vantage of data already stored in the district's geo-
graphical information system (CIS) would be useful. The
stated objective of the watershed/water-body modeling
project was "to select and/or link a watershed(s) and
water-body eutrophication model for use in prioritizing
land-use management and pollution control strategies
and evaluating the effects of implementation of best
management practices (BMPs) on in-lake water quality
and natural systems."
A variety of watershed models exist that make it possi-
ble, within limited degrees of certainty, to evaluate the
effects of land-use practices on receiving waters. These
models are used to prioritize watersheds that contribute
the greatest loading to a water body. When coupled with
an appropriate model of the receiving water body, the
model system can be used to predict how changes in
land use will affect the receiving body, both in terms of
water quantity and quality.
A watershed model is an important planning tool for
evaluating the contributions from existing conditions and
projecting contributions under different scenarios. A wa-
tershed/water-body model system allows those using
them to make decisions regarding alternative land use,
202
-------
zoning, treatment, and BMP options, thus altering con-
stituent loadings to a receiving water body.
Water quality/ecological models are designed to mimic
in-waterbody dynamics as the result of inputs and to
predict trophic state or other conditions of interest.
These models allow the modeler to predict lake condi-
tions based on known or projected inputs, and thus
evaluate how changes in loading will affect the overall
health of a water body. Decisions with regard to how
much of a load reduction is required to produce desired
in-lake effects can be made, and the benefits of imple-
menting a particular corrective strategy can be as-
sessed.
From a water-body management perspective, it is desir-
able to have as a decision tool a linked model that
couples the attributes of both watershed and water-body
models. With such a model, it would be possible to
evaluate how changes in land use will, for example,
affect the trophic state (and other states) of a surface
water body.
Model Attributes
Prior to selecting a consultant, district staff developed a
list of 13 desirable attributes of a linked water-
shed/water-body model (LWWM):
Data can be input directly into the linked model from
the district's CIS (ARC/INFO) database.
The model system should consist of "off the shelf
watershed and water-body models, although some
customizing may be required. (Proprietary software
is not acceptable.)
Calibration and validation data requirements should
not be excessive.
The model can be applied to most Florida aquatic
systems with the watershed component suitable for
estuarine systems.
The model has a storm event or seasonally based
watershed component, yet it is capable of yielding
annualized values.
The output of the watershed model component
should be fully compatible with the input of the water-
body model component.
The model should be user-friendly, menu-driven, in-
teractive, and fully documented.
The water-body model considers the physical, chemi-
cal, and biological parameters and processes neces-
sary to simulate the eutrophication process and
attendant water quality conditions.
The model is sensitive to eolian, sediment, and
ground-water inputs.
The water-body model should consider the temporal
and spatial variation as required to simulate critical
water quality conditions and processes.
The model should be sensitive to trophic dynamics
and exchanges between trophic levels.
The water-body model should predict the trophic
state using existing empirical relationships already
developed for Florida lakes.
Model Selection
Dames and Moore, Inc., was selected to develop the
district's LWWM. The district also established a model-
ing technical advisory committee (TAG) composed of
various recognized modeling and CIS experts from
other agencies, academia, and private consulting firms.
The primary goal of the TAG was to aid the district and
its consultant in finalizing modeling goals and the list of
desirable model attributes to be used in an evaluation
of existing candidate models. One of the initial tasks
accomplished by Dames and Moore was a literature and
model comparison report (1) with recommended models
to be used in the proposed LWWM. This review focused
on model capabilities with regard to the overall LWWM
project objectives and did not include a rigorous inves-
tigation of the background and theory behind each
model.
Dames and Moore, following the examples of Basta and
Bower (2) and Donigian and Huber (3), developed spe-
cific evaluation criteria to objectively review candidate
models consistent with district objectives. Dames and
Moore, with the aid of the TAG and before identifying
available models, developed four criteria to be used on
a preselection basis to identify candidate models for
further consideration:
The models must have written documentation.
The models must be maintained, either formally (i.e.,
funded model caretaker) or informally (through active
use and application).
The models must be PC based or have the capability
of being easily transportable to the PC environment.
The models must be nonproprietary.
Based on the above criteria and considering district
requirements for review of certain specifically named
models, a first-cut list of candidate models was developed
followed by a final list of candidate models (Table 1).
The modeling TAG was relied on heavily to eliminate
models from further consideration and ultimately arrived
at the two selected models, SWMM and WASP4. The
rationale for eliminating certain models is detailed by
Dames and Moore (1); it was decided that the model-
ing system should rely on a single watershed model.
After considerable discussion, certain models were
203
-------
Table 1. List of Final Candidate Models Evaluated by Dames
and Moore, Inc., for Possible Incorporation in the
SWFWMD's Linked Watershed/Water-Body Model
System (1)
Watershed Models
AGNPS
ANSWERS
CREAMS
DR3M-QUAL
EPA-FHWA
EUTROMOD
GLEAMS
HSPF
NPSLAM
STORM
SWMM
SWRRB
Water-Body Models
BATHTUB
BETTER
CE-QUAL-R1
CE-QUAL-W2
HSPF
NUTRIENT LOADING/TROPHIC
STATE (EUTROMOD)
QUAL2EU
WASP4
WQRRS
eliminated because of their primarily rural or agricultural
applicability, other models were eliminated on the basis
of limited maintenance, and considerable in-house de-
bate and discussion centered on the advantages and
disadvantages of "mechanistic" versus "empirical" type
models. Despite its selection, there was concern that
SWMM was too complicated to use without extensive
training and experience and that this would affect the
desirable attribute of being user friendly and easy to
apply (or misapply); this was considered a disadvantage
common to all "mechanistic" models considered.
SWMM is primarily an urban model, and although it has
been applied in nonurban areas successfully, the ero-
sion and sedimentation capabilities are not as detailed
as most rural or agricultural models. Another disadvan-
tage of SWMM is that subbasins must be defined homo-
geneously with respect to land use for the water quality
routines, and this restriction would limit to some extent
the enhancement that could be easily afforded by a CIS
linkage (1). Similar type considerations as those men-
tioned above were used to eliminate candidate water-
body models from further consideration.
Eventually, WASP4 was selected as the appropriate
"mechanistic" model to complement the watershed load-
ing model. The TAG noted that the model was well
maintained, tested, and documented. Although identi-
fied as the most complex of the selected water-body
models, it was also the most flexible because of its
ability to simulate processes, which allows it to be used
at either a screening or predictive level depending on
the availability of data, the experience of the user, and
the objective of the application. Although flexible, the
TAG indicated that WASP4 was still perceived as being
extremely data intensive (1).
Ultimately, SWMM and WASP4 were selected because
these models were determined to be "sufficiently com-
plex to be usable for the most data intensive studies, but
have the capability of 'turning off' or 'zeroing out' com-
ponents such that the model can be made simple. The
models are public domain, and both are supported by
the EPA. In addition, full documentation is available for
both models, and they have each been well tested,
including several applications in the southwest Florida
area" (1). The models selected were not the best for
every application; however, they were considered to be
those that best met the objectives of the SWFWMD.
Linked Watershed/Water-Body Modeling
System Development
The LWWM incorporates three major environmental
modeling components:
Runoff (point and nonpoint)
Hydrodynamic/Hydraulic routing
Time variable water quality modeling
In essence, the LWWM operates as follows:
It obtains land-use and soil-type information from
ARC/INFO coded output.
It incorporates this information into the runoff compo-
nent of SWMM.
SWMM calculates event-driven runoff loads of both
point and nonpoint sources.
The LWWM uses the hydrodynamic model, RIVMOD,
to describe the longitudinal distributions of flow in the
investigated water body.
WASP4 incorporates these loads, flow distributions,
and water quality information and simulates water-
body interactions.
A schematic of the above program linkage is shown in
Figure 1.
The LWWM was developed to allow engineers and sci-
entists to rapidly evaluate the effects of both point and
nonpoint source loads on receiving waters. The LWWM
model obtains land-use information from a CIS that can
be used to swiftly generate land-use and soil-type data
for the runoff component of the LWWM system, SWMM.
The SWMM model calculates event-driven runoff loads
for both nonpoint and point sources. This time series of
loads and water quantity runoff is then used as input for
the receiving water model, WASP4 (EUTRO4). The in-
formation generated by the models will be accessible to
users via interactive graphs and other user-friendly in-
terfaces.
204
-------
Figure 1. Linked watershed/water-body model (LWWM).
Geographical Information System Interface
A CIS is a computer program used for the entry, man-
agement, analysis, and display of geographic or map-
pable information. CIS systems typically include all of
the functions of a computer-aided design (CAD) system,
as well as the powerful analytical and modeling capa-
bilities of a full-featured relational database. The power
of a CIS lies in its ability to derive problem-solving
information from existing data through such techniques
as map overlays and modeling, and to store this infor-
mation in an organized, usable form.
CIS analytical techniques are applied to generate auto-
matically the input data sets for the SWMM watershed
model. The software used for development of these data
sets is ARC/INFO, an industry-standard CIS from Envi-
ronmental Systems Research Institute (ESRI). This soft-
ware is the primary CIS platform in use at the SWFWMD
and at all other water management districts throughout
the state. Several other federal, state, regional, and
local agencies have also adopted ARC/INFO as a stand-
ard and are preparing comprehensive geographic data-
bases in this format. The SWFWMD has compiled an
extensive geographic database of the entire district in
an ARC/INFO format, including detailed coverages for
the U.S. Department of Agriculture Soil Conservation
Service (SCS) soils, land use and cover, and basin
boundaries. These data are compiled using automated
ARC/INFO techniques to generate an input data file for
the LWWM.
and most widely used urban quantity/quality models in
existence today.
SWMM simulates real storm events on the basis of
rainfall hyetographs, land use, topography and system
characterization to predict outcomes in the form of qual-
ity and quantity values. SWMM is composed of various
computational blocks that can be run as stand-alone
programs. The LWWM simplifies this process by select-
ing the appropriate blocks to run. The blocks used by
LWWM and their function are as follows:
Runoff block: Performs hydrologic and water quality
modeling with elementary hydraulic routines.
Combine block: Combines interface files to aggre-
gate results of multiple runs.
Rain block: Processes National Weather Service
(NWS) precipitation data from magnetic tape or disk.
All other computational blocks within SWMM are either
not applicable to the LWWM model or their function is
already incorporated within the LWWM (i.e., graphic and
tabular processing of output).
The LWWM model uses SWMM Version 4.2 but has
been tested successfully with older versions.
RIVMOD Implementation
RIVMOD is a dynamic numerical, hydrodynamic riverine
model that describes the longitudinal distributions of
flows in a one-dimensional water body through time.
The primary criteria for selecting RIVMOD is the need
to describe spatially varying flows in a water body
through time. The model is applicable to rivers, streams,
tidal estuaries, reservoirs, and other water bodies where
the one-dimensional assumption is appropriate.
RIVMOD solves the governing flow equations in a man-
ner that allows prediction of gradually or highly varying
flows through time and space. The model has the capa-
bility of handling flow or head as boundary conditions.
The specification of head as a boundary condition al-
lows use of the model where an open boundary is
required (e.g., an estuary or a river flowing into a lake).
Algorithms are employed in RIVMOD to allow it to pro-
vide WASP4 with flows, volumes, and water velocities.
SWMM
SWMM (4) is a comprehensive mathematical model for
the simulation of urban water quantity and quality in
storm and combined sewer system. All aspects of urban
hydrologic and water quality cycles are simulated.
SWMM was developed between 1969 and 1971 by a
consulting team under contract with the U.S. Environ-
mental Protection Agency (EPA). It was one of the first
such models and has been continually maintained and
updated. The SWMM model is perhaps the best known
WASP Implementation
The WASP4 modeling system (5) was designed to pro-
vide the generality and flexibility necessary for analyzing
a variety of water quality problems in a diverse set of
water bodies. The model considers the hydrodynamics
of large branching rivers, reservoirs, and estuaries; the
mass transport in ponds, streams, lakes, reservoirs,
rivers, estuaries, and coastal waters; and the kinetic
interactions of eutrophication-dissolved oxygen and
sediment-toxic chemicals.
205
-------
WASP4 is a dynamic compartment modeling program
for aquatic systems, including both the water column
and the underlying benthos. The time-varying processes
of advection, dispersion, point and diffuse mass loading,
and boundary exchange are represented in the basic
program. The flexibility afforded by the Water Quality
Analysis Simulation Program is unique. WASP4 permits
the modeler to structure one-, two-, and three-dimen-
sional models; allows the specification of time-variable
exchange coefficients, advective flows, waste loads,
and water quality boundary conditions; and permits tai-
lored structuring of the kinetic processes, all within the
larger modeling framework, without having to write or
rewrite large sections of computer code.
WASP4 simulates the movement and interaction of pol-
lutants within the water using two programs to simulate
two of the major classes of water quality problems:
conventional pollution (involving dissolved oxygen, bio-
chemical oxygen demand, nitrogen, phosphorus, and
eutrophication) and toxic pollution (involving organic
chemicals, metals, and sediment).
Because of WASP4's generalized framework and dy-
namic structure, it is relatively easy to link it to other
simulation models. WASP4 was modified to read loads
from an external file created by SWMM. This allows
WASP4 to update its point and nonpoint source loading
information daily.
SWMM/WASP Interactive Support Program
(SWISP)
SWISP (Figure 2) is an interactive, menu-driven user
environment that allows for the easy execution of the
LWWM system of programs. SWISP allows you to view
and edit WASP/RIVMOD/SWMM input datasets as well
as execute and postprocess the results. SWISP is the
Windows of the LWWMs; once the user executes
SWISP, the user can perform all functions related to all
the simulation models. SWISP provides file manage-
ment, which allows the user to select a file or a set of
File Brouse Edit pReProcess execute Fastrrocess Options
files to activate for manipulation and/or execution.
SWISP automatically loads the correct simulation model
based on the type of input dataset selected; upon exe-
cution of the model, SWISP provides the input data file
names that will be executed. When the simulation is
completed, SWISP is automatically reloaded so that the
results may be postprocessed.
SWMM Runoff Preprocessor (PreRUN)
The PreRUN program (Figure 3) was developed to aid
the user in the development of SWMM RUNOFF block
input datasets (SWMM Version 4.2x and higher). Pre-
RUN provides intuitive data entry forms that successful
guide the user through the development of syntactically
correct datasets. Additionally, the PreRUN program can
import a CIS file that is created before executing the
preprocessor. The CIS interface file provides soil-type
and land-use classifications to the PreRUN program so
that the user can quickly give parameters to the SWMM
Runoff block. PreRUN is designed to work with or with-
out the CIS interface file.
Land Use Delineation
ID
(No)
98881
9BBBZ
98883
9BBB4
98B85
9BB86
38607
98B88
98BB9
98B1S
9BB11
98B1Z
98B13
9BB14
988 15
98816
Area
(acres
§JU£g
117.15
6Z3. 81
748.88
911.72
535.13
76.31
B81.75
235.52
346.85
417.15
BZ3.B1
748. 88
911.72
535.13
76.31
HI lie
1
1 y.
3.46
4.Z7
25.18
12.45
48.44
13.78
8.88
34.88
25. 86
3.46
4.27
Z5.1B
12.45
48.44
13.78
a. BB
lr.1
iidenFfe ResidenLo Res idenAgricultur Indus
2
n
5.78
6.53
13.86
19.56
14.98
16.78
4.87
21.15
15.38
5.78
6.53
19. B6
19.56
14.98
16.78
4.B7
3
/.
33.78
69.86
34.54
38.14
Z7.98
»24.77
45.73
19.89
28.31
33.78
69.86
34.54
38.14
Z7.98
24.77
45.73
4
V.
49.63
9.68
17.94
21.65
7.74
37.83
28.57
21.48
38.18
49.63
9.68
17.94
21.65
7.74
37.83
28.57
5
V,
6.87
8.34
Z.28
7.56
8.53
6.65
8.88
3.31
8.24
6.87
8.34
2.28
7.56
8.53
6.65
8. 88
trial
Total
/.
99.44
89.88
99.88
99.36
99.59
99.73
78.37
99.83
99.17
99.41
89.88
99. 8B
99.36
99.59
99.73
78.37
i
!
'
i1
i!
,i
!i
i!
i,
'[
'!
!'
\
\
Sub-basin area in acres
Edit the selected data rile
Help Sound DOS Spreadsheet Exit ti 15:31:15
Figure 2. SWMM/WASP Interactive Support Program (SWISP).
Figure 3. SWMM Runoff Preprocessor (PreRun).
The power of the PreRUN preprocessor lies in its ability
to import a CIS interface file. The CIS file contains
land-use and soil classification data for user-delineated
watershed subbasins; this information is used by Pre-
RUN to develop area weighted calculations for the
SWMM model.
PreWASP Interactive Preprocessor (PreWASP)
The PreWASP program (Figure 4) aids the user in the
development of a WASP4 eutrophication input dataset.
The preprocessor provides predefined environments
(ponds, lakes, rivers, estuaries) that can be modified to
match site-specific geometries, or the user may elect to
build one from scratch. The PreWASP program allows
the user to rapidly develop an input dataset by providing
forms that can be filled out quickly using several "Quick
Fill" edit functions. The PreWASP program allows the
user to select the level of complexity at which to apply
206
-------
i'l neu.parigj
i1 File Parameterization Transport Kinetics Environment Loading Options Ip
~~
Test Input Dataset
For PreUfiSP
Number uf Systems: B
number of Segments: 29
Start Til
End Tine:
Days
Tine
1 00:00
39 90:00
Restart Option: No restart file Mass Balance Table: NH3
Message Option: Mo messages Negative Solution: N
Time Step Option: 0 Segment Uolumes:
Bed Compaction Option: Q 8.00 Hydraulic Coefficients
Print Interval:
Press for scrolling entry screen
Figure 4. PreWASP Interactive Processor (PreWASP).
the model and provides data forms that are needed to
accomplish that level of complexity.
Linked Water-Body/Watershed Postprocessor
(LWDSPLY)
The interactive graphical postprocessor LWDSPLY allows
the user to rapidly visualize the results of WASP,
RIVMOD, DYNHYD, and SWMM simulations. LWDSPLY
and SWISP are the only software needed to process the
large array of result files that can be produced from
simulations of the models contained in the LWWM.
LWDSPLY allows the user to view the results both
graphically and tabularly and has options for exporting
data to spreadsheets. LWDSPLY has the capabilities to
process more than one simulation result file at a time
(the files must be from the same model), and allows the
plotting of up to four graphs on the screen simultane-
ously. These four plots (view ports) can be manipulated
individually to show different results. As with all the
programs, context-sensitive help is available at anytime
within the program by simply pressing F1 for help or
ALT-H for a listing of the keyboard map (Figure 5).
Wind Kfl
Hydraulic KA
Sed. 02 Demand
CBOD
BODS
Ultimate BOD
Select One Variable
Fl - Help FZ - Sound Alt-H - Key Assgn. Esc - Exit Alt-D - Dos
Figure 5. Linked Water-Body/Watershed Postprocessor (LWDSPLY).
The LWDSPLY program allows the user to view (Figure
6), plot, and export information very rapidly. All simula-
tion results can be plotted or written to an ASCII text
table or exported to a spreadsheet file. LWDSPLY also
provides the algorithms for formatting the output of one
model into the input of another.
Linking SWMM to WASP4
SWMM and WASP4 are linked using the LWDSPLY
program. The linkage is generic and allows the user
to link SWMM results to either the WASP organic or
eutrophication model. This linkage is accomplished by
creating a SWMM combine block interface using the
ASCII combine block option. PreRUN is set to create
this file by default. The user must select the WASP4
(TOXI or EUTRO) model with which the SWMM file is
to be linked; this allows LWDSPLY to configure itself for
the correct output.
Once the appropriate linkage type has been selected,
the user is then required to map the appropriate SWMM
conduit IDs to WASP segments (Figure 7). Note that you
can map more than one conduit's ID to a WASP seg-
ment; LWDSPLY will combine the output. LWDSPLY will
not check any errors regarding the mapping, so the
burden is on the user to fill this table out correctly. The
figure below shows the data entry screen for the basin
to segment mapping. Note that all the conduit IDs do not
need to be mapped out to WASP segments; the user
only needs to be concerned with the conduits that affect
the water body.
Once the conduit to segment mapping has been com-
pleted, the SWMM runoff constituents must be mapped
to the WASP4 state variables. The user must map the
SWMM state variables to the WASP state variables. The
linkage allows the user to fractionate a SWMM state
variable to several WASP state variables. The example
given below shows the mapping of total nitrogen (calcu-
lated by SWMM) into three state variables of WASP's
EUTRO4 (NH3, NO3, and organic nitrogen). To accom-
plish this, the user must specify the percentage of the
total SWMM constituent runoff mass that will go into
each WASP system. This option is presented to the user
because SWMM typically calculates mass runoff for
total nitrogen and total phosphorus, while WASP needs
nitrogen loaded as ammonia, nitrate, and organic nitro-
gen, as well as phosphorus loaded as orthophosphate
and organic phosphorus. There is no error checking
done here. The percentages converted can be less than
or greater than 100 percent.
When the user is completed with the mapping func-
tions, LWDSPLY will prompt the user for a filename
to which to write the nonpoint source interface file.
WASP expects the nonpoint source files to have the ex-
tension .NPS.
207
-------
Segment 3
«
L
ul.Se+00
C8.0e-01
0
O
l.Oe-24
0.0e+001.0e+012.0e+013.0e+014.0e+015.0e+01
e
o
JE.Be+OO
It
L
3d&eff& i
SHVSfXK 2
.11
l.Oe-24
O.OetOQ1.0ei-012.0el-013.0e+014.0e+QI5.0e+01
C
0
a
*^3.9e-i-o6
gi.9e.00
0.0s
Segment 1
/EOC25
Li El Bats HEi
i
f
+001 . Oe+0 12. Oe+0 1 3.0e+0 14. Oe+0 1 5.0e+0 1
T ine
A
C
Os.Se+00
n
LB.Be+OC
*4.3e+00
g
l.Oi
CBOD
,..__^ T i no, a .
+001.4e+001.8e+002.2e+OO2.de+003.0e+00
Segnent
P^^^^^OJ^^^^Slf3j^^^^^M.^li^^^^^^^^M^li^^^ffi^^^^^Msj^ml^^^lSM.^
Figure 6. Viewing data in the LWDSPLY.
Uariable Nunbcr of Systeml Pctl SystemZ Fctl SystenZ
Nane Systens Name Name Name
FLOU CFS
Total N
Total F
Uar3
Uar4
mitTK UARIAB
DUrrti UDHIAB
wntf IJHRIHE
DUItll' UARIAB
DUItV UARIAB
DUITtl' UflRIAH
nunny UARIHB
DUMMY UARIAB
DUrttf UDRIAB
])U|i|if/ UARIflB
Mlltt.' UARIAB
Fctl
e o "** e e A
3 MM
Z OHM
g
6
e
e
e
e
e
e
e
e
e
e
e
10 ND3
15 OP
e
e
0
0
0
0
0
e
e
e
0
e
e
20 ON
65
e
e
e
e
e
e
e
e
e
e
e
e
e
78
e
e
e
0
e
0
e
e
e
e
e
e
e
e
1
j
I
1
i
Number of Uasp Systems to be napped to the Sum Uariable (max. 3)
Figure 7. Mapping SWMM conduit IDs to WASP segments.
Model System ApplicationLake
Thonotosassa, Florida
Sfr/cfy>4rea Descr/pf/on
Lake Thonotosassa is located in northeast Hillsborough
County, Florida (Figure 8). The lake has a surface area
of 813 acres, with a maximum depth of approximately
16 feet. It is tributary to the Hillsborough river system, a
source of water supply for Tampa, and a part of the
Tampa Bay ecosystem providing freshwater to the estu-
ary.
The watershed is approximately 55 square miles and
extends east to Plant City and south to Sydney (Figure
8). Elevation in the watershed ranges from 35 ft National
Geodetic Vertical Datum (NGVD) along the shoreline of
the lake to 145 ft in the eastern section of the catchment.
The area in general has relatively mild slopes but is
steeper on the eastern section when compared with the
southern and western sections. This lake was chosen in
part due to the relatively large database available as a
result of recently completed diagnostic/feasibility stud-
ies (6).
Modeling
Available data included topographic maps, land use,
soils, rainfall, wind, solar radiation, water levels, and
water quality. These data were utilized in the model
setup and calibration processes. Modeling consisted of
developing a database linkage from the CIS, watershed
modeling with the SWMM model, and water-body mod-
eling with RIVMOD and WASP4. The modeling scenar-
ios are described below.
Digitized land use and soils data were obtained from the
SWFWMD on magnetic tapes and downloaded to the
Dames and Moore ARC/INFO system. Drainage divides
that define subbasins were digitized as an additional
overlay. These data provided the basis for developing
the *.GDF file, which was linked with the SWMM model
via the PRESWMM program package. These maps
were directly output from the CIS. In addition, the CIS
was used to provide aggregate maps for soils and land
use.
The CIS identified 42 land uses at up to Level III for the
watershed. SWMM is capable of utilizing five land uses
in its watershed modeling. A decision was therefore
made to aggregate land uses to provide five classes
with similar characteristics. The classes selected were
urban, agriculture, open, wetlands, and uplands. To
maintain flexibility in redefining aggregates during the
208
-------
Railroad
Basin Boundaries
Figure 8. Lake Thonotosassa location map.
modeling process, the unaggregated GIS database
served as model input to PRESWMM. PRESWMM then
provided the aggregated land-use data for modeling
purposes.
SCS soils data on the GIS are more detailed than
required for modeling purposes. These data were ag-
gregated in the GIS to provide mapping of hydrologic
soil groups A, B, C, orD, as provided bytheHillsborough
County soils map and document (7).
The SWMM model (RUNOFF block) was used to simu-
late both water quantity and water quality inflows to the
lake. Before the input file was set up, the watershed was
segmented into 34 subbasins. The subbasins were de-
fined by examining topographic, land-use, and soils
maps.
To set up SWMM, PRESWMM was used to create an
input file consisting of information from the GIS system
and user control input (UCI). The GIS system provided
land-use and soils information, as previously discussed.
These data served as input to PRESWMM, which cre-
ated the input file for SWMM. In addition, UCIs were
input into the PRESWMM interactive program. These
UCIs include data on catchment slopes, overland Man-
nings roughness coefficient (n), evaporation, infiltration
rates, basin widths, percent of directly connecting imper-
vious area (DCIA), depression storage, channel slopes,
channel lengths, channel geometry, and channel Man-
nings roughness coefficient (n). Channel basin linkages
are also defined so that the model can route flows from
the land segment to channels, and from channels to
other channels.
After the model was set up, a data period was selected
for calibration. The period was from June 11, 1991, to
April 24, 1992, and was selected to coincide with avail-
able discharge measurement records. The model was
calibrated by conducting a series of model runs, com-
paring simulated and measured data, and adjusting pa-
rameters.
The calibration was based primarily on data collected at
two stations, LT-4 and LT-5. Station LT-5 is located on
Pemberton Creek just upstream of the Baker Creek
confluence, which represents 40 percent of the total
watershed. The other calibration point is station LT-4,
which covers 98 percent of the lake's watershed. The
difference in flows between these stations is that con-
tributed by Baker Creek draining the southern portion of
the catchment. The final calibration plot for LT-4 is
shown in Figure 9.
250-
200-
g
I/O
D 100-
50-
r
LT-4 measured
SWMM generated
*
11-Jun-91 31-Jul-91 19-Sep-91 OS-Nov-91 28-Dec-91 16-Feb-92 06-Apr-92
Date
Figure 9. Lake inflow calibration.
209
-------
The SWMM water quality setup used the same setup as
for the water quantity except that coefficients that define
buildup/washoff rates and rating curves were added to
the routine. The calibration was performed by compar-
ing water quality concentrations for measured and simu-
lated total nitrogen and total phosphorus. The procedure
was a sequence of model runs, comparing results, and
adjusting parameters.
Water-Body Modeling
Water and pollutant loading inflows generated by
SWMM were used as input to the lake, and the lake
water quality was simulated. The following two models
were used: 1) RIVMOD was used to simulate the dy-
namics of the inflows, outflows, and change storage in
the lake, and 2) EUTRO4 used the simulated hydrody-
namics and relevant quality parameters to simulate the
lake's water quality.
Sources of pollutants to the lake were identified, with
emphasis on nutrient loading. An in-lake model was
applied by utilizing ambient water quality data and flows
and pollutant loadings from the watershed to model
current in-lake processes. The model was calibrated for
nitrogen, phosphorus, and chlorophyll-a. WASP4 was
the lake model used in simulating the in-lake processes.
The lake and inlet channel was subdivided into 10 seg-
ments. Four of the segments were in the inlet channel
(Baker Creek). These segments were included to allow
some flexibility in modification, if necessary, of the nutri-
ent input to the lake during the lake water quality cali-
bration process. The lake had six segments; this was
believed to be adequate considering that there were
only two water quality data collection stations. The final
segment represents the lake outflow point. The segmen-
tation is shown in Figure 10.
The eutrophication water quality model (EUTRO4) was
set up as a system of 10 water column segments (Figure
10) to coincide with the hydrodynamic setup. Model time
step was one day, with simulation for all eight systems
of the WASP4 Intermediate Eutrophication Kinetics
package. The eight systems are ammonia, nitrate+ni-
trate, orthophosphate, chlorophyll-a, biochemical oxy-
gen demand, dissolved oxygen, organic nitrogen, and
organic phosphorus. Water column segments interact
with each other both by advective flows and diffusive
exchange.
The SWMM model generated loads of total nitrogen,
total phosphorus, and biochemical oxygen demand
(BOD). For water quality modeling, data on nitrate-ni-
trate nitrogen, organic nitrogen, ammonia nitrogen, or-
thophosphate, and organic phosphorus were required.
These constituents were estimated by applying
stoichiometric ratios obtained from the data collected
during the extensive data collection period. Loads of
#10
Legend
-£" Contours in Feet
~ #3 Cross Sections
ft) Segments
0 750 1,500 3,000
Scale in Feet
Figure 10. Lake Thonotosassa modeling segmentation and
bathymetric map.
dissolved oxygen were also included in the model.
These were obtained by applying monthly dissolved
oxygen data to SWMM simulated flows.
Seven environmental parameters were included in the
setup. The parameters defined values for salinity, segment
temperature, ammonia flux, phosphate flux, and sediment
oxygen demand. Salinity and temperature were derived
from field measurements. Some of the constants asso-
ciated with the environmental parameters were pointers
used in combination with various time functions to define
time series of water temperature, solar radiation, frac-
tion daylight hours, and wind velocity. Time series of
water temperature, solar radiation, and wind velocity
were derived from the available data discussed above.
Fraction of daylight hours was obtained from latitude-de-
pendent information presented in Chow (8).
Initial constituent concentration was based on the meas-
urements of June 26, 1991, and initial model time. Or-
ganic phosphorus was assumed to be the difference
between total phosphorus and orthophosphate. It is rec-
ognized, however, that organic phosphorus may be
overestimated because of particulate forms of inorganic
210
-------
phosphorus. Organic nitrogen was calculated from total
Kjeldahl nitrogen and ammonia.
The model was set up with the constants required for
eutrophication simulation. Values for these constants
were derived primarily from the literature (9), although
some field measurements were used as guidance to
determine constants. These constants were primarily
calibration factors.
Calibration was accomplished by adjusting constants
within reasonable limits until a satisfactory fit between
measured and simulated data was obtained (Figures 11
to 13). In some instances, although the model fit was by
no means perfect, the model was considered calibrated
within the constraints of the various estimates of inflows
and environmental parameters. Constraints were asso-
ciated with each of the eight systems in the eutrophica-
tion package: ammonia, nitrate-nitrate, orthophosphate,
phytoplankton, BOD, dissolved oxygen, organic nitro-
gen, and organic phosphorus. Ammonia, nitrate-nitrate,
and organic nitrogen are subsystems of the nitrogen
cycle; orthophosphate and organic phosphorus are
subsystems of the phosphorus cycle; and BOD and
dissolved oxygen are subsystems of the dissolved oxy-
gen balance. All systems interact.
0.04
0.035
Figure 11. Ammonia and nitrate-nitrite calibration, Lake
Thonotosassa.
0.0
Organic
Nitrogen
Organic
Phophorus
0.0
Figure 13.
BOD
Chlorophyll-a
BOD and chlorophyll-a calibration, Lake Thono-
tosassa.
Figure 12. Organic nitrogen, organic phosphorus, and PCM
calibration, Lake Thonotosassa.
Summary
The development and model components of the
LWWM system and its user environment, SWISP, have
been described. The LWWM has been applied to Lake
Thonotosassa and its watershed. Water quantity and
quality originating from the watershed were modeled as
pollutant loading to the lake. In-lake processes were
then simulated. Refinements are being made to the
LWWM system in anticipation of project completion in
September 1993. The resultant modeling system will be
tested on two other systems, one a river flowing into an
estuary (i.e., Little Manatee) and one a series of 19 inter-
connected lakes (i.e., the Winter Haven chain of lakes).
It is anticipated that the resultant modeling system will
become the district standard for eutrophication model-
ing of its surface water bodies. The final code and user's
manual for SWISP will be public domain, and it is hoped
that this modeling system will be used by other water
resource managers in developing pollutant load reduc-
tion strategies for their water bodies.
References
1. Dames and Moore, Inc. 1992. Linked watershed/water-body
model development literature review. Prepared for the Surface
Water Improvement and Management (SWIM) Department,
Southwest Florida Water Management District, Tampa, FL.
2. Basta, D.J., and B.T. Bower, eds. 1982. Analyzing natural sys-
tems. Washington, DC: Resources for the Future, Inc.
3. Donigian, A.S., Jr., and WC. Huber. 1990. Modeling of nonpoint
source water quality in urban and nonurban areas. Contract No.
68-03-3513. Prepared for the U.S. Environmental Protection
Agency, Environmental Research Laboratory, Athens, GA.
4. Huber, W.C., and R.E. Dickinson. 1988. Storm Water Manage-
ment Model, Version 4: User's Manual. Prepared for the U.S.
Environmental Protection Agency, Environmental Research
Laboratory, Athens, GA.
5. Ambrose, R.B., T.A. Wool, J.L. Martin, J.P. Connolly, and R.W
Schanz. 1991. WASP4, a hydrodynamic and water quality model:
Model theory, user's manual and programmer's guide. Prepared
for the U.S. Environmental Protection Agency, Athens, GA.
211
-------
6. Dynamac Corporation. 1992. Final report: Lake Thonotosassa 8. Chow, V. 1964. Handbook of applied hydrology. New York, NY:
diagnostic feasibility study. Prepared for the Southwest Florida McGraw-Hill.
Water Management District, Tampa, FL.
7. U.S. Department of Agriculture Soil Conservation Service. 1989. 9. U.S. Environmental Protection Agency. 1985. Rates, constants,
Soil Survey of Hillsborough County, Florida. and kinetic formulations in surface water quality modeling.
212
-------
AUTO_QI: An Urban Runoff Quality/Quantity Model With a GIS Interface
Michael L. Terstriep and Ming T. Lee
Office of Spatial Data Analysis & Information,
Illinois State Water Survey, Champaign, Illinois
Abstract
This paper describes the development and application
of the AUTO_QI model, which the authors developed at
the Illinois State Water Survey in Champaign, Illinois.
The paper includes background information on the Illi-
nois Urban Drainage Area Simulator (ILLUDAS), on
which AUTO_QI is hydrologically based. AUTO_QI
stands for AUTOmated Quality-ILLUDAS. The model is
automated in the sense that it includes an optional
geographic information system (GIS) interface using
ARC/INFO software. The Quality-ILLUDAS portion of
the name indicates that the model simulates quality as
well as quantity of runoff from an urban area.
AUTO_QI uses a continuous simulation of soil moisture
to provide reliable estimates of antecedent moisture
conditions for the simulation of selected runoff events.
The soil moisture simulation requires a continuous pre-
cipitation record for the period of interest. The user may
then specify some base rainfall above which the runoff
volume and pollutant loading are then simulated for
each event in the record. The resulting series of runoff
volumes or pollutant loadings may then undergo statis-
tical analysis. For each catchment in the study area, the
user must provide soils and land cover information as
well as buildup and washoff factors for each pollutant of
interest. The model can simulate multiple drainage out-
fall points for a given rainfall record and group the results
for different receiving waters. The user may incorporate
specific best management practices (BMPs) into the
simulation for comparison of loadings with and without
BMPs. The paper also discusses use of the GIS inter-
face including processing of remotely sensed data.
Introduction
Models for simulation of urban runoff hydrographs such
as the Illinois Urban Drainage Area Simulator (IL-
LUDAS) (1), Stormwater Management Model (SWMM)
(2, 3), and Storage, Treatment, Overflow, Runoff Model
(STORM) (4) have been used for some time. Their wide
usage reflects their reliability for stormwater drainage
design. Models that incorporate urban runoff water qual-
ity are available but are less common. The main reasons
for this are:
The water quality component is less reliable.
The models require extensive input data.
The models lack verification.
The relatively infrequent use of a water quality compo-
nent is unfortunate because urban water quality model-
ing is a convenient tool for assessing pollutant loadings.
Considering the high cost of monitoring and the lack of
extensive data for using a statistical approach, the
proper model with field data verification is a logical and
feasible method for water quality assessments.
The principal investigators have developed an approach
(5) that greatly reduces the cost of applying a determi-
nistic model Q-ILLUDAS (6) to a relatively large area.
This approach incorporates the ARC/INFO geographic
information system (GIS) for data management. The
savings comes from automation of input files. Readily
available automated data include the U.S. Geological
Survey (USGS) LUDA Level II land use data and the
U.S. Census Bureau's DIME or TIGER/LINE file for
population, housing, and street density. The streams,
soils, and other data are also available in the Illinois and
other state and federal GIS databases. This method is
very effective for simulating regional urban runoff load-
ings that involve large databases and multiple outfalls.
The model and GIS interface are known as AUTO_QI.
Literature Review
Shaw (7) describes the special hydrologic problems of
urban runoff as follows. The problem of estimating runoff
from storm rainfall depends on the character of the
catchment surface. The degree of urbanization (extent
of impervious area) greatly affects the volume of runoff
obtained from a given rainfall. Retention of rainfall by
213
-------
initial wetting of surfaces and absorption by vegetation
and pervious areas reduces the amount of storm runoff.
These surface conditions also affect the time distribution
of the runoff. Computational methods used to obtain
runoff from the rainfall should allow for the charac-
teristics of the surface area to be drained. Thus, the first
efforts in urban runoff modeling were to relate runoff
from storm rainfall to the catchment characteristics.
The first stormwater sewer design method was the ra-
tional method by Kuchling (8). Sherman (9) introduced
the unit hydrograph method. After the development of
digital computers, early urban hydrologic models were
developed, such as those by James (10), Papadakis
and Preul (11), Terstriep and Stall (1), and McPherson
and Schneider (12). One characteristic of urban runoff
is that during the early minutes of a storm, urban runoff
mainly derives from the impervious surfaces. Contribu-
tions from the pervious portion of the basin are highly
variable and more difficult to define. Other research
results may be found in Novotny and Chesters (13),
Hann et al. (14), and Shaw (7).
Many conducted early urban runoff water quality mod-
eling research, including Sartor and Boyd (15), Hydro-
logic Engineering Center (4), McPherson (16),
Sutherland and McCuen (17), U.S. EPA (18-20), and
Noel and Terstriep (6). Donigian and Huber (21) pre-
pared a comprehensive review of modeling of nonpoint
source water quality in urban and nonurban areas.
Other reviews that consider surface runoff quality mod-
els include Feldman (22), Huber and Heaney (23),
Kibler (24), Whipple et al. (25), Barnwel (26, 27), Huber
(28, 29), Bedient and Huber (30), and Viessman et al.
(31).
Table 1. Urban Runoff Quality Model
Model Authors
Year
QUAL-II
SWMM
STORM
MUNP
Q-ILLUDAS
QQS
HSPF
Hydrologic Engineering Center
Huber et al.
Hydrologic Engineering Center
Sutherland and McCuen
Noel and Terstriep
Geiger and Dorsh
Johanason et al.
1975
1975
1977
1978
1982
1980
1980
As reported by Sonnen (32), the state of the mathemati-
cal urban water quality model was fairly dismal a decade
ago. Little has changed since then because the physical
processes are so complex that they defy efforts to re-
duce them to mathematical statements. Consequently,
semiempirical methods are often used.
Deposition and Accumulation of Pollutants on
Impervious Surfaces
As described by Novotny and Chesters (13), the primary
sources of pollutants are wet and dry atmospheric depo-
sition, litter, and traffic. Pollutants deposited on the sur-
face during a dry period can be carried by wind and
traffic and accumulate near the curb or median barrier.
Thus, many studies report the street pollutant loading by
unit length of curb.
The street refuse that runoff washes to storm sewers
contains many contaminants. Significant amounts of
organics, heavy metals, pesticides, and bacteria are
commonly associated with street refuse. Factors that
affect the pollutant accumulation rates are atmospheric
fallout, wind, traffic, litter deposition, vegetation, and
particle size distribution.
Pollutant accumulation in an urban area has a signifi-
cant random component; thus, no mathematical model
yields totally reliable results. Consequently, one com-
mon concept used is the storage-input-output schematic
approach, which assumes that the amount of accumu-
lated pollutants on a surface can be described as a
simple mass balance formula:
dP/dt = A - r
where
(Eq. 1)
A = pollutant accumulation rate (Ib/day)
r = pollutant removal rate (Ib/day)
P = amount of street refuse or dust/dirt
present on the street (Ib)
t = time in days
Integrating Equation 1, then:
P = A/r [1 - exp (-rt)] + C (Eq. 2)
where
Table 1 shows a partial list of urban water quality models.
For a detailed description of each of the models, the
reader may review the respective references. This sec-
tion will limit its discussion to the deposition and accu-
mulation of pollutants on impervious surfaces and
removal of solids from the street surface.
exp = exponential function
C = undefined constant
Using the empirical data from U.S. EPA (33), the pa-
rameters were defined for the Washington, DC, area
as follows:
A = (ATMFL +LIT) (SW/2) +1.15 TD
214
-------
r= 0.00116 exp [0.0884 (TS +WS)]
C = 0
where
ATMFL = atmospheric fallout rate (g/m2/day)
LIT = litter deposition rate (g/m2/day)
SW = street width (m)
TD = traffic density (thousand axles/day)
TS = traffic speed (km/hr)
WS = wind speed (km/hr)
Sutherland and McCuen (17) made another attempt by
developing a set of refuse accumulation functions using
average daily traffic volume and pavement condition
expressed by the present serviceability index (PSI). The
results are a set of accumulation equations in terms of
these input factors.
The accumulation of street refuse is the main pollution
source in urban areas. Novotny and Chesters (13) re-
ported on typical urban street refuse. Table 2 also pre-
sents findings from research on this topic.
The Chicago results indicate that multiple-family areas
generate about three times more street dirt than single-
family areas. The commercial and industrial areas gen-
erate about five and seven times more than the
single-family areas.
The street refuse accumulation rate based on the eight
American cities (15, 35) is two to four times higher than
the Chicago dust/dirt accumulation rate. This reflects the
wide variations in pollutant accumulation rates in exist-
ing measured field data for different cities.
Refuse Washoffby Surface Runoff
When surface runoff occurs on impervious surfaces, the
splashing effect of rain droplets and the drag forces of
the flow put particles in motion. Sedimentation literature
includes many hydraulic models that are potentially ap-
plicable to the problem of particle suspension and trans-
port. Two models used frequently in urban runoff
modeling are described below.
Table 2. Street Refuse Accumulation
Yalin Equation
Of numerous equations published in the literature, the
Yalin equation (36) is probably one of the best for de-
scribing suspension and transport of particles by shal-
low flow typical for rills and street gutters. The equation
has been reported in the following form:
p = 0.635 s [1 - In (1 + as) /(as)]
(Eq. 3)
where
p = particle transport per unit width of flow
(g/m/sec)
S = (VYcr) -1
a = 2.45rs-°-4VYcr
In = natural log function
The variables are defined as follows:
Y = particle bed load tractive force =
Kps-1)gD]
ps = particle density (g/c-cm3)
Ycr = the critical tractive force at which sediment
movement begins (newton/m2)
D = particle diameter (m)
u* = sheer velocity (m/sec)
g = gravity acceleration (m/sec2)
Based on Yalin's equation, Sutherland and McCuen (17)
developed a washoff model. The model is based on the
relationship between percentage removal of total solids
in a particle range (0.001 to 1.0 mm) due to a total
rainfall volume of 1/2 in. and correlation factor Kj such
that:
TSj = Kj (TSi)
where
(Eq. 4)
TSj = percentage removal of total solids in a
particle range due to total rainfall volume j,
measured in mm
Solids Accumulation
Land Use
Single family
Multiple family
Commercial
Industrial
Chicago (34)
Dust and Dili
g/curb miles/day
10.4
34.2
49.1
68.4
Ib/acre/day
2.1a
6.8a
9.7a
13.5a
Eight American Cities
Total Solids
g/curb miles/day
48
66
69
127
(15, 35)
Ib/acre/day
9.5a
13.1a
13.7a
25.1 a
aThe curb density in Chicago and eight American cities was assumed by the authors to be 90 m/acre.
215
-------
Kj = factor relating TSj and TSi
TS = percentage removal of total solids in the
particle range due to a total rainfall of 1/2
in.
Sartor et al. Washoff Function
The Sartor et al. washoff function is based on the first-
order washoff function (15, 35):
dP/dt = - Ku r P (Eq. 5)
where
P = amount of solids remaining in
pounds
t = time in days
Ku = constant depending on street
surface characteristics (called
urban washoff coefficient)
r= rainfall intensity (in./hr)
The constant Ku was found independent of particle size
within the studied range of 10 to 1,000 urn The inte-
grated form of the equation can be expressed as:
P, = P0 [1 - exp (-Ku r t)]
(Eq. 6)
where
P0 = initial mass of solids in the curb
storage
P, = mass of material removed by rain
with duration t
exp = exponential function
In spite of the Sartor concept's highly empirical nature
and arbitrarily chosen constants, many urban runoff
models such as SWMM (2, 3) and STORM (4) have
incorporated it.
AUTO_QI Model
Model Overview
AUTO_QI actually comprises three programs known as
HYDRO, LOAD, and BMP. These programs run in se-
ries, each using output from the previous program as
input along with additional information from the user.
HYDRO performs a continuous simulation of soil mois-
ture based on a daily and hourly rainfall record that the
user provides. It also computes runoff volume for each
event above some user-specified rainfall amount. LOAD
uses these runoff volumes along with additional pollut-
ant accumulation and washoff information to calculate
pollutant loadings for each runoff event. The BMP pro-
gram then reduces these loadings in accordance with
user-specified best management practices (BMPs) and
reports the results both with and without BMP condi-
tions. The simulation process may be examined by look-
ing at wet and dry periods.
Runoff
Runoff may only occur during a "wet period," a day
during which rainfall occurs. During these potential run-
off periods, the model requires hourly rainfall amounts.
The basin is assumed to have three types of area:
directly connected paved area, supplemental paved
area, and contributing grassed area. As the name im-
plies, runoff from the directly connected paved area
flows directly to the storm system. Runoff from the sup-
plemental paved area flows first across the grassed
area and is thus subjected to infiltration losses. The
remainder of the basin is assumed to be grassed area,
so all rain falling on this surface is also subjected to
infiltration losses.
Paved Area Runoff
The model distinguishes between directly connected
paved area and supplemental paved area. The losses
from directly connected paved area consist of initial
wetting and depression storage. These losses are com-
bined and treated as an initial loss; they are subtracted
from the beginning of the rainfall pattern. After subtract-
ing these losses from the rainfall pattern, the remainder
of the rainfall appears as effective rainfall and thereby
as runoff from the paved area.
Grassed Area Runoff
Computation of grassed area runoff includes runoff from
the supplemental paved area because both are sub-
jected to infiltration. As in the case of paved area runoff,
rainfall is the primary input for grassed area runoff cal-
culations. The modifications that must change the rain-
fall pattern to grassed area runoff are much more
complex than in the paved area case. The procedure
followed here first adds in supplemental paved area
runoff, then subtracts initial and infiltration losses.
In this model, rainfall on the supplemental paved area
is simply distributed by linear weighting over the entire
grassed area, thereby modifying the actual rainfall for
grassed areas such that:
R' = R (1.0 + SPA/CGA) (Eq. 7)
where
R' = effective rainfall on the grassed area
R = actual rainfall
SPA = supplemental paved area
CGA = contributing grassed area
216
-------
In an urban basin, bluegrass turf most often covers the
area that is not paved. When rain falls on this turf, there
are two principal losses. The first is associated with
depression storage and the second with infiltration into
the soil. In this model, depression storage fills and main-
tains, and infiltration is satisfied before any runoff takes
place. Depression storage is normally considered to be
0.20 in., but the model provides for this to be varied.
The dominant and far more complex loss of rainfall on
grassed areas is caused by infiltration. The theoretical
approach to evaluating infiltration rates uses the physi-
cal properties of the soil to estimate the water storage
available in the soil mantle, and evaluates the role of this
water storage as rain water infiltrates into and through
the soil mantle. The original ILLUDAS manual provides
details of water storage in soil and infiltration rates
through soil. The following text offers only brief descrip-
tions.
Water Storage in Soil
The amount of water that the soil mantle can store
depends on the total pore space available in the soil
between the soil particles. This model divides the total
water stored in the soil mantle into two principal parts.
The first of these is gravitational water. This is the water
that will drain out of soil by gravity. The second is
evapotranspiration (ET) water. This is the water that
plants can remove through evapotranspiration.
Soil moisture storage capacity varies with soil type and
may be classed by hydrologic soil group. This model
considers seven hydrologic soil groups. The U.S. Soil
Conservation Service describes the hydrologic soil
groups as follows:
A = low runoff potential and high
infiltration rate (consists of sand
and gravel)
AB = soil having properties between soil
types A and B
B = moderate infiltration rate and
moderately well drained
BC = soil having properties between soil
types B and C
C = slow infiltration rate (may have a
layer that impedes downward
movement of water)
CD = soil having properties between soil
types C and D
D = high runoff potential and very slow
infiltration rate (consists of clays
with a permanent high water table
and high swelling potential)
Appendix B supplies default values of soil moisture
storage capacity for different soil types. Users can
change these values to suit their own experience. For
further references, see Eagleson (37) and Richey (38).
Infiltration Rate
Knowing the water storage available for infiltration within
a soil mantle makes it possible to compute the infiltration
rate at any time from the Morton equation, as given by
Chow (39):
(fo-fc)exp(-kt)
(Eq. 8)
where
fc = final infiltration rate (in./hr)
f0 = initial infiltration rate (in./hr)
k = shape factor
t = time from start of rainfall (hr)
exp = exponential function
This equation is solved by the Newton-Raphson tech-
nique for given fc and f0 values that depend on soil
properties supplied by the user. A shape factor (k) of 2
was used to provide the shape best reflecting natural
conditions.
The total amount of infiltration during a storm event
depends on the total amount of soil moisture (ET water
and gravitational water) in storage. The higher the
amount of available soil moisture, the lower the amount
of infiltration, and vice versa. This model distributes the
total amount of infiltration among ET storage and gravi-
tational storage in a preassigned 60:40 ratio. AUTO_QI
continuously simulates soil moisture so that a reliable
soil moisture is available at the beginning of any event.
During dry periods, the model operates on two different
time steps: daily if there is no rainfall on the current day
and hourly if there is rainfall at some time during the
current day. During dry periods, depression storage and
soil moisture depend on:
Evaporation, at a user supplied rate, from depression
storage.
Infiltration from depression storage, with the infiltra-
tion volume separated in a 60:40 ratio into ET water
and gravitational water.
Evapotranspiration, at a user specified rate, from ET
water storage.
Percolation, at a constant rate fc, from gravitational
water storage.
Spatial Distribution of Runoff Processes
The model assumes all of the wet period and dry period
processes are spatially distributed, and simulates by the
use of a triangular distribution. Figure 1(a) shows a
distribution assumed to vary linearly from zero to twice
217
-------
&
(a)
2DEPG
DEPG
Q.
-------
P, = load at time t
r = background removal rate
A = daily accumulation rate
Wet Periods
At the start of rainfall, the amount of a particular pollutant
on a surface that produces runoff will be Po, in Ib/acre.
Assuming that the pounds of pollutant washed off in any
time interval, dt, are proportional to the pounds remaining
on the ground, P, the first order differential equation is:
- dP/dt = kP
(Eq. 10)
When integrated, this converts into the exponential wash-
off function for the removal of the surface loads as follows:
Po - P = Po(1 - exp(-kt)) (Eq. 11)
where
P0 - p = washoff load (Ib/acre)
k = proportionality constant
t = storm duration in hours
To determine k, the model uses the same assumption
as the SWMM model. Therefore, k varies in direct pro-
portion to the rate of runoff such that:
k=iB
where
i = runoff (in./hr)
B = constant
To determine B, it was assumed that a uniform runoff of
1/2 in./hr would wash away 90 percent of the pollutant
from paved areas and 50 percent of the pollutant from
grassed areas in 1 hour. That leads to a value for B of
4.6 for paved areas and 1.4 for grassed areas. These
are default values that the user can modify.
To find the washoff load, apply each constituent's load-
ing parameters to the buildup function to determine the
initial load (by land use). Then apply the exponential
washoff equation for impervious and pervious areas.
The event mean concentration (EMC) is determined by
dividing the total washoff loads by the runoff volume for
each land use.
Best Management Practices
BMPs are the measures implemented to reduce pollutants
from source areas, or in streams and receiving waters.
Many factors govern BMP pollutant removal ability.
Schueler (40) outlined three primary interrelated factors:
The removal mechanisms used.
The fraction of the annual runoff volume that is effi-
ciently treated.
The nature of urban pollutants being removed.
The AUTO_QI model does not model specific BMP
processes but represents the effectiveness of BMPs by
a removal efficiency factor. The model can handle one
or more BMPs in a catchment or portion of a catchment.
The pollutant removal factor may be inferred from field
performance monitoring, laboratory experiments, mod-
eling analyses, or theoretical considerations. Most
model users, however, must rely on literature values as
a starting point.
The particulate related pollutants, such as sediment and
lead, are relatively easy to remove by common removal
mechanisms, such as settling. Soluble pollutants, such
as nutrients, are much more difficult to remove. The
settling mechanism has little or no effect on these pol-
lutants. Therefore, biological mechanisms, such as up-
take by bacteria, algae, rooted aquatic plants, or
terrestrial vegetation, are often used. A detailed descrip-
tion of individual BMPs can be found in Schueler (40)
and Novotny and Chesters (13).
The model allows users to test the potential enhance-
ment of water quality by implementing one or more
BMPs in a catchment or group of catchments. The user
specifies what portion, in percent, of a catchment the
desired BMPs will affect and the removal efficiency of
the BMPs. The model output lists the load and EMC
without BMPs, followed by the load and EMC expected
with BMPs. The user may apply this same procedure to
reflect existing conditions if one or more BMPs are
already in place.
Data Preparation
Interfacing the GIS Database andAUTO_QI
Urban runoff quantity and quality are highly depend-
ent on the land use and hydrologic soil type. To tabu-
late the land use/soil complex for a large basin is a
time-consuming process. To simplify the data collect-
ing process, an optional ARC Macro Language (AMI)
program was developed to retrieve the land use/soil
layers in a format suitable for model input.
The AMI includes a menu-driven data review feature
with two windows on the screen. The right window
displays an index map of the whole drainage basin
and the subbasin boundaries. The user can select a
subbasin and display the land use, soil layers, streets,
and storm sewers. If the user wants the land-use input
file of a specific subbasin, the AMI retrieves the at-
tribute data and generates an ASCII file for the model
input.
219
-------
A ML Programs
The AMI programs link and provide the user interface
between the CIS, which runs on a PRIME, and the
AUTO_QI program, which runs on a PC. These pro-
grams process the data that AUTO_QI uses and also
enable the user to view the graphic data at the subbasin
level via a menu. The programs should be used with
ESRI's ARC/INFO software on a PRIME computer and
are grouped into two functions: the preprocessor pro-
grams and the menu system programs. PREPROCES-
SORLANDSOILAML, PREPROCESSORBMPAML, and
RUNITAML are the names of the three main programs.
PREPROCESSORLANDSOIL.AML uses the soil, land
use, and BMP coverages to create a soil/land-use file
for input to the AUTO_QI model. PREPROCES-
SORBMPAML uses land-use and BMP coverages to
create BMP application files for the AUTO_QI model.
RUNITAML accesses the ARC/INFO menu system to
view the coverages and INFO data. This menu also
allows the user to choose and view individual subbasins
and their data layers.
GIS Database Layers
Soil Layer
In 1985, funding from the Illinois Department of Mines
and Minerals (IDMM) allowed for the digitization of the
statewide "General Soil Map of Illinois" for the Illinois
GIS system. This map contains 57 general soil associa-
tions in Illinois. The attribute data include the soil surface
color, surface code, and the hydrologic class (well drained,
moderately well drained, somewhat well drained, and
poorly drained). The AUTO_QI model needs this hydro-
logic soil classification for hydrologic modeling. The source
map scale for the soil associations is 1:500,000.
Land Use Layer
The statewide land-use maps are available from the
U.S. Geographical Survey LUDA digital database (41).
The land uses are classified based on LUDA Level II,
which contains 37 land-use categories (Appendix D).
Digital Landsat image data or scanning aerial photo-
graphs have updated land-use/cover information (42-
44). The Illinois State Water Survey has developed
image analysis capability using the ERDAS image proc-
essing package (45). The results of a classified land use
can easily be transferred to the ARC/INFO system.
Street Layer (DIME file/TIGER/LINE file)
Either the 1980 DIME file or the 1990 TIGER/LINE file,
which were created by the U.S. Census Bureau, can
provide the street coverage. The DIME and TIGER/LINE
files comprise street segment records. A segment is
defined as the length of a street feature between two
distinct vertices or nodes. Other features are political
boundaries and topologic features (e.g., rivers, shore-
lines, canals, railroads, airports). Additional demo-
graphic information is also available in the attribute data.
This includes state, county, and standard metropolitan
statistical area codes, aggregate family income, aggre-
gate rental cost for occupied dwelling units, and numer-
ous other demographic statistics. The data can be
plotted by census tract. The source map scale is
1:100,000. The street layer is valuable for estimating the
pollutant accumulation rate and the imperviousness of
the drainage basin.
Sewer Network
The database may also include an automated storm
sewer network. The AML menu system provides for this
coverage. The coverage is not required by the AML,
however, and is not needed by AUTO_QI.
Model Verification
Overview
Due to the lack of observed data in the Lake Calumet
area, the AUTO_QI model was verified by using the
Boneyard Creek Basin in Champaign-Urbana, Illinois.
The USGS has continuously gauged this station since
1948. The watershed area was reduced from 4.7 to 3.6
square miles in 1960 by a diversion. The basin contains
a portion of Urbana, the commercial center of Cham-
paign, and the University of Illinois campus. The central
business district of Champaign makes up 7.5 percent of
the drainage area and is nearly 100 percent impervious.
Other city properties, including predominantly residen-
tial along with some commercial and light industrial,
constitute an additional 81.2 percent of the basin. The
remaining 11.3 percent of the basin is in parks, open
space, and other land-use classes. Measurements have
found the basin to be approximately 44 percent total
paved area, which includes approximately 24 percent of
direct connected paved area, 13 percent of supplemen-
tal paved area, and 7 percent of nonconnected paved
area. The soils of the basin are predominantly Flanigan
silt loam of hydrologic class B (8).
Runoff Simulation
For runoff simulation, rainfall data for 3 years were
chosen. These years represent low (25 percent), aver-
age (50 percent), and high (75 percent) annual ex-
ceedence of rainfall. Table 3 displays these data.
Land uses in the basin were simplified into two catego-
ries. Table 4 lists the land-use parameters for these
categories which were used to verify the model.
220
-------
Table 3. Selected Years and Total Annual Rainfall
Year
1959
1976
1977
Total Rainfall (in.)
35.94
32.63
42.44
Chance of
Exceedence
(percent)
50
75
25
Comments
Average year
Dry year
Wet year
Table 4. Land-Use Parameters
USGS
Land Use
Level 2
Land Use Code
% PA % SPA DEPI (in.) DEPG (in.)
Residential
Commercial
11
12
15
90
20
5
0.1
0.1
0.1
0.1
% PA = paved area in percent
% SPA = supplemental paved area in percent
DEPI = impervious depression storage depth
DEPG = pervious depression storage depth
Results of Runoff Simulation
The events selected allowed the actual event runoff
volume to be distinguished with reasonable confidence
from the continuous runoff data. Table 5 presents the
actual events for the "average year" of 1959.
Figure 2 shows that AUTO-QI does an acceptable job
of reproducing runoff volumes for dry, average, and wet
years. The simulated runoff/rainfall ratio for these 3
years is approximately 20 percent, which is consistent
with the observed data and with what has been found
previously (1).
Water Quality Simulation
Water quality data for Boneyard Creek were available
for eight events in 1982 from a study by Bender et al.
(46). Simulated water quality data were compared with
those 1982 data.
Water Quality Parameters
Table 6 tabulates the pollutant accumulation/decay pa-
rameters required by the model and used in this study.
The accumulation rate and removal rate were selected
based on typical Midwest urban runoff basins. No at-
tempt was made to adjust these parameters to fit the
observed data.
Table 7 tabulates the comparisons between simulated
washoff and actual washoff of total suspended solids
(TSS), phosphorus (P), and lead (Pb). The results of this
verification are disappointing. They demonstrate, how-
ever, the problems of water quality simulation without
verification and calibration data. The buildup and
washoff factors in the model could be adjusted to "cali-
brate" the model to this data set and produce better
results, but that was not the intent here.
Table 6. Water Quality Parameters
ARi ARp RRi RRp
(Ib/acre/day) (Ib/acre/day) (%) (%)
For residential land use:
Suspended 7.6300
solids
Phosphorus 0.0138
Lead 0.0100
For commercial land use:
9.5500
Suspended
solids
Phosphorus
Lead
0.0100
0.0110
3.9900
0.0070
0.0053
5.5500
0.0053
0.0060
4.50
6.00
6.00
3.00
4.50
6.00
4.50
5.00
5.00
4.50
4.50
5.00
ARi = Accumulation rate for impervious area
ARp = Accumulation rate for pervious area
RRi = Removal rate for impervious area
RR = Removal rate for pervious area
Table 5. Summary of Runoff Simulation for Selected Events in 1959
Date
7/23/59
7/27/59
8/29/59
9/01/59
9/09/59
1 0/1 0/59
11/04/59
11/13/59
1 2/1 0/59
Dry Days
3.21
3.17
6.00
2.63
8.00
3.63
0.08
0.13
5.67
Rainfall (in.)
0.51
0.80
0.23
0.39
0.18
2.52
0.82
1.39
0.68
Event Duration (hr)
6.00
5.00
5.00
6.00
2.00
9.00
8.00
23.00
15.00
Observed Runoff (in.)
0.07
0.15
0.02
0.035
0.024
0.51
0.185
0.32
0.106
Simulated
Runoff (in.)
0.08
0.16
0.03
0.07
0.02
0.60
0.19
0.31
0.11
Simulated
Grass Runoff
(%)
3
4
3
1
2
7
4
2
1
221
-------
Summary
A new comprehensive computer package was developed
on the basis of two proven models for urban water
quantity/quality assessment, ILLUDAS and Q-ILLUDAS.
The package consists of three main parts:
Water quantity/quality model, called AUTO_QI.
A convenient menu system called QIMENU for pre-
paring and editing inputs, viewing the outputs, run-
ning the model, and assisting users.
A CIS interface called RUNIT, and other CIS proc-
essing programs.
The AUTO_QI model, which provides continuous simula-
tion, consists of three main components: HYDRO, LOAD,
and BMP. HYDRO uses a runoff/soil moisture account-
ing procedure, pervious and impervious depression
storage, interception, Horton infiltration curves, and
water storage in the soils to generate runoff volumes for
each event in the record. LOAD is the water quality
simulator that uses the output from HYDRO along with
the pollutant accumulation and exponent washoff func-
tions to generate loads and EMCs. BMP is the best
management practices simulator that handles numer-
ous separate or overlapping BMPs and produces the
model output. The user may simulate the impacts of
pollution reduction at multiple stormwater outfall points.
The results can be viewed at one outfall point or multiple
outfall points.
QIMENU aids users with preparation of input files, se-
lection of parameters, running the model, testing the
BMPs, and viewing the output.
The CIS interface uses the AML and automates the
generation of the major input files for AUTO_QI. It also
provides the user with a menu-driven program to review
CIS coverages on the screen.
Table 7. Washoff Load Simulation for Selected Events of 1982
TSS
The model was verified by using data from the Boneyard
Creek drainage basin in Urbana, Illinois. The three sets
of rainfall data selected represent wet, average, and dry
years. The input data consist of daily and hourly rain-
falls, percent impervious and supplemental paved ar-
eas, depression storage, initial and final infiltration rates,
gravitational and evapotranspiration soil storage, pollu-
tion accumulation and removal rate, and washoff factor.
When comparing the outputs with the observed data for
the Boneyard Creek basin, the results indicated that the
model performed well for runoff volume. The simulations
of pollutant loadings using the uncalibrated model were
poor and indicate the need for further testing and cali-
bration.
Acknowledgments
This research was funded by Region 5 Water Division,
Watershed Management Unit, EPA, Chicago, Illinois,
and the Great Lakes National Program Office, EPA,
Washington, DC. The EPA Project Officer was Thomas
E. Davenport.
The principal investigators of this report were Michael L.
Terstriep and Ming T Lee. Thomas Davenport, EPA
Regional Nonpoint Source Coordinator, reviewed the
early versions of this report and provided a number of
helpful comments and suggestions. Douglas Noel de-
veloped the program for the original Q-ILLUDAS model,
consulted on this project, and provided a general outline
for the revised computer program. M. Razeur Rahman
wrote the LOAD and BMP portion of the model. Evan P.
Mills wrote the menu-driven program QIMENU for han-
dling the inputs and outputs. Amelia V. Greene wrote the
AML program for the CIS interface. John Brother pre-
pared the graphical work.
Phosphorus
Lead
Date
3/19/82
4/02/82
4/15/82
4/16/82
5/15/82
6/15/82
6/28/82
7/18/82
Rainfall (in.)
0.52
0.66
0.12
0.60
0.43
1.17
0.98
1.14
Runoff (in.)
0.08
0.11
0.01
0.10
0.07
0.21
0.16
0.30
Sim. (Ib)
12,312
6,954
2,388
19,549
25,409
3,302
29,808
5,070
Obs. (Ib)
18,777
89,179
3,332
52,087
25,857
30,969
22,931
19,001
Sim. (Ib)
18
10
10
28
36
5
43
8
Obs. (Ib)
18
75
7
46
29
48
31
26
Sim. (Ib)
15
8
3
23
29
5
35
6
Obs. (Ib)
11
77
4
48
15
35
5
11
222
-------
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0.7
0.6
0.5"
° 0.4
ce
T3
s> 0.3
_Q
0 0.2
0.1
0
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
I
Dry Year
(1976)
I I
Average Year
(1959)
I I
Wet Year
(1977)
I
I
I
I
I
I
0.1 0.2 0.3 0.4 0.5 0.6 0.7
Computed Runoff (in.)
Figure 2. Comparison of observed and computed event
runoff volumes in Boneyard Creek basin, Champaign-
Urbana, Illinois.
References
1. Terstriep, M.L., and J.B. Stall. 1974. The Illinois Urban Drainage
Area Simulator, ILLUDAS. Illinois State Water Survey Bulletin 58.
2. Metcalf and Eddy, Inc./University of Florida/Water Resources En-
gineers, Inc. 1971. Stormwater management model, Vols. I-IV.
EPA Report No. 11024 DOC 07/71, 08/71, 09/71, 10/71. Wash-
ington, DC.
3. U.S. EPA. 1975. Stormwater management model user's manual,
Version II. EPA/670/2-75/017. Cincinnati, OH.
4. Hydrologic Engineering Center, Corps of Engineers. 1977. Urban
storm runoff. STORM Computer Program No. 723-S8-L2520.
Davis, CA.
5. Terstriep, M.L., and M.T Lee. 1989. Regional Stormwater model-
ing, Q-ILLUDAS and ARC/INFO. Proceedings of ASCE Sixth
Conference on Computing for Civil Engineering, Atlanta, GA.
6. Noel, D.C., and M.L. Terstriep. 1982. Q-ILLUDAS: Acontinuous
urban runoff/washoff model. Presented at the International Sym-
posium on Urban Hydrology, Hydraulics, and Sediment Control,
University of Kentucky, Lexington, KY.
7. Shaw, E.M. 1983. Hydrology in practice, 2nd ed. London, UK:
International Van Nostrand Reinhold.
8. Kuchling, E. 1889. The relation between the rainfall and the
discharge of sewers in populated districts. Trans. ASCE 20:1-60.
9. Sherman, L.K. 1932. Streamflow from rainfall by unit-graph
method. Engineering News Record 108:501-505.
10. James, L.D. 1965. Using a digital computer to estimate the effects
of urban development on flood peaks. Water Resour. Res.
1 (2):223-234.
11. Papadakis, C.N., and H.C. Preul. 1973. Testing of methods for
determination of urban runoff. J. Hydraul. Div, ASCE HY9:1,319-
1,335.
12. McPherson, M.B., and WJ. Schneider. 1974. Problems in mod-
eling urban watersheds. Water Resour. Res. 10(3):434-440.
13. Novotny, V, and G. Chesters. 1981. Handbook of nonpoint pol-
lution, sources, and management. New York, NY: Van Nostrand
Reinhold Environmental Engineering Series.
14. Hann C.T., H.P. Johnson, and D.L. Brakensiek, eds. 1982. Hy-
drologic modeling of small watersheds. ASAE Monograph No. 5.
St. Joseph, Ml.
15. Sartor, J.D., and G.B. Boyd. 1972. Water pollution aspects of
street surface contaminants. Report No. EPA-R2-72-081. Wash-
ington, DC.
16. McPherson, M.B. 1978. Urban runoff control, quantity, and qual-
ity. Presented at the American Public Works Association, Urban
Drainage Workshop, Omaha, NE (March).
17. Sutherland, R.C., and R.H. McCuen. 1978. Simulation of urban
and nonpoint source pollution. Water Resour. Bull. 14(2):409-
428.
18. U.S. EPA. 1988. Stormwater management model user's manual,
Version 4. EPA/600/3-88/001 a (NTIS PB88-236641/AS). Athens,
GA.
19. U.S. EPA. 1980. Quantity-Quality Simulation (QQS): A detailed
continuous planning model for urban runoff control, Vol. 1. Model
description, testing, and applications. EPA/600/2-80/011. Cincin-
nati, OH.
20. U.S. EPA. 1980. User's manual for hydrological simulation pro-
gram: FORTRAN (HSPF). EPA/600/9-80/015. Athens, GA.
223
-------
21. Donigian, A.S., Jr., and W.C. Huber. 1990. Modeling of nonpoint
source water quality in urban and non-urban areas. Draft report
to U.S. EPA, Environmental Research Laboratory, Office of Re-
search and Development, Athens, GA.
22. Feldman, A.D. 1981. HEC models for water resources system
simulation: Theory and experience. Advances in hydroscience,
Vol. 12. New York, NY: Academic Press, pp. 297-423.
23. Huber, W.C., and J.P. Heaney. 1982. Analyzing residuals genera-
tion and discharge from urban and non-urban land surfaces. In:
Basta, D.J., and B.T. Bower, eds. Analyzing natural systems,
analysis for regional residuals: Environmental quality manage-
ment resources for the future. Baltimore, MD: Johns Hopkins
University Press (NTIS PB-83-223321). pp. 121-243.
24. Kibler, D.F., ed. 1982. Urban stormwater hydrology. Water Re-
sources Monograph 7. Washington, DC: American Geophysical
Union.
25. Whipple, D.J., N.S. Grigg, T. Grizzard, C.W. Randall, R.P. Shu-
binski, and L.S. Tucker. 1983. Stormwater management in urban-
izing areas. Englewood Cliffs, NJ: Prentice-Hall.
26. Barnwell, T.O., Jr. 1984. EPA's Center for Water Quality Modeling.
Proceedings of the Third International Conference on Urban
Storm Drainage, Chalmers University, Goteborg, Sweden, Vol. 2.
pp. 463-466.
27. Barnwell, TO., Jr. 1987. EPA computer models are available to
all. Water Quality International, IAWPRC, No. 2. pp. 19-21.
28. Huber, W.C. 1985. Deterministic modeling of urban runoff quality.
In: Torno, H.C., J. Marsalek, and M. Desbordes, eds. Urban
runoff pollution. NATO ASI Series, Series G. Ecological Sciences
10:167-242. New York, NY: Springer-Verlag.
29. Huber, W.C. 1986. Modeling urban runoff quality: State of the
art. In: Urbonas, B., and L.A. Roesner, eds. Proceedings of
Conference on Urban Runoff Quality, Impact, and Quality En-
hancement Technology. New York, NY: Engineering Founda-
tions, ASCE. pp. 34-48.
30. Bedient, P.B., and W.C. Huber. 1989. Hydrology and floodplain
analysis. Reading, MA: Addison-Wesley Publishers.
31. Viessman, W, Jr., G.L. Lewis, and J.W. Knapp. 1989. Introduc-
tion to hydrology, 3rd ed. New York, NY: Harper and Row.
32. Sonnen, M.B. 1980. Urban runoff quality: Information needs. J.
Tech. Councils, ASCE 106(TC1):29-40.
33. U.S. EPA. 1975. Contribution of urban roadway usage to water
pollution. EPA/600/2-75/004. Washington, DC.
34. American Public Works Association. 1969. Water pollution as-
pects of urban runoff. WP-20-15. U.S. Department of Interior,
FWPCA (present EPA). Washington D.C.
35. Sartor, J.D., G.B. Boyd, and F.J. Agardy. 1974. Water pollution
aspects of street surface contaminant. JWPCF 46:458-667.
Washington, DC.
36. Yalin, M.S. 1963. An expression for bed load transportation. J.
Hydraul. Div., ASCE 89:221-250.
37. Eagleson, PS. 1970. Dynamic hydrology. New York, NY:
McGraw-Hill, pp. 262-264.
38. Richey, C.B. 1961. Agricultural engineer's handbook. New York,
NY: McGraw-Hill.
39. Chow, V.T. 1964. Handbook of applied hydrology. New York, NY:
McGraw-Hill, pp. 14-17.
40. Schueler, T.R. 1987. Controlling urban runoff: A practical manual
for planning and designing urban BMPs. Washington, DC: De-
partment of Environmental Programs, Metropolitan Washington
Council of Governments.
41. Anderson, J.R., E.E. Hardy, J.T Roach, and R.E. Witner. 1976.
A land use and land cover classification system for use with
remote sensor data. U.S. Geological Survey Professional Paper
964. Washington, DC: U.S. Government Printing Office.
42. Hsu, S.Y 1978. Texture-tone analysis for automated land use
mapping. Photogrammetric Engineering and Remote Sensing
44(11):1,393-1,404.
43. Lee, M.T., and Y Ke. 1990. Updating land use classifications of
urbanized areas in Northeastern Illinois by using SPOT and TM
satellite data. Illinois State Water Survey Contract Report 487.
44. Lee, M.T., J.J. Kao, and Y. Ke. 1990. Integration of CIS, remote
sensing, and digital elevation data for a hydrologic model. Pro-
ceedings of 1990 Hydraulic Division Conference, San Diego, CA
(July).
45. ERDAS. 1989. ERDAS user's manual, Version 7.2. Atlanta, GA.
46. Bender, G.M., D.C. Noel, and M.L. Terstriep. 1983. Nationwide
runoff project, Champaign, Illinois: Assessment of the impact of
urban storm runoff on an agricultural receiving stream. Illinois
State Water Survey Contract Report 319.
224
-------
Source Loading and Management Model (SLAMM)
Robert Pitt
Department of Civil Engineering, University of Alabama at Birmingham, Birmingham, Alabama
John Voorhees
Johnson Johnson & Roy/Inc., Madison, Wisconsin
Introduction
The Source Loading and Management Model (SLAMM)
was developed to more efficiently evaluate stormwater
control practices. It soon became evident that to accu-
rately evaluate the effectiveness of stormwater controls
at an outfall, the sources of the pollutants, or problem
water flows, must be known. SLAMM has evolved to
include a variety of source area and end-of-pipe controls
and the ability to predict the concentrations and loadings
of many different pollutants from many potential source
areas. SLAMM calculates mass balances for both par-
ticulate and dissolved pollutants and runoff flow volumes
for different development characteristics and rainfalls. It
was designed to give relatively simple estimates (pollut-
ant mass discharges and control measure effects) for a
very large variety of potential conditions.
SLAMM was developed primarily as a planning level
tool, for example, to generate information needed to
make planning level decisions while not generating or
requiring superfluous information. Its primary capabili-
ties include predicting flow and pollutant discharges that
reflect a broad variety of development conditions and
the use of many combinations of common urban runoff
control practices. Control practices evaluated by
SLAMM include detention ponds, infiltration devices,
porous pavements, grass swales, catchbasin cleaning,
and street cleaning. These controls can be evaluated in
many combinations and at many source areas as well
as the outfall location. SLAMM also predicts the relative
contributions of different source areas (e.g., roofs,
streets, parking areas, landscaped areas, undeveloped
areas) for each land use investigated. As an aid in
designing urban drainage systems, SLAMM also calcu-
lates U.S. Department of Agriculture Soil Conservation
Service (SCS) curve numbers (CNs) that reflect specific
development and control characteristics. These CNs
can then be used in conjunction with available urban
drainage procedures to reflect the water quantity reduc-
tion benefits of stormwater quality controls.
SLAMM is normally used to predict source area contri-
butions and outfall discharges, but SLAMM (1) has also
been used in conjunction with a receiving water model
(HSPF) to examine the ultimate effects of urban runoff.
The development of SLAMM began in the mid-1970s,
primarily as a data reduction tool for use in early street
cleaning and pollutant source identification projects spon-
sored by the U.S. Environmental Protection Agency's
(EPAs) Storm and Combined Sewer Pollution Control
Program (2-4). Much of the information contained in
SLAMM was obtained during EPAs Nationwide Urban
Runoff Program (NURP) (5), especially the early
Alameda County, California (6), and the Bellevue, Wash-
ington (7) projects. The completion of the model was
made possible by the remainder of the NURP projects
and additional field studies and programming support
sponsored by the Ontario Ministry of the Environment
(8), the Wisconsin Department of Natural Resources (9),
and EPA Region 5 (this report). Early users of SLAMM
included the Ontario Ministry of the Environment's
Toronto Area Watershed Management Strategy (TAWMS)
study (8) and the Wsconsin Department of Natural Re-
sources' Priority Watershed Program (9). SLAMM can
now be effectively used as a tool to enable watershed
planners to obtain a better understanding of the effec-
tiveness of different control practice programs.
A logical approach to stormwater management requires
knowledge of the problems that are to be solved, the
sources of the problem pollutants, and the effectiveness
of stormwater management practices that can control
the problem pollutants at their sources and at outfalls.
SLAMM is designed to provide information on the last
two aspects of this approach.
225
-------
Stormwater Problems
Before stormwater control programs can be selected and
evaluated, it is necessary to understand the stormwater
problems in local receiving waters. Table 1 lists typical
receiving water problems associated with both the long-
term accumulation of pollutants and the short-term (event-
related) buildup of pollutants. Many of these problems
have been commonly found in urban receiving waters in
many areas of the United States (10). Because these
problems are so diverse, an equally wide variety of indi-
vidual stormwater controls must usually be used together.
Unfortunately, combinations of controls are difficult to
analyze using conventional stormwater models or the
results of monitoring activities. SLAMM was developed
to effectively examine control practices and land uses
that may affect these receiving water problems.
Table 1. Typical Receiving Water Problems
Long-Term Problems
Associated With
Accumulations of
Pollutants
Short-Term Problems
Associated With
High Pollutant
Concentrations or
Frequent High Flows
(Event Related)
Sedimentation in stormwater
conveyance systems and in
receiving waters.
Nuisance algae growths from
nutrient discharges.
Inedible fish, undrinkable water,
and shifts to less sensitive aquatic
organisms caused by toxic heavy
metals and organics.
Swimming beach closures from
pathogenic microorganisms.
Water quality violations.
Property damage from increased
flooding and drainage system
failures.
Habitat destruction caused by
frequent high flow rates (e.g., bed
scour, bank erosion, flushing of
organisms downstream).
SLAMM Computational Processes
Figure 1 illustrates the development characteristics that
affect stormwater quality and quantity. This figure shows
a variety of drainage systems, from concrete curb and
gutters to grass swales, along with directly connected
roof drainage systems and drainage systems that drain
to pervious areas. "Development characteristics" define
the magnitude of these drainage efficiency attributes,
along with the areas associated with each surface type
(e.g., road surfaces, roofs, landscaped areas). The use
of SLAMM shows that these characteristics greatly af-
fect runoff quality and quantity. Land use alone is usually
not sufficient to describe these characteristics. Drainage
type (curbs and gutters or grass swales) and roof con-
nections are probably the most important attributes af-
fecting runoff quantity and quality. These attributes are
not directly related to land use, but some trends are
obvious; most roofs in strip commercial and shopping
center areas are directly connected, and the roadside
is most likely drained by curbs and gutters, for example.
Different land uses, of course, are also associated
with different levels of pollutant generation. For exam-
ple, industrial areas usually have the greatest pollutant
accumulations.
Figure 2 shows how SLAMM considers a variety of
pollutant and flow routings that may occur in urban
areas. SLAMM routes material from unconnected sources
directly to the drainage system or to adjacent directly
connected or pervious areas, which in turn drain to the
collection system. Each of these areas has pollutant
deposition mechanisms in addition to removal mecha-
nisms associated with them. As an example, uncon-
nected sources, which may include rooftops draining to
pervious areas or bare ground and landscaped areas,
are affected by regional air pollutant deposition (from
point source emissions or from fugitive dust) and other
sources that would affect all surfaces. Pollutant losses
from these unconnected sources are caused by wind
removal and rain runoff wash off, which flows directly to
the drainage system or to adjacent areas. The drainage
system may include curbs and gutters, where there is
limited deposition, and catch basins and grass swales,
which may remove substantial participates that are
transported in the drainage system. Directly connected
impervious areas include paved surfaces that drain di-
rectly to the drainage system. These source areas are
also affected by regional pollutant deposition, in addition
to wind removal and controlled removal processes, such
as street cleaning. Onsite storage is also important on
paved surfaces because of the large amount of partici-
pate pollutants that are not washed off, blown off, or
removed by direct cleaning (2, 4, 6).
Figure 3 shows how SLAMM proceeds through the ma-
jor calculations. There is a double set of nested loops in
the analyses where runoff volume and suspended solids
(particulate residue) are calculated for each source area
and then for each rain. These calculations consider the
effects of each source area control, in addition to the
runoff pattern between areas. Suspended solids washoff
and runoff volume from each individual area for each
rain are summed for the entire drainage system. The
effects of the drainage system controls (catch basins or
grass swales, for example) are then calculated. Finally,
the effects of the outfall controls are calculated.
SLAMM uses the water volume and suspended solids
concentrations at the outfall to calculate the other pol-
lutant concentrations and loadings. SLAMM keeps track
of the portion of the total outfall suspended solids load-
ing and runoff volume that originated from each source
area. The suspended solids fractions are then used to
develop weighted loading factors associated with each
pollutant. In a similar manner, dissolved pollutant con-
centrations and loadings are calculated based on the
226
-------
Figure 1. Urban runoff source areas and drainage alternatives (9).
Controlled Removal
Deposition
Wind Removal
Unconnected
Sources
Deposition
Wind Removal
Street Cleaning and Other
Washoff
Rooftops Draining to
Pervious Areas
Paved Areas Draining
to Pervious Areas
Unpaved Driveways,
Parking Lots, Streets
Undeveloped Land
Bare Ground
Landscaped Areas
i t
Controlled Removal
Directly
Connected
Impervious Areas
On-Site Storage
Catchbasin and
Sewerage Cleaning
Washoff
Rooftops Draining to
Driveways
Driveways
Sidewalks
Streets
t
Gutter and
Sewerage
System
Sedimentation
Outfall
Gutters
Paved or Sealed
Swale Ditches
Inlets, Catch basins,
Manholes
Sewerage
Figure 2. Pollutant deposition and removal at source areas (9).
Degradation
Volatilization
1
Receiving
Waters
Sedimentation
1 Urban Feeder Creeks
1 Small Rivers
1 Large Rivers
1 Ponds
1 Lakes
1 Ocean
227
-------
For Each Rain
For Each
Source Area
Enter Calculation
Module
Calculate Runoff
and Particulate
Loading
Calculate Outfall
Runoff and
Particulate
Loading
Calculate
Pollutant
Loadings
Print/File
Results
End
Figure 3. SLAMM calculation flow chart.
percentage of water volume that originates from each of
the source areas within the drainage system.
SLAMM predicts urban runoff discharge parameters (to-
tal storm runoff flow volume, flow-weighted pollutant
concentrations, and total storm pollutant yields) for
many individual storms and for the complete study pe-
riod. It has built-in Monte Carlo sampling procedures to
consider many of the uncertainties common in model
input values. This enables the model output to be ex-
pressed in probabilistic terms that represent the likely
range of results expected.
Unique Aspects of SLAMM
SLAMM is unique in many aspects. One of the most
important aspects is its ability to consider many storm-
water controls (affecting source areas, drainage sys-
tems, and outfalls) together, for a long series of rains.
Another is its ability to accurately describe a drainage
area in sufficient detail for water quality investigations
without requiring a great deal of superfluous information
that field studies have shown to be of little value in accu-
rately predicting discharge results. SLAMM also applies
stochastic analysis procedures to represent actual un-
certainty in model input parameters to better predict the
actual range of outfall conditions (especially pollutant
concentrations). The main reason SLAMM was devel-
oped, however, was because of problem areas in many
existing urban runoff models. The following paragraphs
briefly describe small storm hydrology and particulate
washoff, the most significant of these problem areas.
Small Storm Hydrology
One of the major problems with conventional stormwater
models concerns runoff volume estimates associated
with small storms. Figures 4 and 5 show the importance
of common small storms when considering total annual
pollutant discharges. Figure 4 shows the accumulative
rain count and the associated accumulative runoff volume
for a medium density residential area in Milwaukee,
Wisconsin, based on 1983 monitored data (11). This figure
shows that the median rain, by count, was about 0.3 in.,
while the rain associated with the median runoff quantity
is about 0.75 in. Therefore, more than half of the runoff
from this common medium density residential area was
associated with rain events that were smaller that
0.75 in. The 1983 rains (which were monitored during
the Milwaukee NURP project) included several very
large storms, which are also shown on Figure 4. These
100
Rain (in.)
Figure 4. Milwaukee rain and runoff distributions (medium-
density residential area).
228
-------
100
Rain (in.)
Figure 5. Milwaukee pollutant discharge distributions (medium-
density residential area).
large storms (3 to 5 in. in depth) distort Figure 4 be-
cause, on average, the Milwaukee area only can expect
one 3.5-in. storm every 5 years. If these large rains did
not occur in most years, then the significance of the
small rains would be even greater.
Figure 5 shows the accumulative loadings of different
pollutants (suspended solids, chemical oxygen demand,
phosphates, and lead) monitored during 1983 in Milwau-
kee at the same site as the rain and runoff data shown in
Figure 4 (11). When Figure 5 is compared with Figure 4,
runoff and discharge distributions appear very similar. This
is a simple way of indicating that no significant trends of
stormwater concentrations were observed fordifferentsize
events. Substantial variations in pollutant concentrations
were observed, but these were random and not related to
storm size. Similar conclusions were noted when all of the
NURP data were evaluated (5). Therefore, accurately
knowing the runoff volume is very important when studying
pollutant discharges. By better understanding the signifi-
cance and runoff generation potential of these small rains,
runoff problems will be better understood.
Figure 6 illustrates the concept of variable contributing
areas as applied to urban watersheds. This figure indi-
cates the relative significance of three major source
areas (street surfaces, other impervious surfaces, and
pervious surfaces) in an urban area. The individual flow
rates associated with each of these source areas in-
crease until their time of concentrations are met. The
flow rate then remains constant for each source area
until the rain event ends. When the rain stops, runoff
recession curves occur, draining the individual source
areas. The three component hydrographs are then
added together to form the complete hydrograph for the
Figure 6. Variable contributing areas in urban watersheds.
area. Calculating the percentage of the total hydrograph
associated with each individual source area enables
estimates of the relative importance of each source area
to be quantified. The relative pollutant discharges from
each area can then be calculated from the runoff pollut-
ant strengths associated with each area.
When the time of concentration and the rain duration are
equal for an area, the maximum runoff rate for that rain
intensity is reached (12). The time of concentration oc-
curs when the complete drainage area is contributing
runoff to the point of concern. If the rain duration ex-
ceeds the time of concentration, then the maximum
runoff rate is maintained until the rain ends. When the
rain ends, the runoff rate decreases according to a
recession curve for that surface. The example shown in
Figure 6 is for a rain duration greater than the times of
concentrations for the street surfaces and other imper-
vious areas, but shorter than the time of concentration
for the pervious areas. Similar runoff quantities origi-
nated from each of the three source areas for this ex-
ample. If the same rain intensity occurs but lasts for
twice the duration (a less frequent storm), the runoff
rates for the street surfaces and other impervious sur-
faces will be the same until the end of the rain, when
their recession curves would begin. The pervious sur-
face contribution would increase substantially, however,
because its time of concentration may be exceeded by
the longer rain duration. If the same rain intensity occurs
229
-------
but only for half of the original duration, the street sur-
faces time of concentration is barely met, and the other
impervious surfaces would not have reached their time
of concentration. In this last example, the pervious sur-
faces would barely begin to cause runoff. In this last
case, the street surfaces are the dominant source of
runoff water. By knowing the relative contributions of
water and pollutants from each source area, it is possi-
ble to evaluate potential source area runoff controls for
different rains.
Figure 7 shows monitored rainfall-runoff results from one
of a series of tests conducted to investigate runoff losses
associated with common small rains on pavement (13).
This figure indicates that initial abstractions (detention
storage plus evaporation losses) for this pavement totaled
about 1 mm, while the total rainfall losses were about
6 mm. These maximum losses occurred after about 20
mm of rain. For a relatively small rain of about 7 mm,
almost one-half of the rain falling on this pavement did
not contribute to runoff. During smaller storms, the ma-
jority of the rainfall did not contribute to runoff. These
rainfall losses for pavement are substantially greater than
commonly considered in stormwater models. Most storm-
water models use rainfall-runoff relationships that have
been developed and used for many years for drainage
design. Drainage design is concerned with rain depths
of at least several inches. When these same procedures
are used to estimate the runoff associated with common
small storms (which are the most important in water
quality investigations), the runoff predictions can be
highly inaccurate. As an example, Figure 8 is a plot of
10 15
Rain (mm)
Figure 7. Rainfall-Runoff plot (example for high-intensity rains, clean and rough streets) (13).
25
0.5
1.0
1.5 2.0 2.5
Rain Depth (in.)
3.0 3.5
4.0
Figure 8. Rainfall-Runoff plot (medium-density area with clayey soils).
230
-------
the observed runoff for different rain depths in Milwaukee
during the 1983 NURP investigations. It was noted pre-
viously that several storms were monitored during this
period that were very large. The volumetric runoff coef-
ficient (the ratio of runoff to rain depth) observed varies
for each rain depth. This ratio can be about 0.1 for
storms of about 0.5 in. but may approach 0.4 for a
moderate size storm of 2.5 in. or greater which is typi-
cally associated with drainage events. The NURP study
(5), however, recommended the use of constant (aver-
age) volumetric runoff coefficients for the stormwater
permit process. Therefore, the runoff volumes of com-
mon small storms would most likely be overpredicted.
Figure 9 shows the calculated SCS (14) CNs associated
with different storms at a medium density residential site
in Milwaukee. This figure shows that the CN values vary
dramatically for the different rain depths that actually
100
occurred at this site. The CN values approach the CN
values that would be selected for this type of site only
for rains greater than several inches in depth. The CN
values are substantially greater for the smaller common
storms, especially for rains less than the 1-in. minimum
rain criteria given by SCS (14) for the use of this proce-
dure. These results are similar to those obtained at
many other sites. In almost all cases, the CN values for
storms of less than 0.5 in. are 90 or greater. Therefore,
the smaller storms contribute much more runoff than
would typically be assumed if using SCS procedures.
The CN method was initially developed, and is most
appropriate, for use in the design of drainage systems
associated with storms of much greater size than those
of interest in stormwater quality investigations.
SLAMM makes runoff predictions using the small storm
hydrology methods developed by Pitt (13). Figure 10
0.5
1.0
3.0
3.5
1.5 2.0 2.5
Rain Depth (in.)
Figure 9. Curve number changes for different rain depths (medium-density area with clayey soils).
4.0
1.00
c
fe
cr.
3
0.10
o.c
/
1
j
x
^
S
fc ~"
\JF
0.1
a
$£
-Q-
d
3
rfi
Of:
-''I
m-
^
n y?
1
1
^
r1
\s
\
s
V
f
.
VT
1
i
1C
Observed Runoff (in.)
Figure 10. Commercial shopping center runoff verification.
231
-------
shows the verification of the small storm hydrology
method used in SLAMM for storms from a commercial
area in Milwaukee. This figure shows that the calculated
runoff for many storms over a wide range of conditions
was very close to the actual observed runoff. Figure 11
shows a similar plot of the predicted versus observed
runoff fora Milwaukee medium density residential area.
These two sites were substantially different from each
other in the amount of impervious surfaces and in the
way these areas were connected to the drainage sys-
tem. Similar satisfactory comparisons using these small
storm hydrology models for a wide range of rain events
have been made for other locations, including Portland,
Oregon (15), and Toronto, Canada (8).
10.00
Particulate Washoff
Another unique feature of SLAMM is its use of a washoff
model to predict the losses of suspended solids from
different surfaces. Figure 12 is a plot of the suspended
solids concentrations for different rain depths for sheet-
flow runoff from paved surfaces during controlled tests
in Toronto (13). This figure shows local "first-flush" ef-
fects, with a trend of decreasing suspended solids con-
centration with increasing rain depth. During the
smallest rains, these concentrations are shown to be
about several hundred milligrams per liter, and as high
as 4,000 mg/L. The suspended solids concentrations dur-
ing the largest events (about 1 in. in depth) decreased
1.00
o:
3
0.01 »
0.01
0.10 .
0.1 1
Observed Total Runoff (in.)
Figure 11. Medium-density residential area runoff verification.
Rain (mm)
Figure 12. Pavement "first-flush" suspended solids concentrations (13).
232
-------
dramatically to about 10 mg/L. These data were obtained
during controlled small storm hydrology and particulate
washoff tests using carefully controlled and constant
rain intensities. A first flush of pollutants, as seen in this
figure, is likely only to occur for relatively small homoge-
neous surfaces subjected to relatively constant rain in-
tensities. First flushes at storm drain outfalls may not be
commonly observed because of the routing of many
different individual first-flush flows that are mixed. Be-
cause the highest concentrations associated with these
individual flows reach the outfall at different times, these
individual first flushes are mixed and lost. More signifi-
cantly, later times during a rain may have much higher
periods of peak rain intensities, resulting in peak washoff
late in a storm. Intermittent periods of high rain intensi-
ties later in rains likely cause localized periods of high
runoff pollutant concentrations that may occur long after
the beginning of the rain. Therefore, first-flush situations
are most likely to occur for homogeneous drainage ar-
eas (such as for large paved areas or roofs) during
relatively constant rain intensities.
SLAMM calculates suspended solids washoff based on
individual first-flush (exponential) plots for each surface.
These plots are derived from observations during rains
and during controlled tests (8). The use of individual
surface washoff plots has been verified using runoff
observations from large and complex drainages (13).
Figures 13 through 15 show washoff plots for total sol-
ids, suspended solids (>0.45 u,m), and dissolved solids
15
(<0.45 urn) during an example controlled street surface
washoff test (13). These plots indicate the accumulative
gram per square meter washoff as a function of rain
0
QL
1
05 10 15
Rain (mm)
Figure 13. Total solids washoff test results (13).
1.3.
1.0
0.8
0.5 .
0.3
LJ- 0.2
0.15
0.1
1.2 g/m2
0 5 10 15 20
Rain (mm)
Figure 14. Dissolved solids washoff test results (13).
10 15 20
Rain (mm)
Figure 15. Suspended solids washoff test results (13).
233
-------
depth. Also shown on these figures are the total street
dirt loadings. As an example, Figure 13 shows that 13.8
g/m2 of total solids were on the street surfaces before
the controlled rain event. After about 15 mm of rain fell
on the test sites, almost 90 percent of the particulates
that would wash off (about 3 g/m2) did, similar to the
rain depth needed for "complete" washoff as reported
by earlier studies by Sartor and Boyd (16). The total
quantity of material that could possibly wash off
(about 3 g/m2), however, is a small fraction of the
total loading that was on the street (13.8 g/m2). If the
relationship between total available loading and total
loading of particulates is not considered (as in many
stormwater models), then the predicted washoff would
be greatly in error.
Figure 14 is similar to Figure 13 but shows the smallest
particle sizes ("dissolved solids," < 0.45 urn) as a function
of washoff. Here, the total loading of the filterable solids
on the streets was only about 1 g/m2, and almost all of
these small particles were available for washoff during
these rains. Figure 15 shows the washoff of largest
particles ("suspended solids," > 0.45 urn) on the street.
Here, the street loading was 12.6 g/m2, with only about
1.8 g/m2 available for washoff. The predicted washoff of
suspended solids could be in error by 700 percent if the
total loading on the street was assumed to be removable
by rains. SLAMM uses test results from Pitt (13) that
measured the washoff and street dirt loading availability
relationships for many street surfaces, rain intensities,
and street dirt loadings to more accurately predict the
amount of washoff.
Another common problem with stormwater models is
the use of incorrect particulate accumulation rates for
different surfaces. Figure 16 shows an example of the
accumulation and deposition of street surface particulates
for two residential areas monitored in San Jose, Califor-
nia (2). The two areas were very similar in land use but the
street textures were quite different. The good condition
asphalt streets were quite smooth, while the oil and screens
overlaid streets were very rough. Immediately after in-
tensive street cleaning, the rough streets still had substan-
tial particulate loadings, while the smooth streets had
substantially less. The accumulation of debris on the
streets also increased the street dirt loadings overtime.
2,500
10 20
Days Since Last Cleaned
30
Figure 16. Deposition and accumulation rates of street dirt (13).
234
-------
The accumulation rates were very similar for these two
different streets having the same land uses. The load-
ings on the streets at any given time, however, were
quite different because of the greatly different initial
loading values (permanent storage loadings). If infre-
quent street dirt loading observations are made, the true
shape of the accumulation rate curve may not be accu-
rately known. As an example, the early Sartor and Boyd
(16) test results that have been used in many stormwa-
ter models assumed that the initial loading values after
rains were close to zero, instead of the actual substantial
initial loadings. The accumulation rates were calculated
by using the slope between each individual loading
value and the origin (zero time and zero loading), rather
than between loadings from adjacent sampling times,
which can easily result in accumulation rates many
times greater than actually occurred.
The street dirt deposition rates were found to be only a
function of the land uses, but the street dirt loadings
were a function of the land use and street texture. The
accumulation rates slowly decreased as a function of
time and eventually became zero, with the loading re-
maining constant after a period of about 1 month of
either no street cleaning or no rains. Figure 16 shows
that the deposition and accumulation rates on the
streets were about the same until about 1 or 2 weeks
after a rain. If the streets were not cleaned for longer
periods, then the accumulation rate decreased because
of fugitive dust losses of street dirt to surrounding areas
by winds or vehicle turbulence. In most areas of the
United States (having rains at least every week or two),
the actual accumulation of material on street surfaces is
likely constant, with little fugitive dust losses (2).
SLAMM includes a large number of street dirt accumu-
lation and deposition rate relationships that have been
obtained for many monitoring sites throughout the
United States and Canada. The accumulation rates are
a function of the land uses, while the initial loadings on
the streets are a function of street texture. The decreas-
ing accumulation rate is also a function of the time after
a street cleaning or large rain event.
Monte Carlo Simulation of Pollutants
Strengths Associated With Runoff From
Various Urban Source Areas
Initial versions of SLAMM only used average concentration
factors for different land-use areas and source areas. This
was satisfactory for predicting the event mean concen-
trations (EMC, as used by NURP [5]) for an extended
period and for calculating the unit area loadings for differ-
ent land uses. Figure 17 is a plot of the event mean
concentrations at a Toronto test site (8). The observed
concentrations are compared with the SLAMM predicted
concentrations for a long-term simulation. All of the pre-
dicted EMC values are close to the observed EMC values.
To predict the probability distributions of the concentra-
tions, however, it was necessary to include probability infor-
mation for the concentrations found in the different source
areas. Statistical analyses of concentration data (attempt-
ing to relate concentration trends to rain depths and sea-
son, for example) from these different source areas have
not been able to explain all of the observed variations in
concentration. The statistical analyses also indicate that
pollutant concentration values from individual source areas
are distributed log normally. Therefore, log-normally dis-
tributed random concentration values are used in SLAMM
I 2
I
o
O
-2
log #/100 ml-
PO,
Phenol
-2
0 2
Model Concentration (log mg/L)
Figure 17. Observed and modeled pollutant concentrations (Toronto industrial site) (8).
235
-------
for these different areas. The results are predictions for
concentration distributions at the outfall. This can pro-
vide estimates of criteria violations for different storm-
water pollutants at an outfall for long, continuous
simulations.
An Example Analysis Using SLAMM To
Identify the Sources of Pollutants and To
Evaluate Different Control Programs
Table 2 is a field sheet that has been developed to assist
users of SLAMM to describe test watershed areas. This
sheet is used to evaluate stormwater control retrofit
practices in existing developed areas, and to examine
how different new development standards effect runoff
conditions. Much of the information on the sheet is not
actually required to operate SLAMM but is very impor-
tant when considering additional control programs, such
as public education and good housekeeping practices,
that are not quantified by SLAMM. The most important
information shown on this sheet is the land use, the type
of the gutter or drainage system, and the method of
drainage from roofs and large paved areas to the drain-
age system. The efficiency of drainage in an area, spe-
cifically if roof runoff or parking runoff drains across
grass surfaces, can be very important when determining
the amount of water and pollutants that enter the outfall
system. Similarly, the presence of grass swales in an
area may substantially reduce the amount of pollutants
and water discharged. This information is therefore re-
quired to use SLAMM.
The areas of the different surfaces in each land use are
also very important for SLAMM. Figure 18 is an example
showing the areas of different surfaces for a medium
density residential area in Milwaukee. As shown in this
Table 2. Study Area Descriptions
Location:
Date:
Photo numbers:
Site number:
Time:
Roll number:
Land-use and industrial activity:
Residential:
Low Medium
Multiple family
Trailer parks
High-rise apartments
High-density single family
Income level:
Age of development:
Institutional:
Commercial:
Industrial:
Open space:
Other:
Maintenance of building:
Heights of buildings:
Roof drains:
Roof types:
Sediment source nearby?
Treated wood near street?
Landscaping near road:
Quantity:
Type:
Maintenance:
Leaves on street:
Topography:
Street slope:
Land slope:
Traffic speed:
Low
<1930
School
Strip
Light
Undeveloped
Freeway
Excellent
1
Underground
Flat
No
No
None
Deciduous
Excessive
None
Flat (<2%)
Flat (<2%)
<25 mph
Medium
'30-'50
Hospital
Shopping center
Medium
Park
Utility ROW
Moderate
2
Gutter
Comp. shingle
Yes (describe):
Telephone poles
Some
Evergreen
Adequate
Some
Medium (2-5%)
Medium (2-5%)
25-40 mph
High
'51-70 71 -'80 New
Other (type):
Downtown Hotel Offices
Heavy (manufacturing) Describe:
Golf Cemetery
Railroad ROW Other:
Poor
3 4+ stories
Impervious Pervious
Wood shingle Other:
Fence Other:
Much
Lawn
Poor
Much
Steep (>5%)
Steep (>5%)
>40 mph
236
-------
Table 2. Study Area Descriptions (continued)
Traffic density: Light Moderate
Heavy
Parking density:
None
Light
Moderate
Heavy
Width of street:
Number of parking lanes:
Number of driving lanes:
Condition of street:
Good
Fair
Poor
Texture of street:
Smooth
Intermediate
Rough
Pavement material:
Asphalt
Concrete
Unpaved
Driveways: Paved
Condition: Good
Texture: Smooth
Gutter material: Grass swale
Condition: Good
Street/Gutter interface: Smooth
Litter loadings near street: Clean
Parking/Storage areas (describe):
Condition of pavement: Good
Texture of pavement: Smooth
Unpaved
Fair
Intermediate
Lined ditch
Fair
Fair
Fair
Fair
Intermediate
Poor
Rough
Concrete
Poor
Uneven
Dirty
Poor
Rough
Asphalt
Unpaved
Other paved areas, such as alleys and playgrounds (describe):
Condition of pavement: Good
Texture of pavement: Smooth
Fair
Intermediate
Poor
Rough
Unpaved
Notes:
6,000 -i
5,000 -
4,000 -
3,000 -
2,000 -
1 ,000 -
4-^
-^F-
-₯-
Figure 18. Source areas: Milwaukee medium-density residential areas (without alleys).
237
-------
example, streets make up between 10 and 20 percent
of the total area, while landscaped areas can make up
about half of the drainage area. The variation of these
different surfaces can be very large within a designated
area. The analysis of many candidate areas may there-
fore be necessary to understand how effective or con-
sistent the model results may be for a general land-use
classification.
Control practices evaluated by SLAMM include infiltra-
tion trenches, seepage pits, disconnections of directly
connected roofs and paved areas, percolation ponds,
street cleaning, porous pavements, catchbasin cleaning,
grass swales, and wet detention ponds. These devices
can be used singly or in combination, at source areas or
at outfalls, or, in the case of grass swales and catchbas-
ins, within the drainage system. In addition, SLAMM
provides a great deal of flexibility in describing the sizes
and other design aspects for these different practices.
One of the first problems in evaluating an urban area for
stormwater controls is the need to understand where the
pollutants of concern are originating under different rain
conditions. Figures 19 through 22 are examples for a
typical medium density residential area showing the per-
centage of different pollutants originating from different
major sources, as a function of rain depth. As an example,
Figure 19 shows the areas where water is originating.
For storms of up to about 0.1 in. in depth, street surfaces
contribute about one-half to the total runoff to the outfall.
This contribution decreased to about 20 percent for storms
greater than about 0.25 in. in depth. This decrease in
the significance of streets as a source of water is associ-
ated with an increase in water contributions from land-
scaped areas (which make up more than 75 percent of
the area and have clayey soils). Similarly, the signifi-
cance of runoff from driveways and roofs also starts off
relatively high and then decreases with increasing storm
depth. Figures 20 and 21 are similar plots for suspended
solids and lead. These show that streets contribute al-
most all of these pollutants for the smallest storms up to
about 0.1 in. The contributions from landscaped areas
then become dominant. Figure 22 shows that the contri-
butions of phosphates are more evenly distributed be-
tween streets, driveways, and rooftops for the small storms,
but the contributions from landscaped areas completely
dominate for storms greaterthan about 0.25 in. in depth.
Rain Depth (in.)
Figure 19. Flow sources for example medium-density residential area having clayey soils.
100
90
80
70
60
w
I 50
c
0
40
Cfl
S
30
20
10
Landscaped Areas
Streets
1.0
4.0
Rain Depth (in.)
Figure 20. Suspended solids sources for example medium-density residential area.
238
-------
0.01
Rain Depth (in.)
Figure 21. Total lead sources for example medium-density residential area.
100
O 70
60
50
5 30
0
en
I 20
O
I 10
Roofs
Driveways
Streets
Landscaped Areas
0.01
0.1
1.0
Rain Depth (in.)
Figure 22. Dissolved phosphate sources for example medium-density residential area.
4.0
Obviously, the specific contributions from different areas
and for different pollutants vary dramatically, depending
on the characteristics of development for the area and
the source controls used. Again, a major use of SLAMM
is to better understand the role of different sources of
pollutants. As an example, to control suspended solids,
street cleaning (or any other method to reduce the
washoff of particulates from streets) may be very effec-
tive for the smallest storms but would have very little
benefit for storms greater than about 0.25 in. in depth.
Erosion control from landscaped surfaces, however,
may be effective over a wider range of storms.
The following list shows the different control programs
that were investigated in this hypothetical medium den-
sity residential area having clayey soils:
Base level (as built in 1961 to 1980, with no additional
controls).
Catchbasin cleaning.
Street cleaning.
Grass swales.
Roof disconnections.
Wet detention pond.
Catchbasin and street cleaning combined.
Roof disconnections and grass swales combined.
All of the controls combined.
This residential area, which was based on actual Bir-
mingham, Alabama, field observations for homes built
between 1961 to 1980, has no controls. The use of
catchbasin cleaning and street cleaning in the area was
evaluated. Grass swale use was also evaluated, but
swales are an unlikely retrofit option and would only be
appropriate for newly developing areas. It is possible,
however, to disconnect some of the roof drainages and
divert the roof runoff away from the drainage system and
onto grass surfaces for infiltration in existing develop-
ments. In addition, wet detention ponds can be retrofit-
ted in different areas and at outfalls. Besides those
controls examined individually, catchbasin and street
cleaning controls combined were also evaluated, in ad-
dition to the combination of disconnecting some of the
239
-------
rooftops and the use of grass swales. Finally, the pros-
pect of using all of the controls together was examined.
The following list shows a general description of this
hypothetical area:
All curb and gutter drainage (in fair condition).
70 percent of roofs draining to landscaped areas.
50 percent of driveways draining to lawns.
90 percent of streets of intermediate texture (remain-
ing are rough).
No street cleaning.
No catchbasins.
About one-half of the driveways currently drain to land-
scaped areas, while the other half drain directly to the
pavement or the drainage system. Almost all of the
streets are of intermediate texture, and about 10 percent
are rough textured. There currently is no street cleaning
or catchbasin cleaning.
The level of catchbasin use that was investigated for this
site included 950 ft3 of total sump volume per 100 acres
(typical for this land use), with a cost of about $50 per
catchbasin cleaning. Typically, catch basins in this area
could be cleaned about twice a year for a total annual
cost of about $85 per acre of the watershed.
Street cleaning could also be used, with a monthly
cleaning effort of about $30 per year per watershed
acre. Light parking and no parking restrictions during
cleaning are assumed, and the cleaning cost is esti-
mated to be $80 per curb mile.
Grass swale drainage was also investigated. Assuming
that swales could be used throughout the area, there
could be 350 ft of swales per acre (typical for this land
use), with swales 3.5 ft wide. Because of the clayey soil
conditions, an average infiltration rate of about 0.5 in./hr
was used in this analysis based on many different dou-
ble-ring infiltrometer tests of typical soil conditions.
Swales cost much less than conventional curb and gut-
ter systems but require increased maintenance. Again,
the use of grass swales is appropriate for new develop-
ment but not for retrofitting in this area.
Roof disconnections could also be used as a control
measure by directing all roof drains to landscaped ar-
eas. The objective would be to direct all the roof drains
to landscaped areas. Because 70 percent of the roofs
already drain to the landscaped areas, only 30 percent
could be further disconnected, at a cost of about $125
per household. The estimated total annual cost would
be about $10 per watershed acre.
An outfall wet detention pond suitable for 100 acres of this
medium density residential area would have a wet pond
surface of 0.5 percent of drainage area for approximately
90 percent suspended solids control. It would need 3 ft
of dead storage and live storage equal to runoff from
1.25-in. rain. A 90-degree V notch weir and a 5-ft wide
emergency spillway could be used. No seepage or
evaporation was assumed. The total annual cost was
estimated to be about $130 per watershed acre.
Table 3 summarizes the SLAMM results for runoff vol-
ume, suspended solids, filterable phosphate, and total
lead for 100 acres of this medium density residential
area. The only control practices evaluated that would
reduce runoff volume are the grass swales and roof
disconnections. All of the other control practices evalu-
ated do not infiltrate stormwater. Table 3 also shows the
total annual average volumetric runoff coefficient (Rv)
for these different options. The base level of control has
an annual flow-weighted Rv of about 0.3, while the use
of swales would reduce the Rvto about 0.1. Only a small
reduction of Rv (less than 10 percent) would be associ-
ated with complete roof disconnections compared with
the existing situation because of the large amount of roof
disconnections that already occur. The suspended sol-
ids analyses shows that catchbasin cleaning alone
could result in about 14 percent suspended solids re-
ductions. Street cleaning would have very little benefit,
while the use of grass swales would reduce the sus-
pended solids discharges by about 60 percent. Grass
swales would have minimal effect on the reduction of
suspended solids concentrations at the outfall. (They
are primarily an infiltration device, having very little fil-
tering benefits.) Wet detention ponds would remove
about 90 percent of the mass and concentrations of
suspended solids. Similar observations can be made for
filterable phosphates and lead.
Figures 23 through 26 show the maximum percentage
reductions in runoff volume and pollutants, along with
associated unit removal costs. As an example, Figure
23 shows that roof disconnections would have a very
small potential maximum benefit for runoff volume re-
duction, at a very high unit cost compared with other
practices. The use of grass swales could have about a
60-percent reduction at minimal cost. The use of roof
disconnection plus swales would slightly increase the
maximum benefit to about 65 percent, at a small unit
cost. Obviously, the use of roof disconnections alone, or
all controlled practices combined, is very inefficient for
this example. For suspended solids control, catchbasin
cleaning and street cleaning would have minimal benefit
at high cost, while the use of grass swales would pro-
duce a substantial benefit at very small cost. If additional
control is necessary, however, the use of wet detention
ponds may be necessary at a higher cost. If close to a
95-percent reduction of suspended solids was required,
then all of the controls investigated could be used to-
gether, but at substantial cost.
240
-------
Table 3. SLAMM Predicted Runoff and Pollutant Discharge Conditions for Example3
Birmingham 1976 rains Runoff Volume Suspended Solids Filterable Phosphate
Total Lead
(112 rams, 55 in. total,
0.01-3.384 in. each)
Base (no controls)
Catchbasin cleaning:
Reduction (Ib or ft3)
Reduction (%)
Cost ($/lb or $/ft3)
($85/acre/yr)
Street cleaning:
Reduction (Ib or ft3)
Reduction (%)
Cost ($/lb or $/ft3)
($30/acre/yr)
Grass swales:
Reduction (Ib or ft3)
Reduction (%)
Cost ($/lb or $/ft3)
($minimal/acre/yr)
Roof disconnections:
Reduction (Ib or ft3)
Reduction (%)
Cost ($/lb or $/ft3)
($10/acre/yr)
Wet detention pond:
Reduction (Ib or ft3)
Reduction (%)
Cost ($/lb or $/ft3)
($130/acre/yr)
CB and street cleaning:
Reduction (Ib or ft3)
Reduction (%)
Cost ($/lb or $/ft3)
($115/acre/yr)
Roof dis. and swales:
Reduction (Ib or ft3)
Reduction (%)
Cost ($/lb or $/ft3)
($10/acre/yr)
All above controls:
Reduction (Ib or ft3)
Reduction (%)
Cost ($/lb or $/ft3)
($255/acre/yr)
Annual Flow-
ft3/acre wtg Rv
59,800 0.3
59,800 0.3
0
0
NA
59,800 0.3
0
0
NA
23,300 0.12
36,500
61
Minimal
56,000 0.28
3,800
6
0
59,800 0.3
0
0
NA
59,800 0.3
0
0
NA
20,900 0.1
38,900
65
0.00026
20,900 0.1
38,900
65
0.0066
CN Flow- Annual
Range wtg mg/L Ib/acre
77-1 00 385 1 ,430
77-100 331 1,230
200
14 14
0.43
77-100 385 1,430
0
0 0
NA
63-1 00 380 554
876
1 61
Minimal
76-1 00 41 0 1 ,430
0
-6 0
NA
77-100 49 185
1,250
87 87
0.10
77-100 331 1,230
200
14 14
0.58
63-1 00 403 526
904
-5 63
0.01
63-1 00 42 55
1,375
89 96
0.19
Flow-
wtg |ig/L
157
157
0
157
0
151
4
156
1
157
0
157
0
139
11
139
11
Annual
Ib/acre
0.58
0.58
0
0
NA
0.58
0
0
NA
0.22
0.36
62
Minimal
0.55
0.03
5
333
0.58
0
0
NA
0.58
0
0
NA
0.18
0.40
69
25
0.18
0.40
69
638
Flow-
wtg |ig/L
543
468
14
543
0
513
6
443
18
69
87
468
14
352
35
36
93
Annual
Ib/acre
2.0
1.7
0.29
14
293
2.0
0.01
0.49
3,000
0.75
1.28
63
Minimal
1.6
0.48
24
21
0.26
1.8
87
73
1.7
0.29
14
397
0.46
1.6
77
6.4
0.05
1.98
97
129
aMedium-density residential area, developed in 1961-1980, with clayey soils (curbs and gutters); new development controls (not retrofit).
241
-------
0.007
0.006
0.005
£=
% 0.004
=3
T3
0)
^ 0.003
E
55
0.002
0.001
0.000
All Controls
Roof Disconnections
Grass Swales
Roof Disc* Swales
0 10
20
30 40 50 60 70 80 90
Maximum Percentage Runoff Volume Reduction
100
Figure 23. Cost-effectiveness data for runoff volume reduction benefits.
0.7
0.6 -
0.5
cc
en ...
T3 04
I
0.3
0.1
0.0
CB& Street Cleaning
1 Catchbasin Cleaning
10
All Controls
'
Wet Detention
Grass Swales Roff Disc. & Grass Swales
I _ [** - ^ - ' - '
20 30 40 50 60 70 80
Maximum Percentage Suspended Solids Reduction
90 100
Figure 24. Cost-effectiveness data for suspended solids reduction benefits.
£=
o
t5
=3
1
-£
JZ
Q.
(f)
O
.£=
Q_
-a
1
§
a
^3
55
600
500
400
300
200
100
n
1 1 1 1 I 1 1 1 1
All Controls
_
-
_
Roof Disconnections
-
-
_
Roof Disc. & Grass Swales
Grass Swales _
i i i i i ' m ' ' '
10 20 30 40 50 60 70
Maximum Percentage Dissolved Phosphate Reduction
90 100
Figure 25. Cost-effectiveness data for dissolved phosphate reduction benefits.
242
-------
500
400
o 300
o:
S 200
100
CB & Street Cleaning
' Catchbasin Cleaning
Roof Disconnections
I * I L
Grass Swales
All Controls
Wet Detention
Roof Disc. & *
Grass Swales
i i I
10
20 30 40 50 60 70
Maximum Percentage Total Lead Reduction
90
100
Figure 26. Cost-effectiveness data for total lead reduction benefits.
Conclusions
This paper has shown how SLAMM can be used to
estimate the relative contributions of different pollutants
from different areas within a complex watershed.
SLAMM can also be used to examine the cost effective-
ness of different individual control programs, or combi-
nations of control programs, that could be located at
source areas or at the outfalls.
SLAMM is unique compared with most stormwater mod-
els. Specifically, the use of small storm hydrology to
predict the contributions of runoff from different source
areas and the use of particulate washoff algorithms have
greatly enhanced the accuracy of SLAMM. In addition,
SLAMM requires a minimum amount of information to
describe the area under consideration and engineering
design parameters for different control practices.
SLAMM is a very useful tool in guiding planners and
watershed managers in devising control strategies. It
has also been used to quantify and justify the benefits
associated with stormwater controls for newly develop-
ing areas.
References
1. Ontario Ministry of the Environment. 1986. Number River Water
Quality Management Plan, Toronto Area Watershed Management
Strategy. Toronto, Ontario.
2. U.S. EPA. 1979. Demonstration of nonpoint pollution abatement
through improved street cleaning practices. EPA/600/2-79/161.
Cincinnati, OH (August).
3. U.S. EPA. 1982. Sources of urban runoff pollution and its effects
on an urban creek. EPA/600/S2-82/090. Cincinnati, OH (Decem-
ber).
4. Pitt, R. 1984. Characterization, sources, and control of urban
runoff by street and sewerage cleaning. Contract No. R-
80597012. U.S. Environmental Protection Agency, Office of Re-
search and Development, Cincinnati, OH.
5. U.S. EPA. 1983. Final report for the Nationwide Urban Runoff
Program. Washington, DC (December).
6. Pitt, R., and G. Shawley. 1982. A demonstration of nonpoint
source pollution management on Castro Valley Creek. Alameda
County Flood Control and Water Conservation District, Hayward,
CA, for the Nationwide Urban Runoff Program, U.S. Environ-
mental Protection Agency, Water Planning Division, Washington,
DC (June).
7. Pitt, R., and P. Bissonnette. 1984. Bellevue Urban Runoff Pro-
gram, summary report. Storm and Surface Water Utility, Bellevue,
Washington (November).
8. Pitt, R., and J. McLean. 1986. Toronto Area Watershed Manage-
ment Strategy StudyHumber River Pilot Watershed Project.
Ontario Ministry of the Environment, Toronto, Ontario (June).
9. Pitt, R. 1986. Runoff controls in Wisconsin's priority watersheds.
In: Urbonas, B., and L.A. Roesner, eds. Proceedings of the
Conference on Urban Runoff QualityImpact and quality en-
hancement technology. New York, NY: American Society of Civil
Engineering (June).
10. Pitt, R. 1983. Effects of urban runoff on aquatic biota. In: Hoffman,
D., ed. Handbook of ecotoxicology. Lewis Publishers.
11. Bannerman, R., K. Baun, M. Bohn, P.E. Hughes, and D.A.
Graczyk. 1983. Evaluation of urban nonpoint source pollution
management in Milwaukee County, Wisconsin, Vol. I. Grant No.
P005432-01-5. NTIS PB 84-114164. Prepared for the U.S. Envi-
ronmental Protection Agency, Water Planning Division (Novem-
ber).
12. McCuen, R.H. 1989. Hydrologic analysis and design. Englewood
Cliffs, NJ: Prentice Hall.
13. Pitt, R. 1987. Small storm flow and particulate washoff contribu-
tions to outfall discharges. Ph.D. dissertation. Department of Civil
and Environmental Engineering, University of WisconsinMadi-
son (November).
14. U.S. Soil Conservation Service. 1986. Urban hydrology for small
watersheds. U.S. Department of Agriculture Tech. Release No.
55revised (June).
15. Sutherland, R. 1993. Portland stormwater quality using SIMPTM.
Draft report. OTAK, Inc., Lake Oswego, OR.
16. Sartor, J.D., and G.B. Boyd. 1972. Water pollution aspects of
street surface contaminants. Report No. EPA-R2-72-081. Pre-
pared for the U.S. Environmental Protection Agency (November).
243
-------
Combining GIS and Modeling Tools in the Development of a Stormwater
Management Program
Chris Rodstrom, Mohammed Lahlou, and Alan Cavacas
Tetra Tech, Inc., Fairfax, Virginia
Mow-Soung Cheng
Department of Environmental Resources, Division of Environmental Management,
Watershed Protection Branch, Prince George's County, Landover, Maryland
Abstract
A geographic information system (GIS) based Water-
shed Simulation Model (GWSM) was developed for
stormwater pollution control in Prince George's County,
Maryland, using the Stormwater Management Model
(SWMM 4.2), ARC/INFO (6.1), and data postproces-
sors. The GWSM was designed to perform planning
level assessments of water quality concentrations and
loadings for 12 water quality parameters in 41 primary
watersheds within the county. The model combines con-
tinuous watershed modeling and the spatial analysis
capabilities of a GIS in a single, integrated system op-
erating on a Sun Spare 2 workstation. The user selects
a watershed to determine daily, monthly, seasonal, or
annual stormwater pollutant loadings using the SWMM
output. Additional routines analyze stormwater control
structures and user-defined subbasins. GWSM output
is saved for watershed comparisons using both graphi-
cal and tabular formats.
GWSM allows county water resources planners to per-
form analyses in the following areas:
Prioritize problem watersheds: Identify where impacts
are most severe based on pollutant-specific data. Both
temporal and spatial problems and trends are identified.
Integration with water quality databases: Data from
national databases, including STORET, WATSTORE,
and Reachfile III streams, are used in characterizing
the water resources of the study area.
Alternative land use assessment: Water quality im-
pacts and trends, based on land use changes or
future master planning scenarios, can be evaluated.
Screening solutions/microscale analysis: Management
assessment tools provide planning level screening of
controls designed to cost-effectively manage the pol-
lutants of concern. This allows determination of which
flows and loads need to be controlled. Smaller, 100-
to 400-acre drainage basins can also be evaluated
with alternative land uses and management prac-
tices.
Introduction
The National Pollutant Discharge Elimination System
(NPDES) municipal stormwater permit regulations re-
sulting from the Clean Water Act Reauthorization of
1987 require large counties and municipalities to de-
velop comprehensive stormwater management pro-
grams. For complex urban fringe areas such as Prince
George's County, Maryland, prioritizing stormwater
problems and developing cost-effective management
techniques is a primary objective if program resources
are to be efficiently allocated. The geographic informa-
tion system (GIS) based Watershed Simulation Model
(GWSM) was designed to support the development and
implementation of the county's stormwater manage-
ment program. GWSM enables planning assessment at
the watershed level through estimation of pollutant loads
and flows for current land use conditions and future
buildout scenarios, with or without structural controls. At
the small basin level, alternative stormwater control sce-
narios can be evaluated for user-defined areas.
Existing Watershed Models
A variety of models are available for simulating water
quantity and quality on a watershed scale (1). These range
from relatively simple empirical models that predict annual
or storm loads to deterministic models that yield flow and
pollutant loads for a variety of flow conditions.
244
-------
Simple models, such as the U.S. Environmental Protec-
tion Agency (EPA) Screening Procedures (2) model and
the Simple Method (3), commonly aggregate the physi-
cal parameters for an entire watershed and calculate
loads on an annual or seasonal time step. Although this
reduces the amount of input data and time required to
apply the model, it does not allow for an examination of
the variations between storm events or water quality prob-
lems occurring over a wide range of hydrologic conditions.
Complex models, such as SWMM (4) and the Hydro-
logic Simulation Program-Fortran (HSPF) (5), simulate
hydrologic processes that generate runoff and pollutant
loads in a continuous manner rather than relying on
simplified rates of change (1). This class of model can
use time series climatic data for continuous simulation
over several years, enabling analysis of not only an-
nual loads and flows but also of single events or a
series of storms.
Previous Studies
CIS is increasingly used for watershed assessment in
support of various water resources programs (6). A re-
view of available literature shows that the use of CIS in
conjunction with hydrologic models comprises a major
part of the current activities. The use of CIS for hydro-
logic modeling can be divided into two general ap-
proaches: 1) performing watershed modeling analyses
directly within a CIS package using empirical or lumped
models and 2) processing input data for use with a
separate or partially linked watershed model.
Empirical modeling within a CIS environment includes an
approach using the modified Universal Soil Loss Equa-
tion (USLE) for evaluating silvicultural activities and con-
trol programs in Montana (7). Tim et al. (8) coupled
empirical simulation modeling with a CIS to identify
critical areas of nonpoint source pollution in Virginia. On
the other hand, linked CIS and hydrologic modeling
approaches include a study by Ross and Tara (9)
using a CIS to perform spatial data referencing and
data processing while traditional hydrologic codes per-
formed the calculations for time-dependent surface-
and ground-water simulations. Terstriep and Lee (10)
developed AUTO_QI, a CIS-based interface for wa-
tershed delineation and input data formatting to the
Q-ILLUDAS model.
Modeling Approach: The Prince George's
County CIS-Based Watershed Simulation
Model
The GWSM developed for the Prince George's County
stormwater program combines results from a watershed
model with CIS analytic routines. Figure 1 illustrates this
modular modeling approach. The GWSM uses a con-
tinuous simulation model to generate single land-use
water quality and quantity time series data. ARC/INFO,
Continuous
Simulation Model
Single Land-Use Water
Quality Time Series
Figure 1. Watershed modeling approach.
combining CIS coverages from various databases, is
used to select a watershed and determine its physical
characteristics, including drainage area and land-use
distribution. The single land-use time series, along with
the land-use and drainage area files, are processed by
a series of Fortran routines to determine watershed
loads and summary statistics (Figure 1). Results can be
interactively displayed for watershed comparisons and
management assessment. As with AUTO_QI (10), the
GWSM modeling approach uses the CIS to furnish data
for use with a continuous simulation model. Unlike other
approaches, however, GWSM uses preprocessed out-
put from a watershed model to calculate storm flows and
pollutant loads for the study watershed.
Although SWMM was used for this application of
GWSM, results from other continuous simulation mod-
els can also be included in the model. This modular
approach enables increased simulation accuracy as
calibrated models become available. Further, several
models can be used within a single application, combin-
ing the strengths of each. For instance, SWMM could
be used for urban areas, while HSPF could be applied
to agricultural lands within a single study area.
Input Data Requirements
GWSM requires both ARC/INFO vector coverages
and continuous simulation model output for each
land-use type modeled. Coverages include watershed
boundaries and current land-use files. Input data for
SWMM include parameters for the rainfall, tempera-
ture, and runoff blocks for each of the nine homoge-
nous land-use basins.
245
-------
A Case Study: Collington Branch
Watershed
Water resource managers face multiple questions on
how best to manage stormwater on a regional, water-
shed, and subbasin scale. In Prince George's County,
an area covering over 480 square miles, there are 41
watersheds of varying size and land-use distribution.
The proximity of the county to the fast-growing metro-
politan Washington, DC, area makes stormwater man-
agement a complex and pressing problem.
An assessment of the predominantly forested and agri-
cultural Collington Branch watershed, covering approxi-
mately 14,820 acres and draining to the Western Branch
and to the Patuxent River, was performed as a demon-
stration of the GWSM. Figure 2 is the watershed selec-
tion screen from the GWSM, including the land-use
distribution for the Collington Branch watershed. This
case study demonstrates how the GWSM can be ap-
plied using a three-step approach: 1) identify and target
problem watersheds, 2) identify pollutant sources and
characterize pollutants, and 3) conceptually identify
control measures and evaluate future land-use
changes.
Watershed Problem Identification and
Targeting
Often, the first questions that water resource managers
ask are, How can problem watersheds be identified, and
how do watersheds compare with each other in terms
of pollutant loads and flows? GWSM enables the rapid
analysis of the relative contributions of each watershed
to the total load, performing a complete assessment and
interpretation of the data within 10 minutes. The results
: Low Density Residential
Medium Density Residential
High Density Residential
a== Commercial
^= Industrial
Open Space
Agriculture
Forest
Barren Land
Figure 2. Watershed selection screen for the Collington Branch watershed, including land-use distribution.
246
-------
include estimates of annual, mean monthly, and monthly
loads for the watershed for 12 parameters. Each con-
stituent may be viewed either as a percentage of the
total load or in actual units (pounds or cubic feet). Figure
3 presents the graphical display from GWSM showing
the total nitrogen load for the Collington Branch, illus-
trating the changes in loads due to climatic variability.
A comparison between two watersheds is easily per-
formed to assess load and flow estimates and review
results graphically. Multiple applications provide a rapid
assessment of all the major watersheds in the county.
This phase of the GWSM analysis provides information
to answer the questions, Which are the likely water
quality impacts, and how significant are they compared
to other watersheds?
Identify Pollutant Sources and Characterize
Pollutant of Concern
Once problem watersheds are identified and targeted
for further analysis, the water quality problems must be
clearly defined. What are the sources of the pollutants
of concern? An analysis of the pollutant contribution by
land use is included in GWSM, calculating constituent
load by land use for each of the 12 parameters. Figure
4 shows total nitrogen contributions for each land use in
the Collington Branch, indicating that agricultural areas
are the primary source. This provides important infor-
mation for targeting control programs throughout the
watershed. Characterizing the pollutant loads is an im-
portant issue for developing management programs.
The following questions are answered at this phase:
What pollutants are of primary concern? What are their
sources and spatial and temporal characteristics? How
do their loads vary seasonally and annually? What are
the temporal variations between pollutants? To answer
these questions, GWSM provides graphical displays of
mean monthly, mean annual, and annual pollutant loads
for each pollutant.
Management Screening
In this phase, implementing the most cost-effective con-
trols is addressed. To address control measures, the
relationship between storm size, runoff volume, and
pollutant load must be assessed. For example, what
storm sizes contribute the largest pollutant loads, and
which storm size should be targeted? The analytic rou-
tines in GWSM provide graphical answers to these
questions. Figure 5 presents lead loads by storm size,
indicating that targeting only a percentage of runoff will
control over half of the total load. Figure 6 illustrates the
rainfall/runoff characteristics of the watershed, with the
majority of storms generating less than 0.05 in. of runoff.
These estimates will vary by watershed and type of
pollutant, but GWSM allows rapid analysis of each pol-
lutant and multiple watersheds.
Management evaluation is done at both watershed- and
site-specific levels. Over an entire watershed, what is
the optimal control level for structural water quality fa-
cilities? GWSM includes a stormwater pond routine that
calculates the pollutant mitigating effects of different
200
150
Si 100
50
jTN(MA = 361,511.6lb)
Low Med High Comm. Indus. Open For. Agri. Bar.
Land Use
Figure 4. Total nitrogen load by land use, Collington Branch
watershed.
.§ 4
3
2
1
Collington Branch
(MA = 6,610.4 Ib)
1980 1981
1982 1983 1984 1985 1986 1987
Year
0-0.05 0.05-0.1 0.1-0.2 0.2-0.3 0.3-0.4 0.4-0.5 0.5-0.75 0.75-1.0 1.0-1.5 1.5+
Flow (in.)
Figure 3. Annual summary of total nitrogen load, Collington
Branch watershed, illustrating changes in loads due
to climatic variability.
Figure 5. Lead distribution by storm size, Collington Branch
watershed.
247
-------
control levels and retention times. At a site-specific level,
such as a proposed new subdivision, similar structures
can also be evaluated to allow optimal design criteria to
be selected. Figure 7 illustrates the phosphorus contri-
bution for a simulated residential subdivision, and the
pollutant reduction from a stormwater pond designed
to control for 0.3 in. of runoff. As seen in Figure 7, the
mean annual phosphorus load was reduced from 453 to
277 Ib by the simulated structure.
Managers must address how future changes will affect
water quality. On the watershed level, what will be the
impact of urbanization on flow and pollutants loads? At
the subbasin level, how will proposed projects change
the runoff characteristics? Both land use scenarios can
be evaluated in GWSM. On the watershed scale, the
current land use can be interactively changed with a
"point-and-click" menu. At the subbasin level, a user-de-
fined basin may be modeled, with the land-use distribu-
tion entered into a pulldown menu. At both watershed
and subbasin levels, once a land-use scenario is se-
lected, GWSM calculates the anticipated pollution
loads. The results can then be compared with preexist-
ing conditions. The following questions are answered
during the final phase of GWSM: How do pollutant loads
relate to rainfall and runoff distribution and intensity?
900
800
500
« 700
o 600
CO
'o
fc 400
£ 300
2 200
100
0
0-0.05 0.05-0.1 0.1-0.2 0.2-0.3 0.3-0.4 0.4-0.5 0.5-0.75 0.75-1.0 1.0-1.5 1.5+
Flow (in.)
Figure 6. Flow (frequency) distribution by storm size, Collington
Branch watershed.
. 3
{/)
11
tr
tt; H
S 1
LDRIOOTS(TP)
No Control: (MA = 453.6)
Control: 0.30 in., 48 hr (MA = 277.4)
0-0.05 0.05-0.1 0.1-0.2 0.2-0.3 0.3-0.4 0.4-0.5 0.5-0.75 0.75-1.0 1.0-1.5 1.5+
Runoff (in.)
Figure 7. Phosphorus contribution for a simulated residential
subdivision with a stormwater pond designed to con-
trol for 0.3 in. of runoff.
What is the optimal control level for structural practices?
What are the likely impacts of future land-use changes
on water quality?
Stormwater Management:
Applications
Future Model
The NPDES stormwater permit regulations have created
new challenges and opportunities for state, county, and
city water resource programs. Water resource manag-
ers are faced with often conflicting stormwater man-
agement objectives and forced to make decisions that
weigh the costs and benefits of each. For instance,
water quality and flood control objectives do not al-
ways coincide. The design and management of re-
gional stormwater ponds will vary depending on
whether water quantity control or water quality control
is the primary objective.
To address the complex array of stormwater issues,
more sophisticated analytical tools and techniques are
needed. Watershed models that effectively evaluate al-
ternative scenarios and allow for optimization routines
for differing management objectives are in demand.
Integrating environmental data, such as wetland areas,
bioassessment information, structural and nonstructural
best management practice (BMP) optimization, and per-
mit and monitoring information will be required in a
user-friendly CIS package.
As the NPDES stormwater regulations are implemented
at the county and local level, unique management pro-
grams will develop according to specific water quality
and resource availability issues. As these programs take
shape, CIS and CIS-based models and information
management systems are likely to play larger roles in
assessing problems and crafting solutions.
Conclusions
The GWSM enabled the rapid assessment of Prince
George's County's stormwater problem areas and the
evaluation of control measures. GWSM was developed to
support the development of the county's stormwater man-
agement program. The model incorporates the strengths of
continuous simulation modeling with the spatial analysis
techniques of CIS in an integrated system. Together, the CIS
and data processing routines allow for further analysis and
interpretation of time series data from the SWMM model.
Combining continuous time series data with georeferenced
watershed/land-use data allows for the further analysis and
interpretation of the results. As additional data from moni-
toring both homogenous land-use basins and in-stream
locations becomes available from the long-term monitoring
program developed as part of the NPDES Part 2 permit,
the accuracy of the model will increase.
As technologies for developing and evaluating stormwa-
ter programs increase in sophistication, the questions
248
-------
asked of water resource managers are likely to become
more difficult. The GWSM will continue to develop to
incorporate more functions designed to assist managers
to make the best, most informed decisions.
Acknowledgments
The authors thank the Prince George's County Water-
shed Protection Branch for its support of the model
development; Prince George's County Park and Plan-
ning forwatershed delineation coverages; and the Mary-
land Department of Planning for 1990 land-use
coverages of Prince George's County.
References
1. U.S. EPA. 1992. Compendium of watershed-scale models for
TMDL development. EPA/841 /R-92/002. Washington, DC.
2. U.S. EPA. 1985. Water quality assessment: A screening proce-
dure for toxic and conventional pollutants in surface and ground-
water, Part 1. EPA/600/6-85/002a. Athens, GA.
3. Schueler, T. 1987. Controlling urban runoff: A practical manual
for planning and designing urban BMPs. Washington, DC: Met-
ropolitan Washington Council of Governments.
4. Huber, W.C., and R.E. Dickinson. 1988. Storm water manage-
ment model, Version 4: User's manual. EPA/600/3-88/001 a
(NTIS PB-88-236641/AS). Athens, GA.
5. Barnwell, T.O., and R. Johanson. 1981. HSPF: A comprehensive
package for simulation of water hydrology and water quality. In:
Nonpoint pollution control: Tools and techniques for the future.
Rockville, MD: Interstate Commission on the Potomac River Basin.
6. DeVantier, B.A., and A.D. Feldman. 1993. Review of CIS appli-
cations in hydrologic modeling. J. Water Resour. Plan. Mgmt.
119(2):246-261.
7. James, D.E., and M.J. Hewitt, III. 1992. To save a river. Geolnfo
Systems, November/December, pp. 37-49.
8. Tim, U.S., S. Mostaghimi, and V.O. Shanholz. 1992. Identification
of critical nonpoint pollution source areas using geographic infor-
mation systems and water quality modeling. Water Resour. Bull.
28(5):877-887.
9. Ross, M.A., and P.O. Tara. 1993. Integrated hydrologic modeling
with geographic information systems. J. Water Resour. Plan.
Mgmnt. 119(2):129-140.
10. Terstriep, M., and M.T Lee. 1990. ARC/INFO CIS interface for
Q_ILLUDAS stormwater quantity/quality model. Proceedings of
the Remote Sensing and GIS Applications to Nonpoint Source
Planning Meeting, Chicago, IL (October).
249
-------
Watershed Screening and Targeting Tool (WSTT)
Leslie L. Shoemaker and Mohammed Lahlou
Tetra Tech, Inc., Fairfax, Virginia
Abstract
Screening-level tools allow managers to understand,
evaluate, and compare the water quality problems of
watersheds so that they can be prioritized. The Water-
shed Screening and Targeting Tool (WSTT) makes it
easier for watershed managers in federal, state, and
local agencies to conduct these evaluations by providing
access to the necessary data and information and facili-
tating the assessment itself. This prototype has been
developed as a cooperative project for the U.S. Environ-
mental Protection Agency (EPA) Region 4 and the Office
of Wetlands, Oceans, and Watersheds in support of the
Total Maximum Daily Load (TMDL) program.
The WSTT is an interactive, user-friendly, two-step
evaluation and targeting process. The first step allows
for preliminary screening based on multiple criteria.
Each criterion can be compared with a default or user-
defined reference value. Data from EPA mainframe da-
tabases allow the user to compare reference values with
land-use and water quality observations from water-
sheds under consideration. The second level of target-
ing, comparative analysis, allows for a more detailed
examination of watersheds. In addition, this analysis
permits the user to include subjective weights and addi-
tional data to the targeting procedure. The algorithms for
this targeting system are based on a hierarchical struc-
ture of objectives and criteria, where a set of up to seven
criteria can be used to describe the comparison objec-
tives. Although the analysis objectives are project spe-
cific, the procedures are developed to use either
user-specified data or information from provided data-
bases. Weights can be entered to give greater or lesser
value to particular criteria. This paper presents exam-
ples of the application of these techniques to sample
watersheds in Alabama.
Introduction
Targeting of watersheds is used to allocate increasingly
scarce water management resources for data collection,
modeling studies, and management assessment, de-
sign, and construction. Water resource managers can
use screening-level evaluations to help assess, com-
pare, and prioritize the water quality problems of water-
sheds within their jurisdictions. The Watershed
Screening and Targeting Tool (WSTT) makes it easier
for watershed managers in federal, state, and local
agencies to conduct these evaluations by providing
easy access to the necessary data and facilitating tar-
geting assessments. A prototype of WSTT has been
developed that allows access to data for Alabama and
Georgia. WSTT operates on a personal computer
(286+) in a DOS environment.
WSTT provides an interactive, user-friendly, two-step
evaluation and targeting process (Figure 1). The first
allows for preliminary screening based on multiple crite-
ria. Each criterion can be compared with a default or
user-defined reference value. Data from U.S. Environ-
mental Protection Agency (EPA) mainframe databases
allow the user to compare reference values with land-
use and water quality observations from watersheds
under consideration. The second level of targeting, com-
parative analysis, allows for a more detailed examina-
tion of watersheds. In addition, this analysis permits the
user to include subjective weights and additional data in
the targeting procedure. The algorithms for this targeting
system are based on a hierarchical structure of objec-
tives and criteria, where a set of up to seven criteria can
be used to describe the comparison objectives. Al-
though the analysis objectives are project specific, the
procedures are developed to use either user-specified
data or information from provided databases. Weights
can be entered to give greater or lesser value to particu-
lar criteria.
Watershed prioritization and targeting involve a mul-
tistep decision-making process using both technical cri-
teria and subjective judgement. The use of formal
targeting procedures throughout this process can assist
in structuring the problem while taking into account all
pertinent and site-specific concerns.
250
-------
Watershed Selection
- Select State
Select Accounting Unit
- Select Catalog Units
Generate Report
- Save
L- Exit
Preliminary Screening
_ Select Criteria
_ Enter Targeting Values
Results
_ Save/Print
L Quit
Comparative Analysis
_ Select Objectives
- Select Criteria for Objectives
Enter Weighting System
- Select Ranking
- Results
- Save/Print
I-Quit
Generate Report
Select Data Bases
Llf WQ, Select Parameters
If PS, Select Report Options
>- Results
I Graphical
Tabular
Save/Print
- Quit
Watershed Screening Model (WSM)
l Format Input Data
"- Run WSM
Figure 1. Schematic of WSTT components.
Multicriteria analysis techniques can aid in processing
available information in a more structured framework,
leading to a rational prioritization of watersheds. These
procedures can be used in the Total Maximum Daily
Load (TMDL) process to identify data sources, retrieve
relevant water quality and watershed data, and analyze
these data within a structured framework to determine
which watersheds require management. The advan-
tages of structured decision-making techniquesespe-
cially when dealing with numerous watersheds where
the ranking in order of priority is not obvious a priori
include analysis directed toward the selection of perti-
nent decision criteria and identification of potential
candidate watersheds; credibility of the selection proc-
ess by the use of demonstrated and valid decision-mak-
ing techniques; reductions in the cost and time for data
collection and processing through a multiphase screen-
ing process; iterative evaluation of watersheds; and in-
creased understanding of the various tradeoffs.
For the incorporation of targeting criteria and the gen-
eration of reports, WSTT is distributed with and relies on
data that were selectively downloaded from EPA's main-
frame computer. The databases that it uses include an
accounting unit/catalog unit (CD) summary table, land
use (U.S. Department of Agriculture Natural Resource
Inventory summary of acres per land-use category),
water quality (EPA STORET ambient water quality data
summarized by CD for 50 parameters), reference levels
(based on EPA water quality criteria), water supplies
(number, flow, location, and type), point sources (num-
ber, flow, location, and type), and water bodies (number
and size). These data, always available to the public,
have traditionally been difficult to access without famili-
arity with EPA's mainframe. Through WSTT, these data
are readily accessible. Using these databases, WSTT
can generate reports, in table or graph form, on land
use, water quality, water supplies, impoundments, and
point source facilities in each of the selected areas.
The data that are distributed with WSTT can also be
used to prepare preliminary data input files for a water-
shed screening model (WSM) which, for this proto-
type version, can be run for CUs within Georgia and
Alabama. The watershed screening methodology per-
mits simple watershed assessments that predict daily
runoff, streamflow, erosion, sediment load, and nutrient
washoff. The WSM relies on observed precipitation and
temperature data from local meteorologic stations, mu-
nicipal point source load estimations from pollutant con-
centrations in the literature, and nonpoint source loading
functions for selected land uses based on literature
values. Users can readily modify or revise the input data
to reflect site-specific conditions. Output data from the
model simulations can be used to augment information
provided by the other databases.
Review of Potential Targeting Procedures
Most multicriteria decision techniques with potential ap-
plication to the prioritization and targeting process can
be grouped into three categories:
Sequential elimination: Techniques to eliminate wa-
tersheds that do not show any need for controls.
Dominance theory: Techniques to eliminate inferior or
dominated watersheds that demonstrate a need for
251
-------
pollution control but do not present a character of
relative urgency.
Ranking procedures: Techniques to prioritize remain-
ing watersheds.
Sequential Elimination
The first group of analytical procedures target nonprior-
ity watersheds or the nonfeasible set of watersheds.
These procedures are typically referred to as sequential
elimination techniques. Each watershed is compared
with a hypothetical watershed using an amalgamation of
standards and criteria. Watersheds that are better than
the hypothetical watershed form the nonfeasible set and
can be eliminated from further analysis. These tech-
niques provide a preliminary filtering system to ensure
the legitimate acceptability of the remaining set of wa-
tersheds. Sequential elimination techniques do not dif-
ferentiate on the basis of relative importance, only on
the ability to satisfy a condition of preset limits. Four
relevant sequential elimination techniques, available for
application in the prioritization process, include the con-
junctive approach, the disjunctive approach, the lexico-
graphic approach, and the compensatory approach (1).
The conjunctive approach screens out watersheds by
establishing minimum cutoff levels for each discriminat-
ing criteria. Depending on the type of criterion and its
method of measurement, the constraint or "cutoff level"
is defined as either a categorical exclusionary or inclu-
sionary limit. The application of a conjunctive scheme
relates the decision criteria and their constraint with the
logical "and" so that all constraints must be satisfied for
a watershed to be eliminated from further consideration.
Evaluation scales do not have to be homogeneous
across criteria and can include logical, numeric, or natu-
ral scales. Because decision criteria and the set of
watersheds are independent, each watershed is com-
pared individually with a hypothetical set of constraints
rather than with other watersheds. In general, decision
criteria in the conjunctive approach should be carefully
selected to focus on criteria with a strict regulatory re-
quirement and technical constraints that cannot be re-
laxed or are not subject to tradeoffs.
The disjunctive approach is similar to the conjunctive
scheme, but it requires that only one criterion be satis-
fied for a watershed to be eliminated from further con-
sideration. Because this process is characterized by the
logical relation "or," problem formulation must be defined
in terms of the level of substitutions among the selected
decision criteria.
Lexicographic screening differs from the previous tech-
niques in that the value of each criterion is compared
across watersheds (2). The criteria are first ordered in
terms of their relative importance, and watersheds are
then screened, starting with the most important criteria.
Watersheds that pass this preliminary test are screened
with the next highest ranked criteria until either all crite-
ria are evaluated or the number of watersheds selected
for further analysis is sufficiently reduced.
Compensatory analysis is a more elaborate form of
conjunctive and disjunctive screening and deals primar-
ily with preferential constraints where the cutoff levels
are set by the objectives rather than by the criteria
themselves (3). The analysis develops constraints on
selected objectives that are represented in the decision
problem by a group of two or more criteria. For each
identified objective, the corresponding criteria are com-
bined into a discriminating model expressing the degree
to which each criterion achieves the objective. The dis-
crimination process can be inclusionary, exclusionary, or
both, depending on the screening model.
Dominance Theory
The second group of analytical tools with potential for
application in the watershed prioritization and targeting
process consist of techniques developed from the domi-
nance approach. This approach serves to identify poorer
watersheds rather than rank them completely. In this
case, when the first watershed that has criterion values
at least as poor as those of a second, as well as one or
more values that are poorer, the first watershed will be
selected for further analysis rather than the second. The
first watershed is said to dominate the second. These
techniques add some capability of determining which
watersheds are worse than others beyond the simple
comparisons offered by the sequential elimination
schemes. Although several techniques have been de-
veloped based on this decision rule, their application to
discrete decision space, such as watershed targeting
applications, may not be effective in eliminating many
watersheds. Among these techniques are the nonin-
ferior curve technique, the indifference map technique,
and fuzzy outranking approaches.
The noninferior curve technique uses the distribution of
the feasible set of watersheds within the decision space
to identify inferior and noninferior sets (4). The curve
defines the level of tradeoff between decision criteria
where any incremental improvement in one criterion
results in a balanced incremental decrease in other
criteria. Application of this technique may require exces-
sive computational time and professional training for
interpretation of the results (5).
The indifference map technique relies on the repre-
sentation of the preference structure to determine the
family of indifference curves (6). An indifference curve
represents points in the decision space for which the
preference is equivalent among all criteria. This ap-
proach can be used in combination with the noninferior
curve technique. Theoretically, if the one indifference
curve tangent to the noninferior curve can be located,
252
-------
then watersheds lying farthest from the point of
tangency form the set with the highest priority for con-
trols.
Outranking techniques analyze sets of watersheds to
derive binary relationships on the set rather than a
function from this set to the real numbers, as in the case
of the classical theory of decision analysis. This binary
relationship also differs from classical decision analysis
in the sense that it does not necessarily require a strict
transitivity condition (7, 8). Outranking procedures can
be used to select one and only one watershed, a set of
acceptable watersheds, or a cluster of watersheds in an
ordered sequence of indifference classes ranging from
best to worst.
Ranking Procedures
The third group of analytical decision techniques ranks
the set of watersheds under consideration. Several al-
gorithms with potential application to discrete situations
include utility theory, compromise programming and dis-
placed ideal techniques, cooperative game theory, and
the analytic hierarchy process.
Decision techniques developed based on the theory of
utilities assign a utility function to each decision criterion,
then compute the expected utility for each watershed
using either an additive or multiplicative model (9). Wa-
tersheds that maximize the expected utilities may be
eliminated from further analysis, and those with low
ranking values form the set to be considered. The diffi-
culties associated with application of the utility models
reside in the development of representative utility func-
tions for each criterion and the insurance that all criteria
satisfy both preference and utility independence axioms.
A utility function refers to a mapping of the values in the
range of natural criteria scale to a bounded cardinal-
worth scale reflecting the preference structures associ-
ated with that criteria as perceived by the decision-
maker(s).
Compromise programming techniques have been ap-
plied extensively to water resources projects. These
techniques attempt to identify watersheds that approach
an ideal case (10), assuming that the watershed located
the closest to the ideal watershed in the decision space
can be eliminated from further consideration. The com-
putation algorithms rank watersheds based on the nor-
malized distance between each watershed and this
ideal point. This approach can also be applied to identify
watersheds that are the closest to an anti-ideal point
using a similar minimization scheme.
Cooperative game theory is a representation of a conflict
situation based on the general concepts of rational be-
havior. Optimization of a set is sought by well-informed
decision-makers with conflicting objectives who are
aware of their preference structure. The objective of
each participant in the game is to identify solutions that
are high on the preference scale. A generic algorithm
based on this theory for an n-person game was devel-
oped by Harsanyi (11). This algorithm was generalized
for a regional ground-water pollution problem (12) and
for the analysis of wastewater management alternatives
(13).
The analytic hierarchy process was developed in an
effort to expand the classical decision models to include
subjective analysis of multilevel or hierarchical systems
(14, 15). The process consists of decomposing the de-
cision problem into smaller subproblems, analyzing
each subproblem individually, and then recomposing the
results to reach a complete ranking of the set of water-
sheds considered. It relies strongly on the structuring of
the decision problem into an intuitively logical hierarchy
of objectives and criteria. The hierarchical structures
express the factual relationships between the decision
elements (objectives, criteria, and alternatives). This de-
cision process parallels the principles of analytical think-
ing (16): constructing hierarchies, establishing priorities,
and logical consistency.
Targeting Techniques in WSTT
The review of decision analysis techniques, briefly de-
scribed above, provided the background for the devel-
opment of the targeting tools used in the WSTT. The
development of decision-making techniques for water-
shed prioritization and targeting was based on the fol-
lowing:
Ability to perform a multicriteria analysis.
Applicability to discrete situations with a limited num-
ber of watersheds.
Applicability to selecting the worst watersheds rather
than the preferred conditions, as is the case in most
decision situations forTMDLs or watershed manage-
ment.
Flexibility of problem structuring, data processing,
and the ability to decompose the problem into smaller
and more homogeneous components.
Stability of the final ranking using simple scaling pro-
cedures.
Ease of interpreting the rankings.
Ability to perform sensitivity analysis and consistency
testing of the value judgment.
These considerations led to the development of a two-
step targeting approach consisting of both a preliminary
screening and a formal comparative analysis. A test
watershed is used for illustrating examples of the two
types of screening techniques (Figure 2).
253
-------
Accounting Unit
031300
Dothan
14
Figure 2. Watersheds selected for preliminary screening.
Preliminary Screening Analysis
The screening level analysis of watersheds at a regional
or accounting unit scale is designed to help users un-
derstand what the water quality conditions are in each
watershed and how the factors governing quality vary
from one watershed to the next. The advantage of this
procedure is its ability to operate under the WSTT envi-
ronment, using automatically retrieved values for the
desired decision criteria, and iteratively screen out wa-
tersheds that do not represent a significant water quality
problem.
The screening algorithm used in WSTT consists of a
sequential elimination scheme adapted from the con-
junctive approach described in the previous section. The
objective of this process is to identify watersheds that
do not represent a significant water quality problem and
consequently reduce the set of watersheds to a work-
able number. The significance of the water quality prob-
lem is, however, indirectly introduced into the analysis
through the selection of screening criteria indicative of
the problem under consideration and the magnitude of
each criterion cutoff level. Figure 3 illustrates this proc-
ess using a single water quality criterion, and Figure 4
presents the case of a two-criteria screening. Based on
the sample cutoff limit shown in Figure 3, six watersheds
(1, 2, 3, 4, 5, and 13) would be selected for further
analysis. In Figure 4, two criteria are examined. In this
case, both acres of urban land and BOD5 concentrations
are selected for examination. Values outside the upper
limits for either of the two criteria would be selected for
4.5
4.0
3.5
3.0
2.5
^2.0
1.5
1.0
0.5
0.0
3.5
3.7
Accounting Unit 031300
Sample
Limit
4.3
2.4
1.7 1 6
1 5
nil il
1 2 34 5 6 7 8 9 10 11 12 13 14
Catalog Unit
BOD5
Figure 3. Preliminary screening example with one criterion.
240 000 Accounting Unit 031300
190/700
160,000
80,000
1
105,400
143,800
TOjOOO 15,700 J
Sample
Limit C1
8,300 4,600 6,400 2l|jf° 5.01
12 3 4 5 6 7 8 9 10 11 12 13 14
Catalog Unit
Urban Land (Top Graph)
BOD5 (Bottom Graph)
Figure 4. Preliminary screening example with two criteria.
further examination. In this case, seven watersheds (1,
2, 3, 4, 5, 8, and 13) would be selected. The user can
select the cutoff limits in an iterative fashion to examine
the differences between the watersheds. Multiple crite-
ria can be selected for evaluation, depending on data
availability and watershed characteristics. This provides
a quick and easy approach for preliminary evaluation of
the differences between the watersheds selected for
examination.
For a multidimensional problem, each criterion is de-
fined in terms of a cutoff limit representing a vector of
threshold values. Depending on the type of criterion and
its measurement scale, each value in this vector may
either represent an upper or a lower limit. Examples of
criteria with an upper limit are water quality parameters
for which the cutoff limit represents a concentration that
should not be exceeded. On the other hand, criteria with
a lower limit include those with ascending scales in
254
-------
which the higher values are better, such as in the case
of dissolved oxygen concentration.
The watershed screening level analysis in WSTT allows
users to retrieve screening criteria and their values auto-
matically from available, preprocessed databases.
When the criteria represent water quality parameters,
watershed rating with respect to each criterion can be
expressed in terms of a mean value, a median, or a
quartile. Multiple databases can be accessed sequen-
tially. Access to the water quality and land-use data-
bases is enabled at the present time. Cutoff limits are
user specified and can be modified in an iterative
scheme by either relaxing the criteria's cutoff limit and
consequently decreasing the set of selected watersheds
or by making them more stringent. Watersheds elimi-
nated during this screening level analysis can still be
considered in the comparative analysis phase. The out-
put of this algorithm generates a list of watersheds that
do not satisfy the criteria's cutoff levels. For these wa-
tersheds, the corresponding input data (payoff-matrix)
can be accessed through the reporting option of the
WSTT. Watersheds that satisfied all user-specified con-
straints are also tabulated. As noted earlier, the screen-
ing analysis does not take into consideration the relative
differences in the exceedence of the observations be-
yond the upper limit. For examination of the relative
importance and actual ranking of the watersheds, the
comparative analysis technique is used.
Comparative Analysis in WSTT
The objective of the comparative analysis is to provide
a system that captures both the importance of the se-
lection objectives and that of the criteria describing
these objectives. Comparative analysis can provide a
complete ordering of watersheds. The process requires
that the targeting problem be formulated in terms of a
decision situation and that judgement values be incor-
porated into each phase of analysis. At this level of
analysis, additional measurable and subjective criteria
are usually incorporated into the analysis; therefore, the
algorithm provides a logical scaling system to evaluate
the importance of these objectives on a common basis.
The algorithm also incorporates a mathematical frame-
work to amalgamate the value judgement and the wa-
tershed observations with respect to each criterion or
objective in terms of a ranking index.
Four subroutines incorporated in the development of the
comparative analysis algorithm in the WSTT are de-
scribed below.
Structuring of the Targeting Problem
The formulation of watershed prioritization problems in
WSTT consists of a multilevel hierarchical structuring of
the selection objectives, the decision criteria, and the
alternative watersheds. This formulation separates the
selection problem into several smaller and homogene-
ous subproblems which can be easily compared. Figure
5 illustrates a generic representation of a clustered hier-
archy in which the project is decomposed into a set of
simple and smaller subprojects. Each subproject can be
analyzed separately, and the results can be reintegrated
to obtained an overall ranking of the watersheds.
Value Judgment
The decision-maker's value judgment is introduced in
terms of the importance weight coefficients of the objec-
tives and criteria. The derivation of criterion importance
weights proceeds according to the hierarchical structure
of the decision problem, starting from the higher level
objectives. This routine takes the decision-maker
through a series of paired comparisons cluster by cluster
in the order shown by the roman numerals in Figure 5.
For each paired comparison between two criteria, the
decision-maker defines which criterion of the pair is
more important and determines the magnitude of the
importance using the integer ratio scale presented in
Table 1. The magnitude of importance is not the desired
importance weight but rather a measure of a pain/vise
ratio defined as follows:
W,
(Eq. 1)
where a represents the ratio of importance weight Wof
criterion i over that of criterion j.
The use of the ratio scale defined in Table 1 generates
a square, positive, and reciprocal matrix in which the
importance weight coefficients consist of the entries of
the eigenvector corresponding to the maximum eigen-
value of the this matrix. The characteristics of the result-
ing comparison matrix are summarized as follows:
(Eq. 2)
for all i and j;
an =
(Eq. 3)
for all i=1 to n where n is the number of criteria; and
aik=aij+aik (Eq. 4)
The rationale for determining the eigenvector corre-
sponding to the maximum eigenvalue as the importance
weight coefficient vector derives naturally from the type
of scale used in the pain/vise comparisons and the as-
255
-------
Hierarchical Levels
Level 1
Overall Objective
Level 2
Targeting Objectives
Level 3
Decision Criteria
n < 7
!
Objective #1 1
Overall
Objective
I
1
Objective #2
1
Objective #3 1
C1 C2 C3 C4 C5 C6 C7 C1 C2 C3 C4 C5 C6 C7 C1 C2 C3 C4 C5 C6 C7
Level 4
Watersheds Evaluated
w, |
W2
W3 | .
w.
wn
!
Figure 5. Generic representation of the watershed targeting problem in WSTT.
Table 1. Evaluation Scale Developed by Saaty (14) for Use
in the Analytic Hierarchy Process
Intensity of
Importance
Definition
Explanation
2, 4, 6, 8
Reciprocal of
above
nonzero
Rational
Equal importance
Weak importance of
one over another
Essential of strong
importance
Demonstrated
importance
Absolute importance
Intermediate values
between the two
adjacent judgments
If activity i has one of
the above nonzero
numbers assigned to
it when compared
with activity j, then j
has the reciprocal
value when
compared with i.
Ratio arising from the
scale
Two activities contribute
equally to the objective.
Experience and judgment
slightly favor one activity
over another.
Experience and judgment
strongly favor one activity
over another.
An activity is strongly
favored, and its
dominance is
demonstrated in practice.
The evidence favoring
one activity over another
is of the highest possible
order of affirmation.
Compromise is needed.
Consistency is forced by
obtaining n numerical
values to span the matrix.
sociated matrix theory used in solving nonlinear sys-
tems, expressed in matrix form as
A W= n W,
(Eq. 5)
where A is the comparison matrix with n entries, n is the
number of criteria, and W is the vector of importance
weight coefficients. The solution of the above eigen-
value problem for each cluster in the order shown in
Figure 5 provides a partial weight coefficient for each
criterion. The overall weight can be derived by multiply-
ing the partial weight of the dominant objective by that
of the criteria:
W, = Wp (objective)- W^criteria^. (Eq. 6)
Consistency of the Preference Structure
When dealing with large numbers of objectives and
criteria, the preference structure tends to lose its transi-
tive character. As intransitive comparisons are intro-
duced, the resulting matrices become less consistent,
and the importance weight coefficients may not repre-
sent the true preference structure.
For a perfectly consistent positive reciprocal matrix, the
maximum eigenvalue should equal the order of the ma-
trix. This suggests that the remaining eigenvalues are
equal to zero. As small inconsistencies are introduced
into the matrix because of intransitive judgements, they
lead to very small perturbations in the original set of
eigenvalues. This represents the fundamental theory of
consistency measurement in positive reciprocal matri-
256
-------
ces (14). The more the maximum eigenvalue deviates
from that of a consistent matrix, the less consistent the
pairwise comparisons are. A consistency index devel-
oped by Saaty (14) was introduced into the targeting
subprogram in WSTT to indicate the degree of consis-
tency at the end of each series of pairwise comparisons.
A consistency index of 0.0 indicates a perfect consis-
tency, and a value of 1.0 indicates a fully inconsistent
matrix. Because of the use of an integer scale in addition
to the nonlinearity of certain subjective judgments, a
slight nonconsistency in developing importance weight
coefficients is common. In fact, a fully consistent com-
parison is not required to reach the desired accuracy.
Analysis of the sensitivity of eigenvalue solution shows
that matrices with a consistency ratio of up to 0.1 are
acceptable (17).
Ranking of Watersheds
The hierarchic representation of the watershed targeting
process is a logical structure for integrating the decision
elements into a single problem and deriving the selec-
tion priorities defined in terms of objectives, criteria, and
their respective weight coefficients. To derive the overall
ranking of the watersheds, a simplified form of the addi-
tive utility model is used. This model is described in
much of the relevant literature as the best known of the
multiattribute utility functions because of its relevance to
a wide range of decision problems, its stability in ranking
alternatives, and its simplicity of application. This model
is also used in most index calculations. Its generic ex-
pression when applied to a hierarchic problem takes the
following form:
7=1
M
k= 1
vk,
(Eq. 7)
where
W = weight coefficient
N = number of objectives
M = number of criteria under each objective
U = ranking of watershed i
V = measurable value of lower level criteria
This model uses normalized values of the criteria in an
ascending scale, meaning that the higher values are
better. The ranking is therefore performed on a descend-
ing scale so that watersheds with the lowest scores are
identified as the priority watersheds.
The results of a sample application are shown in Table 2
below. For illustration purposes, a comparative analysis
was applied, using WSTT, to six watersheds in Alabama.
Three criteria were selected for examinationBODS,
ammonia, and ironbased on the 85th percentile of all
data available on STORET since 1980. The values used
forthe comparative analysis are shown first. Three types
of weights are shown: equal weights and two variable
options. The final section of Table 2 shows how the
changes in weights affect the resulting ranking of the
watersheds. The ability to adjust weights and test a
variety of user- and system-provided criteria allows for
a wide range of flexibility in the assessment of water-
shed ranks. Users can thereby incorporate best profes-
sional judgement and local knowledge into the targeting
procedure in a systematic fashion.
Table 2. Description of Comparative Analysis Application
Values Used for Comparative Analysis (Payoff Matrix)
Catalog Unit
0313001
031 3002
031 3003
031 3004
031 3005
0313013
Criterion 1
BOD5 (mg/L)
4.0
3.5
3.7
2.4
3.0
4.3
Criterion 2
NH4 as N
(mg/L)
0.27
0.62
0.30
0.14
0.66
0.16
Criterion 3
Fe (|ig/L)
1,100
1,600
315
680
2,900
370
Calculated Importance Weight Coefficients
Criteria
1 (BOD5)
2 (NH4 as N)
3(Fe)
Final Watershed
Catalog Unit
0313001
031 3002
031 3003
031 3004
031 3005
0313013
Equal
0.333
0.333
0.333
Ranking
Equal
3
4
5
6
1
4
Variable 1
0.122
0.648
0.230
Variable 1
3
2
4
6
1
5
Variable 2
0.637
0.258
0.105
Variable 2
3
1
5
6
4
2
Application of the comparative analysis requires users
to evaluate which criteria are relevant and significant to
the local watershed conditions. Often, application will be
constrained by the availability of water quality sampling
information or other data. Consideration should also be
given to possible dependence between two criteria se-
lected. Criteria should be independent for accurate as-
sessment of watershed ranking.
Conclusions
The WSTT program and associated databases provide
watershed managers with the tools to effectively target
and assess watersheds on a broad scale. The two levels
of targeting tools included with the WSTT allow for a
257
-------
range of targeting applicationsfrom simple to sophis-
ticateddepending on project needs. The incorporation
of the comparative targeting tool provides the valuable
addition of subjective judgement and user-defined pa-
rameters to the decision-making structure. This powerful
algorithm allows managers to refine decision-making
criteria and evaluate multiple and often conflicting objec-
tives. The incorporation of targeting tools and databases
into a user-friendly PC environment can make these
powerful techniques convenient and accessible to a
wide range of water resources professionals.
Acknowledgments
The authors would like to thank Jim Greenfield, EPA
Region 4, and Donald Brady, Chief, Watershed Manage-
ment Section, EPA Office of Wetlands, Oceans, and
Watersheds, for their guidance and input into the devel-
opment of WSTT Sigrid Popowich and John Craig of
Tetra Tech, Inc., provided significant input into both the
design of WSTT and the preparation of the databases.
Software design and development was performed by
Randy French of Isoceles Software, Inc.
References
1. Lahlou, M., and L.W. Canter. 1993. Alternatives evaluation and
selection in environmental remediation projects. Environmental
Impact Assessment Review. In press.
2. MacCrimmon, K.R. 1973. An overview of objective decision mak-
ing. In: Cochrane, J.L., and M. Zeleny, eds. Multiple criteria deci-
sion-making. Columbia, SC: University of South Carolina Press.
pp. 18-44.
3. Keeney, R.L. 1980. Siting energy facilities. New York, NY: Aca-
demic Press.
4. Church, R.L., and J.L. Cohon. 1976. Multiobjective location analy-
sis of regional energy facility siting problems. BNL-50567. Upton,
NY: Brookhaven National Laboratory.
5. Hobbs, B.F., and A.M. Voelker. 1978. Analytical multiobjective
decision-making techniques and power plant siting: A survey and
critique. ORNL-5288. Oak Ridge, TN: Oak Ridge National Labo-
ratory (February).
6. MacCrimmon, K.R., and M. Toda. 1969. The experimental deter-
mination of indifference curves. Review of Economic Studies
36(2):433-450.
7. Vincke, PH. 1986. Analysis of multicriteria decision aid in Europe.
Eur. J. Research 25(2): 160-168.
8. Roy, B. 1976. Partial preference analysis and decision-aid: The
fuzzy outranking relation concept. Paris, France: SEMA.
9. Keeney, R.L., and H. Raiffa. 1976. Decisions with multiple objec-
tives: Preferences and value tradeoffs. New York, NY: Wiley and
Sons.
10. Zeleny, M. 1982. Multiple criteria decision-making. New York, NY:
McGraw-Hill.
11. Harsanyi, J.C. 1977. Rational behavior and bargaining equilib-
rium in games and social situations. London, England: Cam-
bridge University Press.
12. Szidarovsky, F., L. Duckstein, and I. Bogardi. 1984. Multiobjective
management of mining underwater hazard by game theory. Eur.
J. Operational Research 15(2):251-258.
13. Tecle, A., M. Fogel, and L. Duckstein. 1988. Multicriteria selection
of wastewater management alternatives. J. Water Resources
Planning and Management 114(4):383-713.
14. Saaty, T.L. 1977. A scaling method for priorities in hierarchical
structures. J. Math. Psychol. 15(3):234-381.
15. Saaty, T.L. 1980. The analytic hierarchy process. New York, NY:
McGraw-Hill.
16. Saaty, T.L. 1982. Decision-making for leaders. Belmont, CA: Van
Nostrand Reinhold.
258
-------
Hydrocarbon Hotspots in the Urban Landscape
Thomas Schueler and David Shepp
Department of Environmental Programs,
Metropolitan Washington Council of Governments, Washington, DC
Abstract
This paper reports on a monitoring study that compared
hydrocarbon, polycyclic aromatic hydrocarbons (PAHs)
and trace metal levels in stormwater runoff captured
within standard oilgrit separators (OGSs) serving five
automotive-related land uses in the Maryland Piedmont.
Composite priority pollutant scans and trace metal sam-
ples were collected from the pools and the trapped sedi-
ments of 17 OGSs serving gas stations, convenience
commercial, commuter parking lots, streets, and residen-
tial townhouse parking lots. Previous studies indicated
that OGSs were not effective in trapping sediments over
the long term, based on sediment accumulation rates
overtime. Oily sediments, however, were retained over
a short term, making the OGS sites useful sampling
ports to characterize differences in hydrocarbon and
toxic levels in small, automotive-related land uses.
Gas stations had significantly higher hydrocarbon, total
organic carbon, and metal levels than all other sites in
both the water column and the sediments. Convenience
commercial and commuter parking lots had moderate
levels of contamination, with the lowest levels recorded
for streets and residential townhouse parking lots. Mean
hydrocarbon concentrations of 22 mg/L and 18,155
mg/kg were recorded for the water column and the
sediments at gas station OGS sites. The priority pollut-
ant scan identified 37 potentially toxic compounds in the
sediments and 19 in the pools of gas station OGS sites.
This can be compared with non-gas-station sites, which
had 29 and 7 toxics in the sediment and water column,
respectively. Some of the gas station priority pollutants
included naphthalene, phenanthrene, pyrene, toluene,
xylene, chrysene, benzene, phenols, acetone, and nu-
merous trace metals.
The source of these pollutants appears to be spillage or
leakage of oil, gas, antifreeze, lubricating fluids, cleaning
agents, and other automotive-related compounds. The
study suggests that numerous "hotspots" exist in the urban
landscape that generate significant hydrocarbon and PAH
loadings, particularly where vehicles are fueled, serviced,
and parked for extended periods. Preliminary computa-
tions suggest a possible link between these hotspots
and sediment PAH contamination of a local estuary.
Introduction
Over the past decade, nearly one thousand oil grit sepa-
rators (OGSs) have been installed in the metropolitan
Washington area to treat urban stormwater runoff from
small drainage areas. These structures consist of two
precast chambers connected to the storm drain system
(Figure 1). The first chamber is termed the grit chamber
and is used to trap coarse sediments. The second
chamber, termed the oil chamber, is used to temporarily
trap oil and grease borne in urban runoff so that they
may ultimately adsorb to suspended sediments and set-
tle to the bottom of the chamber.
Most OGSs control runoff from highly impervious sites
of an acre or less and have a storage volume of 0.06 to
0.12 in. of runoff, depending on the local design. As
such, OGSs were never expected to achieve high rates
of pollutant removal (1). Rather, they are intended to
control hydrocarbons, floatables, and coarse sediments
from small parking lots that cannot normally be served
by other, more effective best management practices.
Side View
Reinforced
Stormdrain
Inlet
Permanent Pool:
400 Ft3
of Storage per
Contributing
Acre, 4 Ft Deep
Access Manholes
I
First Chamber Second Chamber Third
(Sediment Trapping) (Oil Separation) Chamber
Figure 1. Schematic diagram of an OGS (1).
259
-------
From a monitoring standpoint, OGSs are interesting in
that they act as a very useful and standardized sam-
pling port to extract runoff samples from very small
areas of differing automotive land use. It was hypothe-
sized that hydrocarbon and trace metal levels might
be greater at sites where vehicles were parked, serv-
iced, or fueled. These potential "hotspots" had never
been systematically monitored in the metropolitan
Washington area before.
Methods
A two-tiered monitoring strategy was employed to test the
effectiveness of OGS systems and to detect hydrocarbon
hotspots. In the first tier of sampling, 110 OGS systems
were surveyed to determine their general characteristics
in the field. Each structure was sampled for the mass
and particle size distribution of trapped sediments, land
use, age, maintenance history, secchi depth, and other
engineering parameters (2).
The emphasis on the second tier of sampling was to
characterize the range of pool and sediment quality
found within OGS and related systems. Nineteen of the
Tier 1 sites were selected for additional detailed sam-
pling of the quality of pool water and trapped sediments.
The sites were grouped into five land-use categories:
townhouse parking lots, streets, all-day parking lots, gas
stations, and convenience store parking lots. Sediment
and pool samples were collected from each chamber
and were subsequently analyzed for nutrients, soluble
and extractable metals, total organic carbon (TOC), and
total hydrocarbons.
In addition, six priority pollutant scans were conducted
based on composite sediment and pool water samples
from gas station sites, non-gas-station sites, and all five
land-use sites combined. The samples were analyzed
for the presence of 128 compounds outlined in the U.S.
Environmental Protection Agency's (EPA's) priority pol-
lutant list. A complete description of the sampling and
analytical protocol is contained in Schuelerand Shepp (3).
Results
(Figure 3), with up to a 50-percent decrease in sediment
depths recorded in a single month. Dye tests indicated
pool residence times of less than 30 min during storms.
Consequently, it is thought that the mass of trapped sedi-
ments contained within an OGS at any given point repre-
sents only a temporary accumulation of pollutants.
General Characteristics of OGS Systems
Trapped sediments within OGSs were coarse-grained,
highly organic, oily in appearance, and interlaced with
litter and debris. Sediments were also quite soupy; only
45 percent total mass of sediment existed as dry weight.
The proportion of volatile suspended solids, a measure
of the general organic content of the sediments, aver-
aged 15 percent of total mass.
OGS pools frequently had a thin oil sheen or surface
scum, and oil stains were present on the chamber walls.
Despite the sheen, the pool water was relatively transpar-
ent, with an average secchi depth of 14 to 22 in. Floatable
trash was present in low to moderate quantities.
Volume of Sediment (ft3)
25
20
15
10
5
n
N = 10£
»
*
. *
*
r = 0.06 * . '
Slope = 0.7 , , * »
.»
* *
. . *.
» »
. * . * *
.
, , , l_!_l I , , , 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 1 I 1 1 1 1 r 1 1 1 T 1 1
o.o
Age (years)
4.9
Figure 2. Relationship of OGS age and volume of trapped sedi-
ments (2).
Retention of Sediments in OGS
The field surveys indicated that OGS systems had
poor retention characteristics. The average wet volume
of trapped sediments in 110 OGSs was 11.2 ft3, with
an average sediment depth of only 2 in. If OGS sys-
tems were highly retentive, the mass of trapped sedi-
ments would be expected to increase with age. No such
relationship was evident, however, in the 110 OGSs
surveyed (Figure 2), suggesting that frequent scour and
resuspension occur.
Monthly sampling of sediment depths in individual OGS
systems revealed sharp fluctuations in depth over time
16
MDE Hydrocarbon Study
Performance Monitoring - Street OGS
Sediment Accumulation Over Time
Average Sediment
Accumulation (in.)
ONJJ^CDODONJJ^
* /"^--^
-^ \ / ^X
\/ \
\ Chamber 1
""^^
^^
' i - Chamber 2
2345 6
Monthly Measurements
Figure 3. Monthly change in depth in OGS (1).
260
-------
Table 1. Characterization of Pollutant Concentrations in the OGS Water Column: Effect of Land-Use Condition (Mean Values)
Sampled Parameter
OP (mg/L)
TP (mg/L)
NH3-N (mg/L)
TKN (mg/L)
OX-N (mg/L)
TOO (mg/L)
Hydrocarbons (mg/L)
TSS (mg/L)
ECD (ng/L)
SCO (ng/L)
ECR (ng/L)
SCR (ng/L)
ECU (ng/L)
SCU (ng/L)
EPB (Mfl/L)
SPB (ng/L)
EZN (ng/L)
SZN (ng/L)
All-Day
Parking
(N = 8)
0.23
0.30
0.20
1.18
0.65
20.60
15.40
4.74
6.45
3.40a
5.37
ND
11.61
8.22a
13.42
8.1 Oa
190.00
106.70
aMean is for all observations in which the indicated
ND = not detected; NA = not applicable.
OP = ortho phosphate phosphorus
TP = total phosphorus
NH3-N = ammonia nitrogen
TKN = total Kjeldahl nitrogen
OX-N = oxidized nitrogen
TOC = total organic carbon
Hydrocarbons = total hydrocarbons
TSS = total suspended solids
ECD = extractable cadmium
Convenience Gas
Commercial Stations
(N = 6) (N = 7)
0.16 0.11
0.50 0.53
1.58 0.11
4.94 2.5
0.01 0.21
26.80 95.51
10.93 21.97
5.70
7.92a 15.29a
ND 6.34a
13.85 17.63a
ND 6.40a
22.11 112.63
ND 25.64
28.87 162.38
ND 26.90a
201 .00 554.00
43.70 471 .00
parameter was actually detected.
SCO = soluble cadmium
ECR = extractable chromium
SCR = soluble chromium
ECU = extractable copper
SCU = soluble copper
EPB = extractable lead
SPB = soluble lead
EZN = extractable zinc
SZN = soluble zinc
Streets
(N = 6)
ND
0.06
0.19
0.84
0.92
9.91
2.86
9.60
ND
ND
5.52a
ND
9.50a
ND
8.23
ND
92.00
69.00
Townhouse/
Garden
Apartments
(N = 6)
0.11
0.19
0.20
1.00
0.17
15.75
2.38
7.07
ND
1 0.34a
ND
4.79a
3.62
2.40
ND
ND
NA
59.00
The influence of contributing land use on the quality of
OGS pool water is evident in Table 1. In general, the
concentration of conventional pollutants such as nutri-
ents and suspended solids was similar to many other
reported urban stormwater runoff datasets (1). The pool
water concentrations of total hydrocarbons, TOC, and
soluble and extractable trace metals, however, were
much higher. In particular, the average concentration of
total hydrocarbons exceeded 10 mg/L in three of the five
land uses studied. Analysis of variance indicated that
gas station OGS sites had significantly greater pool
water hydrocarbon, TOC, zinc, copper, lead, and cad-
mium levels that any other OGS sites.
The influence of contributing site land use was even
more pronounced when sediment quality was analyzed
(Table 2). OGS sediments were all heavily enriched with
hydrocarbons, TOC, nutrients, and metals. The gas station
OGS sites had significantly higher hydrocarbon, TOC,
phosphorus, and metals concentrations compared with
the other four land uses. Convenience commercial and
all-day parking sites generally had higher levels than
streets and townhouse parking lots.
Effects of Automotive Land Use
Previous priority pollutant scans of stormwater runoff and
pond sediments from primarily residential land uses in
the metropolitan Washington area had failed to detect the
presence of polycyclic aromatic hydrocarbons (PAHs) (4).
Numerous PAHs and other compounds on EPAs priority
pollutant list, however, were detected in the automotive-
influenced sites of the OGS study (Tables 3 and 4).
261
-------
Table 2. Characterization of the Quality of Trapped Sediments in OGS: Effect of Land Use
Parameter
(mg/kg)
TKN
TP
TOO
HC
Cadmium
Chromium
Copper
Lead
Zinc
All-Day
Parking
(N = 8)
1,951.0
466.0
37,915.0
7,114.0
13.2
258.0
186.0
309.0
1,580.0
Convenience
Commercial
(N = 6)
5,528.0
1 ,020.0
55,617.0
7,003.0
17.1
233.0
326.0
677.0
4025.0
Gas Stations
(N = 7)
3,102.0
1 ,056.0
98,071.0
18,155.0
35.6
350.0
788.0
1,183.0
6,785.0
Streets
(N = 6)
1,719.0
365.0
33,025.0
3,482.0
13.6
291.0
173.0
544.0
1 ,800.0
Townhouse/
Garden
Apartments
(N = 6)
1 ,760.0
266.7
32,392.0
894.0
13.5
323.0
162.0
180.0
878.0
TKN = total Kjeldahl nitrogen, TP = total phosphorus, TOC = total organic carbon, HC = total hydrocarbons. All metals are extractable.
A total of 19 priority pollutants were detected in pool
water at the gas station OGS sites, compared with
seven detected at non-gas-station sites, most of which
were metals. Thirteen volatile and semivolatile priority
pollutant compounds were detected in pool water at the
gas station OGS sites. Semivolatile compounds in-
cluded phenols, naphthalene, and plasticizers, whereas
the volatile compounds included acetone, benzene,
toluene, xylene, and ethyl benzene. Most, if not all, of
these compounds are linked to gasoline and its deriva-
tives, lubricants, and cleaning agents customarily found
at gas stations (5).
An even greater number of priority pollutants, 26, were
detected in the trapped sediments of gas station OGS
sites. An additional 11 priority pollutants were indi-
cated but were below analytical detection limits. Met-
als and PAHs dominated the list of confirmed priority
pollutants. PAHs found at the highest concentrations
in the sediment included 2-methylnapthalene, naph-
thalene, phenanthrene, fluoranthene, pyrone, and
christen. Three of these PAHs have been listed as
toxics of concern by the EPA Chesapeake Bay Pro-
gram (5). Most of these PAHs are strongly associated
with gasoline and its byproducts. The gas station OGS
sites had the highest sediment metals levels, particu-
larly for cadmium, copper, chromium, lead, and zinc.
Only nine PAHs were recorded at the non-gas-station
OGS sites, and in nearly all cases the concentration in
the sediments was lower. Interestingly, the only pesti-
cides detected in the sampling were discovered at the
more residential non-gas-station sites.
Discussion
The monitoring study has several interesting implica-
tions for urban stormwater runoff and its effective con-
trol, which are discussed below.
Hydrocarbon Hotspots in the Urban
Landscape
The results suggest that hotspots of possible hydrocar-
bon and metal loading do exist in the urban landscape,
and that these are likely to occur where vehicles are
fueled, stored, or serviced. In this study, gas stations
and, to a somewhat lesser degree, frequently used park-
ing lots clearly exhibited greater hydrocarbon and metal
loading potential than more residential sites. Future
monitoring may reveal other potential hotspots such as
bus depots, loading bays, highway rest areas, and ve-
hicle maintenance operations.
The traditional management approach for urban runoff
quality has been to specify a uniform treatment standard
for all impervious areas across the urban landscape
(e.g., the first half inch of runoff). Based on the results
of this study, a more effective strategy might be to
supplement uniform standards with more stringent treat-
ment requirements when a possible hydrocarbon hot-
spot may be involved.
Only nine PAHs were recorded at the non-gas-station
OGS sites, and in nearly all cases the concentration in
the sediments was lower. Interestingly, the only pesti-
cides detected in the sampling were discovered at the
more residential non-gas-station sites.
Possible Link to Estuarine Sediment
Contamination
The bottom sediments of most of the nation's urban
estuaries are frequently contaminated with hydrocarbons,
PAHs, and metals. The sources of the ubiquitous and
pervasive contamination may include air deposition, fuel
spills, leaking underground storage tanks, leachatefrom
landfills or industrial sites, industrial discharges, and
waste oil dumping, among others. This study suggests
262
-------
Table 3. Priority Pollutants
OGS Sediments
Compound (ug/kg)
Semivolatile Organics
Napthalene
2-Methylnapthalene
Acenapthene
Fluorene
Phenathrene
Fluoranthene
Pyrene
Butylbenzylpthalate
Chrysene
bis (2-Ethylhexyl) pthalate
Di-n-octyle pthalate
Benzo (b) flouranthene
Indeno (123-cd) pyrene
Benzo (g,h,i) perylene
Di-n-butyl pthalate
Volatile Organics
Toluene
Ethylbenzene
Total xylenes
Methylene chloride
Pesticides and PCBs
Aldrin
4,4-DDT
Metals
Antimony (mg/kg)
Arsenic
Beryllium
Cadmium
Chromium
Copper
Lead
Nickel
Silver
Zinc
Cyanide and Phenols
Phenol
Cyanide
Detected in Composite
Gas
9,000
24,000
1,800
3,200
11,500
3,400
5,800
3,400
2,200
44,000
2,900
1,400
1,400
1,900
S
6,800
S
6,900
S
5.1
4.1
0.3
6.5
123
126
493
50
953
25.6
S = detected but at concentrations under the
= not present
Nongas
S
1,800
2,000
2,300
S
1,200
13,000
S
S
S
S
1,800
2,300
3,100
13,000
S
29
29
2.6
0.5
0.8
37
36
46
50
261
8.0
detection
Scans of
All Site
S
S
S
20,000
26,000
S
S
220,000
S
S
S
S
S
7,500
950
6.2
1.6
7.2
91.3
132
145
95
2
1,650
76.2
limit
Table 4. Priority Pollutants Detected in Composite Scans
Within the OGS Water Column
Compound (ug/kg)
Semivolatile Organics
Benzyl alcohol
2-Methylphenol
3,4-Methylphenol
2,4-Dimethylphenol
Napthalene
2-Methylnapthalene
bis (2-Ethylhexyl) pthalate
Chrysene
Volatile Organics
Acetone
2-Butanone
Benzene
Toluene
Ethylbenzene
Total xylenes
Pesticides and PCBs
Metals
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Copper
Lead
Nickel
Silver
Zinc
Cyanide and Phenols
Cyanide
Phenol
Gas Nongas All Site
10
22
32
16
100
43
14
12
57 13 46
16
18
140 5
41
230
1.0 1.0
1.2
8
5 6.2 5
72 8.3 15
48 3.3 5
373 65 132
86 10 24
that the washoff of leaked fuels and fluids from vehicles
may also be a key source of sediment contamination.
The significance of runoff from hydrocarbon hotspots
in sediment contamination may be great. For example,
12 out of 13 PAHs present in the sediments of the tidal
Anacostia estuary were also present in the trapped
sediments of gas station OGS sites. On average, the con-
centration in OGS sediments was seven times greater
than that recorded in the tidal estuary. Of even greater
263
-------
interest is the finding that the relative composition of
PAHs in both the river and OGS sediments was quite
similar (3). While the possible link between runoff from
hydrocarbon hotspots and estuarine sediment contami-
nation remains suggestive rather than conclusive at this
point, the subject merits further monitoring and analysis.
Opportunities for Pollution Prevention at
Hotspots
Because leakage, spills, and improper handling and
disposal of automotive products appear to be the key
source of many of the pollutants observed at hydrocar-
bon hotspots, an effective control strategy involves the
use of pollution prevention practices. For small vehicle
maintenance operations, these may include techniques
to run a dry shop, reduce run-on across work areas, use
less toxic cleaning agents, control small spills, store
automotive products in enclosures, and, perhaps most
importantly, train employees to reduce washoff of auto-
motive products from the site (6).
Implications for OGS Cleanout and Disposal
The original purpose of the study was to establish the
characteristics of trapped sediments and pool water
within OGS sites to determine the most appropriate
and safe disposal method. Based on preliminary data,
OGS residuals do not quite meet criteria to be consid-
ered hazardous for landfilling (7). Many local landfills,
however, may set more stringent criteria and will not
accept OGS sediments unless they are fully dewa-
tered. Introduction of OGS residuals into the sanitary
system appears also to be prohibited due to utility
pretreatment requirements.
Regular cleanout of OGS systems appears to be quite
rare. For example, none of the 110 OGS systems sur-
veyed in the field appeared to have been maintained in
the last year (2). Given the poor retention characteristics
of existing OGS designs, a minimum frequency of quar-
terly cleanouts would seem warranted to ensure that the
trapped residuals are removed before they are resus-
pended. The cost to cleanout an OGS system and safely
dispose of the trapped sediments, however, could ex-
ceed $400 per site visit. The need for frequent and costly
cleanouts, coupled with the ambiguities regarding the
possible toxicity of trapped sediments, raises serious
concerns about the effectiveness of the current genera-
tion of OGS systems.
Outlook for Improvements in OGS Design and
Performance
The study indicates that the current generation of
OGS systems does not retain trapped pollutants and
therefore must be maintained at an unrealistically high
frequency. Clearly, the retention characteristics of
OGS must be sharply increased if they are to become
a credible urban best management practice.
Several design improvements have the possibility of
increasing the retention of pollutants. These include
designing the OGS to be fully off-line, so that larger
runoff events bypass the OGS and reduce the frequency
of sediment resuspension; providing larger treatment
volumes; using sorptive media, fabrics, or pads within
chambers; and modifying the geometry of each chamber
to reduce turbulence in the vicinity of trapped sediments.
Until the improved retention of these design modifica-
tions is confirmed in the field, however, it may not be
advisable to use OGS systems on a widespread basis.
Given the possible importance of hydrocarbon hotspots
in the urban landscape and the apparent inadequacy of
the current generation of onsite best management prac-
tices to control them effectively, it is strongly recom-
mended that an intensive research and demonstration
program be started to evaluate alternative small-site
runoff treatment technologies.
Acknowledgments
The study was sponsored by the Maryland Department
of the Environment under an EPA Chesapeake Bay
Implementation grant. Sampling and laboratory analy-
ses were conducted by the Occoquan Watershed Moni-
toring Laboratory.
References
1. Schueler, T. 1987. Controlling urban runoff: A practical manual for
planning and designing urban best management practices. Metro-
politan Washington Council of Governments.
2. Shepp, D., and D. Cole. 1992. Afield survey of oil grit separators.
Prepared for Maryland Department of the Environment, Washing-
ton Metropolitan Council of Governments.
3. Schueler, T., and D. Shepp. 1992. The quality of trapped sediments
and pool water within oil grit separators in suburban Maryland.
Prepared for Maryland Department of the Environment.
4. JTC, Inc. 1982. Washington area NURP priority pollutant scan.
Final report prepared for Washington Metropolitan NURP Project,
Metropolitan Washington Council of Governments.
5. U.S. EPA. 1991. Chesapeake Bay toxics of concern list. Annapolis,
MD: Chesapeake Bay Program.
6. Santa Clara Valley NPS Program. 1992. Best management prac-
tices for automotive related industries.
7. Jordan, B. 1993. Oil-grit separator residual: Potential toxicity
and possible disposal methods. Washington, DC: Metropolitan
Washington Council of Governments, Department of Environ-
mental Programs.
Additional Reading
1. Metropolitan Washington Council of Governments. 1983. Urban
runoff in the Washington metropolitan area. Final NURP project
report prepared for U.S. EPA.
264
-------
Design Considerations for Structural Best Management Practices
Joseph J. Skupien
Somerset County Engineering Department, Somerville, New Jersey
Abstract
Upon selection of the appropriate structural best man-
agement practice (BMP) to address an urban runoff
management need, the design process begins. Suc-
cessful BMP design does not consist merely of achiev-
ing required technical performance levels specified in a
government regulation. To meet both the letter and spirit
of the regulation and to help encourage the public par-
ticipation vital to the future of urban runoff management,
a responsible BMP designer must also acknowledge
and address several other technical and nontechnical
considerations.
This paper emphasizes the need for a strong theoretical
understanding of standard design models and equa-
tions. It also recommends a technique for identifying and
evaluating a structural BMP's inherent maintenance,
safety, and aesthetic needs that may not be readily
apparent when using more conventional design proce-
dures. The paper also identifies the individuals and
agencies that will interact with a structural BMP during
its design and/or following its construction, and empha-
sizes the need to include their interests in the BMP
design process.
Finally, in recognition of the nascent state of nationwide
stormwater management, the paper encourages BMP
designers to contribute to the continued development of
the field by conducting their designs in an open and
objective manner and by continually seeking new and
better responses to the many stormwater management
challenges we face.
Introduction
design \di-zine\ vb 1: to conceive and plan out
in the mind; 2: to devise for a specific function or
end; 3: to conceive and draw the plans for (Mer-
riam-Webster Dictionary)
This definition succinctly describes both the scope and
sequence of activities typically undertaken by the de-
signer of a structural best management practice (BMP).
Having identified a stormwater management problem or
need that can best be solved through the construction
of a structural BMP, the designer then selects the most
appropriate type of BMP, conceptualizes its function and
operation, and determines the specific characteristics
necessary for the BMP to achieve its desired perform-
ance. Having completed this, the designer must then
transform these characteristics into a physical entity.
This is done through the development of detailed con-
struction plans and specifications, which are used to
construct the BMP in the field.
Throughout the entire endeavor, the structural BMP de-
signer must, of course, fulfill certain technical responsi-
bilities if the BMP is to comply with the standards and
requirements of the community's overall stormwater
management program. To do so, the designer must be
familiar with these program requirements as well as the
technical data, equations, and analytic techniques com-
monly used to meet them. If stormwater management is
to grow beyond its traditional concerns for stormwater
quantity to address stormwater quality and nonpoint
source (NPS) pollution, however, such technical compli-
ance is not enough. Instead, the BMP designer must
also recognize his or her unique responsibilities both to
the success of the overall stormwater management pro-
gram and to the people who will live, work, or travel past
the structural BMP they are creating. Only by fulfilling
these larger design responsibilities will stormwater man-
agement be able to achieve and sustain the public
support and participation it needs to effectively address
the complex problems that lie ahead of it. A description
of each of these design responsibilities is presented
below, along with recommendations for fulfilling each.
The Responsibilities of the BMP Designer
As noted above, the effective BMP designer must
fulfill several levels of responsibility. First and fore-
most, the designer is responsible for complying with
the technical requirements and standards of the over-
all stormwater management program of which the
265
-------
BMP will be a part. This typically includes achieving the
required level and range of peak outflow control neces-
sary to prevent or reduce downstream flooding as well
as the detention times and pollutant removal rates nec-
essary for stormwater quality enhancement. Additional
technical requirements contained in the stormwater
management program may include emergency dis-
charge capacity to insure dam or embankment safety,
as well as structural and geotechnical standards to
achieve stability and strength. The BMP designer must
be familiar with the specific technical requirements of
the stormwater management program as well as the
theoretical basis for and use of the various hydrologic,
hydraulic, structural, and geotechnical analyses typi-
cally used to comply with them.
The responsible BMP designer should not only be famil-
iar with the program's technical requirements but also
understand the program's overall intent or goals, for the
designer must recognize that the program's technical
requirements are only the means through which we
hope to achieve the program's goals or ends. As such,
a structural BMP will contribute more towards those
goals if its designer understands, for example, not just
what detention time the BMP should have, but why it
should have one, why it should be a certain duration,
and what will happen if it does not. Such understanding
also produces BMP designs that are better able to
achieve satisfactory results over a much wider range of
real-world conditions than the more limited conditions
that are normally analyzed during the design process.
In addition, due to the inherent complexities of stormwa-
ter quality and nonpoint source (NPS) pollution, we have
not been able in many instances to define the technical
requirements of our stormwater management programs
as well as we have been able to specify their goals. For
example, it is considerably easier to select a goal of 80
percent removal of suspended solids from stormwater
runoff than it is to specify the exact technical measures
that must be implemented to do so. This disparity be-
tween means and ends can be overcome to a great
degree by the responsible designer who, aware of the
disparity, is willing and able to look behind and beyond
the program's somewhat limited technical requirements
and produce designs that do a better job of achieving
the program's goals.
Another BMP design responsibility is based on the fact
that the final product of the designer's efforts will be a
real structure that must be constructed and maintained
and that will occupy space in a real environment. As
such, it is vital that the BMP be both simple and practical
in terms of construction, materials, operation, mainte-
nance, and safety. Such characteristics can only be
achieved by a designer who is aware of their importance
and can define them in physical terms. In addition, such
vital characteristics cannot, at times, be achieved by
strictly adhering to a stormwater management pro-
gram's technical standards and may, in fact, require that
they even be ignored or broken. Such instances de-
mand the involvement of a responsible designerwho will
be able to achieve a more informed, effective balance
between technical compliance and practicality than is
achievable through strict compliance alone.
In the design of any structural BMP, cost must also be
an important factor, and the responsible designer not
only appreciates this fact but also can accurately and
objectively determine both the benefits that a structural
BMP provides and the costs of doing so. A true measure
of a BMP's cost effectiveness can only be achieved by
understanding, quantifying, and comparing both. To do
so, the designer has a responsibility to fully understand
both the cost of BMP construction, operation, and main-
tenance and the relative values or benefits to be gained
from it. This requires, among other traits, a high degree
of objectivity to ensure that the costs and benefits de-
termined by the designer are based on reality and not
the interests or desires of his or her client or supervisor,
or a government regulator.
Finally, the responsible BMP designer understands the
importance of professionalism and will always conduct
the design process in an open, honest, and objective
manner. In view of the nascent state of stormwater
management nationwide, such tenets are particularly
vital if we are to close the present gap between what we
seek to gain from stormwater management and how we
can best achieve it. Such conduct will also enable us to
more quickly identify uncertainties, conflicts, and errors
in our present understanding of stormwater runoff and
NPS pollution and to develop more effective and effi-
cient solutions.
BMP Design Considerations: Points To
Ponder
From the above, it can be seen that the responsible
BMP designer must not merely be concerned with the
technical requirements of a stormwater management
program but, instead, must strive to produce facilities
that also achieve and even advance the program's goals
and intentions. The structural BMPs that result from
such an effort will become assets to the community that
they serve and promote the public interest and involve-
ment necessary for overall program success. The BMP
must also be practical, safe, aesthetically pleasing, easy
to build, and even easier to maintain. Faced with such
a formidable list of requirements, the responsible de-
signer must not only bring competent technical ability to
the design process but also an informed, open attitude
and even a sense of mission or purpose. To help pro-
mote such an attitude and more fully prepare the BMP
designer for the job ahead, the following points regard-
ing BMP design, construction, and operation are of-
266
-------
fered. The BMP designer should consider these points
before undertaking a design effort.
Interested Parties
To produce a BMP design good enough to earn an
"approved" stamp from a stormwater management pro-
gram regulator (who is presumably interested in ensur-
ing compliance with the program's regulations), a BMP
designer must identify with those interests and make
sure they are reflected on the construction drawings. To
further ensure that the BMP will truly be an asset to the
community and will make a positive statement about the
value of stormwater management, the BMP designer
must consider several interested parties.
The Client
Including the client on a list of parties having an interest
in a BMP design should not come as a surprise; how-
ever, a review of what the client's interests really are just
may be. Therefore, the responsible BMP designer will
not automatically assume to know the client's interests
(however obvious they may appear) but will instead fully
discuss them with the client.
The prospect of such a discussion may then lead the
designer to ask the following question: What should the
client's interests be? Does the client have a misin-
formed or misguided attitude towards the goals of storm-
water management? Is this attitude based on a lack of
understanding or information? In such cases, the re-
sponsible designer can, through education (and a touch
of diplomacy), both expand the client's understanding
and improve his or her attitude towards stormwater
management, thereby enhancing the designer's own
chances of producing a positive BMP design.
The Regulator
Similar to the client, the regulator is also an obvious
choice for an interested party list. Once again, the fol-
lowing questions may be raised: What are the regula-
tor's interests, and what should they be? Because a
regulator's review of a BMP design can sometimes stray
from the program's technical standards into more sub-
jective areas (due, in part, to a lack of such standards),
it is often helpful to know what interests the regulator
has stored up in those areas. Are those interests both
in keeping with the goals of the stormwater manage-
ment program and within the program's (and, therefore,
the regulator's) jurisdiction?
For example, a regulator may have a strong interest in
promoting proper land use as a means of achieving a
program's goals. If regulating land use is beyond the
program's scope or authority, however, then such inter-
ests have no rightful place in the regulator's review of
the BMP design. Should such interests become part of
the review, it is the designer's responsibility to point this
fact out and redirect the review back to its proper direc-
tion. In doing so, all of the diplomatic skills the designer
has developed from educating the client will prove in-
valuable.
Similar to the client, a BMP designer may also encoun-
ter a regulator who, through a lack of knowledge or an
abundance of wrong information, either misunderstands
the program's requirements or lacks the ability to fully
ensure their compliance. Once again, the responsible
BMP designer can, through education and a competent,
comprehensive design, expand the regulator's under-
standing and ability so that the designer's intentions can
be better understood.
The Constructor
As noted earlier, one of the key responsibilities of the
structural BMP designer is to transform the BMP from
concept to reality by preparing detailed plans and speci-
fications of how it should be built. It is then up to the
constructor to finish the project by actually building the
BMP from these plans and specifications. Therefore, the
responsible designer appreciates the efforts of the con-
structor and does not see his or her own efforts as an
independent exercise, but rather as an integral part of a
much larger processa process that requires the con-
structor to complete.
As such, the responsible BMP designer recognizes and
responds to the constructor's interests by producing a
well thought-out design that can be constructed as eas-
ily and simply as possible. Because this may not always
be possible, particularly when faced with complex per-
formance requirements or difficult site conditions, the
responsible designer also takes extra care to bring any
difficult or unusual aspects of the design to the construc-
tor's attention before the start of construction, even
consulting with the constructor during the design phase
to mutually devise the best construction technique, ma-
terial, or sequence.
Under ideal circumstances, the BMP designer will also
continue his or her involvement in the project throughout
the construction phase and will work with the constructor
to correct mistakes, address oversights, and develop
revised designs as necessary to overcome problems
that may be encountered in the field.
The Maintainer
Once construction of the BMP has been completed, the
designer's involvement with the process (assuming it
lasted through construction) normally ends. However,
there are interested parties whose involvement with the
BMP is just about to start and whose interests the
designer must also consider. These are the mainte-
nance personnel who will be responsible for mowing the
267
-------
grass, removing the sediment, clearing the debris, man-
aging the habitats, and performing the necessary re-
pairs at the BMP for the rest of its serviceable life.
Similar to the constructor, the maintainer's actions will
be governed by what the designer creates on paper.
Because construction has been completed and the de-
signer has moved on to other projects, however, it is
considerably more difficult for the maintainer to have
deficiencies or oversights in the design corrected.
As such, the designer must understand and address the
interests of the maintainer before it is virtually too late.
As described in more detail in later sections, this can be
accomplished by designing a facility that, optimally, re-
quires a minimal amount of maintenance that can be
performed as easily as practicable.
The Resident
This interested party may also be the worker, commuter,
shopper, student, or local government official who will
interact with the structural BMP on a regular basis. This
interaction may be physical (through the sense of touch,
sight, hearing, or smell) or psychological (as anyone
who has worried about children's safety or the value of
his or her property will understand).
In any case, these are the people who have, perhaps,
the strongest interest in seeing that a positive BMP
design is achieved. These are also the people who will
soon be asked to participate in the community's non-
structural stormwater management programs by chang-
ing some of their aesthetic values and even their
lifestyles. Therefore, the person responsible for produc-
ing the BMP design must be aware of their interests and
incorporate them into the design as well.
Operating Conditions
Just as a wide range of people have an interest in the
BMP design, the BMP must operate under a wide range
of conditions. Just as the BMP designer may fail to
recognize the full range of interests, he or she often fails
to consider all of the real-world conditions that the BMP
will be subject to by focusing solely on those design
conditions necessary for official program approval. This
is unfortunate, because the design conditions that re-
ceived all of the designer's attentions will, in reality, only
occur during a small fraction of the BMP's existence.
However, its performance during the remainder of its
existence, while ignored by the designer, will largely
determine the community's opinion of its value.
Therefore, it is important that the BMP designer be
aware of all of the weather and other site conditions to
which the BMP will be subjected.
Design Conditions
These are obviously the designer's first concern and, as
noted above, are normally established by the commu-
nity's stormwater management program. In the case of
runoff quantity control, these conditions usually include
either a single event or a range of relatively extreme
storm events, the runoff from which must be stored and
released at a predetermined rate. New Jersey's Storm-
water Management Regulations, for example, require
that the runoff from a proposed land development site
for the 2-, 10-, and 100-year storm events be controlled
so that the peak rate of site runoff after development for
each storm does not exceed the peak rate that existed
before development. The Somerset County, New Jer-
sey, standards are stricter, requiring a peak rate after
development that is actually less than existing to ac-
count for development-induced changes in runoff vol-
ume and overall hydrograph shape as well.
In the case of stormwater quality control, typical design
conditions may include the temporary storage and slow
release of the runoff from a much smaller, more frequent
storm event to promote pollutant removal through sedi-
mentation. For example, the New Jersey Stormwater
Management Regulations require the temporary stor-
age of runoff from a 1-year storm event, with release
occurring over 18 to 36 hours depending on the charac-
ter and intensity of the proposed development. The state
of Delaware requires extended storage of the first inch
of runoff from a proposed site, with release occurring
over 24 hours.
Whatever exact design conditions the stormwater man-
agement program may specify, it is vital that the struc-
tural BMP function properly under them or the goals of
the program cannot be met.
Extreme Conditions
In addition to the program's design conditions, which
have been selected with the goal of runoff quantity
and/or quality in mind, the responsible BMP designer
must also recognize that more extreme storms may also
occur. Therefore, due to the inherent dangers of storing
runoff and the exceptionally large quantities of runoff
that can be produced by these extreme events, it is vital
that the BMP designer also address the goal of safety
by ensuring that the BMP will also function properly
under such extreme conditions. This will typically in-
clude the provision of an emergency spillway or other
auxiliary outflow device that will safely convey the ex-
treme event runoff that exceeds the capacity of the
BMP's normal outflow structure. It will also include pro-
tection of critical portions of any embankment, dam, or
discharge points that may be subject to scour or erosion
from the high flow velocities generated by the storm
event.
268
-------
Dry Weather Conditions
While design and even extreme storm conditions can be
expected to occur periodically, the most common oper-
ating condition at a structural BMP will be dry weather
with various seasonal temperatures, winds, humidities,
and periods of daylight. While dry weather may be the
most prevalent operating condition, it is also the one that
is most frequently overlooked by the BMP designer. As
a result, how the BMP will look, smell, and even sound
during the majority of its operating life is then left to
chance. This can be particularly unfortunate forthe BMP
maintainer and, more critically, the resident, worker, or
commuter who will interact most often with the BMP
during dry weather conditions. Therefore, the responsi-
ble BMP designer will not only address extreme storm
events but will also make sure that the BMP performs
satisfactorily when it isn't raining at all.
Design Methodologies
Before starting the actual design process, the responsi-
ble designer will have an adequate understanding of the
selected design methodologies. These methodologies
can cover such aspects as rainfall-runoff computations,
hydrograph routings, infiltration and ground-water
movement, structural design, and geotechnical issues.
In doing so, the designer's understanding should in-
clude the methodology's theoretical basis, assumptions,
limitations, and applicability. In addition, the responsible
designer will also have an understanding of both the
accuracy needed to perform the design and the accu-
racy of the method he or she has selected to do it. From
this, the responsible designer will neither waste time
producing unneeded accuracy nor attempt to achieve a
level of accuracy beyond the limits of the method. Fi-
nally, the responsible designer will understand the sen-
sitivity of each of the method's input variables and will
appropriately allocate his or her time and resources in
developing each one.
Facility Type
The final point for the BMP designer to ponder before
beginning the actual design process is the type of struc-
tural BMP to be used. Presently, a wide range of facili-
ties are available for use, ranging from relatively simple
vegetated filter strips and swales to large ponds and
constructed wetlands. Selection of the appropriate BMP
depends on several factors, including program require-
ments, BMP location, site conditions, maintenance
needs, safety, cost, and performance characteristics.
Similar to BMP operating conditions, the BMP designer
may often consider only a few of these factors, most
notably program requirements (keep the regulator
happy) and cost (keep the client happy, too), in making
his or her selection. The responsible designer, how-
ever, will recognize the performance, needs, uncertain-
ties, and risks inherent in each type of BMP and will then
select (or help influence the selection of) the most ap-
propriate type of BMP for the site. This process typically
begins with the identification of the fundamental charac-
teristics of each type of BMP, along with the project's
physical, economic, social, and regulatory constraints.
The process then becomes one of comparison and
analysis, with the best match found by eliminating the
worst.
For example, a site with porous soils, low ground-water
table, and close proximity to residences may not be best
suited for a wet pond or constructed wetland, while the
active recreational needs of the residents may benefit
from a dry, extended detention basin that can also serve
as an athletic field. Although perfect matches rarely
occur, comparisons and analyses such as this will help
reduce the number of potential BMPs, improve the thor-
oughness and objectivity of the overall selection proc-
ess, and ideally produce the optimal facility type. This
process can even help identify inherent weaknesses in
or problems with the selected type, which will enable the
responsible BMP designer to devote additional time and
effort towards correcting them during the design phase.
To undertake such a selection process obviously re-
quires a designer who understands the fundamental
characteristics and needs of each BMP and who can
objectively assess all of the pertinent site constraints.
Such a designer must also be willing and able to confront
the differing opinions of other, less objective or informed
parties (including the regulator and client) to ensure that
the best BMP is selected. As noted throughout this
paper, achieving an optimal BMP design is a complex
and demanding process that must incorporate numerous
interests and requirements. Starting the process with the
wrong facility type, however, transforms a complex and
demanding process into an impossible one.
BMP Design Considerations: A Checklist
Having completed the BMP selection process with
honor, idealism, and design contract still intact, and
armed with both the necessary technical and regulatory
knowledge and economic and social sensitivity, the re-
sponsible BMP designer is ready to begin the actual
design process. Presented below is a checklist of six
key design considerations to help guide this effort. Ide-
ally, these six items have or will become an integral part
of the designer's thought process and will automatically
be included in each design effort. These items can also
serve as guidelines for those responsible forthe review
and approval of specific BMP designs as well as goals
for those developing new stormwater management pro-
grams.
269
-------
Safety
For several reasons, the safety of the structural BMP
must be the primary concern of the designer. Due to its
"structural" nature and, in many instances, the fact that
it will impound water either permanently or temporarily,
the structural BMP will inherently pose some degree of
safety threat.
Those at risk include people living, working, or traveling
downstream of the BMP whose safety and/or property
will be jeopardized if the BMP were to fail and release
stored runoff. Because this is a risk that has been cre-
ated solely by the BMP, the designer must ensure that
the probability of such a failure is acceptably small.
Also at risk at a structural BMP are maintenance per-
sonnel, inspectors, mosquito control personnel, and
equipment operators, who must work in and around the
facility. Typical hazards include deep water, excessively
steep slopes, slippery or unstable footing, limited or
unsafe access, and threats posed by insects and ani-
mals. As noted above, the responsible BMP designer
understands the importance of facilitating BMP mainte-
nance. Providing a safe working environment for the
BMP maintainer is one important way to do it.
Finally, those living, working, attending school, or play-
ing in the vicinity of a structural BMP may also be at
risk, particularly if the BMP serves both as a stormwater
management and recreational facility. Once again,
such things as standing water, steep slopes, unstable
footings, and insect and animal bites must be ad-
dressed by the designer to avoid creating a facility that
is a detriment to the community it is intended to serve.
Failure to do so will only alienate those members of the
community who will be asked to play a vital role in
future stormwater management efforts.
Performance
Having made a strong commitment to safety, the BMP
designer must then consider facility performance. This
normally includes achieving the necessary stormwater
detention times, flow velocities, settling rates, peak flow
attenuation, and/or ground-water recharge for the
range of storm events to be managed. Again with a
commitment to safety, the designer must also ensure
that the BMP performs adequately under emergency
conditions, most notably when the peak rate and/or
volume of runoff flowing into the basin exceeds the
discharge capacity of the BMP's principal outlet. This
will require the inclusion of emergency or auxiliary
outlets in the BMP to safely convey this excessive
inflow through the BMP without jeopardizing its struc-
tural integrity.
In most instances, the performance standards that the
BMP design must meet will be specified in the storm-
water management program's regulations. Experience
has shown, however, that these performance standards
may, at times, be vague, contradictory, or even impos-
sible to meet. For example, many BMP designers have
been confronted with a requirement to reduce both the
peak rate and total runoff volume from a developed (or
developing) watershed to predeveloped levels. This has
often lead to much head scratching, for the solution
normally requires the use of an infiltration or recharge
basin which, due to site constraints, may either be im-
practical or impossible. Faced with such circumstances,
the responsible designer looks beyond the written regu-
lations and investigates their origins and true intent with
regulatory personnel. Direct inclusion of these individu-
als in the design process will also help ensure more
positive overall results.
Constructability
Up until now, the designer's efforts to achieve adequate
BMP safety and performance levels have been
achieved only on paper or computer disk. Because the
ultimate goal of the design process is to actually create
a BMP, the BMP designer must also give careful consid-
eration to how it is to be constructed. Achieving excep-
tional safety and performance characteristics in a BMP
that cannot actually be built solves nothing and wastes
much. Achieving required levels of safety and perform-
ance in a BMP that can be reconstructed with relative
ease using readily available materials, equipment, and
skills is commendable and not only solves a specific
stormwater management problem, but also helps to
advance the community's overall program. "Constructa-
bility" can be defined as a measure of the effort required
to construct a structural BMP. A BMP that is highly
"constructable" utilizes materials that are readily avail-
able, relatively inexpensive, and do not require special
shipping or handling measures. They will be both dura-
ble and easily modified in the field to meet specific site
conditions. Similarly, the construction techniques and
equipment required to construct the BMP will also be
relative simple, straightforward, and familiar to the peo-
ple who will be performing and operating them.
It is important to note that the above description is not
intended to discourage the use of new or innovative
materials or construction techniques, nor to inhibit crea-
tivity in the BMP design process. In fact, innovation in
design and construction is vital to the future growth of
stormwater management. Instead, the above descrip-
tion of "constructability" is intended to remind design-
ers that they must consider the construction aspects
of the BMP in the design process and strike a balance
between performance and safety requirements,
constructability, and innovation for each design they
undertake.
270
-------
Maintenance
The same reminder stated above for constructability
must also be said for BMP maintenance. Similar to
construction, the degree of effort and expense required
to adequately maintain a structural BMP will help deter-
mine the overall success of its design. A BMP with
manageable maintenance needs can be expected to
remain in reasonably good condition and has a
stronger chance of becoming an asset to the sur-
rounding community. On the other hand, a BMP with
excessive maintenance needs is likely to be ne-
glected and will quickly become a community liability.
As such, BMP maintenance can directly effect the
overall success of the community's stormwater man-
agement program.
The BMP designer can help determine a BMP's main-
tenance needs by considering several aspects of that
maintenance in the design process. First, the BMP
design should include the use of durable materials
that are able to withstand the many and varied physi-
cal conditions that the BMP will experience over its
lifetime. Secondly, suitable access to key BMP com-
ponents and areas is vital if required maintenance
levels are to be achieved. This will include provisions
for walkways, staging and disposal areas, access
hatches and gates, and safe, stable working areas.
The frequency of maintenance has a large impact on
both maintenance cost and quality, and it is the de-
signer's responsibility to achieve an appropriate level.
Finally, the BMP designer should always strive to
minimize the overall amount of maintenance at the
BMP and to make that amount as easy as practicable
to perform.
Cost
Inclusion of a BMP's cost in a list of design considera-
tions is not surprising. Once again, however, a review of
the full costs associated with a structural BMP may yield
a few surprises that may increase designers' under-
standing and encourage them to give BMP costs the full
consideration they deserve.
The most obvious BMP cost is its construction. This can
be estimated with reasonable accuracy and is the cost
most directly borne by the designer's client. As such,
designers most often focus on this cost during the de-
sign process to the exclusion of all others.
What other costs may be overlooked? One may be the
designer's own fee, which is part of the overall BMP
cost but which has probably been excluded from con-
sideration because it has already been determined.
The designer's fee, however, has a direct impact on
the BMP design because it determines the effort and
resources the designer uses to produce it. The level
of effort expended during the BMP design can have a
similarly direct effect on the effort and cost of both
construction and maintenance. The greater cost of a
more thorough BMP design can ultimately result in cost
savings to the client during subsequent project stages.
Therefore, while this is not a signal to BMP designers to
raise theirfees, it is meant to remind designers that their
fee is part of the overall BMP cost and that it is their
responsibility to determine what level of design effort
and cost represents the best investment for both the
client and the community.
Another portion of total BMP cost that is frequently
overlooked is the cost associated with its maintenance.
While this cost on an annual basis is usually a small
percentage of the construction and even the design
cost, it must be remembered that, unlike construction or
design, maintenance costs are recurring and must be
paid throughout the life of the BMP. Therefore, while a
maintenance cost savings may appear to be insignifi-
cant on a per-operation basis and not worth the extra
investment in design or construction required to achieve
it, its value may be viewed quite differently when multi-
plied by the numerous times it will be realized. As such,
an added investment in design to produce a trash rack
that will require less frequent cleaning or an added
investment during construction to reduce the frequency
of repairs may quickly yield a positive return in the form
of reduced maintenance costs. Similar conclusions can
be reached for many other design and construction
efforts, such as providing better access, using more
durable materials, and selecting a BMP that best suits
site conditions.
Comm unity A cceptance
The final recommended design consideration once
again involves those people who may have the greatest
interest in the structural BMP. Not coincidentally, these
are the same people who will have the greatest role in
the various nonstructural programs intended to augment
and even replace structural BMPs in the future. To pro-
tect those interests and encourage assumption of that
role, it is up to the designer to help achieve a structural
BMP that will be reviewed as a community asset rather
than a liability.
As discussed above, this can be achieved by consider-
ing the aesthetic value of the BMP, preventing the crea-
tion of nuisances and safety threats, as well as
achieving required performance levels. Through all
three, stormwater management gains the under-
standing and credibility it requires within the community.
Suggested Design Review Techniques
Throughout this paper, the BMP designer has been
encouraged to consider a wide range of interests,
operating conditions, costs, and other responsibilities
271
-------
throughout the design process. Presented below are
two recommended techniques to help accomplish it.
They can either be used as review techniques following
completion of a preliminary BMP design or, ideally, be
incorporated into the overall design process and used
continually throughout it.
Spend a Mental Year With the BMP
To use this technique, the BMP designer simply imag-
ines conditions at the constructed BMP throughout a full
year. This should not only include rainy and sunny
weather, but also light rain showers (with little or no
runoff), light and heavy snowfalls, and frozen ground
conditions. Other site conditions may include late
autumn, when trees have lost their leaves and the BMP
has found them, and hot, dry weather or even drought,
when the turf or other vegetation is stressed or even
killed. Finally, the designer may wish to imagine what
the BMP will be like at night.
As these conditions are visualized, the designer should
also imagine how those conditions may effect not only
the operation of the BMP itself but also the people that
will interact with it. Can blowing snow completely fill the
BMP, leading the unsuspecting pedestrian to think that
the grade is level? Will the outlet structure's trash rack
be particularly prone to clogging by fallen leaves, par-
ticularly from the trees the designer just specified forthe
BMP's bottom?
What about the ice that will form on the surface of a pond
or constructed wetland? Can someone fall through?
Could that someone be a child taking a shortcut home?
How will people be warned not to? How will they be
rescued if they do anyway? What about night condi-
tions? Will the constructed wetland next to the office
parking lot that is so attractive during summer lunch
hours become a safety hazard to workers walking to
their cars in the winter darkness? Or will that same
summer sun and a lack of rainfall produce some of the
wonderful aromas of anaerobic decomposition?
At first, it may be exasperating to realize that the possi-
ble site conditions and circumstances can be as numer-
ous and varied as the number of possible BMP uses.
But then again, that is the point of the exercise. It is
intended to help the designer consider and design for
all conditions at the BMP, not just the 1- or 100-year
storm event required by the regulations. In doing so, the
BMP designer will not only meet the letter of the regu-
lations but will raise the spirit of the entire stormwater
management program.
Who, What, When, Where, and How?
The second recommended review technique a BMP de-
signer may employ is to simply focus on one or more
characteristics or functions of the BMP and then ask (and
attempt to answer) the above questions. For example,
let's consider BMP maintenance and then ask:
Who will perform it? Does the BMP design require
specialists, or will someone with general mainte-
nance equipment and training be able to do the job?
What needs to be maintained? Preparing a list of all
the BMP components included in a design that will
need attention sooner or later may prompt a revised
design with a shorter list.
When will maintenance need to be performed? Once
a day? A week? A year? Remember, the recurring
costs of BMP maintenance can be substantial. In
addition, can maintenance only be performed during
dry weather? If so, what happens during 2 or 3 weeks
of wet, rainy weather? What happens when repairs
need to be made or debris removed during a major
storm event? In terms of effort and possible conse-
quences, it is easier for the designer to find answers
to these questions now than for maintenance or
emergency personnel to scramble for them later.
Where will maintenance need to be performed? Will
the maintainer be able to get there? Once there, will
he or she have a firm, safe place to stand and work?
In addition, where will such material as sediment,
debris, and trash removed from the BMP be disposed
of? Before answering that question, do you know
what is in it? Are there toxics or hazardous materials
in the sediment or debris? If so, is the place you
originally intended to use still suitable? Once again,
it is easier to address these questions now then when
the dump truck is loaded and the engine's running.
How will maintenance be performed? The simple in-
struction to remove the sediment or harvest the vege-
tation can become rather complicated if no provisions
have been made to allow equipment to get to the
bottom or even into the site. "Mowing the grass" can
become "risking your limbs" on long, steep slopes.
How will you explain to your client why the BMP in
which he or she has invested has become a liability
to themselves and their community?
Similar exercises can be performed with constructors,
inspectors, and residents as the object of inquiry. For
example, where will the nearest residence be? How will
the constructor build the emergency spillway? When will
the inspector need to visit to check for mosquitos?
Similar to the "mental year" review technique, the ques-
tions raised in this technique are intended to make the
designer more aware of all the possible impacts the
BMP may have and, further, to encourage the designer
to address those impacts now, during the design phase,
rather than leave them for others to cope with later. Even
if the designer cannot completely answer all of the ques-
tions, he or she will be able to advise the others of any
272
-------
unavoidable needs or problems that will be inherent in
the BMP and allow them to adequately prepare.
Summary
Stormwater management must still be considered a
relatively new endeavor, particularly on a nationwide
basis. Despite its nascent state, it has been charged
with the responsibility of addressing some very complex
environmental problems. For stormwater management
to grow to the level demanded by this charge, the de-
signers of structural BMPs must be willing to assume a
degree of responsibility for that growth. BMP designers
can fulfill that responsibility by producing BMP designs
that do not merely meet official regulations and stand-
ards, but help inspire new, better, and more comprehen-
sive ones. BMP designers must also incorporate a wide
range of interests into the BMP design, including those
held by stormwater program regulators, BMP construc-
tors and maintainers, and all those members of the
community who will interact with the BMP over its life-
time. During the design process, BMP designers must
not only consider the BMP's performance but also its
cost, durability, ease of construction, and maintenance
needs. Finally, BMP designers must always recognize
the BMP's impacts both on the community around it and
on the stormwater management program with which the
community has entrusted them.
273
-------
Targeting and Selection Methodology for Urban Best Management Practices
Peter Mangarella, Eric Strecker, and Gail Boyd
Woodward-Clyde Consultants,
Oakland, California, and Portland, Oregon
Abstract
Selecting best management practices (BMPs) to imple-
ment as part of a stormwater management plan is quite
difficult and controversial because of a variety of tech-
nical, regulatory, institutional, and financial factors and
constraints. Specifically, the nature and sources of
stormwater-borne pollutants and the water quality and
ecological problems these pollutants cause are not well
understood. The cost, effectiveness, and applicability of
many BMPs are also not well understood, although
several BMP manuals summarize existing information.
The federal National Pollutant Discharge Elimination
System (NPDES) stormwater regulations provide flexi-
bility in selecting BMPs to control urban pollutants. EPA
gives only general guidance on the types of BMP pro-
grams that are desirable and does not require the im-
plementation of specific BMPs. Several other factors
contribute to difficulties in selecting and implementing
BMPs. In many cases, institutional jurisdictions do not
correspond to watershed boundaries, and water man-
agement institutions' roles and responsibilities are frag-
mented for effectively dealing with the myriad nonpoint
sources of pollution associated with stormwater drain-
age systems. Finally, the availability of funds, which are
currently very limited, significantly determines BMP
implementation.
This paper provides guidance on the selection of BMPs
given this current environment and based on experience
in developing stormwater management plans forareawide
programs, individual municipalities, industries, develop-
ments, and government facilities. The paper describes
the current tools available for BMP selection, a 10-step
"model" selection process, and case studies for a large
areawide municipal program and for an industrial facility.
Introduction
In October 1990, the U.S. Environmental Protection
Agency (EPA) issued regulations requiring certain mu-
nicipalities and industries to select and implement best
management practices (BMPs) to control pollution as-
sociated with stormwater runoff and dry weather dis-
charges into storm drain systems. Such BMPs would be
selected and described in stormwater management
plans and implemented in compliance with an NPDES
permit. The specific regulatory language in Section
402(p) of the Clean Water Act is "Permits for discharges
from municipal storm sewers shall require controls to
reduce the discharge of pollutants to the maximum ex-
tent practicable . . .." The maximum extent practicable
(MEP) standard has a legal definition; however, consid-
erable uncertainty exists in the regulated community
about what constitutes technical compliance with the
MEP standard.
Other existing and proposed regulations require BMP
selection. Section 303 of the Clean Water Act requires
that delegated states and EPA establish total maximum
daily loads (TMDLs) for designated "water quality lim-
ited" water bodies. The TMDL process considers both
point and nonpoint sources. For nonpoint sources, water
quality management plans must be developed to meet
load allocations for urban and other land uses. The 1990
Coastal Zone Act Reauthorization Amendments (CZARA)
require the development of state nonpoint source con-
trol plans for the coastal zone using BMP guidance
recently released by EPA and the National Oceanic and
Atmospheric Administration (NOAA).
Finally, watershed planning is gaining favor as a way of
meeting water quality goals for the nation's waters. The
watershed planning approach requires examination of
all land uses and activities in a watershed and develop-
ment of BMPs to protect water quality. EPA is consider-
ing the watershed approach for the phase II portion of
the NPDES program.
This paper describes our experience in selecting BMPs
for clients complying with the NPDES stormwater regu-
lations; the process would also be applicable to TMDL,
coastal zone, and watershed planning. We discuss
types of BMPs and sources of information on BMPs
274
-------
available for developing management plans. Based on
our experience, we also describe the attributes of a
good selection process and describe the steps involved
in a "model" selection process. Because of numerous
site-specific conditions that enter into any selection
process, the actual process chosen must be adapted to
each situation. To illustrate how such a process might
be adapted to different circumstances, we describe two
case studies, one for a large areawide municipal pro-
gram and one for multiple federal facilities regulated as
industrial dischargers.
Best Management Practices
Although BMPs may be organized in many ways, it is
useful in the selection process to distinguish controls
based on how they function. BMPs based on function
are often considered as source controls, treatment con-
trols, and hydraulic controls.
Source controls are intended to prevent pollution in
the first place (i.e., pollution prevention) or to inter-
cept the pollutants before they enter the storm drain-
age system. Preventing pollution in the first place
often involves behavior modification, which requires
public information and education, an important
source control BMP. Street sweeping and catch basin
cleaning are examples of source controls that inter-
cept pollutants before stormwater carries them into
receiving waters.
Treatment-based controls are controls that remove
pollutants from stormwater, usually through some struc-
tural means such as a detention basin or grassy swale.
Hydraulic controls are structural controls that reduce
the volume of runoff (or otherwise alter the runoff
hydrograph) or divert flows away from source areas.
Examples of hydraulic controls are infiltration sys-
tems.
In general, the effectiveness of these types of controls
are not well understood. The effectiveness of treatment
and hydraulic controls generally can be measured
through monitoring, and there is an increasing body of
literature regarding the effectiveness of treatment and
hydraulic controls under limited conditions. Federal,
state, and local agencies have developed numerous
BMP guidance manuals to help identify, select, and
design BMPs. The following is a partial list of manuals,
starting with design manuals that contain detailed con-
trol selection and design information.
U.S. EPA. 1993. Handbook: Urban runoff pollution
prevention and control planning. EPA/625/R-93/004.
City of Austin Environmental Resource Management
Division. 1991. Environmental criteria manual. Envi-
ronmental and Conservation Services Department
(February 19).
Metropolitan Washington Council of Governments
(MWCOG). 1987. Controlling urban runoff: A practi-
cal manual for planning and designing urban BMPs.
Prepared for Washington Metropolitan Water Re-
sources Board (July).
State of Florida Department of Environmental Regu-
lation. 1988. The Florida development manual: A
guide to sound land and water management (June).
State of Washington Department of Ecology. 1992.
Stormwater management manual for the Puget
Sound Basin (the technical manual) (February).
Urban Drainage and Flood Control District. 1992. Ur-
ban storm drainage criteria manual. Denver, CO
(September).
Metropolitan Washington Council of Governments
(MWCOG). 1992. Design of stormwater wetland sys-
tems. Prepared for the Nonpoint Source Subcommit-
tee of the Regional Water Committee (October).
The following documents primarily discuss control effec-
tiveness and do not contain control selection and design
information:
City of Austin Environmental Resource Management
Division. 1990. Removal efficiencies of stormwater
control structures. Environmental and Conservation
Services Department (May).
Metropolitan Washington Council of Governments
(MWCOG). 1992. A current assessment of urban best
management practices. Prepared for the U.S. Envi-
ronmental Protection Agency (March).
U.S. EPA. 1990. Urban targeting and BMP selection:
An information and guidance manual for state
nonpoint source program staff engineers and man-
agers. Region 5, Water Division, Chicago, IL 60604
(November).
Metropolitan Washington Council of Governments
(MWCOG). 1992. Analysis of urban BMP perform-
ance and longevity.
U.S. EPA. 1993. Guidance specifying management
measures for sources of nonpoint pollution in coastal
waters. EPA/840/B-92/002. Washington, DC (Janu-
ary). (Includes costs.)
California State Stormwater Task Force. 1993. Cali-
fornia BMP handbooks for municipal, construction,
and industrial/commercial (April).
Finally, the following document addresses BMP costs:
Southeastern Wisconsin Regional Planning Commis-
sion. 1991. Costs of urban nonpoint source water
pollution control measures. June.
These manuals describe BMP function, requisite site
conditions, existing performance information, and cost
275
-------
ranges. In general, these manuals are well written and
provide a good starting point for developing an under-
standing of the advantages and disadvantages of many
treatment-based controls. For some BMPs, there is lim-
ited information on effectiveness and cost; for these, pilot
testing may be helpful under site-specific conditions.
Treatment-based controls are especially applicable in
construction and new developments, where structural
measures may be incorporated into the construction
process and site design. The cost of constructing and
maintaining treatment-based controls is a major con-
cern to municipal and industrial dischargers.
In contrast to treatment-based controls, source control
effectiveness in terms of water quality improvement can-
not easily be measured, if at all. For example, the effect
of a public education program on improving water qual-
ity cannot be determined, although some public educa-
tion activities obviously are more effective than others.
The effectiveness of street sweeping and catch basin
cleaning on water quality requires careful and expensive
paired catchment types of studies. Source controls are
generally considered the most cost-effective long-term
solution because they address the cause of the prob-
lem; thus, we see many programs focusing on source
control measures.
Attributes of a Good Selection Process
The following sections describe some attributes of a
good selection process.
Keep It Simple and Straightforward
BMP selection for nonpoint source controls is in its infancy
compared with point source controls, for which treat-
ment technologies and associated costs are well under-
stood. Instead of traditional cost benefit analysis, nonpoint
source BMP selection is more of an art and requires
experience, sound judgement, and common sense.
Though the process of selection may involve several
steps, the process itself must be easily understandable
and accepted by the various interest groups involved,
including public agency staff and decision-makers, envi-
ronmental groups, and regulatory personnel.
Document the Process
It is essential to carefully document the process by
which BMPs were selected and the various assump-
tions and considerations made during the selection
process. In other words, the process, even though it
may be subjective in part, should not be "arbitrary and
capricious." The selection process must be clear to
reviewers in evaluating the adequacy of the process in
meeting the intent of the regulations. Also if the process
is clear, it can be improved or modified in the future as
more information becomes available or policies change.
Be Comprehensive
The federal regulations require a comprehensive ap-
proach such that a broad range of controls are evalu-
ated for various land uses and activities. The selection
process must evaluate a comprehensive list of BMPs to
address pollutants of concern and their sources.
Plan for Implementation
Human nature being what it is, effectively implementing
many BMPs at once is difficult. The solution to this
dilemma is to minimize the number of BMPs chosen,
prioritize or phase their implementation, and/or group
related BMPs into a few categories, sometimes called
program elements.
Involve Affected Parties in the Process
A second element of human nature is adverseness to
implementing someone else's plan. Therefore, BMPs
are selected ideally by those who have to implement
them (with guidance, of course). A second alternative is
that the process heavily involves those who will imple-
ment the BMPs in a review and approval role. If neither
of these approaches are followed, the plan is not likely
to be well implemented.
Indeed, involvement of the affected parties in the selec-
tion process is probably more important to the success
of the program than the exact nature of the process
itself. Through this process, the parties become edu-
cated regarding problems, possible solutions, and the
need for teamwork in implementing solutions.
Model of a Good Selection Process
There is no one correct selection process as the process
must be tailored to local institutional, political, and regu-
latory conditions. Figure 1 is a schematic showing six
steps in a BMP evaluation, selection, and planning proc-
ess that are generally applicable. The following is a
somewhat expanded discussion of BMP selection steps
appropriate for most areawide municipal programs.
Step 1: Establish Program Goals and
Objectives
The clients must agree on a compliance strategy from
which will stem goals and objectives for the program.
The strategy should address such issues as organiza-
tion and administration, decision-making, coordination
with other interest groups, and degree of proactiveness.
Step 2: Identify Receiving Waters, Problems,
Pollutants, and Resources
The ultimate intent of the regulations is to protect and
improve the water quality and ecology of receiving wa-
ters, and this goal should drive the BMP selection proc-
276
-------
1. Compile Comprehensive 2. Prescreen Candidate 3. Evaluate BMPs Using 4. Adjust BMP
List of Candidate BMPs BMPs Selection Factors Categories
Re-insert "duds"
as necessary 1w
compliance
5. Group BMPs Into Program
Elements
Continued
from 4. above
Continued
below
6. Develop BMP Implementation
Programs
Pubtlc In vorvemBfit and Educallon
BMPs
" Eli?
/ Playors
IliE:::
, Budget
E:
i
i
i
aw
/ Scrtedule
_"~
Struct ma! Controls
BMP*
^ ==
ssir
/ Player*
ZZSSXLm
/
s
Bud,
1
!--;
el
3Sj
, Sch-dul.
-
OtMtallonBand Maintenance Controls
BMPi
\ZZ'"£.
/ Player*
si
/ Budgel
=
w°
!°
, Schedule
-
"
Figure 1. BMP evaluation, selection, and planning process.
ess to the extent possible. Ideally this step identifies
water resources of particular value that are especially
critical to protect, as well as impaired water bodies (e.g.,
304(L) segments) that are currently not meeting water
quality objectives appropriate for the beneficial uses.
Where data are available, pollutants to be controlled
should be identified. Without this step, much work and
resources may be focused on activities that do not
necessarily translate into an improved aquatic environ-
ment. Many programs find that a nontechnical one- or
two-page "fact sheet" on receiving water problems, pol-
lutants, sources, and management implications helps to
develop support from taxpayers and decision-makers.
Step 3: Identify Sources and Pathways
Given the problems, the next step is to try to identify the
important point and nonpoint sources of pollutants that
are causing the problems. This is an essential step, because
control of nonpoint sources only makes sense to the
extent that it is a major source of the problem pollutant.
For nonpoint sources, try to describe the pathway from
source to receiving water, because this helps identify the
BMPs that can most effectively intercept the pollutant
along the pathway. For example, dumping waste oil into
catchbasins can be mitigated by labeling storm drain
inlets and/or requiring a monetary deposit at the point of
purchase. It should also be pointed out that some sources
may be quite difficult to control (e.g., natural erosion).
Step 4: Prioritize Sources (Areas) for Control
Targeting sources for BMP application is the next step.
Focusing resources on selected areas is important, oth-
erwise resources tend to be spread too thin to be effec-
tive. This is particularly important in municipal programs,
where some early "successes" encourage the participa-
tion and financial support of local citizens.
277
-------
A systematic targeting scheme using a ranking process
based on stream size, beneficial uses, pollutant loads,
and ease of implementation of the BMP is provided in
U.S. EPA (1) and U.S. EPA (2). Use of these manuals
might be appropriate after an areawide plan is devel-
oped; for example, a BMP might be to begin basin
planning for selected basins within a city. The targeting
manual (1) could be used to identify the basin and
subbasins for BMP selection.
Step 5: Identify and Evaluate Existing BMPs
Compile a list of existing BMPs that are currently being
conducted and organize them according the sources
identified in Step 5. Identifying existing measures is often
very difficult. Some municipalities do not know their
system very well and are organized into departments in
such a way that no one department is aware of what
stormwater measures are currently being implemented.
Carefully crafted questionnaires work quite well at de-
veloping information on existing practices that affect
stormwater quality. Evaluate the effectiveness of these
measures and improve or discontinue as appropriate.
This step also involves identifying existing environmental
programs that are conducting activities that relate to storm-
water pollution control and with whom cooperation should
be sought. Examples include pretreatment programs,
HAZMAT programs, solid waste control and recycling
programs, and public information programs.
Step 6: Compile Candidate BMPs
Compile a comprehensive list of candidate BMPs that
may be appropriate. This list should contain both
source- and treatment-based controls and include such
things as regulatory authority. Attach attributes to each
BMP, including (if available) pollutant type controlled,
cost, and effectiveness. (Recall that such information
is generally not available for source controls.) Note
dependencies or synergistic relationships between
BMPs. For example, some BMPs may be more effective
if or may require that another BMP is implemented
before or at the same time.
Step 7: Develop Selection Criteria
In addition to the obvious criteria that the BMP address
the problems and sources identified in Steps 2 and 5,
developing a list of additional criteria that can be used
to assist in the selection process is helpful. Such criteria
include regulatory requirement compliance, effective-
ness, reliability and sustainability, implementation and
continuing costs, equitability, public and agency accept-
ability, risk and liability, environmental implications, and
synergy with existing or other BMPs.
Step 8: Apply Criteria for Selection of
Baseline Measures
Selection criteria may be applied in numerous ways. For
example, applying different criteria in multiple screening
"passes" is a common procedure. BMPs may be re-
quired to meet "critical criteria" such as obtain co-per-
mittee acceptance, address the problem pollutants and
sources, and meet regulatory requirements. Then, in a
second "pass," those BMPs that met the critical criteria
are further evaluated by applying additional criteria that
would help to select preferred BMPs. Such criteria could
include effectiveness, cost, and reliability. Often the sec-
ond pass allows the municipality to help determine what
is financially feasible. In the second pass, qualitative
(e.g., high, medium, low) or simple quantitative (e.g., 1,
2, 3) scoring might be used to help rank preferred BMPs.
Unequal weighting can be assigned to each criteria as
appropriate.
BMP selection should also anticipate the evolution of the
program. For example, we often recommend that a set
of "baseline" BMPs be selected that fully exploits the
existing control measures and focuses on additional
source control. The selection process can then be used
to select the baseline measures and also candidates for
a reserve list of BMPs that could be implemented at a
later time based on experience with the baseline BMPs.
Step 9: Implement Baseline Measures
Implement the baseline measures with appropriate
phasing to allow for planning, pilot testing, etc., prior to
full scale implementation. For each BMP, develop meas-
ures of effectiveness. As described above, baseline
measures tend to be source controls.
Step 10: Monitor Effectiveness and
Reevaluate BMPs
Monitor the effectiveness of each BMP and, based on
monitoring, annually reevaluate each BMP. As appropri-
ate, delete or select additional BMPs. Annual evaluation
should also include any new information obtained
through monitoring receiving waters and/or source iden-
tification studies.
Case Study 1: Areawide Municipal
Program
The following describes a case study of the BMP selec-
tion process that multiple agencies who were part of an
areawide stormwater program conducted.
County X is 200 square miles in area and contains 20
co-permittees consisting of municipalities, the county,
and a special district. The county population is 1 million
people. The municipalities cover a wide range of sizes
and land uses, from one city of 100,000 population with
major industrial facilities down to small residential cities
278
-------
of 10,000 population. At the behest of the state environ-
mental agency, the co-permittees elected to form a
countywide stormwater pollution control program to
comply with the federal NPDES stormwater regulations.
During the Part I application, the co-permittees compiled
a list of existing BMPs.
The co-permittees were very concerned that their man-
agement plans reflected local conditions and resources
and insisted that they each conduct the BMP selection
process themselves. We refer to this approach as the
"bottom up" approach, in contrast to the "top down"
approach in which BMP selection is conducted by the
program and then distributed to the co-permittees for
their review and approval. Woodward-Clyde Consult-
ants (WCC) acted as facilitators by designing a process
for BMP selection that included development of guid-
ance documents, workshops for all co-permittees, and
meetings with individual cities. Program representatives
and WCC met with the individual jurisdictions three
times throughout the process to provide assistance or
clarification. The process from start to finish took about
9 months.
The following guidance documents were developed:
1. Description of Management Plan Development
Process
2. Selecting the "Right People" To Participate in the
Process
3. Source Identification
4. BMPs for Industrial Facilities
5. BMPs for Agency Activities
6. Transportation BMPs
7. Illicit Discharge Elimination BMPs
8. Commercial Area BMPs
9. Construction and New Development BMPs
10. Public Education and Industrial Outreach BMPs
11. How To Complete Your Stormwater Management Plan
The guidance documents included tables that each
co-permittee was asked to complete based on guidance
provided. The tables formed the basis of each entity's
plan. A key element in the process was a problem and
source identification step (Guidance Document 3), in
which each entity identified receiving water problems,
water resources of special interest, and pollutant sources.
Based on this problem identification, cities selected
BMPs to address source areas in their jurisdictions.
Guidance Documents 4 through 9 described a menu of
individual BMPs from which the cities could select. In
addition, WCC recommended a basic list of BMPs ap-
plicable for most jurisdictions. The co-permittees chose
to participate in a countywide public education program
involving various BMPs described in Guidance Docu-
ment 10. Guidance Document 11 explained how to "put
it all together."
An example of a BMP description is given in Table 1.
The information provided consisted of a BMP name and
identifier, description, steps for implementation, meth-
ods to assess effectiveness, and remarks. For those
BMPs selected, co-permittees were asked to show
when tasks would be completed, and the budget for
each BMP over the 5-year permit period.
The BMP information was intended for guidance only,
and some jurisdictions revised or created new BMPs
that better addressed their circumstances. Some juris-
dictions showed real creativity and enthusiasm in devel-
oping BMPs. This participatory process results in a
much more implementable, practical, and effective
stormwater management plan.
Case Study 2: Industrial Facility
Selection of BMPs for industrial facilities is more site
specific and tends to be guided by the types of activi-
ties being conducted at the facility. The process of
BMP selection then involves identifying industrial ac-
tivities that could potentially generate sources, identi-
fying the types of pollutant releases associated with
Table 1. Best Management Practices for Agency Activities
and Facilities
Number AA-11
Best Management
Practice
Description
Steps for
Implementation
Methods To Assess
Effectiveness
Remarks
Reduce agency use of herbicides and
pesticides.
Reduce the use of herbicides and
pesticides on city streets, landscaping
in parks, flood control channels,
municipal golf courses, etc.
1) Assess current herbicide and
pesticides uses (e.g., types,
amounts, areas used).
2) Research areas where less toxic
substances could be substituted or
usage could be eliminated
altogether (e.g., use of
mosquitofish rather than pesticides).
3) Develop implementation programs
for various areas.
Compare amounts and types of
herbicides and pesticides currently
used with amounts and types used
after implementation of the program(s)
to demonstrate overall reduction
and/or transition to less toxic
substances.
Coordinate with public education and
industrial outreach component
for public education in the area of
residential herbicide and pesticide use.
279
-------
each source, identifying optional BMPs that would pre-
vent oreliminatethatsource, and selecting the preferred
option. The following describes a pared-down process
of BMP selection that we have used on several industrial
projects.
Step 1: Identify Drainage System and
Receiving Water
Define the drainage system and receiving waters, in-
cluding water quality and other concerns in receiving
waters. Ensure plant personnel (particularly nonenviron-
mental personnel) understand the receiving water and
regulatory issues when they are involved in the BMP
selection process.
a sit-down brainstorming session with plant personnel.
Table 2 shows the result of this step for a steam plant.
Indicated in the table are the source activities, drainage
areas within the facility where these sources are lo-
cated, potential pollutants associated with the source,
and a relative measure of the importance of the source
for creating receiving water problems.
Contamination potential:
1 = high
2 = medium
3 = low
Step 2: Identify Industrial Activities and
Associated Pollutant Sources
Discuss what industrial activities are conducted at the
facility and how these activities might lead to discharges
into storm drain systems. This can best be accom-
plished through a combination of a site investigation and
Step 3: Develop Candidate Control Measures
Develop candidate control measures for consideration
that address each of the potential and known sources
of pollutants. The last column in Table 2 shows these
measures.
Table 2. Example of Source and Pollutant Identification and BMP Selection for Industrial Facility
Source Areas
Parking lots
Loading docks
Construction
equipment parking
Materials storage
Curing oil storage
Vehicle fueling
Aboveground fuel
storage
Utility pole storage
Vehicle rinse area
Steam cleaner
Drainage Potential
Areas Pollutant
1 , 2, 4 Oil and grease
TSS
1,2 Oil and grease
Toxics
1,2 Oil and grease
TSS
1 TSS
Metals
Toxics
1 Oil and grease
2 Fuel
Oil and grease
2, 3 Fuel
2 PCP1
Creosol
Metals
Oil and grease
2 TSS
Oil and grease
2 TSS
Oil and grease
Detergents
Toxics
Contamination
Potential
2
2
3
3
2
2
2
2
3
2
3
3
3
1
1
1
1
2
1
3
3
3
3
Recommended Control Measure
Inspect and clean catchbasins
Conduct good housekeeping practices
Provide mats to cover catchbasins if
spill occurs while raining
Inspect and clean catchbasins
Conduct good housekeeping practices
Sweep after loading and unloading
materials from concrete vaults
Place materials with greatest
contamination potential under
Ferry St. overpass
Move drums inside or to a bermed
area that is covered
None
None
Determine feasibility of moving poles
under Ferry St. overpass
Clean sediment trap more often
Consider adding oil/water separator
Enlarge pad area
Post signs providing employees with
proper instructions
Rinse pad after cleaning
Clean oil/water separator more often
280
-------
Table 2. Example of Source and Pollutant Identification and BMP Selection for Industrial Facility (Continued)
Source Areas
Transformer cleaner
Drainage
Areas
2
Potential
Pollutant
Mineral oil
PCBs
Contamination
Potential
3
3
Recommended Control
None
Measure
Sodium hypochlorite
storage
NaOCI
Relocate drums inside or to a bermed
area that is covered
Hogged fuel pipe
Sulfuric acid storage
Oil drum storage
Ash handling area
3 Tannin and lignin
BOD
COD
3 H2SO4
Oil and grease
4 TSS
PH
Toxics
3
2
3
3
3
2
2
2
Sweep street after heavy winds
Clean catchbasin more often
None
None
Enlarge the loading area
Improve the loading procedure
Clean the catchbasins in the
immediate area more often
Step 4: Conduct BMP Evaluation and
Selection
Conduct a BMP evaluation and selection session with
plant personnel. Just as in a municipality, involving the
right plant personnel in the process is very beneficial.
Such involvement allows the plan to reflect their exten-
sive knowledge of the site and industrial activities, and
encourages the plant staff to take ownership of the
management plan. Often, we have found that personnel
have been trying to implement some of the BMPs, and
the NPDES permit requirements now give them the
impetus to get them more fully implemented. In these
sessions, we have sometimes used a formal decision
process, while at other times a less formal, but still
documentable, discussion of the potential BMPs was
used to select BMPs. The focus of BMPs at industrial
sites where we have worked has been source control.
Compared with municipalities, however, industries tend
to be more willing to consider installing or retrofitting
structural controls.
Step 5: Prioritize BMPs and Develop
Monitoring Program
Prioritize BMPs and develop and implement "monitor-
ing" programs for assessment of effectiveness.
References
1. Woodward-Clyde Consultants. 1990. Urban targeting and BMP
selection: An information and guidance manual for state NPS
program staff and managers. Prepared for U.S. Environmental
Protection Agency (May).
2. U.S. EPA. 1993. Handbook: Urban runoff pollution prevention
and control planning. EPA/625/R-93/004.
281
-------
A Catalog of Storm water Quality Best Management Practices for Heavily
Urbanized Watersheds
Warren Bell
City of Alexandria, Alexandria, Virginia
Abstract
Various federal and state environmental programs re-
quire the use of onsite structural best management
practices (BMPs) to control the quality of stormwater
discharges from development sites. Space constraints,
extremely high property values, soil conditions, and the
proximity of other building foundations often preclude
the use of conventional stormwater BMPs for infill con-
struction or redevelopment in the intensely builtup cen-
ters of major cities, where pollutant loads are usually the
greatest. Unconventional solutions must be applied in
these heavily urbanized environments.
Alexandria, Virginia, has adopted and published design
criteria for several nonconventional BMPs, many of
which employ intermittent sand filter technology; some
of these BMPs were developed by pioneering jurisdic-
tions throughout the United States; the city's engineer-
ing staff devised others:
Stormwater sand filter basins in widespread use in
Austin, Texas, are readily adaptable for large devel-
opment projects.
Underground vault sand filters employed in the Dis-
trict of Columbia (DC) allow full economic use of
surface areas.
Double-trench sand filters adopted by the state of
Delaware can be placed either in or adjacent to
paved areas.
Simple trench and modular sand filters developed
by the city of Alexandria are suitable for small or
medium-size sites.
A peat-sand filter adapted from a Metropolitan Wash-
ington Council of Governments design is applicable
to situations where high pollutant removal is required.
Water quality volume detention tanks for use in Alex-
andria's combined sewer areas capture the most
polluted stormwater for later treatment in the waste-
water treatment plant.
The Heavily Urbanized Environment
The U.S. Environmental Protection Agency (EPA) pro-
gram for National Pollutant Discharge Elimination System
(NPDES) permits for stormwater discharges envisions
the use of onsite structural best management practices
(BMPs) to control the quality of runoff from development
sites. Many state programs already impose the require-
ment for onsite BMPs on developers. Under the Virginia
Chesapeake Bay Preservation Act (VCBPA), no net
increase in pollutants in stormwater runoff is allowable
from previously undeveloped sites in Chesapeake Bay
Preservation Areas (CBPAs). Runoff from redevelop-
ment sites in CBPAs must contain 10 percent fewer
pollutants than existed before redevelopment. In devis-
ing a local program to meet these pollutant removal
performance requirements, Alexandria confronted the
dilemma of which structural BMPs to employ. The entire
city is designated as a CBPA. Most of the land is already
developed, and large areas are heavily built up, in many
cases with lot-line to lot-line structures. Property values
are also extremely high. Such conditions exist in the
central business districts of most metropolitan areas.
Use of conventional structural BMPs is often impractical
in the heavily urbanized environment. Space and cost
constraints severely inhibit the use of dry detention
ponds and wet ponds. Soil conditions and high water
tables in the river valleys where most older cities are
located frequently preclude the use of infiltration devices
because of the prevalence of marine clays. Unconven-
tional solutions had to be found to remove the pollutants
from stormwater runoff created by development activity.
Research by the engineering staff of Alexandria's Trans-
portation and Environmental Services Department
revealed that very little information is available on how
to remove pollutants from runoff in heavily urbanized
environments.
282
-------
BMP Design Criteria for Heavily
Urbanized Areas
The Alexandria engineering staff consulted with jurisdic-
tions throughout the United States where BMPs ad-
dressing heavy urbanization are being investigated,
then synthesized the information obtained into compre-
hensive design criteria for local developers. The staff
also developed several additional BMPs for use in the
city. Design criteria forthese BMPs for heavily urbanized
areas were published in the Alexandria Supplement to
the Northern Virginia BMP Handbook in February 1992
(1). The publication is being used by the Virginia
Chesapeake Bay Local Assistance Department as a
guide for other urban stormwater programs within the
commonwealth.
The Concept of BMPs for Heavily
Urbanized Areas
Stormwater quality management in the heavily urban-
ized environment involves the following activities for the
most polluted runoff:
Collection
Pretreatment to remove sediments
Storage
Treatment to remove pollutants of a specific quantity
In Virginia, the minimum quantity of stormwater to be
treated is the first 1/2 in. of runoff from the impervious
areas on the sitethe water quality volume (WQV). The
WQV for each impervious acre is just over 1,800 ft3.
Capturing the WQV
Atypical approach for achieving isolation of the WQV is
to construct an isolation/diversion weir in the stormwater
channel or pipe such that the height of the weir equals
the height of the water in the BMP when the entire WQV
is being held. When additional runoff greater than the
WQV enters the stormwater channel or pipe, it will spill
over the isolation/diversion weir, and the extent of mix-
ing with water stored in the BMP will be minimal. The
overflow runoff then enters a peak flow rate reducer or
exits directly into the stormwater collection system. Fig-
ure 1 illustrates this approach.
Pretreatment Requirements
Several conventional BMPs, such as buried infiltration
devices, and most unconventional BMPs require some
type of pretreatment system to remove excessive sedi-
ments, which would result in premature failure of the
BMP. Pretreatment mechanisms may be installed either
at the point of collection or after separation of the WQV.
These mechanisms may be either separate devices or
an integral part of the BMP itself.
Outflow to Heavily Urbanized
Best Managment Practice
Manhole Access
for Maintenance
Inflow
Runoff
Diversion Weir
Overflow to Quantity
Detention or Storm Sewer
Figure 1. Typical isolation/diversion structure.
Water quality inlets (WQIs), or oil-grit separators
(OGSs), have been employed for several years for the
removal of grit and oil, which are found in large quanti-
ties in parking lots and other areas where vehicular
traffic is significant. Recent studies by the Metropolitan
Washington Council of Governments (MWCOG), how-
ever, have established that WQIs provide little or no
pollutant and questionable hydrocarbon removal (3).
Sedimentation basins have traditionally been the first step
in water or wastewater treatment. Where site conditions
allow, presettling basins may provide a low cost ap-
proach to removal of sediments, which can clog infiltra-
tion devices or filter systems. In situations where space
is not a problem, presettling basins may be built directly
into the ground. In the heavily urbanized environment,
where space utilization is an important economic con-
sideration, underground presettling chambers in vaults
or pipe galleries may provide a more feasible solution.
Alexandria sizes sedimentation basins using a method-
ology based on the Camp-Hazen equation, published by
the State of Washington Department of Ecology (4).
Grassed filter strips are a common method employed in
northern Virginia for removing sediments from stormwa-
ter to be treated in infiltration systems. To be effective,
the strip must be at least 20 ft wide, have a slope of 5
percent or less (5), and be stabilized.
Storage of the WQV
Following isolation of the WQV and pretreatment to
remove sediments and other pollutants, water must be
stored until it can be processed in the primary treatment
device (up to 40 hours in Alexandria). Creating over
1,800 ft3 of water storage per impervious acre on the
site is often the most costly item in the overall BMP
system. In some cases, as with sedimentation basins,
storage may be combined with pretreatment. In others,
283
-------
separate storage galleries of round or arched-section
pipe may be required. Some BMPs for heavily urbanized
areas combine pretreatment, storage, and primary treat-
ment in a single underground vault.
Treatment of the WQV
Most of the BMPs described in this paper employ inter-
mittent sand filters. Originally developed during the
1800s for treating both water supplies and wastewater,
intermittent sand filters have regained popularity for use
in the treatment of small wastewater flows (6).
Austin, Texas, and the state of Florida pioneered the use
of sand filters in the treatment of stormwater runoff.
Alexandria uses the Austin sand filter equation derived
from Darcy's Law by the Austin Environmental and Con-
servation Services Department to size sand filters (2):
Af = laHdf/k(h+df)tf
where
Af = surface area of sand bed (acres or square feet)
la = impervious drainage area contributing runoff to
the basin (acres or square feet)
H = runoff depth to be treated (feet)
df = sand bed depth (feet)
k = coefficient of permeability for sand filter (feet
per hour)
h = average depth (feet) of water above surface of
sand media between full and empty basin
conditions (half maximum depth)
tf = time required for runoff volume to pass through
filter media (hours)
Based on long-term observation of existing sand filter
basins, Austin uses k values of 3.5 ft/day for systems
with full sedimentation pretreatment and 2.0 ft/day for
systems with only partial sedimentation pretreatment.
Alexandria has also adopted these values. Both Austin
and Alexandria use a BMP drawdown time (tf) of 40
hours. With these constants, the equation for sand filter
systems with full sedimentation protection reduces to
= 310ladf/(h+df),
where Af is in cubic feet and la is in acres.
For sand filter systems with partial sedimentation pro-
tection, the equation reduces to
Af(PS) = 545ladf/(h+df),
where Af is in cubic feet and la is in acres.
Descriptions of BMPs for Heavily
Urbanized Areas
The BMPs discussed below should not be thought of
merely as drainage structures. They are low technology
treatment works that use water and sewage treatment
technology from the late 19th century. Treatment works
cannot always be made to function by gravity flow,
although it is usually desirable from a cost-effectiveness
standpoint.
Surface Sand Filter Basin Systems
Austin, Texas, was a pioneer in the use of intermittent sand
filtration systems for treating stormwater runoff. The Austin
program is managed by the Environmental and Conser-
vation Services Department, which has published de-
sign criteria in their Environmental Criteria Manual (2).
Typical intermittent sand filters employ an 18- to 24-in.
layer of sand as the filter media underlain by a collector
pipe system in a bed of gravel. A layer of geotechnical
cloth separates the sand and gravel to keep the sand
from washing into voids in the gravel. Austin pretreats
the stormwater runoff in a sediment trapping structure to
protect the filter media from excessive sediment loading.
Figure 2 is a centerline cutaway of one Austin sand filter
configuration. In this system, the sedimentation struc-
ture is a basin designed to hold the entire WQV, then
release it to the filtration basin over an extended draw-
down period. An alternate design allows use of a smaller
sedimentation chamber but requires increasing the filter
size to compensate for increased clogging of the filter
media. While the system shown uses concrete basins,
a sediment pond and a geomembrane-lined filter built
directly into the ground may be used where terrain and
soil conditions allow. The Austin sand filter systems are
most appropriate for large developments covering sev-
eral acres.
Austin has monitored the performance of their sand
filters for several years and currently recognizes up to
60 percent phosphorus removal efficiency based on
these studies (7). Alexandria is currently recognizing a
40 percent phosphorus removal rate pending further
sand filter monitoring results by Austin and the District
of Columbia. (Phosphorus is the "keystone pollutant"
used to measure compliance with the VCBPA.)
Underground Vault Sand Filter Systems
Truong developed a stormwater quality sand filtration
system in an underground vault (8). Over 70 of the
structures have been installed since 1987. Figure 3 is a
centerline cutaway of the original concrete vault DC
sand filter. DC sand filters may be placed underneath
parking lots, alleys, or driveways, taking up no usable
space on the surface. This is an important advantage in
the heavily urbanized environment. Truong believes that
284
-------
Designed for Load and Soil
Conditions ^>^ J-
Perforated Riser
With Trash
Rack
Hatch to Access Ramp
for Cleaning
Sedimentatio
Basin
Perforated
Collector Pipes
in Gravel Bed
Under Sand
First 1/2 In. of Runoff
(WQV) From Flow Separator
Sediment Trap
With Underdrain
18-24 In. Sand
Filter Underlain
With Geotechnical
Filter Cloth
Figure 2. Austin basin sand filter system.
this system works best on watersheds with 1 acre or
less of impervious cover.
The DC sand filter is a three-chamber gravity-flow sys-
tem. The first chamber and the throat of the second
chamber contain a permanent pool that traps grit and
floating organic material, such as oil, grease, and tree
leaves. A submerged rectangular opening at the bottom
of the first dividing wall connects the two parts of the
pool. The second chamber also contains a 24-in. deep
sand filter underlain by a layer of geotechnical fabric and
collector pipes in gravel. A top layer of plastic-reinforced
geotechnical filter cloth held in place by a 1-in. layer of
gravel is provided above the sand to compensate for the
smallness of the sedimentation chamber.
New runoff entering the structure causes the pool to rise
and overflow onto the filter. After percolating through the
sand, the treated water enters the underdrains and flows
out into the third chamber, or clean/veil. The clearwell
conveys the treated water to the storm sewer or drain-
age system. If possible, this BMP should be configured
to allow gravity outflow; however, in instances where
filters must be placed below the storm drainage system
elevation, such as under the entrance driveway to a
parking garage, a sump pump must be used.
The trash and hydrocarbon water trap in the first cham-
ber must be pumped out and refilled with clean water
every 6 months for proper functioning. Every 3 to 5
years, the top filter cloth layer and gravel must be re-
moved and replaced because of fine sediment clogging.
Placement of the second chamber manhole directly
above the center of the filter allows the corners of the
cloth to be peeled up and bound together to form a bag
that can be lifted out as a unit.
The District of Columbia Environmental Regulation Ad-
ministration is conducting a program of monitoring to
establish the actual removal rates of this system. As of
this writing, no data are available.
The Austin partial sedimentation sand filter may also be
placed in underground vaults. Figure 4 shows a modi-
fied vault design developed by Alexandria from both
Austin and District of Columbia methodologies. The
Austin approach uses a gabion wall to separate the
partial sedimentation chamber from the filter area. The
gabion absorbs energy and provides initial filtration.
Heavy sediments are deposited in this first chamber to
dry out between storms. The filter is exactly like that
used in the DC sand filter system.
Double Trench Sancf Filter Systems
Shaver developed a surface sand filter system for use
in Delaware (9). The Delaware sand filter is intended to
be an in-line facility processing all stormwater exiting the
site until it overflows.
285
-------
Access Manhole
Structural Concrete Vault
Designed for Load and Soil
Conditions.
6-ln. PVC Dewatering
Drain With Gate Valve
Outflow to
Storm Sewer
Clean/veil Chamber
6-ln. Perforated PVC Collector
in 8-ln. Gravel Bed (3 Required)
2-Ft Sand Filter Between Geotechnical
Filter Cloth Layers
Inspection Well/Cleanout Pipe With
Waterproof Cap (3 Required)
First 1/2 In. l Sediment Chamber
of Runoff (WQV) with Water Seal to
From Flow Separator Trap Hydrocarbons
Figure 3. DC underground vault sand filter.
Outflow to
Storm Sewer
-ln. Gravel Ballast
Over Geotechnical
Filter Fabric
6-ln. Perforated Collector
Pipes in Gravel Bed
Beneath Geotechnical
Filter Fabric (3)
Chamber Wall
and Energy
Inflow From
Flow Splitter
Figure 4. Dry vault stormwater sand filter.
Structural:
Designed for Soil
and Load Conditions
286
-------
Figure 5 is a schematic drawing of the Delaware sand
filter system. The concept uses two parallel waterproof
trenches connected by close-spaced wide notches in
the top of the wall between them. The trench adjacent
to the site being served is the sedimentation chamber.
Polluted stormwater must be conveyed to the chamber
in enclosed storm-drain pipes. The permanent pool in
the sedimentation chamber inhibits resuspension of par-
ticles that were deposited in earlier storms and prevents
the heavier sediments from being washed into the filter
chamber. As new stormwater enters the system, the per-
manent pool overflows through the weir notches and onto
the filter as sheet flow to prevent scouring out the sand.
The second chamber contains an 18-in. sand filter that
is always fitted with a solid cover. No underdrain piping
is provided. Water percolates through the sand and
escapes from the filter through a geotechnical cloth-cov-
ered grate at the downhill end of the filter chamber.
Four Delaware sand filters were constructed in Alexan-
dria during 1992. The first two systems served small
parking lots and were built according to the original
Delaware design. The third application, involving two
separate filters, was used to treat runoff from a large (1.7
acre) parking lot. The high cost of steel grates and covers
led the developer's consultant to propose moving the
filter off the lot and providing slotted curb ingress and
precast concrete lids. Premature failure of one of the
filters led the owner to install a collector pipe in gravel
below the sand layer. This design is shown in Figure 6.
Although the filters illustrated are contained in reinforced
concrete shells, these systems may be installed in any
waterproof container that will bear the wheel loads or
soil pressures involved with the particular application;
molded fiberglass or other plastic materials would work
well. Delaware sand filters made of timber lined with
rubberized roofing material have been proposed for use
on temporary parking lots for development sales offices.
Delaware does not rate these systems for nutrient re-
moval efficiency. Delaware has made a determination,
however, that when treating the first 1 in. of runoff, this
sand filter provides 80-percent suspended solids removal,
as required by state environmental regulations (9).
Stone Reservoir Trench Sand Filter Systems
The filter system concepts embodied in the Austin and
District of Columbia designs may be readily adapted for
small and less complex applications. Alexandria's engi-
neering staff has developed a simple trench sand filter
for use on such projects as townhouses or small com-
mercial developments in areas where infiltration devices
are not practicable.
Figure 7 is a schematic drawing of a stone reservoir
trench sand filter. The system is constructed in an exca-
vation lined with impervious geomembrane (such as
EPDM roofing material) sandwiched between protec-
tive layers of filter cloth. The bottom of the trench con-
tains a simple sand filter that is connected to the storm
sewer. The upper part of the system is built the same as
an infiltration trench designed to treat the first 1/2 in. of
runoff. Placement of perforated pipes in the stone res-
ervoir greatly increases the voids available for storage.
Dispersed overland sheet flow is treated in a grassed
filter strip before entering the system. The reservoir is
further protected from sediment clogging by a layer of
geotechnical filter cloth 6 in. beneath the top surface of
Steel Plate
Cover
Steel Grate
Filtration Chamber
18 In. of Sand
Outfall Pipe
Grate (Fabric Wrapped
Over Entire Grate Opening)
Parking Lot
Pavement
Sediment Chamber
(Heavy Sediments, Organics,
Debris)
Figure 5. Delaware sand filter with grated inlets.
287
-------
Cleanout and Observation Well
Under Hatch With Waterproof Cap
Storage for Treatment of
Water Quality Volume
4-ln. Gravel Beneath-
Geotechnical Fabric
Layer
Figure 6. Slotted curb Delaware sand filter.
Chamber Separation Weir
s^sm&M^iiS^- Permanent Sediment Pool
4-ln. PVC Pipe Semisubmerged in Concrete
Floor Perforated in Top Half
Parking Lot
Pavement
Observation Cap
Reservoir
Stone
Perforated
PVC Pipe
Flow to Storm
Sewer
Liner Sandwich Edges Tucked To Prevent Any Bypass
Replaceable Filter Cloth Sediment Barrier
Permanent Side Wall
30-Mil. Geomembrane Sandwiched
Between Layers on Filter Cloth
Filter Cloth Between Layers
Sand Filter (Min. 18 In. Above Gravel)
Perforated Gravel Layer
Collector Pipe
Figure 7. Stone reservoir trench sand filter.
the aggregate. The WQV flows into the reservoir until
the voids in the rock and perforated pipes are com-
pletely full. Any overflow is directed to the storm sewer.
Runoff collected in the reservoir filters down through the
sand to the collector pipe, from which it is conveyed to
the storm sewer.
Trench sand filter systems should have the same re-
moval efficiency as an Austin sand filter.
Peat-Sand Filter Systems
Because of their high pollutant removal capabilities,
simple design, low-maintenance, and affordability, peat-
sand filters (PSFs) are potentially effective in heavily
urbanized areas. A stormwater "end-of-pipe" PSF sys-
tem was scheduled to be constructed in Montgomery
County, Maryland, in the summer of 1993. MWCOG staff
participated heavily in the development of this project.
288
-------
Figure 8 is a centerline cutaway of a stormwater PSF
system concept developed by the Alexandria engineer-
ing staff. It combines features of the Austin sand filtration
system with the PSF design proposed by John Galli of
MWCOG for use in the Montgomery County application
(10). The Alexandria concept is intended to operate as
an off-line system treating the WQV from each storm.
Any additional detention required for stormwater quan-
tity restrictions should be provided separately down-
stream of the PSF system. PSFs would be appropriate
for commercial developments for which a high pollutant
removal is required or for end-of-pipe treatment of entire
storm sewer watersheds.
The sedimentation basin design is essentially the same
as that of the Austin sand filter. Because PSF systems
cannot normally operate during the more severe winter
months of the mid-Atlantic region, however, a gate-valve
equipped bypass is provided to divert flow from the
basin directly to the storm sewer. The invert of this pipe
is placed at an elevation that will detain a permanent
pool in the basin averaging at least 4 ft deep. In effect,
this configuration converts the sedimentation basin to a
small extended detention/wet pond during the winter
months. As with the Austin sand filter, the basins may
be either walled with concrete, as shown, or, if soil
conditions permit, be constructed as soil structures.
The filtration basin is basically the Austin design with the
sand filter enhanced by adding a 12- to 18-in. thick
surface layer of hemic or fibric peat, a layer of calcitic
limestone (for greater phosphorus removal), and a 4-in.,
50:50 well-mixed layer of peat and fine-medium grain
sand atop the normal filter sand and collector under-
drains. A nutrient-removing grass-cover crop must be
planted and maintained in the top of the peat layer.
(PSFs will not function in underground applications be-
cause anaerobic conditions would develop.)
The system shown is designed for gravity flow. In situ-
ations where the terrain does not provide sufficient
relief, pumps must be added to move the stormwater
between basins.
Based on information provided by MWCOG (10), the
Alexandria engineering staff estimates that their PSF
design should have a phosphorus-removal efficiency
approaching 90 percent during the months in which the
filter is in operation. Assuming that the filter would be
bypassed from mid-December to mid-March in the mid-
Atlantic region, the annual phosphorus removal efficiency
of the overall system, including the small extended de-
tention/wet pond, is estimated at 70 percent.
Water Quality Volume Storage Tanks
This concept involves the collection and storage for later
treatment in the wastewater treatment plant of the WQV
from each storm. WQV storage tanks are used on all
developments or redevelopments that require a BMP
within Alexandria's combined sewer watersheds. Figure
9 shows a centerline cutaway of a WQV storage tank.
The stored water is released into the combined or sani-
tary sewer system by telemetry-controlled pumps or
automatic valves that ensure that none of the WQV
escapes while combined sewer overflows into streams
,, <* * *' *,. y //A* $< /^Ai 'i&ii*** '*+
\ /-,vA+*,v^t*i?i*^/A f? "V *«.* y//^/7*'ir*^
^ Gate Valve Bypass to Storm Sewerv ^ ^,V /^ ^ * ^^ ^ ^^^ / * ^ Jt ^ ^^^/^V^^^
. . ' , .''''^K^J'*^,^'^,, RvnaRft'f^inp RptainR^ I^^S^ ^C^X.1^ i. ,.v iX ., Pira?;?; Onvpr Ornn i^y
r* * * r- ' 'A*
Grass Cover Crop
Energy Dissipators
Runoff (WQV) From
Flow Separator
Sediment Trap
With Underdrain
12-ln. Min. Peat (Hemic
pi or Fibric)
I/ 4-ln. 50/50 Peat/Sand
* - Mix
18-ln. Min. Sand
Collector Pipes
Filtered Outlet
6-ln. Washed Bank-
Run Gravel
Perforated Collector Pipes
in Gravel Bed
Geotextile Fabric
Figure 8. Stormwater peat-sand filter system.
289
-------
To Power Feed To Telemetry to
for Pump City Control Point
Access A A
Manhole
Access Manholes
Pump Outflow
to Combined (or
Sanitary) Sewer
Remotely Controlled
Pump
First 1/2 In. of Runoff
From Isolation/Diversion
Chamber
. Oil Trap Chamber With Inverted
Elbow Outflow
Sediment Chamber With Trash Rack
Figure 9. Water quality volume storage tank.
are occurring or in periods when inflow and infiltration
are taxing the capacity of the wastewater treatment
plant. This approach conforms to EPA's August 19,
1989, National Combined Sewer Overflow Strategy,
which requires establishment of a high-flow manage-
ment plan that maximizes the capacity of the combined
sewage system for storage and treatment.
The tank shown in Figure 9 has a water quality inlet to
provide sediment and petroleum hydrocarbon removal
before the runoff is allowed to enter the storage tank.
The inlet must be pumped out and refilled with clean
water every 6 months for proper functioning.
WQV storage reservoirs may be either prefabricated
tanks or vaults fabricated on site from such materials as
Portland cement. Either single or multiple tanks may be
employed. Although originally developed for use in com-
bined sewer watersheds, WQV storage tanks may be
applied in other situations where WQV runoff will not be
routed into the storm sewer (e.g., landscaping irrigation
systems or "gray water" toilet flushing systems).
When WQV water is discharged directly into a combined
or sanitary sewer or used in gray-water flushing sys-
tems, the pollutant removal efficiency of the system
becomes that of the receiving wastewater treatment
plant. The phosphorus removal capacity of such plants
is typically in the 95- to 100-percent range. When the
WQV water is reused and retained on site for landscape
irrigation, pollutant removal may approach 100 percent
if the water is not allowed to escape from the site.
Challenges in Development and Use of
BMPs for Heavily Urbanized Areas
The field of BMPs for heavily urbanized areas is in its
infancy. The next few years must bring much wider use
of this technology if the pollutant removal objectives of
the NPDES stormwater program and other federal and
state clean water initiatives are to be met. Several sig-
nificant challenges need to be addressed.
Reduce Construction Effort and Costs
The construction cost for Austin sand filters serving
projects with approximately 1 acre of impervious cover
ranged from $13,000 to $19,000 in 1990 (1). The cost
of DC sand filters was approximately $35,000 per im-
pervious acre when the filters were first introduced but
has since fallen to approximately $12,000 to $16,000
through the introduction of precasting and the maturity
of the design (11). The large, slotted-curb Delaware
sand filters recently constructed in Alexandria cost ap-
proximately $40,000 to serve 1.7 acres of impervious
cover. This was, in essence, a prototype facility, and
costs are expected to fall in a manner similar to the DC
sand filter costs as contractors and engineers become
familiar with the technology.
Applying prefabrication and modular concepts, espe-
cially for smaller projects, should further reduce con-
struction effort and costs. Alexandria and the District of
Columbia are exploring the rationalization of sand filter
vaults in circular sections with manufacturers of alumin-
ized corrugated pipe and fiberglass underground tanks.
The pipe manufacturer has indicated that filters that
would serve up to 1 acre of impervious cover could be
prefabricated in a shop and delivered as a unit to a job
290
-------
site. The District of Columbia has also developed a sand
filter in a standard precast sewer manhole. By introduc-
ing the runoff through a large catch basin with a hooded
outlet, the addition of a 6-ft manhole with a sand filter in
the bottom makes a BMP suitable for treating the
runoff from approximately 5,200 ft2 of impervious cover;
8-ft manhole filters can serve approximately 10,000 ft2.
Alexandria is examining the feasibility of adapting stand-
ard large highway precast curb inlets as the shells of
both Delaware sand filters and underground vault sand
filters. Storage of runoff awaiting filtration in arched
corrugated-pipe galleries appears to be a promising
approach in areas where storm sewers are too shallow
to employ vault filters without pumping. Much more
innovation is still needed for heavily urbanized areas.
One of the major costs of BMPs for heavily urbanized
areas is creating a container to store the runoff before it
undergoes treatment. More studies need to be per-
formed characterizing different types of runoff to deter-
mine whether all sites need similar treatment. For
instance, pollutants in runoff strictly from roofs may be
concentrated in a smaller amount of "first flush." Pollu-
tion concentration versus time studies of roof water
might well establish that treatment of a smaller amount
of runoff would meet pollutant removal performance
requirements. This development would likely have a
significant impact on costs.
Reduce Maintenance Requirements and Costs
All BMPs for heavily urbanized areas require significant
maintenance. Permanent pools require pumping out on
a periodic basis (currently twice per year in Alexandria)
to remove accumulated sediments and trapped hydro-
carbons. As discussed above, sand filters require the
replacement of the top few inches of sand or overlying
layers of geotechnical cloth every 3 to 5 years. Trash
must be removed from all BMPs as it accumulates to
prevent premature clogging. Special care must be taken
to ensure that sand filter systems are not placed in
service before all open areas are stabilized with vege-
tation. Otherwise, the filters might quickly clog with top-
soil, as occurred with one of Alexandria's Delaware sand
filters. Trash screens need to be included in all designs
to preclude the intrusion of materials into filter chambers
that can cause premature failures. The provision of
ready maintenance access is an absolute necessity.
The initial cost/maintenance cost tradeoff must be care-
fully examined during the BMP design process.
Enhance Removal of Pollutants
The 1990 Austin report on removal of pollutants by that
city's sand filters is the only scientific data available at
present on long-term monitoring of such systems (7).
The reported results are encouraging, but more moni-
toring data is needed to assess the impact of such
factors as acid rain and variations in chemical content
of the filter media on performance before the Austin
experience can be generalized for application to other
regions of the country.
While Austin reports very promising phosphorus re-
moval values, enhancing the nitrogen and perhaps the
heavy-metal removal efficiencies of BMPs may develop
as a more pressing need as NPDES runoff monitoring
data become available. One avenue that appears prom-
ising is the employment of a wet gravel filter component
to introduce biological activity in the treatment process,
an approach that is already being used to treat individual
home sewage in Anne Arundel County, Maryland (12).
The District of Columbia is considering adding a layer of
activated carbon to a sand filter to assess the benefits
through monitoring. BMPs for heavily urbanized areas
represent a field that is ripe for additional innovation.
Universities should take a more active role in developing
BMP technologies for these areas.
Spread the Technology
Currently, the use of BMPs for heavily urbanized areas
is limited to a relatively small area in the mid-Atlantic
states, the Austin area in Texas, and the state of Florida.
The technology is applicable to all areas of the country
where pollution in stormwater runoff must be controlled
under the NPDES permit program. Information on these
BMPs needs to be disseminated throughout the country
by EPA and other environmental agencies so that the
technology is available to all parties who are wrestling
with the problems of attaining NPDES compliance. This
paper was written to facilitate that process.
References
1. City of Alexandria. 1992. Unconventional BMP design criteria. In:
Alexandria supplement to the Northern Virginia BMP handbook.
Alexandria, VA: Department of Transportation and Environmental
Services.
2. City of Austin. 1988. Water quality management. In: Environ-
mental criteria manual. Austin, TX: Environmental and Conser-
vation Services Department.
3. Galli, J. 1992. Analysis of urban BMP performance and longevity
in Prince George's County, Maryland. Washington, DC: Metro-
politan Washington Council of Governments.
4. Washington State Department of Ecology. 1992. Stormwater
management for the Puget Sound Basin (the technical manual),
Vol. III. Runoff controls. Olympia, WA. pp. 4-42.
5. Northern Virginia Planning District Commission. 1992. BMP plan-
ning considerations. In: Northern Virginia BMP handbook. Annan-
dale, VA: Northern Virginia Planning District Commission.
6. Anderson, D.L., R.L. Siegrist, and R.J. Otis. [No date]. Technology
assessment of intermittent sand filters. Municipal Environmental
Research Laboratory, Office of Research and Development, U.S.
Environmental Protection Agency, Cincinnati, OH.
7. City of Austin. 1990. Removal efficiencies of stormwater control
structures. Austin, TX: Environmental and Conservation Serv-
ices Department.
291
-------
8. Truong, H.V. 1989. The sand filter water quality structure. Wash- 11. Truong, H.V. 1993. Application of Washington, DC, sand filter for
ington, DC: District of Columbia Environmental Regulation Ad- urban runoff control. Washington, DC: District of Columbia Envi-
ministration. ronmental Regulation Administration.
9. Shaver, E. 1991. Sand filter design for water quality treatment. 12- piluk> RJ-> and OJ- Hao- 1989- Evaluation of onsite waste dis-
Dover, DE: State of Delaware Department of Natural Resources Posal system for nitrogen reduction. J. Environ. Eng. 115(4).
and Environmental Control.
10. Galli, J. 1990. Peat-sand filters: A proposed stormwater manage-
ment practice for urbanized areas. Washington, DC: Metropoli-
tan Washington Council of Governments.
292
-------
Postconstruction Responsibilities for Effective
Performance of Best Management Practices
Joseph J. Skupien
Somerset County Engineering Department, Somerville, New Jersey
Abstract
Effective performance of best management practices
(BMPs) is vital to achieving the high goals and justifying
the equally high estimated costs of urban runoff man-
agement. This paper identifies inspection, maintenance,
and performance monitoring as three key postconstruc-
tion activities for ensuring correct and continued per-
formance of BMPs. These activities are equal in
importance to planning, design, and construction BMPs.
The paper demonstrates how failure to meet inspection
and maintenance BMP responsibilities not only leads to
diminished BMP performance but may also create new
health and safety threats that exceed those the BMPs
were intended to prevent. It further demonstrates how
such a result represents both a failure to realize a gain
on the resources already invested in BMPs and the
cause of significant additional expenditures.
The paper also describes the key components of a
successful postconstruction inspection and mainte-
nance program, including the need for self-evaluation
and feedback components to inform planners, design-
ers, construction contractors, and maintenance person-
nel about ways to reduce or facilitate future
maintenance. Additionally, the paper emphasizes the
importance of a stable source of program funding and
discusses various methods for achieving it.
Finally, the paper emphasizes the need for accurate,
scientific monitoring and reporting of BMP performance
to achieve optimal BMP designs and expand the ability
to address urban runoff impacts on a regional or water-
shed basis.
Introduction
One of the top priorities of any stormwater manage-
ment program is the effective performance of structural
best management practices (BMPs). Effective BMP
performance not only helps ensure that the program's
goals are accomplished but also represents a positive
return on the time, effort, and materials invested in the
structural BMP's planning, design, and construction. To
achieve such performance, however, everyone involved
with the stormwater management program must fulfill
several key responsibilities before, during, and after
construction.
Before construction, these responsibilities include the
development by program managers of design standards
and practices that are both accurate and practical. De-
signers must use these standards and practices to pro-
duce construction drawings that accurately convert their
ideas into a tangible structure. Using these drawings,
construction contractors must create a durable structure
that meets the designers' requirements and is true to the
regulators' intentions.
While stormwater management remains a relatively
new field, the results to date of these relatively short-
term preconstruction activities have been greatly im-
proved by several factors, including the maturation of
older flood control programs; the continued growth of
hydrologic and hydraulic databases, design methods,
and training programs; and the implementation of for-
mal construction inspection programs. Other factors
that have assisted in the improvement of regulatory,
design, and construction activities include the contin-
ued development and greater availability of computer
software and hardware and the greater level of con-
struction experience and capability. As a result, the
ability of program managers, designers, and construc-
tion contractors to meet their responsibilities for effec-
tive BMP performance has increased significantly in
recent years. Furthermore, these improvements have
helped to kindle further interest and involvement in
stormwater management.
In addition to planning, design, and construction respon-
sibilities, however, three key areas of responsibility must
be met once construction has been completed and the
293
-------
structural BMP has been put into operation. These re-
sponsibilities consist of the inspection, maintenance,
and monitoring of the structural BMP. For the purposes
of this paper, these three activities are defined briefly as
follows:
Inspection: Periodic observation and evaluation of a
structural BMP and its individual components by
qualified personnel to determine maintenance needs.
Maintenance: Periodic preventative and corrective
measures taken by qualified personnel to ensure
safe, effective, and reliable BMP performance.
Monitoring: Extended observation and evaluation of
BMP performance by qualified personnel to deter-
mine effectiveness and improvement needs.
Of the three activities, inspection and maintenance are
the most well established in terms of BMPs, while moni-
toring represents a somewhat more recent aspect of
stormwater management. More complete descriptions
of each activity and their growing importance is pre-
sented in later sections of this paper. For now, it is
important to note that each activity represents a long-
term, ongoing responsibility carried out after the shorter
term planning, design, and construction efforts have
been completed. It is also important to note that BMPs
for inspection, maintenance, and monitoring have not
received the same level of attention typically devoted to
planning, design, and construction. While lack of ade-
quate funding may be a cause, the reasons for this
imbalance are generally unclear. This is unfortunate,
because such an imbalance may critically affect the
long-term success of stormwater management pro-
grams and regulations. Possible reasons include the
ongoing, long-term, and somewhat routine nature of
inspection and maintenance in particular, which may not
offer either the intellectual and creative challenge of
planning and design or the immediacy of construction.
Additional reasons may be an unacknowledged reluc-
tance to confront the reality of current planning, design,
and regulatory efforts (particularly the negative aspects
of that reality), or the failure to fully appreciate the
importance of BMPs in regard to inspection, mainte-
nance, and monitoring and the serious consequences
of their prolonged neglect.
Regardless of the reasons, it is apparent that BMPs for
inspection, maintenance, and monitoring have suffered
the neglect typical of long-term, ongoing activities. As
noted above, this neglect has critical implications for the
long-term success of efforts to manage stormwater, par-
ticularly through the use of structural BMPs. In an effort
to correct this problem, this paper presents information
emphasizing the importance of and need for BMPs in
inspection and maintenance and describes the key com-
ponents of a comprehensive inspection and mainte-
nance program. Additionally, the paper highlights the
increasing need for monitoring as a means to improve
BMP performance and effectiveness and to reduce re-
quired inspection and maintenance efforts.
The Importance of BMP Inspection and
Maintenance
A common requirement of virtually all stormwater struc-
tures, particularly those that encounter various weather
conditions, is their need for periodic inspection and
maintenance. While these needs may be obvious in a
general sense, the particular importance of inspection
and maintenance for structural BMPs needs to be
stressed.
Perhaps the most recognizable reason is the need to
reliably and consistently achieve the performance levels
required by the stormwater management program and
designed into the BMP. For example, a BMP that relies
on the temporary storage of stormwater runoff to
achieve required peak outflow or pollutant removal rates
must be periodically cleaned of accumulated sediment
and debris to maintain required storage capacity and
prevent re-suspension of captured pollutants. The outlet
structures at these facilities must also be periodically
cleared of accumulated debris to maintain discharge
rates at required levels. Maintenance of vegetation is
also important, particularly for those BMPs that use the
vegetation for pollutant filtration and/or uptake. This
maintenance can range from mowing, seeding, and fer-
tilizing turf grass areas to ensure stability and prevent
erosion to harvesting wetland vegetation to promote and
manage growth.
The maintenance described can also be viewed as an
effective means of ensuring a positive return on the time,
effort, and materials invested in the planning, design,
and construction of a BMP. The total amount of this
investment for a single BMP can be considerable, with
total construction costs exceeding $50,000 and total
project costs exceeding $100,000. Failure to adequately
inspect and/or maintain such a facility can lead to inef-
fective performance, structural failure, and, conse-
quently, a failure to realize a return on the investment.
It is generally recognized that the cost of providing
comprehensive water quality protection may be consid-
erably greater than our present ability to pay for it. In
such cases, we must strive to achieve the greatest
possible return on the resources we do invest in such
protection.
Perhaps the most important need for BMP inspection
and maintenance is the need to avoid the health and
safety threats inherent in their neglect. The foremost of
these threats is the potential for structural failure, which
can rapidly release stored waters and flood downstream
areas, causing property damage, injury, and even death.
The fact that this flooding threat would not exist if the
BMP had not been constructed further highlights the
294
-------
need for proper inspection and maintenance to prevent
it from ever occurring. Another health and safety threat
from maintenance neglect is mosquito breeding, which
can threaten a broad area in the general vicinity of the
BMP. Other undesirable insects, animals, and odors can
also result from maintenance neglect, adversely affect-
ing those who must live or work nearby. In all such
cases, the BMP can actually have worse environmental
impacts than those it was originally constructed to prevent.
A final reason for effective BMP inspection and mainte-
nance lies in preserving and nurturing the community
and political support that stormwater management ef-
forts have gained to date. Such continued support is
vital to the success of our stormwater management
efforts, particularly because much of the solution to
stormwater pollution lies in source controls and lifestyle
changes that the public will be asked to adopt. We
cannot count on even passive public support, however,
let alone active public involvement in nonstructural pro-
grams, if we are unable to create and maintain structural
BMPs that are community assets rather than liabilities.
Any support that we now have or hope to generate in
the future will quickly be lost if we allow structural BMPs
to become aesthetic nuisances or safety hazards due to
a lack of adequate inspection and maintenance.
Comprehensive Inspection and
Maintenance: An Overview
The key components of a comprehensive inspection
and maintenance program for structural BMPs are de-
scribed below. The exact character of each component
and the manner in which it is implemented depends on
the specific economic, political, environmental, and so-
cial characteristics of the community in which the pro-
gram functions.
Official Inclusion of Inspection and
Maintenance in Overall Stormwater
Management Program
BMP inspection and maintenance should not be an
afterthought but should be included from the beginning
in the community's overall stormwater management pro-
gram. As the overall program develops, determining
how (and how often) inspections and maintenance ef-
forts are performed is as important as determining al-
lowable peak outflow rates and extended detention
times. To ignore this fact is to invite eventual program
failure through diminishing BMP performance and in-
creasing health and safety threats. To ensure a secure
role for inspection and maintenance in the overall storm-
water management program, both the importance of
inspection and maintenance and the ways in which they
are achieved should be officially included in any imple-
menting ordinances, resolutions, or laws establishing
the overall program.
Sufficient and Stable Funding
Because BMP inspection and maintenance requires
specific actions by qualified personnel, the availability of
sufficient and stable funding may be the single most
important component of a comprehensive program. The
best intentions, talent, and equipment cannot overcome
a paucity of funds, nor can regular, consistent inspec-
tions and maintenance be achieved if funding levels are
erratic and/or uncertain.
Therefore, during the development of the overall storm-
water management program, a stable source of funding
for inspection and maintenance must be identified and
formalized. This may include the use of general or spe-
cialized tax revenues, dedicated contributions from land
developers or owners, and/or permit fees from those
creating the need for the structural BMP. Funding may
also be secured through the creation of a stormwater
utility, which would provide BMP inspection and mainte-
nance services funded by fees paid by those within the
utility's service area. While the creation of a stormwater
utility requires a significant amount of effort to organize and
operate, several successful stormwater utilities have been
created throughout the country in recent years.
Adequate Equipment and Materials
Having sufficient equipment and materials is particularly
important for BMP maintenance efforts, which involve
the regular performance of preventative maintenance
activities such as grass mowing and debris removal and
the prompt execution of emergency repairs and restora-
tions. The long-term, repetitive nature of the preventa-
tive activities, in particular, demonstrates how a positive
return can be quickly achieved from investments in
equipment that expedite maintenance efforts and in ma-
terials that prolong the life of BMP components.
Fortunately, due in part to the basic nature of stormwater
and its management, the character of the equipment
necessary to conduct most maintenance efforts is not
particularly complex or specialized. Instead, standard
and relatively simple equipment such as lawn mowers,
shovels, rakes, compressors, and trimmers can be used
to perform the majority of maintenance tasks. This helps
simplify the selection and acquisition process and keeps
costs at more manageable levels.
Trained and Motivated Staff
Similar to equipment needs, many BMP maintenance
tasks are not particularly complex or specialized. This
means that, under most circumstances, program staff
can be assembled from a relatively large labor pool,
either directly by a public agency performing mainte-
nance in house or by a contractor hired to provide such
services. These factors, however, should not diminish
the need for thorough training of maintenance staff. This
295
-------
has become increasingly true in recent years as the role
of structural BMPs expands to provide higher levels of
stormwater treatment and more comprehensive control
of runoff rates. This has led to increasingly sophisticated
facilities containing specialized vegetation and diverse
habitats that require management as well as mainte-
nance. This trend is expected to continue, further em-
phasizing the need for thoroughly trained staff.
The importance of motivation and enthusiasm must also
be emphasized. Unfortunately, the repetitive and rela-
tively simple nature of many BMP maintenance tasks
can lead to indifferent staff performance. In addition to
poor overall results, this indifferent attitude can also be
dangerous, particularly for those staff members operat-
ing mowing or cutting equipment that, however simple,
demands concentration and care. Indifference and a
lack of enthusiasm can also stifle creativity, which is
essential if improved and/or less costly maintenance
techniques are to be honed from existing ones. Finally,
experience has shown that the vegetated, "living" char-
acter of most structural BMPs requires a certain interest
and concern on the part of maintenance staff (qualities
that are evident in most successful gardeners) if proper
maintenance, performance, and aesthetic levels are to
be achieved.
Therefore, it is essential for maintenance staff to have
an interest in the overall success of the BMP. One way
that this may be accomplished is by having the long-
term maintenance of a given BMP performed by the
same maintenance crew, which then becomes the sole
group responsible for its success or failure. Such "own-
ership" of the BMP helps promote more direct interest
in its condition and a greater effort to maintain it.
In addition, competent BMP inspection, particularly of
larger, more complex structures and dams, requires a
high degree of skill, experience, and knowledge. Often,
such levels require that some of the inspections be
conducted by a licensed professional engineer who has
a background in geotechnical and structural engineer-
ing. Other necessary skills may include biology or plant
sciences, particularly if the BMP includes diverse vege-
tation and habitats. Obviously, the training required for
such inspection personnel is more rigorous and the
number of qualified personnel available to the program
will be less. Finally, the training provided to maintenance
workers should, in part, be directed at making them
informal inspectors as well. When maintenance workers
are trained and motivated to spot and report such prob-
lems as sloughing or settling of embankments, surface
erosion, animal burrows, and structural cracks, repairs
can be performed more promptly and with less expense
and effort.
Regular Performance of Routine Maintenance
Tasks
The essence or core of any facility maintenance pro-
gram is the regular, consistent performance of the actual
maintenance tasks that the remainder of the program
has identified, planned, and scheduled, and for which
staff, equipment, and funding have been provided. The
competent and consistent performance of these routine
tasks is the single greatest factor in determining the
success of the overall BMP inspection and maintenance
program. These routine tasks normally include grass
mowing and trimming, trash and debris removal, soil
fertilization, and sediment removal. Experience has
shown that the regular, frequent (e.g., monthly or less)
performance of these tasks often requires less overall
time and effort on an annual basis than if the tasks are
performed only a few times a year.
In addition, a flexible and informed definition of "regular"
should be adopted when scheduling routine mainte-
nance tasks. For example, while it will be easier to
schedule maintenance at a given BMP for the first week
of every month, the actual performance of the work
should instead be based on weather conditions and
maintenance need. This is particularly true of turf grass,
which may be damaged by a regularly scheduled mow-
ing during dry or drought conditions. During wet condi-
tions, attempts to perform maintenance tasks may result
in rutting and other ground disturbances, causing more
facility damage. The ability to perform "regular" mainte-
nance tasks on a somewhat "irregular" basis is one of
the greatest challenges of a comprehensive inspection
and maintenance program.
Timely Performance of Emergency
Maintenance Tasks
Despite the best efforts of any inspection and mainte-
nance program, emergency maintenance measures may
be necessary at a structural BMP from time to time for a
variety of causes, ranging from excessive rainfall to van-
dalism. As a result, the successful inspection and main-
tenance program must be ready to respond to this need
in a timely and comprehensive manner. To do so, it is
best to plan ahead for emergencies by developing an
emergency response plan that identifies potential emer-
gency problems and ways to address them. This may
include the preparation of a list of typical repair materials,
which then can be either stockpiled in house or quickly
acquired through designated suppliers. The plan may
also identify individuals and organizations that can pro-
vide technical input or services on short notice to assist
in the emergency repair effort. Finally, a designated num-
ber of staff personnel should be available on a 24-hour
basis to respond to maintenance emergencies.
296
-------
Regular, Competent Inspections
One of the keys to program efficiency and overall BMP
safety is the performance of competent BMP inspection
on a regular basis. In view of the increasingly complex
nature of structural BMPs and the wide range of techni-
cal aspects inherent in each, the need for competent
inspectors should be obvious. In fact, a team of inspec-
tors may be necessary to adequately review the
geotechnical, environmental performance, structural,
hydraulic, and biological aspects of many BMPs. In-
spections must be performed on a regular basis to
identify problems and special maintenance needs
quickly and efficiently. This allows repairs to be per-
formed promptly without the need for major remedial or
emergency action.
The frequency of inspections varies with the size and
complexity of a given BMP. Regular inspections by
qualified personnel may range from once a year for
large facilities with high damage potential to every 2 to
5 years for smaller, less complex sites. Additional in-
spections should also be performed as appropriate fol-
lowing major rain storms and other extreme
climatological events such as droughts, extreme snow-
falls, or high winds. It should also be noted that the
growing complexity and technical range of structural
BMPs is expected to require more frequent inspections
covering a wider range of BMP features.
Finally, the formal inspections described above should
be supplemented by informal inspections conducted by
maintenance personnel during each of their site visits.
This further enhances the program's ability to quickly
identify and respond to special maintenance needs be-
fore they can become costly emergencies. As noted
above, such informal inspections require further training
of maintenance personnel.
Performance Guarantees and Defaults
In many BMP inspection and maintenance programs,
the owners of the property on which the BMP is located
are responsible for performing maintenance tasks. Such
properties may range from single-family residences to
major industrial or commercial complexes. Under such
conditions, the governmental agency responsible for the
overall success of the program must obtain some form
of guarantee that the maintenance will in fact be per-
formed. This guarantee is acquired through several
steps. First, the property owner's responsibilities should
be specified in a written agreement between the owner
and the agency. This agreement should also grant the
agency the right to enter the property and inspect the
BMP to ensure that the stipulated maintenance is, in
fact, being performed satisfactorily. In addition, the
agreement should also provide a method by which the
agency can perform both emergency and regular main-
tenance tasks in the event of default by the owner,
including a provision to charge the owner for the cost.
Finally, such an agreement should be binding on all
future owners of the property to ensure continuity.
Accurate Recordkeeping
In view of the large number of tasks, equipment, and
materials that may be involved in a comprehensive
BMP inspection and maintenance program, accurate
records of the maintenance effort should be kept. This
includes logs of time and manpower, records of mate-
rial quantities and costs, and the type and frequency of
the various maintenance tasks performed. In addition,
accurate records should also be kept of any complaints
received from community residents regarding the ade-
quacy and/or frequency of the various maintenance
tasks as well as all reports of potentially hazardous
conditions. The time and expense of such recordkeep-
ing, including the need for staff training in the proper
procedures, can be quickly offset if the recorded infor-
mation is used to improve scheduling, task perform-
ance, and purchasing practices. Additional details of
such use is described below.
Productive Self-Evaluation and Interaction
To achieve improved levels of efficiency, a BMP inspection
and maintenance program should conduct regular reviews
and self-evaluations. The availability of thorough program
records is of great assistance in this effort. The program
review should include input from all program personnel
and should address such aspects as maintenance fre-
quency, the sequence of facility visits, equipment suitabil-
ity, staffing levels, and training needs. In addition,
establishing a positive dialogue with stormwater regula-
tors, designers, and contractors is highly desirable be-
cause all of these people are responsible for creating the
structural BMPs that the inspection and maintenance pro-
gram must ultimately (and forever) maintain. Studies and
experience have shown that many of the problems en-
countered during BMP maintenance are actually the result
of poor or misinformed regulations, designs, or construc-
tion efforts. Therefore, maintenance personnel need to
identify such problems and be given a means to inform
those responsible. Such interaction can be achieved
through conferences and meetings with professional so-
cieties, industry groups, and governmental agencies and
departments. Public input should also be sought through
individual contacts (using the complaint records noted
above) and community meetings.
The Growing Need for BMP Performance
Monitoring
More than just grass mowing, BMP inspection and main-
tenance represent a broad range of integrated technical
activities. In fact, this can also be said for the entire field
of modern stormwater management, which requires
297
-------
technical interaction between regulators, designers,
contractors, maintenance personnel, and the public to
truly achieve the goal of comprehensive runoff manage-
ment. Unfortunately, due to the random and, at times,
unpredictable behavior of storm events and the inherent
complexity of the rainfall-runoff process, it is often diffi-
cult to determine how well our runoff goals are being
met, regardless of the proficiency of design, construc-
tion, and maintenance efforts. For this reason, BMP
performance monitoring should also be included in any
stormwater management program.
By closely and accurately monitoring BMP performance
through field monitoring, sampling, and laboratory
analysis, BMP monitoring can enable us to better define
the "problem" of runoff pollution and allow regulators
and designers to gain a better understanding of both
BMP function and performance. This information can be
used more conclusively to identify those runoff goals
and management functions that either can or cannot be
realistically achieved by structural BMPs. This will fur-
ther allow regulators and designers to improve those
functions that are viable and to develop alternatives to
those that are not, both through enhanced design stand-
ards and techniques and updated regulations. BMP per-
formance monitoring can also provide information
regarding construction and maintenance practices that
may have an effect on facility performance, which can
in turn lead to improved or new practices or equipment.
In overview, BMP performance monitoring can be seen
as a means of achieving greater return on the time,
materials, and property invested now and in the future
in our stormwater management programs. And because
these amounts are expected to grow considerably as we
expand our programs to address more complex storm-
water problems, the importance of such improved re-
turns will certainly increase.
In addition, BMP performance monitoring can also be
seen as a way to help ensure overall program credibility
and achieve stronger community acceptance. In recent
years, much attention has focused on the need to ex-
pand traditional stormwater management programs be-
yond structural measures to also include nonstructural
measures in order to achieve more comprehensive re-
sults. To do so, we must achieve greater community
involvement in our stormwater management efforts,
both through lifestyle changes (involving a wide scope
of activities, from pet care to car washing to home
landscaping) and through participation in various non-
structural stormwater programs (ranging from house-
hold waste disposal to carpooling to resource
preservation). With the real data obtained through BMP
performance monitoring, it will be easier to convince the
community of both the need for and the promise of
stormwater management.
Such data will also lend greater credibility to our con-
cerns over runoff pollution and will enable us to credibly
demonstrate the value of both structural and nonstruc-
tural measures. Such credibility is vital if we are to
expect the public to make the changes and sacrifices
demanded by both the structural and nonstructural
BMPs we now have or hope to implement in their
communities (and even their backyards) in the future.
Finally, BMP performance monitoring will help us to
more closely monitor our progress and more quickly
identify program problems and shortcomings. This will
help us to develop and implement program modifica-
tions and improvements in a manner that will not
threaten community acceptance. As noted earlier, we
will not be able to rely on public support for nor par-
ticipation in vital nonstructural stormwater programs if
we are unable to create and maintain aesthetically
pleasing structural BMPs. We can also expect similar
results if we discover that those same BMPs simply
do not work.
Summary
To achieve comprehensive success in our stormwa-
ter management efforts, it is vital that inspection,
maintenance, and monitoring be considered as
equal in importance to structural BMP planning, de-
sign, and maintenance.
The neglect of BMP inspection and maintenance can
actually result in worse environmental impacts to a
community than the ones that the BMP was intended
to prevent. This result can threaten the viability of the
entire stormwater management program.
BMP inspection and maintenance must be an official
component of a comprehensive stormwater manage-
ment program, with adequate staffing, equipment, and
funds.
Self-evaluation and interaction with regulators, design-
ers, constructors, and members of the community are
vital to reducing overall maintenance needs, efforts, and
costs.
BMP performance monitoring is increasingly impor-
tant to the continued effectiveness and growth of
stormwater management programs.
298
-------
Guidance Specifying Management Measures for Sources ofNonpoint Poiiution in
Coastal Waters
Rod Frederick
Office of Water, U.S. Environmental Protection Agency, Washington, DC
Abstract
This paper describes the technology-based manage-
ment measures developed under Section 6217(g) of the
Coastal Zone Act Reauthorization Amendments to con-
trol sources of nonpoint pollution in the coastal zone.
The implementation of state coastal nonpoint source
control programs, including the development of enforce-
able policies and mechanisms, is the subject of other
papers. The management measures, and the various
practices that can be implemented cost-effectively to
achieve conformity with the management measures, are
the subjects of this paper. The U.S. Environmental Pro-
tection Agency document Guidance Specifying Man-
agement Measures for Sources ofNonpoint Pollution in
Coastal Waters (1) contains most technical information
available on the effectiveness of practices to control
nonpoint source pollutants and the costs of these prac-
tices. Nonpoint sources addressed in the document in-
clude agriculture, forestry, urban areas, marinas, and
hydromodification (dams, shorelines, and channels).
Practices include nonstructural methods such as plan-
ning, pollution prevention, and source reduction alterna-
tives in addition to structural methods such as detention
ponds and composting facilities. A separate chapter of
the document contains information on the protection and
restoration of wetlands with nonpoint source pollution
abatement functions and the use of vegetated treatment
systems in nonpoint source control programs.
Introduction
Section 6217 of the Coastal Zone Reauthorization Amend-
ments of 1990 (CZARA) requires the development of
coastal nonpoint source (NPS) control programs to protect
and restore coastal waters. States with coastal zone man-
agement plans that the National Oceanic and Atmos-
pheric Administration (NOAA) has already approved will
develop the new NPS control programs by implement-
ing management measures found in the U.S. Environ-
mental Protection Agency (EPA) document Guidance
Specifying Management Measures for Sources of Non-
point Pollution in Coastal Waters (1). The development
process, including determination of program content, use
of alternative management measures, and development
of additional management measures to meet water qual-
ity standards, is described in a separate document (2) and
is the subject of other papers. This paper focuses on the
development of the management measures and their
basisthe structural and nonstructural practices that
can be used to cost-effectively control NPS pollution and
achieve conformity with the management measures. The
value of the management measures guidance as a com-
prehensive technical reference should not be underes-
timated because it was developed as guidance for
coastal state programs; the management measures guid-
ance contains detailed information on the cost and effec-
tiveness of a wide variety of methodologies and
technologies that have proven effective in controlling
nonpoint sources of pollution in both coastal and non-
coastal areas.
Legislative Background
Congress enacted CZARA on November 5, 1990. A
major focus of this law is the control of NPS pollution to
avoid impacts on coastal waters. Congress showed
concern in section 6202(a) that growing populations in
the coastal zone are endangering wetlands and marine
resources. Section 6217 addresses this concern by re-
quiring that each state with an approved coastal zone
management program develop a coastal NPS control
program and submit it to NOAA and EPA for approval.
The purpose of the coastal NPS control program is to
develop and implement management measures for
NPS pollution to restore and protect coastal waters,
working closely with other state and local agencies.
Simply stated, EPA develops the management meas-
ures and publishes them as guidance, and the states
develop and implement programs in conformity with the
management measures and program guidance.
299
-------
Section 6217(g) of CZARA defines management meas-
ures as the best available controls that can be economi-
cally achieved to reduce pollutants from existing and
new categories and classes of NPS pollution. The
charge is clearly to develop technology-based controls
to reduce pollution from nonpoint sources. In addition,
Section 6217(b) of CZARA requires the implementation
of additional water-quality-based management meas-
ures to protect impaired and critical coastal areas if
implementation of the measures developed under Sec-
tion 6217(g) is not effective at improving water quality.
Guidance Development
To develop the guidance, EPA formed work groups
composed of more than 250 people recognized as
knowledgeable in the control of NPS pollution. The
work groups corresponded to the six technical chap-
ters of the management measures guidance and were
cochaired by EPA staff and a combination of staff from
NOAA, the U.S. Department of Agriculture (USDA)
and the U.S. Forest Service. Other work group mem-
bers included staff from state agencies, interstate agen-
cies, research agencies, universities, and other federal
agencies including the Bureau of Land Management,
Fish and Wildlife Service, Army Corps of Engineers,
Federal Highway Administration, National Park Serv-
ice, and Geological Survey.
Work group members provided references, literature
reviews, and advice as EPA worked with its own con-
tractors and experts to pull together, analyze, and sum-
marize information on management practices and their
effectiveness. EPA released the proposed management
measures guidance in May 1991. EPA and NOAA also
published a proposed program implementation guid-
ance in October 1991.
Input on the proposed management measures guid-
ance was solicited from the public during a 7-month
comment period. The major problems identified in the
public comments on the technical chapters were a
lack of cost information and a perceived "East Coast
bias" in the practices identified. There were, however,
many positive comments on the usefulness of the
guidance as a compendium of structural and non-
structural control alternatives for NPS pollution in all
areas of the country.
The final management measures guidance was re-
leased in January 1993. That document incorporated
most suggested improvements and additional informa-
tion received from the public comments, as well as 1) a
more thorough literature review; 2) additional focus on
regional differences in climate, weather, and geomor-
phology; 3) additional cost information; and 4) informa-
tion on economic achievability. The final management
measures guidance is more than twice the size of the
May 1991 proposed guidance and, hopefully, twice as
useful. There are more alternative practices, better de-
scriptions, additionalsource reduction and pollution pre-
vention programs, and examples of successful
implementation of cost-effective practices undera vari-
ety of site conditions. Based on the favorable response
to date on the final management measures guidance,
the guidance is a valuable technical reference for iden-
tifying NPS problems and cost-effective solutions.
Description of the Final Management
Measures Guidance
Problem Identification
Each chapter contains a discussion of NPS pollutants
and problems as a rationale for the management meas-
ures and controls to be implemented as part of state
coastal NPS control programs.
Agricultural Runoff
Coastal waters are affected by NPS pollution result-
ing from the erosion of crop land; from the manure
and other wastes produced in confined animal facilities;
from the application of nutrients, pesticides, and irriga-
tion water to crop land; and from physical disturbances
caused by livestock and equipment, particularly in and
along streambanks.
Urban Runoff
Urbanization in the form of new development changes
the natural hydrology of an area and increases runoff
volumes, erosion, sediment loadings to surface waters,
and loadings of sediment, nutrients, oxygen-demanding
substances, pathogens, metals, hydrocarbons, and
other pollutants. These changes and increases can im-
pair water quality, alter habitats, close and destroy fish-
eries and shellfish beds, and close recreational areas
such as beaches. Decreases in base flows caused by
impervious areas can also adversely alter habitat and
impairwaterquality. Existing urban activities such as the
use of onsite disposal systems, improper disposal of
household wastes, turf and lawn management, pets
wastes, and road maintenance can also cause water
quality problems.
Silvicultural (Forestry) Operations
Forestry operations can degrade water quality in water
bodies receiving drainage from forest lands. Sediment
concentrations can increase because of accelerated
erosion; water temperatures can increase because of
removal of the overstory riparian shade; slash and other
debris can deplete dissolved oxygen; and organic and
inorganic chemical concentrations can increase due to
harvesting and the use of fertilizers and pesticides. In-
creased stream flow can also result from the removal of
trees and vegetation.
300
-------
Marinas and Recreational Boating
Because marinas are located at the water's edge, a variety
of nonpoint effects are associated with poor flushing of
boat basins, spills from refueling areas, bilge pumping, and
wastes produced by the cleaning and repair of boats.
Hydromodification
Hydromodification activities have been separated into
three categories:
Channelization and channel modification frequently
diminish the suitability of instream and streamside
habitat for fish and wildlife, and alter instream pat-
terns of water temperature and sediment transport.
Hardening of banks, in particular, can increase the
speed of movement of NPS pollutants from the upper
reaches of watersheds into coastal waters.
Dams can affect the hydraulic regime, the quality of
surface waters, and the suitability of instream and
streamside habitat for fish and wildlife.
Shoreline and streambank erosion is a natural proc-
ess that can have either beneficial or adverse im-
pacts on surface water quality and on the creation
and maintenance of coastal habitat. Eroded shoreline
sediments help maintain beaches and replenish the
substrate in tidal flats and wetlands. Excessively high
sediment loads, however, can smother submerged
aquatic vegetation, cover shellfish beds, fill in riffle
pools, and contribute to increased levels of turbidity
and nutrients.
Wetlands and Vegetated Treatment Systems
Wetlands and riparian areas reduce NPS pollution by
filtering pollutantsespecially sediment, nitrogen, and
phosphorusfrom surface waters. Wetlands and ripar-
ian areas can also attenuate flows from higher-than-av-
erage storm events, thereby protecting receiving waters
from peak flow hydraulic impacts such as channel scour,
streambank erosion, and fluctuations in temperature.
Degraded wetlands lose this important set of NPS con-
trol functions. Also, degradation of wetlands and riparian
areas can cause these areas to become sources of
nonpoint pollution because they will then deliver in-
creased amounts of sediment, nutrients, and other pol-
lutants to adjoining water bodies.
Management Measures and Practices
The management measures are major subheadings within
each chapter. The coastal NPS control programs that
states are to develop must be in conformity with these
measures. An applicability section for each measure con-
tains information on the activities and locations to which
each measure applies. A description section is included
for each measure to illustrate goals and objectives and
provide more detail on what the measures mean. The
selection section provides the rationale used in select-
ing the management measure. Usually, selection is
based on widespread use of a management practice or
combination of practices that can be used to achieve the
management measure. The economic achievability of
the management measures was evaluated separately
(3). If this evaluation affected the selection of a measure,
the effect is described in the selection section.
Management practices are described in a separate
section under each management measure for illustra-
tive purposes. State programs do not have to specify
or require the implementation of any of these manage-
ment practices. EPA does expect, however, that one
or a combination of these practices appropriate to
local conditions can be used to achieve conformity
with the management measures. For example, the
management measure for runoff from new develop-
ment calls for 80 percent reduction in the average
annual total suspended solid (TSS) loadings. Several
management practices such as sand filters or ex-
tended detention wet ponds can be used to achieve
the required TSS removal. If local conditions are not
appropriate for one of those practices, however, a
combination of vegetated filter strips, grass swales,
wet ponds, or constructed wetlands could also be
used to achieve the measure. The costs and effective-
ness of the management practices are usually in-
cluded within the description of each practice or in a
separate summary section at the end of each man-
agement measure chapter. An economic impacts
study (3) was prepared based on representative prac-
tices and combinations of practices and their costs.
Management Measures by Chapter
Presented below are brief synopses of the major man-
agement measures presented in each of the technical
chapters. The discussion below is not comprehensive,
and the management measures guidance should be
consulted to establish the exact requirements and appli-
cability of the management measures.
Agriculture
Sediment and erosion control: Rely on USDAs con-
servation management system to promote practices
such as conservation tillage and strip-cropping.
Animal facilities (large units): Contain runoff and ani-
mal waste in storage structures.
Animal facilities (small units): Use less-stringent re-
quirements for economic reasons.
Nutrient management: Develop and implement com-
prehensive nutrient management plans that involve
fertilizer application rates, timing, and use efficiency.
301
-------
Pesticide management: Evaluate the problem and
site, use integrated pest management (IPM) where
possible, and apply pesticides properly and safely.
Livestock grazing: Protect sensitive areas through ap-
propriate grazing management techniques (e.g., pro-
viding alternative water, salt, and shade sources away
from sensitive areas and providing livestock crossing
areas).
Irrigation: Optimize water use and use chemigation
safely.
Forestry
Preharvest planning: Consider the timing, location,
and design of harvest activities.
Streamside management areas (SMAs): Establish
SMAs to protect against soil disturbance and delivery
of sediment and nutrients from upslope activities; re-
tain canopy species to moderate water temperature.
Road construction/reconstruction and road manage-
ment: Reduce the generation and delivery of sediment.
Timber harvesting: Protect waters during harvesting,
yarding, and hauling.
Site preparation and forest regeneration: Confine on-
site potential NPS pollution and erosion resulting
from these activities.
Management of fire, chemicals, and forested wetland
areas: Reduce NPS pollution of surface waters.
Revegetation of disturbed areas: Prevent sedimenta-
tion from harvest units or road systems.
Urban
Runoff control for new development: Reduce runoff lev-
els of TSS by 80 percent, and maintain natural hydrology.
Watershed protection/site development: Use compre-
hensive planning to protect areas that are ecologi-
cally sensitive, provide water quality benefits, or are
prone to erosion.
Construction erosion/sediment and chemical control:
Reduce construction-related erosion, retain sediment
onsite, and properly manage chemical use.
Runoff management for existing development: Iden-
tify and implement runoff quality controls as appro-
priate and feasible.
New and operating onsite disposal systems (OSDSs):
Select, site, and operate OSDSs to reduce OSDS
impacts on coastal waters.
Pollution prevention for urban areas: Target and imple-
ment NPS reduction and public education programs.
Roads, highways, and bridges:Site, construct, operate,
and maintain roads, highways, and bridges properly.
Marinas
Marina siting and design:
- Allow for maximum flushing of the marina basin.
- Perform water quality and habitat assessments to pro-
tect against adverse impacts on shellfish resources,
wetlands, and submerged aquatic vegetation.
- Control stormwater runoff (additional controls exist
for hull maintenance areas).
Fueling station design: Design to allow for ease of
cleanup, and develop spill contingency plans.
Sewage facilities: Ensure availability of pumpouts and
pump stations, and develop maintenance procedures.
Operation and maintenance: Establish marina opera-
tion and maintenance programs to control and to
provide for proper disposal of solid waste, fish waste,
liquid materials, petroleum products, and boat clean-
ing byproducts.
Public education: Develop public education programs
for marina users.
Hydromodification
Channelization and channel modification: Evaluate
effects of new projects on physical and chemical
characteristics of surface waters and on instream and
riparian habitats
Dams: Control erosion/sediment and chemicals dur-
ing and after construction; develop and implement an
operation and maintenance plan to protect surface
water quality and instream and riparian habitat.
Eroding shorelines and streambanks: Stabilize stream-
banks and shorelines where erosion is a nonpoint prob-
lem; vegetative methods are strongly preferred over
engineering structures where vegetation will be cost-
effective. Protect streambanks and shorelines from ero-
sion from the use of the shore and adjacent waters.
Wetlands, Riparian Areas, and Vegetated
Treatment Systems
Protection: Protect wetlands and riparian areas serv-
ing a NPS pollution abatement function to maintain
water quality benefits and ensure that they do not
become a source of nonpoint pollution.
Restoration: Promote the restoration of damaged
and destroyed wetlands and riparian systems
where they will have a significant NPS pollution
abatement function.
302
-------
Vegetated treatment systems: Promote the use of
constructed wetlands and filter strips where they
will serve a significant NPS pollution abatement
function.
Next Steps
1993
NOAA and EPA began meeting with states and other
interested parties to assist in program development and
determine their needs for future technical assistance.
Activities included:
Regional workshops with state coastal zone manage-
ment and NPS control agencies.
Briefings of other federal agencies and interest groups
(e.g., trade associations and environmental groups).
Presentations at meetings of other interested parties
(e.g., International Marina Institute, National Associa-
tion of Conservation Districts, Water Environment
Federation, and Coastal State Organization).
1994
NOAA and EPA formulated and implemented a technical
assistance program using information on needs obtained
from state and local government, industry, trade organiza-
tions, and others. Elements of this program include:
Publishing several guidance documents, including
State and Local Government Guide to Environmental
Program Funding Alternatives and Developing
Successful Runoff Control Programs for Urbanized
Areas (4, 5).
Providing funds to help produce additional technical
guidance, including Urbanization and Water Quality,
Watershed Protection Techniques, and Fundamen-
tals of Urban Runoff Management (6-8).
Conducting workshops on such topics as stream res-
toration, NPS monitoring, and marina NPS controls.
Developing educational curricula and sponsoring
train-the-teacher programs on runoff NPS pollution.
Developing an expert system for identifying and se-
lecting agricultural NPS controls.
References
1. U.S. EPA. 1993. Guidance specifying management measures for
sources of nonpoint pollution in coastal waters. EPA/840/B-
92/002. Washington, DC.
2. U.S. Department of Commerce/NOAA/U.S. EPA. 1993. Coastal
nonpoint pollution control program, program development, and
approval guidance. Washington, DC: U.S. Department of Com-
merce, National Oceanic and Atmospheric Administration, and
U.S. EPA.
3. RTI. 1992. Economic analysis of coastal NPS pollution controls.
Project 5990-91. Prepared by Research Triangle Institute for U.S.
EPA Nonpoint Source Control Branch, Washington, DC.
4. U.S. EPA. 1994. State and local government guide to environ-
mental program funding alternatives. EPA/841/K-94/001.
5. U.S. EPA. 1994. Developing successful runoff control programs
for urbanized areas. E PA/841/K-94/003.
6. Terrene Institute/U.S. EPA. 1994. Urbanization and water quality.
Washington, DC.
7. Center for Watershed Protection. Watershed protection tech-
niques. Quarterly bulletins (February and Summer).
8. Horner, Skupien, Livingston, and Shaver/Terrene Institute/U.S.
EPA. 1994. Fundamentals of urban runoff managment: Technical
and institutional issues. Washington, DC.
303
-------
Biotechnical Streambank Protection
Don Roseboom, Jon Rodsater, Long Duong, Tom Hill, Rich Offenback,
Rick Johnson, John Beardsley, and Rob Hilsabeck
Illinois State Water Survey, Peoria, Illinois
Abstract
Streams in areas of intense residential and commercial
development have high rates of surface water runoff, so
bank erosion and downstream flooding become more
common and severe. Throughout the greater Chicago
area, this has resulted in destabilized streams lacking
habitat for fish, wildlife, and people. The Illinois Environ-
mental Protection Agency and the U.S. Environmental
Protection Agency funded the urban stream restoration
projects on Glen Crest Stream and the Waukegan River.
During the spring and summer of 1992, stabilization
sites were completed on Glen Crest Stream, in Glen
Ellyn, and in Washington Park of the Waukegan Park
District. The lunker technique was chosen for its low
cost of installation and ability to resist the high-velocity
runoff while increasing instream habitat for gamefish
and the stream side habitat for the urban population. At
Glen Ellyn, lunkers were constructed of recycled plastic
lumber for increased longevity. Low-cost vegetative sta-
bilization incorporated an initial matrix of grasses and
willows, plus rooted stock of redosier dogwood near the
water's edge, followed by appropriate riparian trees on
the upper bank that the landowner chose. Both projects
trained senior members and staff personnel of the park
district and the city in the application of lunkers and
vegetative stabilization.
Introduction
This paper describes methods of biotechnical stabiliza-
tion and instream habitat enhancement that have been
field trialed in Illinois. These practices have been author-
ized and funded by the U.S. Fish and Wildlife Service,
the Soil Conservation Service, the U.S. Environmental
Protection Agency, and all Illinois state agencies respon-
sible for stream modification permits. The following
methods are described: willow post bank stabilization,
lunker instream habitat enhancement with vegetative
bank stabilization, and A-jack structural and vegetative
bank stabilization (Figures 1, 2, and 3).
In rural Illinois areas, bank erosion is not addressed
because of limited financial resources. In agricultural
states, U.S. Army Corp of Engineers district offices
receive many requests for assistance on bank erosion
protection. Wthin recent years, the need for bank
erosion control has been coupled with the need for
environmental protection of the stream habitat and
riparian areas for wildlife and fisheries. Keeping costs
low while considering various environmental issues
has made bank erosion control a difficult challenge for
the Corps.
In Illinois, stream channel erosion increased when
prairies were converted to rowcrop agriculture and
residential development, thereby increasing surface
water runoff rates. Man has become a dominant geo-
morphic factor in the watershed hydrology of both
rural and urban watersheds. In most urban and agri-
cultural areas, streams were channelized to move
floodwaters away from valued lands, to maximize the
size and uniformity of land holdings, even to decrease
channel erosion (1). One result of increased water
runoff rates and poorly designed channelization ef-
forts has been massive bank erosion in the flood-
plains of Illinois streams.
Watershed studies by the Illinois State Water Survey
have documented the channel erosion damages to
floodplain fields and the consequent increased sedi-
ment yield. Channel erosion contributed 40 to 60 per-
cent of the sediment yield in two monitored Illinois
watersheds (2). Within these watersheds, increased
runoff rates and stream channelizations caused the
streambed to be downcut at first and then erode lat-
erally to regain a meander shape (Figure 4). This
process was hastened by channel incision into ex-
tremely unstable glacials and gravel deposits below
an 8- to 20-ft layer of loess clays. The Crow Creek
watershed study demonstrates both the bridge dam-
ages from channel incision and the field damages
from bank erosion (3).
304
-------
Willow Stakes
Predisturbance
Graded
Slope
Figure 1. Willow posts installed below depth of streambed
scour.
Prairie Grasses -^
Figure 2. Lunker with riprap below baseflow stage. Rebar is
driven below bed scour depth.
Willow Stakes V ay
U (//
Graded Slope
Figure 3. A-jack bank structures.
Willow Post Bank Stabilization
The willow post method differs from most European
bioengineering techniques (4, 5) in that individual wil-
lows are positioned vertically below the depth of
channel scour. Most biotechnical bank stabilization
techniques have used vegetation with a riprap men-
tality. Layers of horizontally bundled woody vegetation
are entrenched in the bed and bank. This type of earth
Figure 4.
Recovery
Incision and recovery process. Vegetative bank sta-
bilization can be applied during the widening phase.
moving and hand labor often doubles installation
costs and installation times.
Willows and most woody riparian vegetation do not
naturally extend root systems very deeply below the
water table. The posts are implanted much deeper than
native seedlings would grow. Lateral root growth rapidly
binds adjacent posts together in the bank soil. Lateral
branch growth also interlocks adjacent posts to slow
flow velocity near the bank.
The willow post method was mentioned by Scheichtl (4)
as a method of ravine stabilization in Germany during
the 1800s. Both the Corps of Engineers and the Soil
Conservation Service used large willow poles in the
1930s (6, 7). In most cases, the posts or poles were
laid as a layer along the sloped bank. York (8) placed
willow posts in vertical holes to protect the base of
levees in Arizona.
Willows are cut into 10- to 14-ft posts when the leaves
have fallen and the tree is dormant. At this time, growth
hormones and carbohydrates are stored in the root
system and lower trunk. Dense stands of 4- to 6-year-
old willows make the best harvesting areas. These
stands are commonly found on the stream deltas in
lakes or in old stream channel cutoffs. The willow posts
are 4 to 6 in. in diameter and may be stored up to 1
month if kept wet.
The eroding streambank is shaped to a 1:1 slope with
the spoil placed in a 6-in. deep layer along the top of the
305
-------
bank. In major erosion sites, post holes are formed in
the bed and bank so that the end of the post is 2 ft below
maximum streambed scour. The posts are placed 4 ft
apart in rows up the streambank. The posts in one row
are offset from the posts in adjacent rows.
While the steel ram and excavator is more efficient at
depths of 6 ft in clay soils, a hydraulic auger and
excavator unit forms deeper and longer lasting holes in
stony or sand streambeds. Large stone layers of
streambed material cause damage to the excavator
when the steel ram is used. In fine sand layers, ram
holes collapse before the post reaches the bottom of
the holes. In highly fluid sands, even auger holes fill but
the post can be pushed deeper with the bucket or
boom. In streams with sand or gravel beds, the hydrau-
lic auger places posts 9 to 11 ft deep in the bed. Almost
all contractors in Illinois currently use an excavator and
hydraulic auger unit.
In larger streams with noncohesive sand banks, large
cedar trees are cabled to the willow posts along the toe
of the bank. The cedars not only reduce bank scour
while root systems are growing but also retain moisture
during drought periods. In larger streams, such as Illi-
nois's only designated scenic river, the Middle Fork,
large rounded boulders were used as additional bank
protection with the willow posts.
In Illinois, the contractor slopes 15-ft banks on a 1:1
grade for 80 cents per linear foot. Each post hole is
augered 10 ft deep for $2.90. Each willow post costs $1
to $2. With a five-man crew at $10.00 per hour per man,
bank sites are estimated to cost between $5 and $8 per
linear foot.
Bank Erosion Site Assessment
The following questions should be asked when deter-
mining the applicability of willow bank post stabilization:
1. Does sunlight fall directly on the eroding bank?
(Willows must have sun.)
2. Is bedrock close to the surface? (Streambed mate-
rial should be 4 ft deep; check with a tile probe.)
3. Are lenses of fine sand exposed in the eroding bank?
4. Is the stream channel stable upstream of the ero-
sion site? (If the stream cuts behind the upper end
of willow posts, the entire bank will erode.)
5. How deep is the stream along the eroding bank?
(Willow posts must be 2 ft deeper than the deepest
water or the posts will be undercut below the root
zone. The length of the willow posts depends on the
water depth. In sand or cobble streams, a hydraulic
auger forms a deeper and more stable hole.)
6. How wide is the stream channel at the erosion sites
compared with stable channels upstream and
downstream? (If the channel is wider at the erosion
site, vegetation will not choke the stream channel
and cause other erosion problems.)
7. Do you have a source of large willows close to
the site? (Your costs are small when the willows
are close.)
8. Will the site be wet during dry summers? (Willow
posts require a lot of water while the roots are
regrowing; willow posts should only extend 1 to 2 ft
aboveground in dry sites.)
9. Can you keep cattle away from the posts during the
first summer? (Willows must be able to produce
leaves for photosynthesis and regrowth.)
10. Have debris jams forced floodwater into the eroding
bank? (Large debris jams must be removed ac-
cording to guidelines established by the American
Fisheries Society (9).)
The willow post method of bank stabilization is the lowest
cost bank stabilization method that provides both wildlife
and fisheries benefits. This method has received wide-
spread support by both the agricultural and environmental
communities: Farm Bureau, soil and water conservation
districts, American Fisheries Society, and Nature Conser-
vancy. The willows serve only as a pioneer plant on the
disturbed soils. Succession to wooded or grass banks is
speeded by additional trees or grass plantings with active
site management if the landowner desires.
Lunker Instream Habitat Structures
Lunkers are constructed of 2-in. oak planks (10). The
planks form upper and bottom layers so that the inte-
rior is open to water flow at both ends and on the
stream side of the structure (Figure 2). A series of
lunkers are placed along the base of the eroding bank.
When necessary, the lunkers are placed into an exca-
vated trench, especially on the upper and lower ends
of the sites. Each lunker is held with nine lengths of
rebar, which are driven 5 ft into the streambed. In the
Illinois adaptation, riprap was placed only on lunkers
behind the blocking log.
In rural areas and in state parks, the bank above the
lunkers was stabilized with willow posts. The bank was
steeply sloped to keep the lunkers scoured (11) and to
prevent silt deposition in the lunkers. In Court Creek, the
upper bank was seeded with prairie grasses. During the
second year, the posts were cut down so that only a
narrow fringe of willow grew along the water's edge. By
the third year, with active burn management, the prairie
grasses had become established.
At Franklin Creek State Park, the banks were seeded
with cool season grasses because the erosion site was
located beside the equestrian corral. Once again, the
willow posts were to be cut during the second year. A
306
-------
large population of protected beavers sped up the
postcutting schedule. A spray of Ropel, an unpleasant-
tasting liquid, mixed with a tackifer (to decrease water
solubility) gave protection until the grasses became es-
tablished. When Ropel applications were discontinued,
the large posts were quickly cut down. Even with heavy
browsing, however, the willow stubs regrew branches
because the root systems were not damaged.
While the cool season grasses became established
more quickly than the prairie grasses, the root systems
of cool season grasses are shallow and therefore more
susceptible to scour during high velocity flows. While
damages have been minor after 4 years, two 9 ft2 areas
were seeded with grasses and 18-in. willow cuttings in
April 1993. Adult smallmouth bass populations in-
creased over 50 percent. Of more importance to stream
bass populations, the yearling bass survival increased
300 percent at the lunker site (12).
Costs of lunker installation were $25 to $35 per linear
foot, with prairie grass seeding and maintenance ac-
counting for higher costs at the Court Creek site. Labor
was 45 percent of costs, contractual equipment was 30
percent, and materials were 25 percent. A 300-ft site is
estimated to cost $8,000 to $10,000.
Urban Lunkers
In northeastern Illinois near Chicago, urban streams
respond quickly to rainfall events so that floods are
extremely erosive. Damage to homes and the higher
cost of lands allow more intensive stream management.
Often this has led to concrete or heavily riprapped
stream channels with acute environmental damages.
While necessary in some urban settings, the value of
residential homes and parks can be increased if stream
channel stabilization can be made more environmen-
tally sensitive. In the smaller stream, the lunkers were
constructed from recycled plastic lumber so that lunkers
would not dry rot during lowflow drought periods. In
larger stream segments, deeper pools allowed the use
of wooden lunkers.
In urban streams, the higher cost of materials, the higher
cost of contractual equipment as excavators, and the
very high cost of landscape repairs to private lawns
substantially increase the cost of lunker installation. The
lunker installations are $45 to $55 per linear foot of bank.
Summer scheduling of stream restoration required the
use of rooted and therefore smaller willow saplings.
Additional rooted stock as redosier dogwood played a
greater role in riparian revegetation of urban sites. Tree
corridors were preserved as sound barriers to traffic
noises and visual privacy barriers between homes. The
resulting shade, however, denied the use of willows in
some areas. In these shaded areas, redosier dogwood
were planted with very good survival.
These urban sites were only 1 year old at the time this
paper was presented, but the Chicago area had just
undergone an extremely wet fall and spring. Two fall
floods and three spring floods did not damage the urban
lunkers sites.
A-Jack Structures With Willow and
Dogwood Bank Revegetation
A-jacks look like small versions of the World War II tank
traps (see Figure 3). The A-jacks can be placed so that
each A-jack will interlock within each row and with A-
jacks in adjacent rows. The lowest rows of A-jacks are
trenched along the base of the eroding bank, with the
excavated sediment placed along the top of the bank.
In the Glen Crest Stream and the Waukegan River, 2-ft
diameter A-jacks were used.
Fibredam, a geotechnical fabric that locks the curled
wood fibers in excelsior blankets, was placed between
the rows of A-jacks and the bank soils to reduce soil
movement through the A-jacks. Fibredam is easily torn
apart and molded into crevices between A-jacks.
Willow cuttings were driven into the streambed between
A-jacks and behind the last interior rows of A-jacks. The
fluid sediment was placed on the rows and allowed to
fill the interior spaces. The vertical streambank was then
sloped over the A-jacks.
This structure ran $45 to $50 per linear foot of bank
when composed of two base rows and one upper row.
The cost of materials was $25 per foot. Ease of handling
and suitability for transport by small marsh vehicles are
advantages of this system. Each A-jack is composed of
two halves that lie flat on pallets during transport.
A-jacks are assembled at the bank site.
When the willows and dogwood are fully grown, root
systems lock the entire structure together while giving a
natural appearance to the streambank. Small stone is
added to A-jack rows near the waterline to give a more
natural appearance.
References
1. Keller, E.A. 1976. Channelization: Environmental, geomorphic,
and engineering aspects. In: Coates, D.R., ed. Geomorphology
and engineering. Stroudburg, PA: Dowden, Hutchinson and Ross,
Inc. pp. 115-140.
2. Roseboom, D.P., and W. White. 1990. The Court Creek restoration
project. In: Erosion control: Technology in transition. Proceedings
of the XXI Conference of the International Erosion Control Asso-
ciation, Washington, DC. pp. 25-40.
3. Roseboom, D., W. White, and R. Sauers. 1991. Streambank and
habitat strategies along Illinois River tributaries. In: Proceedings
of the Governor's Conference on the Management of the Illinois
River, pp. 112-122.
4. Schiechtl, H. 1980. Bioengineering for land reclamation and con-
servation. Edmonton, Alberta: University of Alberta Press, p. 400.
307
-------
5. Gray, D.H., and AT. Leiser. 1989. Biotechnical slope protection
and erosion control. Malabar, FL: R.E. Krieger.
6. Lester, H.H. 1946. Streambank erosion control. Agricultural En-
gineering. September: 407-410.
7. Foster, A.B. 1959. Approved practices in soil conservation. Dan-
ville, IL: Interstate Press.
8. York, J.C. 1985. Dormant stub planting techniques. Proceedings
of the First North American Riparian Conference, University of
Arizona, Tucson, AZ.
9. AFS. 1983. Stream obstruction removal guidelines. American
Fishery Society, 5410 Grosvenor Lane, Bethesda, MD 20814.
10. Ventrano, D.H. 1988. Unit construction for trout habitat improve-
ment structures for Wisconsin coulee streams. Administrative Re-
port No. 27. Madison, Wl: Wisconsin Department of Natural
Resources.
11. Nunnally, N.R. 1978. Stream renovation: An alternative to chan-
nelization. Environ. Mgmt. 2(5):403-411.
12. Roseboom, D., R. Sauer, D. Day, and J. Lesnack. 1992. Value of
instream habitat structures to smallmouth bass. Misc. Pub. 139.
Peoria, IL: Illinois State Water Survey.
308
-------
The Use of Wetlands for Stormwater Pollution Control
Eric W. Strecker
Woodward-Clyde Consultants, Portland, Oregon
Abstract
This paper presents the results of a literature review that
summarizes the current state of knowledge regarding
the use of wetlands forstormwater pollution control. The
paper reviews the primary removal mechanisms in wet-
lands, including sedimentation, adsorption, precipitation
and dissolution, filtration, biochemical interactions, vola-
tilization and aerosol formation, and infiltration. The re-
sults from 26 wetlands are reviewed and contrasted
regarding their ability to remove pollutants from storm-
water. The systems range from salt marshes to high-
elevation riverine wetlands. The study sites are
reviewed in relation to the type of wetlands system,
including design features and upstream watershed
characteristics. The wetlands receive stormwater from
different land uses, including residential, commercial,
highway, golf courses, and open. The observed pollut-
ant removal efficiencies are quite variable but generally
show good removals of phosphorus (median of 46 per-
cent average removal) and the heavy metals cadmium,
copper, lead, and zinc (median of 70, 40, 83, and 42
percent average removal, respectively) from stormwa-
ter. Constructed wetlands generally perform better and
with greater consistency. In general, larger wetlands
perform better than their watershed areas as well. Nev-
ertheless, some carefully planned constructed systems
with a small area performed quite well compared with
their watershed areas. Because there is little information
on noted impacts to biota, these are just briefly re-
viewed. Finally, the paper suggests collecting additional
information in new studies. This would make compari-
sons among different sites more useful in assessing the
factors that affect the abilities of constructed wetlands
to remove pollutants from stormwater.
Introduction
Constructed wetlands are receiving increasing attention
as attractive systems for removing pollutants from storm-
water runoff. Other potential benefits that such systems
provide include flood control and habitat. Wetlands have
long been used for the treatment of wastewaters from
municipal, industrial, and agricultural sources (1). The U.S.
Environmental Protection Agency (EPA) encourages the
use of constructed wetlands for water pollution control
through the innovative and alternative technology provi-
sions of the construction grants program (2).
The purpose of this paper is to assist EPA, state, and
local technical personnel in assessing the capabilities
and limitations of using wetlands as a control measure
to reduce the environmental impacts of stormwater pol-
lution on downstream water bodies. The paper summa-
rizes a report prepared for EPA by Strecker et al. (3) that
reviewed published literature and documented reports
on aspects of stormwater wetland design, operation,
and performance. An appendix that accompanied the
published report included a one- to six-page summary
of each pertinent study reviewed for the report. The
summaries covered influent and effluent water quality,
the effectiveness of the system, flows and volumes,
wetland and watershed areas, and the biological char-
acteristics of the system.
Table 1 presents a list of selected reports with which
researchers have documented the ability of wetland sys-
tems to remove pollutants from stormwater. The table
includes some general characteristics of the wetland sys-
tems. Figure 1 shows the wetlands' geographic locations.
The wetlands differed widely in location and wetland type
(e.g., Florida's southern swamplands, Minnesota's north-
ern peatlands, California's brackish marshlands, and
Puget Sound's palustrine wetlands). Each of these loca-
tions differs in climate, vegetation, and soil types.
Wetland Stormwater Pollutant Removal
Mechanisms
Wetlands can combine various actions to remove pol-
lutants from stormwater:
Incorporation into or attachment to wetland sediments
or biota.
Degradation.
309
-------
Table 1. Literature Researched To Investigate Performance Characteristics of Wetlands
Study/
Reference
Martin and
Smoot (4)
Harper
et al. (5)
Reddy
et al. (6)
Blackburn
et al. (7)
Esry and
Cairns (8)
Brown, R.
(9)
Wotzka
and
Oberts (10)
Hickok
et al. (11)
Barten (12)
Meiorin
(13)
Morris
et al. (14)
Scherger
and Davis
(15)
ABAC (1 6)
Jolly (17)
Oberts
et al. (18)
Reinelt
and
Horner
(19, 20)
Rushton
and Dye
(21)
Hey and
Barrett (22)
Year of
Publication
1986
1986
1982
1986
1988
1985
1988
1977
1987
1986
1981
1982
1979
1990
1989
1990
1990
1991
Location
Orange County, FL
FL
Orange County, FL
Palm Beach, FL
Tallahassee, FL
Twin Cities Metro
Area, MN
Roseville, MN
MN
Waseca, MN
Fremont, CA
Tahoe Basin, CA
Ann Arbor, Ml
Palo Alto, CA
St. Agatha, ME
Ramsey-Washington
Metro Area, MN
King County, WA
Tampa, FL
Wadsworth, IL
Name/I.D.
Orange County
Treatment
System
Hidden Lake
Lake Apopka
Palm Beach
PGA Treatment
System
Jackson Lake
Twin Cities Metro
McCarrons
Treatment
System
Wayzata
Clear Lake
DUST Marsh
Tahoe Basin
Meadowland
Pittsfield-Ann
Arbor Swift Run
Palo Alto Marsh
Long Lake
Wetland-Pond
Treatment
System
Tanners Lake,
McKnight Lake,
Lake Ridge, and
Carver Ravine
B3I and PC12
Tampa Office
Pond
Des Plaines
River Wetland
Demonstration
Project
Detention
Pond/Wetland
Detention
pond and
wetland
Wetland
Wetland
Wetland
Detention
pond and
wetland
Wetlands
Detention
pond and
wetland
Wetland
Wetland
Wetland
Wetland
Detention
pond and
wetland
Wetland
Detention
pond and
wetland
Detention
pond and
wetland
Wetland
Wetland
Wetland
Constructed/
Natural
Constructed
Natural
Constructed
Constructed
and natural
Constructed
Natural and
constructed
Constructed
Natural
Constructed
Constructed
Natural
Constructed
and natural
Natural
Constructed
Constructed
Natural
Constructed
Constructed
Wetland
Classification
Hardwood
cypress dome
Hardwood
swampland
Cattail marsh
Southern
marshland
Southern
marshland
Northern
peatland
Cattail marsh
Northern
peatland
Cattail marsh
Brackish
marsh
High
elevation
riverine
Northern
peatland
Brackish
marsh
Cattail marsh
Cattail marsh
Palustrine
Cattail marsh
Freshwater
riverine
310
-------
C3ittsfield-Ann Arbor
Swift Run
f Ramsey-
| Washington
I Metro
Figure 1. Location of wetlands researched for their ability to treat stormwater runoff.
Export of pollutants to the atmosphere or ground
water.
Both physical and chemical pollutant removal mecha-
nisms probably occur in wetlands. These mechanisms
include sedimentation, adsorption, precipitation and dis-
solution, filtration, biochemical interactions, volatiliza-
tion and aerosol formation, and infiltration. Because of
the many interactions between the physical, chemical,
and biological processes in wetlands, these mecha-
nisms are generally not independent. Sedimentation is
usually the most dominant removal mechanism. The
large variation in wetland characteristics (e.g., hydrol-
ogy, biota) may cause the dominant removal mecha-
nisms to vary from wetland to wetland. Variations in
wetland characteristics can also help explain why wet-
lands differ so widely in their pollutant removal efficien-
cies. Following is a brief description of the principal
removal mechanisms.
Sedimentation
Sedimentation is a solid-liquid separation process using
gravitational settling to remove suspended solids. It is
considered the predominant mechanism for the removal
of many pollutants from the water column in wetland and
other flow detention systems. Sedimentation of sus-
pended material, along with pollutants that are highly
adsorbed, has been documented as the primary re-
moval mechanism in wetlands by many study authors,
including Martin and Smoot (4) and Oberts (23). The
most significant factors affecting settling of suspended
material pertain to the hydraulic characteristics of the
wetland system, including the detention time, inlet-outlet
conditions, turbulence, and depth. The opposite of sedi-
mentation is flotation. Floatable pollutants such as oil
and grease, litter, and other pollutants can accumulate
in the surface microlayer. These pollutants can be re-
moved by adsorption.
Adsorption
Adsorption of pollutants onto the surfaces of sus-
pended particulates, sediments, vegetation, and organic
matter is a principal mechanism for removing dissolved
or floatable pollutants. The literature suggests that these
processes remove pollutants such as phosphorus, dis-
solved metals, and other adsorbents (including colloidal
pollutants) (5, 11, 16). Adsorption occurs through three
main processes:
Electrostatic attractions.
Physical attractions (e.g., Van der Waals forces and
hydrogen bonding).
Reactions.
311
-------
The rates by which these processes occur are thought
to be inversely related to the particle size and directly
related to the organic content of the particles in the wetland
soils (5). Increasing the contact of stormwater with the
underlying soils and organic matter can enhance adsorp-
tion processes. In addition, high residence times, shal-
low water depths, and even distribution of influent
enhance the interactions of water with soil and plant
substances, thereby increasing the adsorption potential.
Precipitation and Dissolution
Many ionic species (e.g., metals) dissolve or precipitate
in response to changes in the solution chemistry of the
wetland environment. Metals such as cadmium, copper,
lead, mercury, silver, and zinc can form insoluble sul-
fides under the reduced conditions commonly found in
wetlands (24). Decaying organic matter releases fulvic
and humic acids that can form complexes with metal
ions. In addition, decreased pH can promote the disso-
lution of metals, thereby making them available for
bonding to inorganic and organic molecules (25).
Filtration
Filtration occurs in most wetlands simply because vege-
tation acts like a sieve to remove pollutants and sedi-
ments from the water column. Dense vegetation can be
very effective at removing floatables (including oil and
grease) and litter from stormwater. Filtration can also
take place in the soil matrix when infiltration occurs.
Brown (9) and Wotzka and Oberts (10) also noted that
increased density of vegetation slows the velocity and
wave action, thereby allowing increased settling of sus-
pended material.
Biochemical Interactions
Vegetative systems possess a variety of biochemical
interaction processes that can remove nutrients and
other material from the water column. In general, these
processes are:
High plant productivity and associated nutrient uptake
Decomposition of organic matter
Adsorption
Bacterially aerobic or anaerobic mediated processes
Through interactions with the soil, water, and air, plants
can increase the assimilation of pollutants within a wet-
land system. Plants provide surfaces for bacterial
growth and adsorption, filtration, nutrient assimilation,
and the uptake of heavy metals (26).
Volatilization and Aerosol Formation
Volatilization (or evaporation) can remove volatile
pollutants from wetlands. Air and water temperature,
wind speed, subsurface agitation, and surface films can
affect the rate of volatilization. Surface films may act as
a barrier for the volatilization of some substances.
Alternatively, evaporation may be a key mechanism for
exporting substances such as chlorinated hydrocar-
bons or oils, which are often found in the surface films
of water bodies receiving urban stormwater runoff (26).
Aerosol formation may play only a minor role in remov-
ing pollutants in wetlands and occurs only during strong
winds (26).
Infiltration
For wetlands with underlying permeable soils, infiltration
can remove pollutants. Stormwater percolates through
the soil, eventually reaching ground water. Passage
through the soil matrix can provide physical, chemical,
and biological treatment depending on the matrix thick-
ness, particle size, degree of saturation, and organic
content. Infiltration is also dependent on the ground-
water level at a site. In some instances, seasonal
fluctuations in ground-water levels may cause some
wetlands to discharge ground water during part of the
year and recharge to ground water during other times of
the year. The potential of pollutants to migrate to ground
water depends highly on the type of pollutant, the soil
type and properties, the hydrology, and the charac-
teristics of the aquifer. Contamination of unconfined
aquifers by stormwater is likely to be more significant
from upland infiltration than from recharge through wet-
lands because of the high filtering action of typical wet-
land soils (27).
Wetland Stormwater Pollutant Removal
Efficiencies
Only a limited number of studies have investigated the
effectiveness of wetlands to treat stormwater runoff (Fig-
ure 1), and those have primarily focused on a few geo-
graphical locations (e.g., Florida, Minnesota, and
California). The studies that this paper summarizes rep-
resent a wide diversity of wetland types, ranging from
southern cypress swamplands and northern peatlands
to brackish marshlands and high-elevation meadow-
lands. This section presents a discussion of wetland
stormwater pollutant removal efficiencies found in the
literature.
Table 2 summarizes reported removal efficiencies for
total suspended solids (TSS) and selected nutrients and
metals. The broad ranges of pollutant removal efficien-
cies were not surprising because wetlands vary in their
hydraulic conditions, climate, and vegetation, and be-
cause the studies employed various monitoring and
reporting procedures. Figure 2 presents histograms of
pollutant removal efficiencies reported for TSS, total
phosphorus (TP), ammonia (NH3), and lead (Pb).
312
-------
w
Table 2. Average Removal Efficiencies for Total Suspended Solids and Nutrients in Wetlands Reported in the Literature
Pollutant Removal Efficiency (Percent)3 Lead
Zinc
Copper
Chromium
Study
Martin and
Smoot (4)
Harper
et al. (5)
Reddy
et al. (6)
Blackburn
et al. (7)
Esry and
Cairns (8)
Brown (9)
Wotzka
and Oberts
(10)
Hickock et
al. (11)
Barten (12)
Meiorin (13)
Morris et
al. (14)
System Name
Orange County
Treatment
System
Hidden Lake
Lake Apopka
Palm Beach
PGA Treatment
System
Jackson Lake
Fish Lake
Lake Elmo
Lake Riley
Spring Lake
McCarrons
Wetland
Treatment
System
Wayzata
Wetland
Clear Lake
DUST Marsh
Basin A
Basin B
Basin C
System
Angora Creek
Tallac Creek
System Type
Detention
pond*
Wetland*
Entire system
Wetland
Reservoirs
Flooded fields
System
System
Wetland/pond
Wetland
Wetland
Wetland
Detention
pond*
Wetland*
System
Wetland
Wetland
Wetland*
Wetland*
Wetland
Wetland*
Wetland
Wetland
TSS
65
66
89
83
50
96
95
88
-20
-300
91
87
94
94
76
63
40
51
76
54
36
NH3
60
54
61
62
57.5
51.9
17
37
0
50
25
-86
-44
55
-8
-5
18
16
20
33
N03
-17
40
9
80
68.1
64.2
33
70
60
22
63
32
2
12
29
50
35
Dis.
TP P COD BOD Total Dissolved Total Dissolved Total Dissolved Total Dissolved
33 76 7 39 29 15 -17
17 -30 18 73 54 56 75
43 21 17 83 70 70 65
7 81 55 56 41 57 40 29 73 75
60.9 75.1
7.3 16.7
62 35
90 78
37 28
27 25
-43 -30
-7 -10
78 57 90 85
36 25 79 68
78 53 93 90
78 94 82 80
54 40
46 -25 30 42 -20 55
-4 -46 27 24 -60 47
36 -18 83 -29 17 13
58 -57 88 42 -19 66
5
-120
-------
Table 2. Average Removal Efficiencies for Total Suspended Solids and Nutrients in Wetlands Reported in the Literature (continued)
Pollutant Removal Efficiency (Percent)3 Lead Zinc
Copper
Chromium
Study
Scherger
and Davis
(15)
ABAC (16)
Jolly (17)
Oberts et
al. (18)
w
Reinelt and
Horner (19,
20)
Rushton
and Dye
(21)
Hey and
Barrett (22)
System Name
Pittsfield-Ann
Arbor Swift Run
Palo Alto Marsh
Long Lake
Wetland-
Pond
Treatment
System
Tanners Lake
McKnight Lake
Lake Ridge
Carver Ravine
B3I
PC12
Tampa Office
Pond
Des Plaines
River Wetland
EWA3
EWA4
EWA5
EWA6
System Type
Detention
pond*
Wetland
Wetland
Entire system
Detention
pond*
Detention
ponds*
Wetland
Wetland-pond
system
Wetland
Wetland
Wetland
Wetland
Wetland
Wetland
Wetland
Median pollutant efficiency for wetland systems
(without *)
TSS
39
76
87
95
63
85
85
20
14
56
64
72
76
89
98
76
NH3 NO3
1
11
17
9
4
20
70
42
70
95
33 46
Dis.
TP P COD BOD Total Dissolved Total Dissolved Total Dissolved Total Dissolved
23 61
49 83
-6 54
92
7 -14 59
34 1 2 63
37 8 52
1 1 6
-2
-2
55 34
59
55
69
97
46 23 55 45 83 63 42 61 40 29 70 75
aNegative removal efficiencies indicate net export in pollutant loads.
COD = chemical oxygen demand
BOD = biochemical oxygen demand
-------
(a) Measured TSS Removal by Indicated Wetland
(c) Measured Nh|3 Removal by Indicated Wetland
_ ro
ro §rf
o 2
I O 0
(b) Measured TP Removal by Indicated Wetland
(d) Measured Pb Removal by Indicated Wetland
o E
5l
"- ro
£3>
100
80
60
40
20
20
40
60
80
inn
4
f] nnnnnn
J nil
P
n
-
J2 ~
Q- C
°1
ro 15
>
-------
Table 3. Wetland Geographic and Hydraulic Characteristics
Study
Martin and
Smoot (4)
Harper
et al. (5)
Reddy et
al. (6)
Blackburne
et al. (7)
Esry and
Cairns (8)
0)
Brown (9)
System
Name
Orange
County
Treatment
System
Hidden Lake
Lake Apopka
Palm Beach
PGA
Treatment
System
Jackson
Lake
Fish Lake
Lake Elmo
Lake Riley
Watershed
Land Use
Residential
Highway
Forest
Residential
Agriculture
Residential
Golf course
Urban
Residential
Commercial
Agricultural
Open
Residential
Commercial
Agricultural
Open
Residential
%
Land
Use
33
27
40
NA
100
NA
NA
NA
NA
30
5
12
53
12
1
34
53
13
System
Type
Detention
pond
Wetland
System
Wetland
Reservoirs
Flooded
fields
Wetland
Wetland
Detention
pond
Wetland
Wetland
Wetland
Wetland
Constructed/
Natural
Constructed
Constructed
Constructed
Natural
Constructed
Constructed
Constructed
Constructed
+ natural
Constructed
Constructed
Natural
Natural
Natural
Wetland
Size
(acres)
0.2
0.78
0.98
2.5
0.9
0.9
89
296
20
9
16
225
77
Watershed
Size
(acres)
41.6
NA
55.2
NA
NA
2,350
2,350
2,230
2,230
700
2,060
2,475
Wetland/
Watershed
Ratio
0.5%
1.9%
2.4%
4.5%
NA
NA
3.8%
12.6%
0.9%
0.4%
2.3%
1 0.9%
3.1%
Average
Flows
(ft3/sec)
2.5
NA
0.22
0.56
0.23
NA
NA
NA
NA
0.001-
0.01
0.001-
0.65
0.004-
1.35
Basin
Volume
(acre-ft)
1 .2-1 .9
0.5-2.8
NA
2.6
0.6
NA
NA
150
13.5
64
900
231
Detention
Time (hr)
7.5
8
NA
9.4 days
4.8 days
NA
NA
NA
NA
NA
NA
NA
Depth
(ft)
8-11
0-5
NA
3.3
0.7
NA
NA
7.5
1.5
4
4
3
Inlet
Condition
Discrete
Discrete
Diffuse
Diffuse
Diffuse
Diffuse
Diffuse
Diffuse
Diffuse
Discrete
Discrete
Discrete
Comments
a
b
c
c
d
c
e
f
g
Commercial 2
Agricultural 30
Open 55
Spring Lake Residential 5
Commercial 1
Agricultural 57
Open 37
Wetland
Constructed 64
5,570
1.1%
0.008-4 256
NA
Discrete
-------
Table 3. Wetland Geographic and Hydraulic Characteristics (continued)
w
Study
Wotzka
and
Oberts
(10)
Hickok et
al. (11)
Barten
(12)
Meiorin
(13)
Morris et
al. (14)
Scherger
and Davis
(15)
ABAC (16)
Jolly (17)
Oberts et
al. (18)
System
Name
McCarrons
Wetland
Treatment
System
Wayzata
Wetland
Clear Lake
DUST Marsh
Angora Creek
Tallac Creek
Pittsfield-Ann
Arbor Swift
Run
Palo Alto
Marsh
Long lake
Wetland-Pond
Treatment
System
Tanners
Lake
McKnight
Lake
Lake Ridge
Watershed
Land Use
Urban
Residential
Commercial
Urban
Urban
Agricultural
Residential
Forest
NA
Residential
Commercial
Agriculture
Open
Residential
Commercial
Open
Agriculture
Residential
Residential
Residential
Land
Use
NA
NA
NA
93
7
NA
NA
NA
45
19
13
23
62
12
26
100
NA
NA
NA
System
Type
Detention
pond
Wetland
System
Wetland
Wetland
Wetland:
A
B
C
System
Wetland
Wetland
Detention
pond
Wetland
Wetland
Wetland/
pond
Pond
Pond
Wetland
Constructed/
Natural
Constructed
Constructed
Constructed
Natural
Constructed
Constructed
Constructed
Constructed
Constructed
Natural
Natural
Constructed
Natural
Natural
Constructed
Constructed
Constructed
Constructed
Wetland
Size
(acres)
29.7
6.2
35.9
7.6
52.9
5
6
21
32
NA
NA
25.3
25.5
613
1.5
0.07
5.53
0.94
Watershed
Size
(acres)
600
600
65.1
1,070
2,960
2,960
2,816
2,781
4,872
1,207
17,600
18
1,134
5,217
531
Wetland/
Watershed
Ratio
5.0%
1.0%
6.0%
11.7%
4.9%
0.7%
1.1%
NA
NA
0.5%
2.1%
3.5%
8.3%
Neg.
0.1%
0.2%
Average
Flows
(ft3/sec)
0.05-0.2
0.08
1.5
10-250
8.46
8.68
0-2,916
0-166
1 50-320
0.01
NA
NA
NA
Basin
Volume
(acre-ft)
2.3-9.7
NA
10
150
NA
NA
21-176
15-60
400-750
1.5
0.1
13.2
2.0
Detention Depth Inlet
Time (hr) (ft) Condition Comments
24 days 2.5 Diffuse h
NA NA Discrete i
3-5 days 0.5 Diffuse
4-40 days 4.7 Diffuse j
NA NA Diffuse k
NA NA Diffuse
4-105 0-6 Discrete f
1 2-82 0-3 Discrete
30 1-6 Discrete I
m
NA 0.5-8 Diffuse n
NA 3.0 Discrete o
NA 4.9 Discrete
NA 4.8 Discrete
-------
Table 3. Wetland Geographic and Hydraulic Characteristics (continued)
oo
Study
Reinelt
and
Horner
(19, 20)
Rushton
and Dye
(21)
Hey and
Barrett
(22)
System
Name
Carver
Ravine
B31
PC12
Tampa
Office Pond
Des Plaines
River Wetland
Demonstration
Project
Watershed
Land Use
Residential
Urban
Rural
Commercial
Agriculture
Urban
%
Land
Use
NA
NA
NA
100
80
20
System
Type
Wetland/
pond
Wetland
Wetland
Wetland
Wetland:
3
4
5
6
Constructed/
Natural
Constructed
Natural
Natural
Constructed
Constructed
Constructed
Constructed
Constructed
Wetland
Size
(acres)
0.37
4.9
3.7
0.35
5.6
5.6
4.5
8.3
Watershed
Size
(acres)
170
461.7
214.8
6.3
Wetland/
Watershed
Ratio
0.2%
1.1%
1.7%
5.6%
Average
Flows
(ft3/sec)
NA
1.5
0.7
NA
5
0.6
4
1
Basin
Volume
(acre-ft)
1.0
0.03-
0.43
0.05-0.60
0.32
NA
NA
NA
NA
Detention
Time (hr)
NA
3.3
2.0
NA
NA
NA
NA
NA
Depth
(ft)
2.0
NA
NA
0-1.5
1
1
1
1
Inlet
Condition
Discrete
Discrete
Discrete
Discrete
Discrete
Discrete
Discrete
Discrete
Comments
P
q
r
s
NA = Not available
System = summary information
a Short-circuiting was observed during several storms.
b The wetland is not a basin but similar to a grassy swale.
c Design configuration suggests little short-circuiting occurred.
d Generally sheet flow exists within the artificial wetland.
e The major influent to these natural wetlands is discrete channelized flow.
f The schematic suggests large areas of dead storage exist.
g Short-circuiting was not discussed by the author.
h Three discrete inlets help to minimize short-circuiting and dissipate surface water energy.
i Design configuration suggests minimal short-circuiting existed regardless of a single discrete inlet.
j Design configuration suggests little short-circuiting occurred due to long and narrow wetland basins.
k Flow occurs as channelized flow until the storm volume is large enough to force sheet flow through the meadowlands.
I Water level and volume are controlled by the tidal cycle.
m Channelized flow exists until the tide increases, causing the surrounding marsh to become inundated.
n Entire system consists of a sedimentation basin, grass filter strip, constructed wetland, and deep pond.
o Monitoring occurred during a dry period.
p Storm flows reduce detention times.
q Channelization reduced effective area in wetland.
r Overflow from adjacent wetlands occurred during extremely high water; leak and breach problems occurred during study.
s Water is pumped to the system from the river (drainage area of 210 square miles) for 20 hr/wk.
-------
mulates stormwater flows within the system and dis-
charges effluent slowly over days or weeks, depending
on the interval between storms. Thus, the water col-
lected at the discharge from the DUST marsh is prob-
ably a mixture of water that entered from the previous
storms.
The type of inlet structure and the flow patterns through
wetland areas also can significantly affect pollutant re-
moval efficiencies. Morris et al. (14) found that sheet flow
(as opposed to channelized flow) was the most critical
factor in the effectiveness of meadowland treatment. This
finding is consistent with the theory that shallow, vegetative
overland flow decreases velocities and increases sedi-
mentation. In addition, close contact with the soil matrix
was found to increase assimilation of nutrients and bacte-
ria. Brown (9) found that an undefined inflow (multiple input
locations) to the wetland, which results in better dispersion
of incoming load, was critical in the effectiveness of the
wetland. An undefined inflow reduced short-circuiting
and increased mixing and contact of the stormwater with
the soil and plant substrates.
The change in seasons has been considered another
important factor in the effectiveness of wetland treat-
ment of storm runoff. Typical factors of seasonality are
evapotranspiration rates and seasonal productivity and
decay of plant and animal life. Removal efficiencies in
wetlands located in areas with strong seasonal variation
may vary significantly between seasons. For example,
Meiorin (13) reported that high summer evapotranspira-
tion rates caused a 200- to 300-percent increase in the
total dissolved solids concentrations within the DUST
marsh. Furthermore, high productivity during warm pe-
riods can lead to decreases in nutrients and increases
in biochemical oxygen demand (BOD) and suspended
solids. Morris etal. (14) reported that flushing and leach-
ing effects of spring snowmelt caused an increase in
total Kjeldahl nitrogen and organic carbon in flows leav-
ing the Tahoe Basin meadowland. Harper et al. (5)
reported that detention times greater than 2 days
caused an increase in the export of orthophosphorus
from the Hidden Lake wetland.
Hickoketal. (11) described microbial activity as the most
important factor affecting phosphorus removal. Other
factors that probably cause variations in the reported
pollutant removal effectiveness include maturity of the
wetland, the buildup of nutrients and heavy metals in a
wetland system, particle-size distribution (which affects
the settling of suspended sediments), and maintenance
practices performed at a wetland.
Comparison of Factors Affecting
Reported Treatment Efficiencies
This study reviewed data on removal efficiencies for 26
different wetland systems. The study evaluated the fol-
lowing factors regarding their effects on wetlands pollut-
ant removal performance:
Constructed versus natural systems.
Vegetation types found in the wetland.
Land-use types draining to the wetland.
Area of the wetland system compared with the con-
tributing watershed.
Estimated average storm-flow quantities draining to
the wetland.
Inlet types.
These factors affected only a few meaningful direct
relationships. This was because of the limited amount
TSS
Pollutant
NH3 I TP
TPb
100 .
80 .
60 .
ro
>
1 40
0
o 20
0)
ro
0 n
s.
n -20
S
£ -40
0.
-60
N=
'
11
1
ii.|.mii.|.iii
P»-ft.^l
I' l|l
L^^| Jl
1
1
1
N-14
N
p
=7
y
'?%%&&
N=6
I
|
I__^_ im
> "^":
. . iTTs* sr-^%4':
i* ,V*
I 1
N~4
- '
' N=15 N=5
Natural
Constr.
Natural
Constr. Natural
Constr. Natural Constr
Figure 3. Box plot percentiles comparison of site average pollutant removals for natural and constructed wetland systems: TSS :
total suspended solids, NHs = ammonia, TP = total phosphorus, TPb = total lead, and N = number of wetland sites.
319
-------
of data available to determine these relationships as
well as the multiple factors that affect performance.
Without a large database, a meaningful multiple regres-
sion analysis was not possible.
Several trends, however, were noted. First, constructed
systems generally had a higher average removal perfor-
mance than natural systems, with less variability. Second,
larger wetlands compared with their tributary watershed
areas also showed the same trend: a higher average
removal performance, with less variability. Figure 3 pre-
sents TSS, TP, NH3, and TPb in a percentile box plot for
the constructed and natural systems. Note that, in all cases
100
90
80
"ro
o 70
cc 60
oj
D) 50
2
> 40
<
0) ^n
D) OU
B
| 2°
fc -in
D_ lu
0
-10
-20
A. All Wetlands
TSS
TP
for the pollutants summarized, constructed systems
showed a higher average and median performance level.
More significant, however, is the difference in variability
between the two types of wetlands. Constructed sites
were much less variable. This is not a surprising finding,
given that constructed systems have generally been
designed to handle expected incoming flows and to
minimize short-circuiting. They should generally show a
higher performance level with more consistency.
Investigators also looked at the area of the wetland system
compared with the size of its contributing watershed.
Regression of the wetland to watershed area ratio
TPb
T
N=13
N=6
N=13
I
'x-'^^fc' **:
N=6
N=5
N=3
10
B. Constructed Wetlands
100
90
on
ra
o 70
a: 60
-------
(DAR) to pollutant removal performance did not reveal
good direct relationships. Grouping sites according to a
greater than or less than 2 percent DAR, however, did
result in some general trends. Figure 4 presents perfor-
mance results for all wetland systems with reported tribu-
tary watershed areas. In general, the larger DAR wetlands
had higher performance levels, with less variability. This
analysis included all wetland sites, natural and con-
structed. To separate out the effects of natural versus
constructed systems, Figure 4 also presents a similar
analysis for constructed sites only. Generally, for
constructed sites the trends are the same, although the
differences in performance levels and variability in per-
formance are much less. The data indicate that carefully
constructed systems can probably mitigate the impor-
tance of DAR as a factor in determining performance.
Therefore, at this time we are not suggesting that 2
percent minimum DAR is a proper design criteria for
constructed wetlands.
The Jackson Lake wetland is an example of a wetland
with a small DAR that still achieved excellent perform-
ance (85 percent TSS removal). The DUST marsh and
the Lake Ridge wetlands also showed high performance
levels (76 and 85 percent TSS removals, respectively).
One factor that explains the DUST marsh performance
is that it is an "off-line" device: it only receives flow
volumes up to a certain flow rate, then bypasses high
flows. This type of design is particularly appropriate for
wetlands receiving stormwater from larger catchments
relative to wetland size.
To better measure the capacity of a wetland to treat runoff
from a given watershed would entail evaluating average
storm runoff volumes of wetland tributary areas with wet-
land storage volumes and/or contact surface areas. The
data from the studies, however, did not consistently in-
clude data on rainfall statistics, percent impervious for land
uses, specific percentages for land uses in a catchment,
flow volumes to the wetland, capacity of the wetland sys-
tem, and surface areas for contact with stormwater (includ-
ing soils and plants). Therefore, we were not able to
analyze the wetland systems using this approach. The
summary of this paper contains some recommendations
regarding reporting information for future studies, so that
such analyses can be completed.
Finally, no good studies or documentation exists regard-
ing maintenance activities in wetlands that are treating
stormwater. In addition, the need for maintenance and
level of maintenance are not well understood or docu-
mented. These activities could affect performance char-
acteristics of wetlands, particularly over the long term.
Assessment of the Reliability of Wetland
Data
There are various difficulties in comparing one wetland
study to another. Table 4 presents a list of the selected
literature, including information on the sampling charac-
teristics that each study employed. The table shows that
the studies identified generally lasted a year or less.
There was quite a variation in the number of samples
collected (from 3 to about 150), as well as in the sam-
pling methods used (i.e., grab sample or samples ver-
sus composite sample for an event). These factors all
contribute to the difficulty of comparing results from the
different studies. Another complication in comparing the
performance of wetlands involves the method of quan-
tifying their effectiveness.
Noted Impacts of Stormwater Runoff on
Wetland Biota
Many researchers have expressed concern over the
impact of the quantity and quality of stormwater runoff
on wetland biota, especially in natural wetlands (27,28).
The quantity of stormwater runoff determines the hydro-
logic characteristics of a wetland, including the average
and extreme water levels and duration and frequency of
flooding. Stormwater runoff also contains pollutants
that can adversely affect wetland biota if accumulated
in high concentrations. The hydrology of a wetland is
one of the most important factors in establishing and
maintaining specific types of wetlands and wetland
processes (29). Hydrology is a key factor in wetland
productivity, vegetation composition, nutrient imports,
salinity balance, organic accumulation, sedimentation
transport, and soil anaerobiosis.
Few of the reports reviewed indicated concern regarding
the effects of contaminants in urban stormwater on wet-
land systems. Many of the reports referenced studies
performed in wetlands receiving sewage effluents or
industrial discharges of some type. Urban runoff, espe-
cially from residential watersheds, frequently has much
lower concentrations of pollutants than do sewage efflu-
ents or industrial discharges.
Sediments typically constitute the most significant store
of toxic substances available to organisms in a wetland
(29). Plants can take up metals and toxic organic com-
pounds from the sediments, thus introducing them into
the food web (30-32). Both metals and organics tend to
be adsorbed to finely divided solids, depending on con-
ditions such as pH, oxidation-reduction potential, and
salinity (33). The way a metal is complexed determines
its availability to plants (33).
Water resides longer in wetlands compared with more
swiftly moving waters because of the flatness of wet-
lands and the filtering action of the vegetation. This
longer residence time allows suspended solids to drop
out and be retained (32, 33). Woodward-Clyde Consult-
ants (34) found that the greatest concentration of metals
in sediments occurred at the location nearest the storm-
water inlet and declined with distance from the inlet.
They found the sediment concentration and bioavailabil-
321
-------
Table 4. Sampling Characteristics From the Wetlands Reviewed
Time of Length
Study Location Study of Study
Martin and Smoot (4)
Harper et al. (5)
Reddy et al. (6)
Blackburn et al. (7)
Esry and Cairns (8)
Brown (9)
Wotzka and Oberts
(10)
Hickoket al. (11)
Barten (12)
Meiorin (13)
Morris et al. (14)
Scherger and Davis
(15)
ABAC (1 6)
Jolly (17)
Oberts et al. (18)
Reinelt and Horner
(19, 20)
Rushton and Dye (21)
Hey and Barrett (22)
Orange County, FL
FL
Orange County, FL
Palm Beach, FL
Tallahassee, FL
Twin Cities Metro
Area, MN
Roseville, MN
MN
Waseca, MN
Coyote Hills,
Fremont, CA
Tahoe Basin, CA
Ann Arbor, Ml
Palo Alto, CA
St. Agatha, ME
Ramsey-Washington
Metro Area, MN
King County, WA
Tampa, FL
Wadsworth, IL
1982-1984
1 984-1 985
1977-1979
1985
1985
1982
1984-1988
1974-1975
1982-1985
1984-1986
1977-1978
1979-1980
1979
1989
1987-1989
1988-1990
1989-1990
1990
2 years
1 year
2 years
1 year
NA
1 year
2 years
10 months
3 years
2 years
1 year
8 months
3 months
5 months
2 years
2 years
12 months
8 months
Type of
Sample
7 multigrab,
6 composite
Composite
Single grab
Single grab
NA
Composite
Composite
NA
Composite
Composite
Single grab
Composite
Composite
Composite
Composite
Composite
Composite
Discrete
Number of
Storms
Monitored
13
18
Approx. 150
36
1
5-7
25
NA
27
11
Approx. 75
7
8
11
7-22
13
3-8
Continuous
Method of
Computing
Efficiencies
ROL
ER
MC
MC
NA
SOL
ROL
SOL
ER
SOL
MC
SOL
ER
SOL
SOL
SOL
ER
SOL
ER = event mean concentration
MC = mean concentration
NA = not available
ROL = regression of event loads
SOL = sum of event loads
ity of copper, lead, and zinc to be at or near background
levels in the downstream marsh area.
Plants take more metals from the sediment than from
the water column. Phytoplankton, however, can remove
metals directly from the water, releasing them to the
sediments or to the water upon death (35). In general,
far greater amounts of metals remain in the sediment
than are taken up by plants (36-39). Some plants are
apparently able to exclude toxic metals selectively. Or-
ganic compounds undergo many of the same processes
in wetlands as metals, including adsorption to sediments
and plant uptake. In addition, they can be biodegraded.
The uptake of toxic materials by plants can introduce
these materials into the grazing and detrital food chains,
with potentially deleterious effect. Metals from sewage
effluents introduced to wetlands tend to accumulate in
the food chain (32). Finally, the relative responses of
plants and animals to toxic metals and organic com-
pounds indicate that these contaminants are more likely
to affect animals negatively.
Comparison of Wetland and Detention
Basin Performance
Detention facilities have traditionally been constructed
to control stormwater runoff quantities. These facilities
temporarily store stormwater runoff and later release the
water at a lower flow rate. Design of detention basins
and ponds can provide for water quality enhancement
by including a permanent pool of water and inlet and
outlet structures to maximize detention. Quiescent ve-
locities within the basins allow sediments to settle out of
the stormwater and undergo chemical and biological
removal processes. Detention basins usually do not
have vegetation within the permanent pool, but the
banks may be planted with grasses for erosion control.
322
-------
Detention basin/constructed wetland treatment systems
have been recommended for stormwater treatment (4,10,
40). Typically in these systems, stormwater runoff dis-
charges to the detention basin, which then releases the
water to the wetland for additional treatment. The deten-
tion basin can provide pretreatment for the wetland,
reducing the sediment and pollutant loads to the wetland.
In other instances, detention basins and constructed
wetlands are competing alternatives under considera-
tion for stormwater treatment. To make a decision, the
designer or planner requires knowledge of the relative
pollutant removal efficiencies, environmental impacts,
maintenance requirements, and costs of the two alterna-
tives.
To further illustrate how those systems compare, the
following discussion focuses on the results from a case
study of the McCarrons treatment facility system, which
compared the performance of wetlands with that of de-
tention basins through simultaneous monitoring of both
systems. Wotzka and Oberts (10) presented a paper
discussing the combined detention-wetland stormwater
treatment facility. The McCarrons treatment facility consisted
of a 30-acre detention basin with an average depth of 1.2 ft
and a 6.2-acre constructed wetland with an average depth
of 2.5 ft. The detention basin received stormwater and
then discharged to the wetland. The contributing water-
shed consisted of 600 acres of primarily urban land use.
The predominant vegetation in the wetland consisted of
cattails with other emergent plant species.
Overall, the system produced very good results. The de-
tention basin proved to be more effective than the wetland
in reducing several pollutants. For example, Table 5 lists
removal efficiencies for the detention basin and wetland.
Wotzka and Oberts (10) discussed some of the possible
explanations for the good results of the detention basin
and for its differences from the wetland. In general, they
believed that the treatment efficiencies were lower in the
wetland due to pretreatment by the detention basin.
They stated that the inflows into the detention basin
spread equally around the perimeter of the detention
basin, thus dissipating the entry velocities of the storm
runoff. Dissipation of inflow energy probably promoted
settling and minimized short-circuiting.
Table 5. Removal Efficiencies (%) for Detention Basin and
Wetland
Parameter
Detention Basin
Wetland
TSS
TP
TN
TPb
91
78
85
85
87
36
24
68
Wotzka and Oberts (10) suggested that the percentage
of phosphorus in the dissolved and particulate phases
affected the reduction potential. They found that more
than 80 percent of the phosphorus was in the particulate
form, resulting in high removal efficiencies due to set-
tling. Apparently, the wetland did not perform as well as
the detention basin because of the periodic release of
nutrients from decaying vegetation and the fact that
significant pretreatment had occurred. The authors also
suggested that the high removal of phosphorus was due
in part to the newly exposed soils on the bottom of the
detention basin. They explained that the newly exposed
soils probably had more adsorption capacity available
than the soils in the wetland further downstream. They
also suggested that saturated soil conditions could lead
to a reduction in phosphorus removal.
In conclusion, this study indicated that the detention
basin performed better than the wetland system. This
may be misleading, however, because the wetland re-
ceived pretreated waters from the detention basin. The
detention basin removed the fraction of pollutants that
were more readily settled and treated, leaving the wet-
land with the more difficult-to-treat, finer particulates and
dissolved pollutants.
Summary
Wetlands have a good capability for removing pollutants
from stormwater runoff. Several factors contribute to
and influence removal efficiencies, including sedimenta-
tion, adsorption, precipitation and dissolution, filtration,
biochemical interactions, volatilization and aerosol for-
mation, and infiltration. The reported removal efficien-
cies are, as expected, quite variable. For the wetlands
systems reviewed, removal efficiencies for TSS had a
median of 76 percent. TSS removal is a good indicator
of pollutant removal potential for heavy metals and
phosphorus, as well as other pollutants associated with
fine particulate matter. Constructed wetlands tended to
be more consistent than natural wetlands in their re-
moval of TSS and the other analyzed parameters. Wet-
lands have also shown the ability to remove dissolved
metals. Nutrient removal in wetlands is variable, depend-
ing on both wetlands characteristics and seasonal effects.
Because many dissimilarities exist between the wet-
lands studied, wetlands stormwater pollutant removal
efficiencies vary widely. Properly designed, constructed,
and maintained wetlands, however, can be effective
pollution control measures. Examining additional wet-
lands in a variety of geographical areas, as well as
long-term pollutant removal efficiencies, is definitely
necessary.
A significant issue, however, involves whether storm-
water control measures should include natural wetland
systems. In general, natural wetlands have been found
to be somewhat less predictable than constructed wet-
323
-------
lands in terms of pollutant removal efficiency. This dif-
ference may be due to the fact that constructed wet-
lands have generally been engineered specifically to
provide favorable flow capacity and routing patterns. As
a result, they tend to detain inflows for longer periods
and have less short-circuiting than many natural systems.
People often question the appropriateness of using a
natural, healthy wetland for such purposes. Their con-
cern is whether the modified flow regime and the
accumulation of pollutants will result in undesirable en-
vironmental effects. There are many situations where
natural wetlands have been receiving urban runoff for
years. Some of these wetlands reflect significant degra-
dation because of many factors, including urban runoff,
whereas others have been less affected. A general con-
sensus from the literature is to discourage the use of a
healthy natural wetland for stormwater pollution control.
In the case of rehabilitating a natural but degraded
wetland, modifications should ensure that the applied
runoff receives sufficient pretreatment. One pretreat-
ment technique would be to use pond areas to provide
an opportunity for suspended materials to settle out
before the flows enter the wetland. Other possible op-
tions include routing inflows to the wetlands through
upstream grass swales, oil/water separators, heavily
vegetated areas (e.g., thick, shallow cattail areas), and
overland flow areas.
These techniques would not only act on solids but also
on floatables such as oil and water. Although little evi-
dence exists of problems in wetlands that have been
receiving stormwater runoff, the available data are quite
limited, and developing additional information on im-
pacts is critical. Additional studies on the impacts to
biota should be undertaken.
In addition, the maintenance needs of wetland systems
that treat stormwater merit further study. Such mainte-
nance activities could include sediment removal and
plant harvesting. Further studies should address the
need for and the frequency and appropriateness of
maintenance.
Gathering more information on wetland effectiveness
would benefit design development procedures for sizing
wetland treatment facilities. There is currently not
enough information in the existing literature to develop
design guidelines for constructed wetland treatment
systems. Additional studies are needed to broaden the
type of wetland systems reviewed, develop information
on long-term performance, and evaluate seasonal char-
acteristics of wetland performance.
A review of the data available on wetland stormwater
treatment effectiveness revealed that most studies did
not contain enough information on study and wetland
characteristics to analyze in detail the factors affecting
treatment performance among different wetlands. Table
Table 6. Suggested Reporting Information for Studies That
Assess the Ability of Wetlands To Treat Stormwater
Pollution
Wetland classification
Constructed or natural or combination wetland?
Vegetation species
Vegetation density (percentage open and vegetated)
Vegetation types (submerged, emergent, floating)
Wetland size
Wetland aspect (length-to-width ratio)
Side slopes
Soil type and depths
Watershed size (acres)
Watershed land use (percent residential, industrial, agricultural,
undeveloped, etc.)
Watershed percent impervious (percent impervious area)
Rainfall data/statistics
Average rainfall during study (in./year)
Average number of storms per year
Average storm intensity (in./hr)
Average storm duration (hr)
Average time between storms (days)
Low flow inflow rate(s)
Ground-water interaction?
Total flow from average storm
Wetland volume (maximum storage volume)
Average detention time for average storm (hours)
Water depth (minimum, maximum, average)
Inflow condition (discrete or diffuse inlets)
Pretreatment of inflow (settling forebays, overland flow, detention
basin, grassed swales, etc.)
Maintenance practices (including frequency)
Plant harvesting?
Flushing?
Sediment removal?
Chemical treatment?
Other maintenance?
Provide hydrology and water quality data for all storms
monitored
Type of samples (grab or composite)
Number of storms monitored
Method used to compute pollutant removal efficiencies
Dominant removal mechanisms (sedimentation, adsorption,
filtration, biochemical, etc.)
6 presents a summary of the information that would
hopefully provide a better means to compare wetland
324
-------
designs and treatment effectiveness from different wet-
land systems. This type of information could be useful
when comparing watershed to wetland characteristics
regarding performance.
This paper compared watershed to wetland size ratios.
A comparison of average storm volume to wetland vol-
ume would have made a better analysis of the effect of
wetland "sizes" on treatment abilities. The currently
available data, which predominantly present areas of
wetlands and watersheds, did not allow for this kind of
comparison. Percent impervious factors and therefore
runoff volumes could be very different in different
watersheds. Data such as percent imperviousness, land
use information, and rainfall statistics, along with wet-
land volume information, would have allowed us to com-
pare average runoff volumes, wetland volumes, and
resulting performance characteristics.
Acknowledgments
The author would like to thank EPA and specifically
Thomas Davenport, the project officer, for his support
and guidance on the original project. The author grate-
fully acknowledges the coauthors of the study report on
which this paper is based, Joan Kersnar, Eugene Dris-
coll, and Richard Horner.
References
1. Hammer, D.A., ed. 1989. Constructed wetlands for wastewater
treatment, municipal, industrial, and agricultural. Chelsea, Ml:
Lewis Publishers.
2. Bastian, R.K., RE. Shanaghan, and B.P. Thompson. 1989. Use
of wetlands for municipal wastewater treatment and disposal
regulatory issues and EPA policies. Constructed wetlands for
wastewater treatment, municipal, industrial, and agricultural.
Chelsea, Ml: Lewis Publishers.
3. Strecker, E.W., J.M. Kersnar, E.D. Driscoll, and R.R. Horner.
1992. The use of wetlands for controlling stormwater pollution.
Prepared by Woodward-Clyde Consultants for U.S. EPA. Wash-
ington, DC: Terrene Institute.
4. 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. Report 85-4310. U.S. Geological Sur-
vey Water Resources Investigation.
5. Harper, H.H., M.P. Wanielista, B.M. Fries, and D.M. Baker. 1986.
Stormwater treatment by natural systems. Report 84-026. Florida
Department of Environmental Regulation.
6. Reddy, K.R., P.O. Sacco, D.A. Graetz, K.L. Campbell, and L.R.
Sinclair. 1982. Water treatment by aquatic ecosystem: Nutrient
removal by reservoirs and flooded fields. Environ. Mgmt.
6(3):261-271.
7. Blackburn, R.D., PL. Pimental, and G.E. French. 1986. Treat-
ment of stormwater runoff using aquatic plants. West Palm
Beach, FL: Northern Palm Beach County Water Control District.
8. Esry, D.H., and D.J. Cairns. 1988. Effectiveness of the Lake
Jackson restoration project for treatment of urban runoff. Pre-
sented at the joint meeting of the American Society of Civil Engi-
neers (ASCE), Florida/South Florida Section.
9. Brown, R.G. 1985. Effects of wetlands on quality of runoff enter-
ing lakes in the Twin Cities metropolitan area, Minnesota. Inves-
tigation Report 85-4170. U.S. Geological Survey Water
Resources.
10. Wotzka, P., and G. Oberts. 1988. The water quality performance
of a detention basin-wetland treatment system in an urban area.
In: Nonpoint pollution: Policy, economy, management, and ap-
propriate technology. American Water Resources Association.
pp. 237-247.
11. Hickok, E.A., M.D. Hannaman, and N.C. Wenck. 1977. Urban
runoff treatment methods, Vol. 1. Nonstructural wetland treat-
ment. EPA/600/2-77/217. Wayzata, MN: Minnehaha Creek Wa-
tershed District.
12. Barten, J.M. 1987. Nutrient removal from urban stormwater by
wetland filtration. Lake Line 3(3):6-7, 10-11. North American Lake
Management Society.
13. Meiorin, E.G. 1986. Urban stormwater treatment at Coyote Hills
marsh. Association of Bay Area Governments.
14. Morris, FA., M.K. Morris, T.S. Michaud, and L.R.Williams. 1981.
Meadowland natural treatment processes in the Lake Tahoe Ba-
sin: Afield investigation. Final report. EPA/600/54-81/026.
15. Scherger, D.A., and J.A. Davis, Jr. 1982. Control of stormwater
runoff pollutant loads by a wetland and retention basin. In: Pro-
ceedings of the International Symposium on Urban Hydrology,
Hydraulics, and Sediment Control, Lexington, KY. pp. 109-123.
16. Association of Bay Area Governments (ABAC). 1979. Treatment
of stormwater runoff by a marsh/flood basin. Interim report. Oak-
land, CA.
17. Jolly, J.W. 1990. The efficiency of constructed wetlands in the
reduction of phosphorus and sediment discharges from agricul-
tural watersheds. M.S. thesis. Orono, ME: University of Maine.
18. Oberts, G.L., PJ. Wotzka, and J.A. Hartsoe. 1989. The water
quality performance of urban runoff treatment systems, Part 1.
Report to the Legislative Commission on Minnesota Resources.
St. Paul, MN: Metropolitan Council of the Twin Cities Area.
19. Reinelt, L.E., and R.R. Horner. 1990a. Characterization of the
hydrology and water quality of palustrine wetlands affected by
urban stormwater. Seattle, WA: King County Resource Planning
Section.
20. Reinelt, L.E., and R.R. Horner. 1990b. Urban stormwater impacts
on the hydrology and water quality of palustrine wetlands in the
Puget Sound region. In: Ransom, T.W., ed. Proceedings of the
Puget Sound Research Conference, Seattle, WA (January).
Olympia, WA: Puget Sound Water Quality Authority.
21. Rushton, B.T, and C.W Dye. 1990. Tampa office wet detention
stormwater treatment. Southwest Florida Water Management
District annual report for stormwater research program: Fiscal
year 1989-1990. Brooksville, FL: Southwest Florida Water Man-
agement District, pp. 39-59.
22. Hey, D.L., and K.R. Barrett. 1991. Hydrologic, water quality, and
meteorologic studies. In: The Des Plaines River Wetlands Dem-
onstration Project, Draft Final Report to the Illinois Department of
Energy and Natural Resources. Chicago, IL: Wetlands Research.
23. Oberts, G.L. 1982. Impact of wetlands on nonpoint source pollu-
tion. Presented at the International Symposium on Urban Hydrol-
ogy, Hydraulics, and Sediment Control, Lexington, KY.
24. Benforado, J. 1981. Ecological considerations in wetland treat-
ment of wastewater. In: Richardson, B., ed. Selected proceedings
of the Midwest Conference on Wetland Values and Management,
St. Paul, MN (June), pp. 307-323.
25. U.S. EPA. 1984. The removal of heavy metals by artificial wet-
lands. EPA/600/D-84/258. Ada, OK.
325
-------
26. U.S. EPA. 1981. The use of wetlands for water pollution control.
EPA/600/S2-82/086. Cincinnati, OH.
27. Stockdale, E.G. 1991. Freshwater wetlands, urban stormwater,
and nonpoint pollution control: A literature review and annotated
bibliography, 2nd ed. Olympia, WA: Washington Department of
Ecology.
28. Newton, R.B. 1989. The effects of stormwater surface runoff on
freshwater wetlands: A review of the literature and annotated
bibliography. Publication 90-2. Amherst, MA: The Environmental
Institute at the University of Massachusetts.
29. Mitsch, W.J., and J.G. Gosselink. 1986. Wetlands. New York, NY:
Van Nostrand Reinhold.
30. Kadlec, R.H.,and J.A. Kadlec. 1979. Wetlands and water quality.
In: Greeson, P.E., J.R. Clark, and J.E. Clark, eds. Wetland func-
tions and values: The state of our understanding. Minneapolis,
MN: American Water Resources Association, pp. 436-456.
31. Kreiger, R., D. Kreiger, K. Tomson, and W Warner. 1986. Lead
poisoning in swans in the lower Coeur d'Alene river valley. Paper
No. 651. Presented at the Annual Meeting of the Society of
Toxicology, New Orleans, LA (March).
32. Kadlec, R.H., and D.L. Tilton. 1979. The use of wetlands as a
tertiary wastewater treatment alternative. CRC Crit. Rev. Environ.
Control 9:185-212.
33. Horner, R.R. 1986. A review of wetland water quality functions.
In: Strickland, R., ed. Wetland functions, rehabilitation, and crea-
tion in the Pacific Northwest: The state of our understanding.
Olympia, WA:
33-50.
Washington State Department of Ecology, pp.
34. Woodward-Clyde Consultants. 1991. Sediment and storm runoff
concentrations of copper, zinc, and lead in the Crandall Creek-
DUST marsh system. Prepared for County of Alameda, Public
Works Agency, Hayward, CA.
35. Hart, J.T 1982. Uptake of trace metals by sediments and sus-
pended particulates: A review. Hydrobiologia 91:299-313.
36. Dubinski, B.J., R.L. Simpson, and R.E. Good. 1986. The retention
of heavy metals in sewage sludge applied to a freshwater tidal
wetland. Estuaries 9(2):102-111.
37. Banus, M.D., I. Valiela, and J.M. Teal. 1975. Lead, zinc, and
cadmium budgets in experimentally enriched salt marsh ecosys-
tems. Estuarine Coastal Mar. Sci. 3:421-430.
38. Teal, J.M., A. Giblin, and I. Valiela. 1982. The fate of pollutants in
American salt marshes. In: Gopal, B., R.E. Turner, R.G. Turner,
R.G. Wetzel, and D.F. Whigham, eds. Wetlands ecology and
management. Jaipur, India: National Institute of Ecology and
International Scientific Publications, pp. 357-366.
39. Horner, R.R. 1988. Long-term effects of urban stormwater on
wetlands. In: Roesner, L.A., B. Urbonas, and M.B. Sonnen, eds.
Design of urban runoff quality controls. New York, NY: American
Society of Civil Engineers, pp. 451-466.
40. Meyer, J.L. 1985. A detention basin/artificial wetland treatment
system to renovate stormwater runoff from urban, highway, and
industrial areas. Wetlands 5:135-146.
326
-------
Constructed Wetlands for Urban Runoff Water Quality Control
Richard Horner
University of Washington, Seattle, Washington
Abstract
Like all options for urban runoff water quality control,
constructed wetlands have their advantages, disadvan-
tages, and limitations. To realize their advantages, avoid
problems, and use them appropriately requires recogni-
tion and adherence to certain principles. A hallmark of
true constructed wetlands is their structural diversity,
which yields the substantial advantages of breadth in
treatment capabilities and potential for ancillary benefits
as well as the disadvantage of larger land requirements
for equivalent service than alternative measures. Pre-
requisites for success are functional objectives for the
project to achieve and a corresponding design concept
based on the structural characteristics of natural wet-
lands that are responsible for effective performance of
the identified functions. Critical implementation consid-
erations are proper siting, sizing, configuring of design
features, construction, and various aspects of opera-
tions. Careful site-specific hydrologic analysis must be
performed to ensure a sufficient water supply to sustain
a wetland. The basis for sizing is limited at present, but
application of climatological statistics and existing
knowledge of needed hydraulic residence times for
given treatment objectives provide some foundation.
Equal in importance to planning, siting, and sizing are
shaping, contouring, vegetating, and following up with
short- and long-term maintenance, for which specific
guidance is offered.
Background
Scope
Wetlands specifically constructed to capture pollutants
from stormwater runoff draining urban and agricultural
areas are gaining attention as versatile treatment op-
tions. Several recent major pieces of work have covered
constructed wetland treatment, including those by Ham-
mer (1), Strecker et al. (2), Olson (3), and Schueler (4).
This paper draws on these resources and is intended to
offer a concise summary of the current state of storm-
water treatment using constructed wetlands and the
methods for developing projects. The paper was derived
from a 1-day continuing education course on the subject
at the University of Washington, for which a course
manual is available (5). In particular, this paper empha-
sizes the fundamental concepts on which successful
application is based.
More than 150 wetlands have been constructed in the
United States to treat municipal and industrial, espe-
cially mining, wastewaters (2). No complete accounting
of stormwater constructed wetlands exists, but their
number is certainly fewer.
The two basic types of municipal and industrial systems
are both forms of attached growth biological reactors:
free water surface (FWS) and subsurface flow (SF), or
vegetated submerged bed (VSB) (6). The first type is
similar to natural wetlands, with a soil base, emergent
vegetation, and water exposed to air. The second type
has a soil base overlain by media, emergent vegetation,
and a water level below the media surface. The majority
of municipal and industrial applications, most of small
scale, are of this type. The advantages of a submerged
system in these applications are reduced odor, insect
problems, and land requirements because of the greater
surface area for biological growth offered by the media.
The FWS type is generally more appropriate for storm-
water applications, where usually no odor problem ex-
ists, flows vary widely, and often there is a desire to
integrate the treatment system with the landscape and
to provide ancillary benefits. This paper covers only the
FWS type of system.
Legal and Regulatory Considerations
From a legal and regulatory standpoint, "constructed
wetlands" are designed, built, and continually main-
tained for the purpose of waste treatment. In this status,
they are not regarded under the Clean Water Act as
"waters of the United States." Accordingly, no regula-
tions apply to water quality within, but the discharge is
regulated in the same way as any treatment system.
327
-------
This designation is in contrast to wetlands built for such
purposes as mitigation of wetland losses under Clean
Water Act Section 404 or to develop waterfowl habitat,
known as "created wetlands." These systems have the
same legal protections as natural wetlands, including
prohibition on using them for the conveyance or treat-
ment of waste. They usually have multiple functions,
with any water quality improvement benefit being only
incidental; entering water must be managed to prevent
damage to any intended function. A constructed wetland
also differs in purpose and legal status from a wetland
"restoration," the purpose of which is to return a de-
graded system with reduced acreage or functional ability
to the condition preceding degradation. If the wetland is
not completely restored but one or more functions are
increased, it is termed an "enhanced wetland." Restored
and enhanced wetlands also have the same legal pro-
tections as natural wetlands.
A somewhat fuzzy issue with respect to constructed
wetlands is their regulatory status if the principal pur-
pose is waste treatment but ancillary benefits (e.g.,
wildlife habitat) are gained by design or incidentally. This
situation is subject to interpretation by state and federal
agencies. Such benefits are often among the objectives
of project developers and are certainly possible to attain
along with stormwater treatment in many circumstances;
this paper provides advice on pursuing these objectives
in a judicious way.
Constructed Wetlands in Relation to
Alternative Methods
Alternatives to constructed wetlands for general-
purpose stormwater treatment include wet ponds, ex-
tended-detention dry ponds, infiltration basins and other
devices that drain into ground water, filtration, and
"biofiltration" through terrestrial or hydrophytic plants in
swales or on broad surface areas. Constructed wetlands
have both advantages and disadvantages relative to
these other options. Principal advantages are:
More diversity in structure than any alternative, which
offers the potential for relatively effective control of
most types of pollutants.
Wider range of potential side benefits than any
alternative.
Relatively low maintenance costs.
Wider applicability and more reliable service than
infiltration.
Disadvantages of constructed wetlands include:
Larger land requirements for equivalent service than
wet ponds and other systems, especially if intended
to serve quantity as well as quality control purposes.
Relatively high construction costs.
Delayed efficiency until plants are well established.
Uncertainty in design, construction, and operating
criteria, a drawback also hampering competitive
methods.
Public concern about nuisances that can develop
with stormwater constructed wetlands without care in
siting, design, construction, and operation.
Functioning of Constructed Wetlands
Pollutant Removal Mechanisms
Numerous physical, chemical, and biological mecha-
nisms can potentially operate in constructed wetlands to
trap and transform entering pollutants. Understanding
these mechanisms is the basis for determining effective
treatment systems. That understanding can inform the
entire process, from conception of the project, through
preliminary planning and all phases of implementation,
and, finally, to the long-term operation of the system.
Table 1 summarizes the various mechanisms, the pol-
lutants that they affect, and features that can promote
their operation.
Some beneficial features are controllable through
choices made during the project development process,
while others are largely outside of the designer's influ-
ence, especially in a stormwater application. As can be
seen in Table 1, some features are helpful in achieving
multiple treatment objectives, but others are more spe-
cialized. Features that are largely under the project
developer's control and help achieve any objective are
1) increasing hydraulic residence time (HRT); 2) provid-
ing an environment that creates flow at a low level of
turbulence; 3) propagating fine, dense, herbaceous
plants; and 4) establishing the wetland on a medium-fine
textured soil, or amending soils to attain that condition.
Somewhat more specialized features, still mostly con-
trollable, include 1) circumneutral Ph, which advances
microbially mediated processes such as decomposition
and nitrification-denitrification and avoids the mobility of
certain pollutants at extreme pH; 2) a relatively low level
of toxic substances in the site soils and entering flow,
also needed for microbes; and (3) high soil organic
content, which advances adsorption and decomposition
and can be attained by site selection or soil amendment.
Even more specialized are measures that can aid phos-
phorus capture, one of the most difficult treatment ob-
jectives to achieve. High soil exchangeable aluminum
and iron contents have been found to enhance phos-
phorus reduction (7) but would require special soil
amendments where naturally lacking, which thus far is
an undemonstrated option in a full-scale wetland sys-
tem. Addition of precipitating agents is an active treat-
ment measure that is difficult to apply in passive
328
-------
Table 1.
Constructed Wetland Pollutant Removal
Mechanisms
Mechanism
Pollutants Affected
Promoted By
Physical
Sedimentation
Filtration
Soil
incorporation
Chemical
Precipitation
Adsorption
Solids, BOD,
pathogens; participate
COD, P, N, metals,
synthetic organics
Solids, BOD,
pathogens; particulate
COD, P, N, metals,
synthetic organics
All
Dissolved P, metals
Dissolved P, metals,
synthetic organics
Ion exchange Dissolved metals
Oxidation
Photolysis
Volatilization
Biological
Microbial
decomposition
COD, petroleum
hydrocarbons, synthetic
organics
COD, petroleum
hydrocarbons, synthetic
organics
Volatile petroleum
hydrocarbons and
synthetic organics
BOD, COD, petroleum
hydrocarbons, synthetic
organics
Plant uptake P, N, metals
Natural die-off Pathogens
Nitrification NH3-N
Denitrification NO3 + NO2-N
Low turbulence
Fine, dense
herbaceous plants
Medium-fine textured
soil
High alkalinity
High soil Al, Fe (P);
high soil organics
(met.); circumneutral
PH
High soil cation
exchange capacity
Aerobic conditions
High light
High temperature and
air movement
High plant surface
area and soil organics
High plant activity and
metabolism and
surface area
Plant excretions
Dissolved oxygen
>2 mg/L, low toxics
temperature >5-7°C
circumneutral pH
Anaerobic, low toxics,
temperature >15°C
Al = aluminum, BOD = biochemical oxygen demand, COD = chemical
oxygen demand, Fe = iron, N = nitrogen, NH3 = ammonia, NO2 =
nitrite, NO3 = nitrate, P = phosphorus.
stormwater treatment systems subject to unpredictable
and variable flow conditions.
Also outside the control of the designer and operator in
a stormwater wetland is exploitation of the nitrification-
denitrification processes to achieve nitrogen removal
ultimately through evolution of nitrogen gas to the at-
mosphere. Full operation of the several steps in the
bacterially driven processes requires alternating aerobic
and anaerobic conditions at favorable temperatures, the
first condition to permit oxidation to nitrate and the sec-
ond to allow nitrate reduction to free N2 gas. While these
processes can be brought under some control in munici-
pal and industrial treatment applications through timing
of flow introduction, that degree of management is usu-
ally not possible in stormwater cases.
Expected Performance of Constructed Wetlands
Strecker et al. (2) conducted a full literature review of
the use of both natural and constructed wetlands for
controlling stormwater pollution. This review considered
more than 140 papers and reports and assembled de-
tailed information on 18 locations throughout the United
States. Median pollutant removals in constructed wet-
lands were 80.5 percent for total suspended solids
(TSS), 44.5 percent for NH3-N, 58.0 percent for total
phosphorus (TP), 83.0 percent for lead (Pb), and 42.0
percent for zinc (Zn). Coefficients of variation (standard
deviation/mean) for these contaminants ranged from
27.7 to 56.1 percent, pointing out that both substantially
higher and lower performances than median levels were
reported. Pollutant reductions in constructed wetlands
were overall higher than in natural wetlands, which was
attributed to the specific design features and more in-
tensive management of the constructed systems.
Schueler (4) estimated the performance potential of
wetlands designed as he recommended based on the
overall literature (Table 2). He considered these efficien-
cies to be provisional pending monitoring of the new
systems.
Table 2. Projected Long-Term Pollutant Removal Rates for
Wetlands Constructed as Recommended by
Schueler (4)
Pollutant
Removal Rate (percent)3
TSS
TP
Total nitrogen (TN)
BOC, COD, total organic carbon
Pb
Zn
FC
75
45b
25C
15
75
50
Two orders of magnitude
a Lower by an unknown amount for pocket wetlands (see below for
description of wetland types).
b 65 percent in pond/marsh system.
c 40 percent in pond/marsh system.
The Constructed Wetland Design and
Implementation Process
Developing a constructed wetland treatment system
should proceed carefully through a number of steps, as
follows:
1. Planning the project.
329
-------
2. Selecting the site.
3. Sizing the facility.
4. Configuring the facility, and incorporating design fea-
tures that promote pollution control.
5. Designing for ancillary benefits.
6. Selecting vegetation and developing a planting plan.
7. Constructing the facility and establishing vegetation.
8. Developing and implementing an operation and
maintenance plan.
The remainder of this paper explains these steps.
Project development for a constructed wetland must be
a team effort, with a number of skills and specialties
represented, including:
Hydrology
Water quality
Soils
Botany
Wildlife ecology
Landscape architecture
Design engineering
Construction engineering
Stormwater facility maintenance
It bears emphasizing that a high level of hydrologic
expertise should be employed to ensure that the most
essential needwater supplyis met.
Planning and Site Selection
Preliminary Planning Considerations
Constructed wetland projects should be planned sys-
tematically and on a watershed scale as much as pos-
sible. This comprehensive analysis should start with
consideration of management and source control prac-
tices that can prevent pollutant release. Another general
consideration that should receive attention is the overall
place of constructed wetlands and how they can best be
used in conjunction with other treatment practices.
If the constructed wetland option is pursued, project
objectives should be stated in functional terms, for
example:
The type of protection to be provided to the receiving
water, pollutants to be controlled, and levels of control
to be achieved (if possible).
Benefits to be provided in the areas of, for example,
open space, aesthetics, and recreation.
Animals and life stages for which habitat is to be
provided.
The potential for constructed wetlands to play a key role
in stormwater management has developed from the
understanding of natural wetland functioning gained
during the past 20 years. Natural wetlands serve their
recognized functions, which include providing flood flow
control, water quality improvement, and ecological
benefits, as a consequence of their structure and the
interactions among their component parts. Mimicking
these functions in an engineered system can best be
done with reference to natural models. Therefore, using
nearby natural wetlands as reference models for the
configuration and planting of the wetland to be designed
is strongly recommended. The reference system(s)
should be characterized through formal observations
and measurements of its hydrology, water quality, soils,
vegetation, and, if appropriate, animal habitat and spe-
cies. It is not necessary to mimic the reference plant
community entirely, but studying it provides an idea of
how the constructed system is likely to evolve.
With the natural model(s) in mind, a design concept can
be developed. Schueler (4) proposed four basic storm-
water wetland designs:
Shallow marsh: A system with a relatively large land
requirement that generally is used in larger drainage
basins.
Pond/Marsh: A two (or more) cell arrangement with
a land requirement that is reduced by a relatively
large deep pool.
Extended-detention wetland: A more highly fluctuat-
ing hydrologic system in which the land requirement
is reduced by adding high marsh to the shallow
marsh zone.
Pocket wetland: A design for smaller drainage basins
(0.4 to 4 hectares) that may provide insufficient base-
flow for permanent pool maintenance and cause
greater water level fluctuations.
Figure 1 illustrates the pond/marsh type design. For
diagrams of the other designs, see Schueler (4). Table
3 summarizes some of the principal selection criteria for
the respective wetland types.
To complete preliminary planning, the design process
and its aftermath should be organized. The following list
of general principles for project design and implementa-
tion, derived from the various comprehensive refer-
ences cited earlier, provides guidance for these steps:
Design and implement with designated objectives
constantly and clearly in mind.
Design more for function than for form. Many forms can
probably meet the objectives, and the form to which
the system evolves may not be the planned one.
330
-------
Aquatic Bench
High Marsh
Maximum Safety
Storm Limit
Figure 1. Two-cell pond/marsh design concept (4).
Table 3. Design Concept Selection Criteria (adapted from
Schueler [4])
Attribute
Extended-
Shallow Pond/ Detention Pocket
Marsh Marsh Wetland Wetland
Minimum 0.02
wetland-to-
watershed area
ratio
Minimum 10
watershed area
(hectares)
Dry weather Yes
baseflow
Relative High
potential for
ecological
benefits
0.01 0.01 0.01
10 4 0.4
Yes Not Not
necessarily necessarily
High Moderate Low to
moderate
Design relative to the natural reference system(s),
and do not over-engineer.
Design with the landscape, not against it (e.g., take
advantage of natural topography, drainage patterns).
Design the wetland as an ecotone. Incorporate as
much "edge" as possible, and design in conjunction
with a buffer and the surrounding land and aquatic
systems.
Design in structural complexity for beneficial distribu-
tion of water (e.g., its contact with vegetation and
soils) and for biological advantages, as appropriate
to objectives.
Design to protect the wetland from potential high
flows and sediment loads.
Design to avoid secondary environmental and com-
munity impacts.
Plan on sufficient time for the system to develop
before it must satisfy objectives. Attempts to short-
circuit ecological processes by overmanagement
usually fail.
Design for self-sustainability and to minimize mainte-
nance.
Constructed Wetland Site Selection
Prospective constructed wetland sites should be evalu-
ated carefully and a selection made after analyzing a
number of conditions. Brodie (8) presented a general-
ized site screening procedure, which is reproduced in
Figure 2. Table 4 summarizes the major considerations
that should enter into this analysis. Application of these
recommendations implies a significant data-gathering
effort, which is essential at this sensitive stage in project
development.
The need for a sufficient water supply to sustain a
wetland is an especially important consideration; ne-
glect of this consideration has led to constructed and
created wetlands that are not viable. Thus, a water
balance should be carefully established using the fol-
lowing formula to ensure that water availability and in-
puts at least balance outputs at all periods throughout
the year:
I + P + D + SXD + E + R
where
I = surface inflow
P = precipitation
D = ground-water discharge
331
-------
WAtTEWATER MANAGEMENT
Objectives
NON-TREATMENT
ALTERNATIVES
Inteaslble
TREATMENT
1
t
NON- WETLAND
TREATMENT
t
Further studies
CONSTRUCTED WETLAND
ni
PREUMINARY OFFICE STUDY
Land use/Ownership
Topography
Geology/Soils
Hydrology
Regulatory
Water quality
Infeaslble
PRELIMINARY DESIGN
AIR PHOTO INTERPRETATION
Hydrology
Access
Lind availability
Cultural resources
Geology/Soils
Land UM hlitory
Infeaslble
FIELD SURVEY
Surtec* nwtoriil*
Acc*t>
Lind/wittr ut*
Q*n*ril «IM ch»r»et»rl»tkf
Local Inqulrl«»/S«rvle»t
Hydrology
Inf.tilbU
REFINED DESIGN
LIMITED DATA COLLECTION
Soils classification
Augarlng
Percolation testa
Test pits
Water quality
Social condition*
Legal accaas
Inaufflclant Data
DETAILED DATA COLLECTION
Soils mapping
Geologic mapping
Bedrock profiles
Hydrologlc studies
Gaotachnlcal testing
Water quality
Cultural reaourcea
Wildlife surveys
1
DATA EVALUATION
EnvlroMMntal affacta
Ragulatory raqulramants
Public Involvamant
Insufficient Data
Infaaalbla
FINAL WETLAND DESKSN
PILOT TESTING
DESIGN MODIFICATIONS
CONSTRUCTION
Figure 2. A generalized methodology for screening sites for constructed wetlands (8).
332
-------
Table 4. Considerations in Constructed Wetland Site
Selection
Category
Considerations
Land-use and Land availability
general factors Existing site use and value
Site problems (e.g., previous dumping, utility
lines)
Adjacent land use and value
Connection to wildlife corridors and potential for
adjacent areas to be biological donors
Public opinion
Accessibility for construction and maintenance
Ability to control public access according to
project objectives
Environmental Federal, state, and local laws and regulations
and regulatory Avoidance of archaeological and cultural
factors resources
Avoidance of critical wildlife habitat areas
Hydrology and Water supply reliability
water quality Low potential for disruptive flooding
factors Water supply of adequate quality to sustain biota
Low potential for the project to adversely affect
downstream water bodies and adjacent
properties and their water supplies
Need for lining to retain water or avoid
ground-water contamination
Geology Preferably flat or gently sloped topography
factors Adequate soil development
Sufficient depth to bedrock
Soil characteristics consistent with pollution
control objectives
Suitability of site materials for use in construction
S = wetland storage at beginning of calculation
period
O = surface outflow
E = evapotranspiration
R = ground-water recharge
All units are expressed in terms of volume or water
depth over the wetland surface.
The water balance should be estimated during site se-
lection and checked after preliminary design. In areas
with pronounced seasonal drought (e.g., most of the
western United States), the calculation should definitely
be performed for this period. Ground-water terms are
difficult to establish with assurance, but they should at
least be estimated as closely as possible by a hydro-
geologist familiar with the location. As demonstrated by
the fact that natural wetlands often dry below the soil
surface, permanent standing water is not required for a
wetland to be viable. Research on natural wetlands in
Washington State has found that plant community rich-
ness declines substantially when drying extends longer
than 2 months, compared to wetlands with shorter dry
periods (9). Hence, the water balance should at least
demonstrate that drying will never extend longer than 2
months.
Brodie (8) and Mitsch (10) have discussed positioning
constructed wetlands in watersheds. Brodie (8) listed
advantages and disadvantages of locating wetlands in
upper reaches, on slopes, and in lowlands. No single
setting is clearly optimal; thus, location from this stand-
point depends on project objectives and the relative
importance of the advantages and disadvantages at the
specific site under consideration. Some possibilities for
locating constructed wetlands in the overall landscape
include:
Just off stream channels, for baseflow supply by
diversion.
In stream floodplains, separated from the low-flow
channel by a natural levee, with periodic water sup-
plied to the wetland when the levee is topped.
Several small wetlands in upper reaches of the
watershed.
One large wetland in lower reaches.
Several small wetlands in lower reaches.
Terracing into the landscape in steep terrain.
Constructing several small wetlands in the upper water-
shed provides some advantages relative to locating one
large wetland in the lower reaches, such as better sur-
vival of extreme events, closer proximity to pollutant
sources, and local flood protection. In contrast, the sin-
gle large lowland wetland can provide overall greater
flood reduction capability, if that is an objective. An
alternative is the multiple lowland wetland plan, under
which each can take a portion of high flows with less
vulnerability to any one.
Sizing Constructed Wetlands
Establishing Volume
Possible arrangements of a constructed wetland in re-
lation to runoff quantity and quality control requirements
are:
Place a runoff quantity control device "on line" and a
constructed wetland "off line" to treat all runoff up to
a certain volume.
Construct a wetland with a permanent pool ("dead
storage") zone for treatment and a "live storage" zone
and discharge control sized for peak runoff rate
control.
Construct a wetland only for treatment (for situations
where quantity control is not required).
The first arrangement takes advantage of the fact that
most of the pollutant mass loading over time is trans-
ported by runoff from the more frequent, smaller storms
and the "first flush" from the less frequent, larger storms.
This is the recommended arrangement where runoff quan-
tity control is required because 1) the relatively shallow
depths needed to maintain wetlands are somewhat
333
-------
inconsistent with the large storage volume needed for
quantity control and 2) large surges of water can dam-
age the wetland.
Basic sizing decisions involve the dead storage volume,
surface area, depth contouring, and live storage volume,
if runoff quantity control will be provided. There are three
fundamental ways to calculate the treatment volume of
a constructed wetland:
Compute the volume needed to provide the required
HRT for achieving a desired effluent concentration of
the limiting pollutant (the hardest to remove), given
a certain influent concentration, by using a mecha-
nistic equation.
Compute according to maximum allowable loading
rates of water or specific pollutants established em-
pirically from measurements on operating systems.
Compute on the basis of a hydrologic criterion.
The first two approaches are employed in municipal and
mining industry wastewater applications, where para-
meterized mechanistic equations or allowable loading
rates exist for BOD and nitrogen in sewage and iron and
manganese in mining effluents (6). Similar relationships
do not exist for stormwater and will be difficult to
develop, given the variability of flows and pollutant
concentrations.
Therefore, stormwater wetland sizing must be deter-
mined using some form of the third approach. One
version calls for choosing a volume sufficient to hold all
runoff from a set percentage of the annual storms (e.g.,
90 percent) or to hold a set depth of runoff generated by
the contributing catchment (e.g., the first 2.5 cm = 1 in.).
Schueler (4) presents several sizing rules of this type.
Equivalent to this version is an approach for using a
"water quality design storm" of a selected recurrence
frequency and duration. The Washington State Depart-
ment of Ecology (11) has taken this approach, selecting
the 6-month, 24-hour rainfall event, which in Seattle is
approximately equivalent to the first 3 cm of runoff, for
stormwater treatment design in the Puget Sound basin.
A third version of the hydrologic basis is the method
developed from wet pond performance data collected
during the Nationwide Urban Runoff Program by the
U.S. Environmental Protection Agency (EPA) (12). Us-
ing this method implicitly assumes that constructed wet-
lands will perform at least as well as wet ponds of
equivalent treatment volume, which seems to be a safe
assumption given the treatment advantages offered by
a more structurally complex, vegetated system. The
data exhibited an association between treatment effi-
ciency and the ratio of permanent pool volume to runoff
volume associated with the mean storm, termed the
"volume ratio." The mean storm is the average rainfall
quantity over all storms in a long-term record at a gaging
station. TSS loading reduction is typically around 75
percent at a volume ratio of 2.5, which is a common
design basis. Obtaining increasingly better performance
levels requires exponentially increasing basin size be-
cause the contaminants hardest to capture are those
still in suspension or solution.
With this means of sizing constructed wetlands, the task
almost entirely involves hydrologic analysis. This is an-
other point at which hydrologic expertise is important to
the design effort. Unless actual data are available from
gaging the catchment that will contribute to the con-
structed wetland, the hydrologic analysis must be per-
formed using a model. Modeling options include, in
order of preference, a well-calibrated continuous simu-
lation computer model, such as EPAs SWMM and
HSPF, an event-based model such as the Soil Conser-
vation Service's curve number method, and, where ade-
quate data exist, a locally derived empirical model of the
rational method type.
Once the hydrologic analysis is complete, the perma-
nent pool volume (VP) calculation can be made very
simply by using the equation:
VP = C * VR * AC
where
C = unit conversion factor
VR = runoff volume from hydrologic analysis
AC = contributing catchment area
Schueler (4) recommended a minimum VP of 1.6 cm/ha
of contributing catchment area, which will increase the
wetland size over that calculated by the equation in
small catchments.
This procedure is used for general runoff pollution con-
trol purposes. Knowledge is inadequate at present to
perform detailed sizing calculations for such specific
purposes as control of metals and nutrients. These spe-
cial objectives can be advanced in part by installing
appropriate design features (addressed later in this pa-
per). It is known that the maximum potential to remove
dissolved pollutants, which include certain nitrogen and
phosphorus forms and some metals, is reached with a
long HRT in the dead storage (2 to 3 weeks) (13, 14).
The average residence time can be checked as follows:
1) perform the hydrologic analysis to determine the rate
of flow to the wetland associated with the mean storm
(Q), and 2) calculate HRT = VP/Q. If HRT is less than 2
to 3 weeks and dissolved pollutant removal is an objec-
tive, increase VP to obtain HRT in that range.
If the wetland has live storage for peak runoff rate
control, the volume of that zone and the discharge ori-
fice size will also have to be calculated. These calcula-
334
-------
tions require hydrograph simulation and routing analysis
and are beyond the scope of this paper. They should be
performed by a qualified hydrologist.
Permanent Pool Surface Area and Depth
Contouring
A larger surface area for the same volume provides
better treatment by allowing more light penetration for
photosynthetic activity by plants and algae, more aera-
tion for aerobic chemical and biological processes, and
a shorter settling distance for particles. A straightforward
way of establishing the wetland surface area (AW) is to
start by selecting a trial mean depth (D) from the follow-
ing approximate ranges (after Schueler [4]):
Shallow marsh:
Pond/marsh:
Extended-
detention
wetland:
Pocket wetland:
0.30 to 0.45 m
0.60 to 0.85 m
0.25 to 0.30 m (permanent pool)
1.0 m (extended-detention zone)
0.15 to 0.40 m
Using the trial mean depth, calculate surface area by
AW = VP/D. Determine the wetland to contributing
catchment area ratio (AW/AC), and compare it with the
guidelines in Table 3.
Once satisfactory basic dimensions are determined, al-
locate depths to the different wetland zones according
to the design concept. Schueler (4) recommended the
following zones to obtain diversity in structure and treat-
ment capabilities:
Deep areas (30 to 180 cm deep, no emergent vege-
tation)forebay, micropools, deep water pools, and
channels.
Low marsh (15 to 30 cm below normal pool).
High marsh (0 to 15 cm below normal pool).
Irregularly inundated zone (above normal pool).
Schueler went on to supply approximate depth alloca-
tions for the various zones and design concepts, and the
reader is referred to his guidelines for these details. For
example, he recommended allocating 40 percent of the
surface area to the high marsh and 40 percent to the
low marsh in the shallow marsh design, with 5 percent
each given to the forebay, micropools, deep water, and
irregularly inundated zones.
Recommended Constructed Wetland Design
Features
Adequate size is a necessary but not sufficient condition
for good treatment performance. The theoretical HRT
provided by the volume will not be achieved in practice
if the layout permits water to traverse the wetland faster.
Many of the features presented in this section are rec-
ommended to reduce the tendency of flow to short-
circuit the wetland and fail to achieve an actual HRT as
long as the theoretical HRT. Given that natural wetlands
generally exhibit the recommended features, the
selected reference system(s) should be employed as a
model for designing these features. The recommenda-
tions are presented here in an abbreviated list format;
consult the comprehensive sources referenced earlier
for more detail.
Shaping the Wetland
Create a complex microtopography to lengthen the edge
and flow path by using high marsh peninsulas and
islands. Create at least two distinct cells by restricting
the flow to a narrow passageway using the following
features:
Make the wetland relatively wide at the inlet to facili-
tate distribution of the flow well.
Maximize the distance between the inlet and outlet.
The effective length to width ratio should be 5:1, prefer-
ably, and at least 3:1.
Slopes
The longitudinal slope (parallel to the flow path) should
be less than 1 percent.
The wetland should be carefully constructed to have no
lateral slope (perpendicular to the flow path) to avoid
concentration of the flow in preferred channels, which
reduces actual HRT and risks erosion.
Side slopes should be gradual (e.g., 5:1 to 12:1 horizon-
tal to vertical), as in natural wetlands. Nowhere should
the side slope be greater than 3:1.
Forebay
A forebay is a relatively deep zone placed where influent
water discharges. It traps coarse sediments, reduces
incoming velocity, and helps to distribute runoff evenly
over the marsh.
Install a forebay in shallow marsh and extended-deten-
tion wetlands. In the case of a pond/marsh system, the
pond serves this purpose. The restricted size of pocket
wetlands generally does not allow for a forebay. Make
the forebay 1.2 to 1.8 m deep. The forebay should be a
separate cell set aside by high marsh features.
Provide maintenance access for heavy equipment
(4.5 m wide and a maximum 5:1 slope) directly to the
forebay. The forebay bed should be hardened to prevent
disturbance during cleanout.
335
-------
Flow Channeling
Create sheet flow to the maximum extent possible.
Where flow must be channeled, use multiple, meander-
ing channels rather than a single straight one. Inter-
sperse open water areas with marsh, rather than
connecting along the flow path. Minimize velocity in
channels to prevent erosion and expand habitat
opportunities.
Outlet Design
Place a micropool 1.2 to 1.8 m deep at the outlet. Install
a reverse-sloped pipe 30 cm below the permanent pool
elevation. This outlet design has been found to avoid
clogging, to which constructed wetland outlets are prone
(4).
Install a drain capable of dewatering the wetland in 24
hours to allow for maintenance. Control the drain with a
lockable, adjustable gate valve. Place an upward-fac-
ing, inverted elbow on the end of the drain to extend
above the bottom sediments.
Soils
Medium-fine textures, such as loams and silt loams, are
optimal for establishing plants, capturing pollutants, re-
taining surface water, and permitting ground-water dis-
charge. Circumneutral pH (approximately 6 to 8) is best
for supporting microorganisms, insects, and other
aquatic animals.
A relatively high content of highly decomposed organics
("muck") is favorable for plant and microorganism
growth and the adsorption of metals and organic pollut-
ants. Muck soils are preferred to peats (less decom-
posed organics), which tend to produce somewhat
acidic conditions, to be low in plant nutrients, and to offer
relatively poor anchoring support to plants.
Vegetation becomes established more quickly and ef-
fectively in constructed wetlands when soils contain
seed banks or rhizomes of obligate and facultative wet-
land plants. Attempt to obtain any available soils that
offer these resources.
Soil characteristics recommended for specific pollution
control objectives are:
High cation exchange capacityfor control of metals.
High exchangeable aluminum and/or ironfor con-
trol of phosphorus.
Liner
An impermeable liner is required when infiltration is too
rapid to sustain permanent soil saturation, when there
is a substantial potential of ground water being contami-
nated by percolating stormwater, or both. Infiltration
losses are insignificant at most sites with Soil Conser-
vation Service Class B, C, and D soils. Also, sediment
deposition is likely to seal the bottoms of constructed
wetlands. Generally, therefore, a liner is likely to be
needed only in Class A soils.
Emergency Spillway
An emergency spillway is required when the wetland will
be used for runoff quantity control (and any other situ-
ation in which it would be possible for runoff to enter
from a larger storm than the largest storm the facility is
sized to handle).
Buffer
A buffer should be provided around the wetland both to
separate the treatment area from the human community
and, if development habitat is an objective, to reduce the
exposure of animals to light, humans, and pets. The
buffer requirement can be waived for pocket wetlands
without wildlife habitat objectives and adjacent struc-
tures. The minimum buffer width should be 8 m, meas-
ured from the maximum water surface elevation, plus 5
m to the nearest structure. The buffer should be in-
creased to at least 16 m when developing wildlife habitat
is an objective. It should be sloped no more than 5:1
(horizontal to vertical).
Preserve existing forest in the buffer area if at all possi-
ble. At least 75 percent of the buffer should be forested
to avoid attracting geese and to provide better protection
and habitat for other wildlife.
Pretreatment
The constructed wetland is expected to serve as the
primary treatment device. Nevertheless, some pretreat-
ment can prevent problems in the wetland, produce a
more self-sustaining system, and increase the potential
for ancillary benefits. Pretreatment mechanisms that
should be considered include:
Catch basins, for trapping the largest solids.
A presettling basin or biofilter, when the watershed
produces relatively high solids loadings.
Oil-water separators.
Designing for Ancillary Benefits and
Avoidance of Problems
Ancillary Benefits
Potential ancillary benefits of constructed wetlands
include:
Wildlife habitat.
Aquaculture for harvest.
Primary production for food-chain support.
336
-------
Biological diversity.
Open space for recreational, educational, and other
human uses.
This paper focuses on creating wildlife habitat, which
also helps achieve the latter three benefits. The preced-
ing recommendations on configuring the wetland were
also designed in part to contribute to these benefits.
An issue, of course, is the attraction of wildlife to a
wastewater treatment area that might be contaminated.
It is thought, but not proven, that levels of contamination
hazardous to wildlife are a relatively rare problem
restricted to watersheds with very high vehicle traffic,
proportions of impervious surface, and/or population
densities. It is also thought that such problems can be
addressed at least partially by reversing the recommen-
dations to attract wildlife; that is, install features that
discourage wildlife colonization. In either case, a quali-
fied wildlife biologist is needed to design the features.
For now, the best course seems to be using and study-
ing constructed wetlands for many applications, but
avoiding their use in areas with a high potential for toxic
contamination.
The main factor in designing for wildlife habitat is com-
plex structure that provides a variety of possible niches
to support feeding, nesting, breeding, and refuge re-
quirements of desired species. Fortunately, many fea-
tures that promote pollution control also enhance wildlife
habitat. Figure 3 illustrates several suggested features
for habitat development, and the comprehensive refer-
ences provide other illustrations.
Following is a summary of features that enhance wildlife
habitat drawn from Figure 3 and the references:
Irregular shorelines.
A wide range of depth zonesdeep zones provide
habitat for invertebrates, amphibians, and possibly
fish; higher marsh areas offer feeding grounds to
birds; and various nesting opportunities are provided
in the different zones.
Perimeter forest buffer at least 16 m wide.
Connect wetland to corridors (e.g., streams and pas-
sages to forests and other wetlands) that allow wild-
life movement.
Increase wetland size if very smallresearch has
shown that wildlife use is not strongly correlated to
the size of natural wetlands of 0.5 to several hectares
in area (15), but is low in very small natural and
constructed wetlands (less than 0.1 hectares).
Select plants that offer refuge, nesting, feeding, and
breeding habitat.
Install other features providing for nesting and refuge,
such as:
- Islands (protection for ground-nesters) (minimum
3 m2 for a waterfowl pair, above maximum water
surface elevation, densely vegetated, positively
drained).
Wood Duck
Nesting Box
Log raft feature for
provide hiding places secure nesting chained
and homes for small to concrete anchor
creatures who are
part of food chain ^l-?,^ High Water Level
Rock island for secure nesting
provides good fish habitat, too
Neting Platform
Brush heap will be
used by creatures
in and out of water
Island left during
grading of pond
Some of these features will cause
a small loss of water storage capacity; size
pond accordingly
Islands should be higher than
pond edges for safety during
flooding
Figure 3. Suggested constructed wetland habitat features (11).
337
-------
- Snags (dead tree trunks installed for cavity-
n esters).
- Nest boxes and platforms (unique designs for
cavity-nesters).
- Buffer trees (for foil age-n esters).
- Logs, stumps, and brush (for bird perches and
small-mammal refuge).
Avoiding significant water level fluctuationsthis is
an inherent disadvantage of stormwater wetlands
relative to wildlife. The best remedy is to precede the
wetland with runoff quantity control. Otherwise, the
configuring recommendations stated earlier provide
the best situation obtainable in stormwater applications.
Avoidance of Problems
Potential problems associated with constructed wet-
lands include:
Mosquito breeding.
Aesthetic drawbacks.
Safety concerns.
Attraction of geese and ducks, which can constitute
a nuisance.
Development of a monoculture of undesirable
vegetation.
Accumulation of toxicants.
The extent of actual occurrence of these problems and
managing to avoid or minimize them is addressed briefly
in this paper.
Mosquitoes are actually rarely a problem in well-
designed and operated constructed wetlands; thus,
education of the concerned public is part of the solution.
A problem with mosquitoes can best be prevented by
providing diverse habitats that support predatory in-
sects. Mosquito fish (Gambusia) have been used suc-
cessfully to control mosquitoes in permanent ponds, but
the introduction of the fish in areas to which they are not
native must be carefully assessed.
Aesthetic problems can be avoided with careful atten-
tion to construction and vegetation establishment. The
buffer and tall emergent vegetation can be used to
conceal such wetland characteristics as water level fluc-
tuation and films on the water.
Constructed wetlands are inherently safer than deeper
ponds, but some degree of potential hazard to children
is associated with deep zones. Hazards can be avoided
by establishing gradual side slopes and a shallow marsh
safety bench (5 m wide) where the toe of the side slope
meets any deep pool, by concealing outlet piping, and
by providing lockable access. In general, fencing should
only be needed on the embankment above large out-
falls, where they exist.
Nuisance waterfowl can be discouraged in several
ways. Maintain the buffer largely with forestland (at least
75 percent), and avoid the growth of turf grass around
the wetland. Also, maintain a variety of depths, espe-
cially high marsh not favored by geese and mallards.
Another important measure is to educate citizens and
place signs to discourage feeding.
The tendency for wetlands to develop undesirable plant
monocultures can be limited by maintaining structural
diversity and a range of depths, especially shallower
areas. A diverse selection of native flora should be
planted shortly after the wetland is constructed.
Regarding toxicant accumulations, evidence suggests
that metals and organics are tightly bound in sediments
and do not tend to become mobilized over long periods.
When maintenance is performed, disposal of spoils be-
comes an issue. Current knowledge indicates that spoils
pass hazardous waste tests and can be safely land
applied or landfilled (4). Plan an onsite application area
if possible to save costs of disposal.
Vegetation Selection and Establishment
As experience with wetland creation, restoration, and
construction projects accumulates, it is becoming in-
creasingly clear that the plant community develops best
when the soils harbor substantial vegetative roots, rhi-
zomes, and seed banks. Its development is also en-
hanced by the opportunity for volunteer species to enter
from nearby donor sites; however, volunteers cannot be
relied on for vegetation establishment. Transplants may
be supplanted by more vigorous resident and volunteer
stock under these circumstances and may actually con-
stitute a minor component of the eventual community.
Nevertheless, transplanting is generally a wise strategy,
and most of the specific guidance available for estab-
lishing wetlands concerns this source; thus, it is fully
covered below.
Hydric soils containing vegetative plant material col-
lected for establishing new wetlands are becoming
known as "wetland mulch." It appears that ample use of
this mulch enhances diversity and the speed of vegeta-
tion establishment, but the mulch content is somewhat
unpredictable and donor sites are limited. Also, guide-
lines for extracting, handling, and storing the material
are limited. A danger with the use of mulch is the possi-
ble presence of exotic, opportunistic species that will
out-compete more desirable natives. Therefore, at least
the donor sites that obviously support such plant spe-
cies should be avoided in obtaining material. Preferred
donor material includes wetland soils removed during
maintenance of highway ditches, swales, sedimentation
ponds, retention/detention ponds, and clogged
338
-------
infiltration basins and during dredging, or from natural
wetlands that are going to be filled under permit (al-
though these soils are best used for mitigating the loss).
It is recommended that the upper 15 cm of donor soils
be obtained at the end of the growing season, if possi-
ble. The best way to hold soils until installation is some-
what uncertain but must include keeping soils moist in
conditions that will maintain vital dormancy. Efforts are
underway to establish repositories for mulch reclaimed
in maintenance operations.
The reliability of transplanting and the instant partial
cover it provides make it necessary regardless of the
potentials offered by wetland mulch and volunteer spe-
cies recruitment. Commercial wetland plant nurseries
now operate in many places in the nation to provide
material. The following list of general vegetation selec-
tion principles was compiled from Garbisch's (16) rec-
ommendations for creating wetlands and from the
comprehensive constructed wetland works:
Base selections more on the prospects for successful
establishment than on specific pollutant uptake ca-
pabilities (plant uptake is a highly important mecha-
nism only for nutrients, much of which are released
upon the plant's death; nutrient removal is more the
result of chemical and microbial processes than of
plant uptake).
Select native species, and generally avoid natives
that invade vigorously.
Use a minimum of species adaptable to the various
elevation zones; diversification will occur naturally.
Select mostly perennial species, and give priority to
those that establish rapidly.
Select species that are adaptable to the broadest
ranges of depth, frequency, and duration of inunda-
tion (hydroperiod).
Match the environmental requirements of plant selec-
tions to the conditions to be offered by the site. Con-
sider especially hydroperiod and light requirements.
Give priority to species that have been used success-
fully in constructed wetlands in the past and to com-
mercially available species.
Avoid specifying only species that are foraged by
wildlife expected to utilize the site.
Phase the establishment of woody species to follow
herbaceous ones.
Consider planting needs to achieve designated ob-
jectives other than pollution control.
Although excessive emphasis on vegetation selection
based on pollution control capabilities should be
avoided, considerable information on that subject has
been compiled. Kulzer (17) prepared a summary of the
demonstrated capabilities of plants for the various com-
mon classes of pollutants. The most versatile genera
that have species representatives in most parts of the
nation are Carex, Scirpus, Juncus, Lemna, and Typha.
Schueler (4) and Garbisch (16) have assembled a
considerable amount of specific guidance on the con-
struction and vegetation establishment process for
constructed wetlands and created wetlands, respec-
tively. The course manual by Horner (5) also incorpo-
rates this guidance. Given the available literature, these
topics are not addressed in this paper.
Operating Constructed Wetlands
Relative to retention/detention ponds, constructed wet-
lands pose a relatively significant routine operating bur-
den. Operated properly, however, they should not
require periodic expensive sediment cleanouts. From
the outset, the project should include a formal operation
and maintenance plan that covers the following ele-
ments: 1) inspection, 2) sediment management, 3)
water management, and 4) vegetation management.
There are two levels of inspection: routine and compre-
hensive. Rapid, routine inspections should be made by
a qualified observer to identify and take action on any
problems that would damage the wetland's function.
Recommended scheduling for these inspections is
monthly and after each storm totaling more than 1.25 cm
(0.5 in.) of precipitation. Comprehensive inspections
should take place twice yearly the first 3 years, once in
the growing season and again in the nongrowing sea-
son. Conditions that should be noted during these in-
spections include:
Dominant plants and their distributions in each zone.
Relative presence of intentionally planted and volun-
teer invasive and noninvasive species.
Plant conditionlook for signs of disease (yellowing,
browning, wilting), pest infestations, and stunted
growth.
Depth zones and microtopographic features com-
pared with the original plan.
Normal pool elevation compared with the original
plan.
Sediment accumulations (locations and approximate
quantities).
Outlet clogging.
Buffer condition.
The objective of sediment management is to trapand
when necessary removesediments before they reach
the shallow zones. Forebays will probably have to be
drained and dredged every 2 to 5 years. The pond in a
339
-------
pond/marsh system is, in part, a large forebay and
should not need dredging as frequently.
If water levels do no conform to plans, orthere is another
reason to change them, regulation can be accomplished
by installing a flash board at the desired height at the
outlet weir or by adjusting the gate valves (if provided).
Remove clogging debris from around the outlet as
necessary.
In vegetation management, provide extra care during
the first 3 years to plantings, especially trees, including
watering, supporting, mulching, and removing weeds.
Reinforcement plantings will probably be required after
1 or 2 years and should be added as necessary. Manu-
ally remove undesirable species with a high potential to
invade and dominate, if they will subvert achievement
of the designated objectives. Cut or dig out woody,
unwanted species in marsh zones before they cause
damage and become too difficult to remove.
Harvesting the wetland for nutrient control can be per-
formed but has many drawbacks, including cost, dis-
posal, and damage to the system. It is generally only
possible to cut aboveground biomass, which will not
adequately control the release of nutrients.
References
1. Hammer, D.A., ed. 1989. Constructed wetlands for wastewater
treatment. Chelsea, Ml: Lewis Publishers.
2. Strecker, E.W., J.M. Kersnar, E.D. Driscoll, and R.R. Horner.
1992. The use of wetlands for controlling stormwater pollution.
Chicago, IL: U.S. EPA.
3. Olson, R.K., ed. 1992. Created and natural wetlands for control-
ling nonpoint source water pollution. Boca Raton, FL: Lewis Pub-
lishers.
4. Schueler, T.R. 1992. Design of stormwater wetland systems.
Washington, DC: Metropolitan Washington Council of Govern-
ments.
5. Horner, R.R. 1992. Constructed wetlands for storm runoff water
quality control. Seattle, WA: University of Washington, Engineer-
ing Continuing Education.
6. Water Pollution Control Federation. 1990. Natural systems for
wastewater treatment. Alexandria, VA: Water Pollution Control
Federation.
7. Richardson, C.J. 1985. Mechanisms controlling phosphorus re-
tention capacity in freshwater wetlands. Science 228:1,424-
1,427.
8. Brodie, G.A. 1989. Selection and evaluation of sites for con-
structed wastewater treatment wetlands. In: Hammer, D.A., ed.
Constructed wetlands for wastewater treatment. Chelsea, Ml:
Lewis Publishers, pp. 307-318.
9. Azous, A. 1991. An analysis of urbanization effects on wetland
biological communities. M.S. thesis. Seattle, WA: University of
Washington, Department of Civil Engineering.
10. Mitsch, WJ. 1992. Landscape design and the role of created,
restored, and natural riparian wetlands in controlling nonpoint
source pollution. In: Olson, R.K., ed. Created and natural wet-
lands for controlling nonpoint source water pollution. Boca Raton,
FL: Lewis Publishers, pp. 27-48.
11. Washington State Department of Ecology. 1992. Stormwater
management manual for the Puget Sound basin. Olympia, WA:
Washington Department of Ecology.
12. U.S. EPA. 1986. Methodology for analysis of detention basins for
control of urban runoff quality. EPA/440/5-87/001. Washington,
DC.
13. Walker, WW. 1987. Phosphorus removal by urban runoff deten-
tion basins. Lake and Reservoir Management 3:314-326.
14. Hartigan, J.P. 1989. Basis of design of wet detention BMPs. In:
Roesner, L.A., B. Urbonas, and M.B. Sonnen, eds. Design of
urban runoff quality controls. New York, NY: American Society of
Civil Engineers, pp. 122-144.
15. Martin-Yanny, E. 1992. The impacts of urbanization on wetland
bird communities. M.S. thesis. Seattle, WA: University of Wash-
ington, College of Forest Resources.
16. Garbisch, E.W 1986. Highways and wetlands: Compensating
wetland losses. FHWA-IP-86-22. McLean, VA: Federal Highway
Administration.
17. Kulzer, L. 1990. Water pollution control aspects of aquatic plants:
Implications for stormwater quality management. Seattle, WA:
Municipality of Metropolitan Seattle.
340
-------
Stormwater Pond and Wetland Options for Stormwater
Quality Control
Thomas R. Schueler
Metropolitan Washington Council of Governments, Washington, DC
Abstract
In this paper, 10 designs for Stormwater wetland and pond
systems used for effective urban runoff quality control
are surveyed. Each design is based on a different allo-
cation of deep-pool, marsh, and extended detention
storage. The comparative pollutant removal capability of
the 10 designs are reviewed based on a national survey
of 58 performance monitoring studies. In addition, the
reported longevity, maintenance requirements, and en-
vironmental constraints of each design is assessed.
A team approach for selecting the most appropriate
design at the individual development site is strongly
recommended. Key selection factors, such as space,
drainage area, and permitability, are discussed. A
seven-stage design/construction process is outlined to
ensure the team selects and builds the most appropriate
and effective design.
The paper points out that the uncertain regulatory status
of pond/wetland systems should be resolved so that
this effective runoff control technology can be appropri-
ately used.
Introduction
The use of Stormwater ponds to control the quality of
urban Stormwater runoff has become more widespread
in recent years. At the same time, designs have become
more sophisticated to meet many environmental objec-
tives at the development site. Today, the term Stormwa-
ter pond can refer to any design alternatives in a
continuum that allocates different portions of runoff
treatment volume to deep pools, shallow wetland areas,
and temporary extended detention storage. This paper
provides a broad review of the comparative capabilities
of pond and wetland systems.
In an operational sense, these systems can be classified
into one often categories:
1. Conventional dry ponds (quantity control only)
2. Dry extended detention (ED) ponds
3. Micropool dry ED ponds
4. Wet ponds
5. Wet ED ponds
6. Shallow marsh systems
7. ED wetlands
8. Pocket wetlands
9. Pocket ponds
10. Pond/marsh systems
Table 1. Comparative Storage Allocations for the 10
Stormwater Pond/Wetland Options (% of Total
Treatment Volume)
Pond/Wetland
Alternative
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Conventional dry
ponds (quantity
control only)
Dry ED ponds
Micropool dry ED
ponds
Wet ponds
Wet ED ponds
Shallow marsh
systems
ED wetlands
Pocket wetlands
Pocket ponds
Pond/marsh systems
Deep Pool
0
0
30 (f, m)
80
50
40 (f, m, c)
20 (f, m)
20 (f)
80
70
Marsh
0
10 (Is)
0
20 (b)
10(b)
60
30
80
20 (b)
30 (b, m)
ED
0
90
70
0
40
0
50
0
0
0
Note: The storage allocations shown are approximate targets only.
Is = lower stage of ED pond often assumes marsh characteristics
f = forebay
m = micropool
c = channels
b = aquatic bench
341
-------
7
6.
j'V /ftM It'/IY Q' t ri yfl i Vt
2. V
7.
f f «TtiK mrft*\/*i tarns
Quantity
Figure 1. Stormwater pond options.
ED
Pool
Marsh
Each of these designs (shown in cross-sectional view in
Figure 1) can be distinguished by how it allocates the
total treatment volume to deep pools, shallow wetlands,
and temporary extended detention storage. As can be
seen, most designs incorporate two and sometimes
three runoff treatment pathways. Comparative storage
allocations are shown in quantitative terms in Table 1. It
is important to note that these allocation targets are
approximate and relative, and individual systems may
not always conform to the target.
Stormwater pond systems can also be configured in
many different ways, as shown in Figure 2. Ponds can
be located "on-line" or "off-line" and can be arranged in
multiple cells. On-line ponds are located directly on
streams or drainage channels. Off-line ponds are con-
structed away from the stream corridor. Runoff flow is
split from the stream and diverted into off-line ponds by
a flow splitter or smart box.
The total treatment volume need not be provided within
only one cell. Stormwater ponds can contain multiple
storage cells, and these often enhance the perform-
ance, longevity, and redundancy of the entire system.
All pond designs provide additional storage to control
the increased quantity of Stormwater produced as a
consequence of urban development. This "quantity con-
trol" storage is usually defined as the storage needed to
keep postdevelopment peak discharge rates equivalent
to predevelopment levels for the 2-year storm. The
quantity control storage is in addition to, and literally on
top of, the quality control runoff storage.
a. On-Line
b. Off-Line
c. Multiple Cells
Figure 2. Stormwater pond configurations.
Comparative Pollutant Removal of
Stormwater Pond Designs
Each of the three basic treatment volume allocations (pool,
marsh, and ED) use different pollutant removal pathways.
Therefore, it is not surprising to find considerable vari-
ability in the projected removal rates for each of the 10
Stormwater pond designs (Table 2). The table is based
342
-------
Table 2.
Comparative Pollutant Removal Capability of
Stormwater Pond/Wetland Alternatives
Pollutant Removal Rate
Pond/Wetland Alternative
TSS TP
TN Reliability
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Conventional dry ponds
Dry ED ponds
Micropool dry ED ponds
Wet ponds
Wet ED ponds
Shallow marsh systems
ED wetlands
Pocket wetlands
Pocket ponds
Pond/marsh systems
10
30
70
70
75
75
70
60
60
80
0
10
30
60
65
45
40
25
30
70
0
10
15
40
40
25
20
15
20
45
Moderate
Low
Moderate
(projected)
High
High
High
Moderate
Moderate
(projected)
Moderate
(projected)
High
TSS = total suspended solids
TP = total phosphorus
TN = total nitrogen
on a review of 58 pond and wetland performance studies
conducted across the United States and Canada (1).
While seven of the ten pond designs have been moni-
tored in the field, the performance of three designs
(pocket ponds, pocket wetlands, and micropool dry ED
ponds) can only be projected based on design infer-
ences and field experience.
Two of the pond designs possess limited capability to
remove pollutantsthe conventional dry pond and the
dry ED pond. These pond systems seldom have been
observed to reliably remove sediment and have shown
virtually no capability to remove nutrients. The perform-
ance of dry ED ponds is expected to improve if micro-
pools are added at the inlet and the outlet. Micropools
help to pretreat incoming runoff, prevent resuspension,
and reduce clogging.
When properly sized and designed, wet ponds can reli-
ably remove sediments and nutrients at relatively high
rates. The deep pool of the wet pond allows for gravita-
tional settling. Removal rates for wet ponds can be
incrementally improved if the deep pool is combined
with extended detention, as in the wet ED pond system.
The removal capability of wetland systems (designs 6,
7, and 8) is generally comparable to that of wet ponds
of similar size. Sediment removal often is slightly higher
in wetland systems, but nutrient removal appears to be
somewhat lower and less reliable. Shallow marsh sys-
tems exhibit slightly higher removal rates than either the
ED wetland or the pocket wetland systems, which may
be explained by the greater surface area and complexity
of shallow marsh systems (2).
Ponds and wetlands that do not have a reliable source
of base flow, and that have a water level that frequently
fluctuates, are termed pocket ponds and wetlands.
These systems typically serve very small drainage ar-
eas and are excavated to the local water table. Conse-
quently, pocket facilities are often less than a quarter
acre in size and possess few of the design features of
their larger counterparts. Therefore, pocket wetlands
are thought to have lower pollutant removal capability,
especially for nutrients.
Pond-marsh systems appear to possess the greatest
overall pollutant removal capability of all the designs
monitored. The permanent pool and the shallow wetland
provide complementary and redundant removal path-
ways, and reduce remobilization of pollutants.
It should be noted that while differences in removal
capability do exist among the 10 designs, other key
design factors also must be present if these rates are to
be achieved. First, the system must be capable of cap-
turing at least 90 percent of the annual runoff volume
delivered. Second, incoming runoff must be pretreated
in a forebay or deep pool. Third, the system must meet
minimum criteria for internal geometry (flow path, micro-
topography, surface-area-to-volume ratio). Clearly, a
poorly conceived or designed pond system will not
achieve the rates shown in Table 2.
Comparative Ability To Protect
Downstream Channels
Pond systems that combine ED storage with stormwater
quantity storage appear to provide the best measure of
protection for downstream channels exposed to the
erosive potential of bankfull and subbankfull floods. Re-
cent field research has demonstrated that control of the
2-year storm quantity exacerbates, rather than reduces,
downstream channel erosion problems. Modeling stud-
ies suggest that extended detention (e.g., 6 to 24 hours)
of relatively small treatment volumes may have some
potential to alleviate downstream channel erosion prob-
lems. Additional field research is needed to confirm the
value of ED in protecting channels.
Comparative Physical, Environmental,
and Maintenance Constraints
Each of the 10 pond systems are subject to many different
constraints that may limit their use at a particular site. Some
of the more common constraints are outlined in Table 3.
Physical constraints include available space, climate, dry
weather base flow, and contributing drainage area. Main-
tenance constraints may involve susceptibility to clogging
and the frequency and difficulty of sediment cleanout.
343
-------
Table 3. Comparative Capability of 10 Pond/Wetland AlternativesPhysical, Environmental, and Maintenance Constraints
Pond/Wetland Alternative
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Conventional dry ponds
Dry ED ponds
Micropool dry ED ponds
Wet ponds
Wet ED ponds
Shallow marsh systems
ED wetlands
Pocket wetlands
Pocket ponds
Pond/marsh systems
Minimum
Drainage
Area3
5
10
15
25+
25+
25+
10+
1-5
1-5
25+
Space
lndexb
0.5
1.0
1.0
1.0
1.0
2.5
1.5
2.0
1.0
1.5
Water
Balance
No
restrictions
No
restrictions
May require
base flow
Climate
Climate
Climate,
base flow
Climate,
base flow
Climate,
ground
water
Climate,
ground
water
Climate,
base flow
Clogging
Risk
Moderate
High
Low
Low
Low
Low
Low
Moderate
Moderate
Low
Waters
Sediment of U.S.
Cleanout (404)
Basin ?
(1 0-20 yr)
Basin Yes
(1 0-20 yr)
Forebay Yes
(2-5 yr)
Forebay Yes
(2-5 yr)
Forebay Yes
(2-5 yr)
Forebay Yes
(2-5 yr)
Forebay ?
(2-5 yr)
Basin No
(5-1 Oyr)
Basin No
(5-1 Oyr)
Pool Yes
(1 0-1 5 yr)
Stream
Warming
Low
Moderate
Moderate
High
High
High
High
Moderate
Moderate
High
Safety
Risk
Low
Low
Low
High
High
Moderate
Moderate
Moderate
Moderate
High
aMaximum of 400 acres in most cases.
bSpace consumption index (1 = space required for wet pond).
Perhaps the most restrictive constraints, however, are
of an environmental nature. Recent research has indi-
cated that on-line pond and wetland systems can have
serious impacts on the local and downstream environ-
ment, if they are not properly located and designed (2).
The most serious include the modification or destruction
of high-quality forests and wetlands as a consequence
of construction, and downstream warming. Conse-
quently, the siting of ponds and wetlands in the mid-At-
lantic region has become a major focus of federal and
state regulatory agencies. Presently, both a Section 404
(wetlands) and a Section 401 water quality certification
permit must be obtained for the construction of any
on-line stormwater pond or wetland.
A Team Approach for Selecting the Most
Appropriate System
Selecting and designing a pond system has become a
complex and lengthy process. An effective approach is
to assemble a design team consisting of a stormwater
engineer, landscape architect, environmental consult-
ant, and the construction contractor. The combined
expertise of the design team, along with early and
frequent coordination with local plan reviewers, is an
essential ingredient for implementing the most appro-
priate system for the development site and the down-
stream community.
The design team works together throughout the plan-
ning, design, approval, and construction process, which
can take as long as 2 years. Building an effective and
appropriate pond system consists of seven general
steps, as outlined below:
1. Evaluation of the Feasibility of the Site
The design team has two major tasks. The first task is
to define, in consultation with local planning and re-
source protection agencies, the primary watershed pro-
tection objectives for the particular site and stream. The
objectives may include specific targets for pollutant re-
duction, flood control, channel protection, wetland
creation, habitat protection, protection of indicator
species (e.g., trout), or preservation of stream corri-
dors. Careful identification of realistic and achievable
objectives early in the process is critical for allowing the
design team to incorporate them into the design and
construction process.
The second task is to analyze the physical and environ-
mental features of the development site to determine if
a pond system is feasible, appropriate, and can meet
the primary watershed protection objectives. This typi-
cally involves a thorough delineation of the wetlands,
forests, and catchments within the development, as well
as the collection of geotechnical data to define soil
properties and water balances. The design team also
344
-------
should assess both the site and downstream aquatic
conditions during a site visit.
2. Development of the Initial Concept Plan
The task for the design team in this stage is threefold:
1) select the most appropriate pond design option, 2)
identify the most environmentally suitable location for it,
and 3) compute the size and geometry of the facility. The
design team assembles a concept plan and then submits
it to the local stormwater review agency and other regula-
tory agencies for preliminary review and approval. Early
input from the permitting agencies is essential, and a joint
field visit is often a useful means of securing it.
3. Development of the Final Design
In final design stage, the team adds engineering details
to the concept plan and responds to the comments
made by the local permitting authorities. The team works
together to ensure that all standard pond design fea-
tures are incorporated into the final design plans (e.g.,
benches, forebays, buffers, gate valves). (See Schueler
[2] for a full list.) In addition, the plan should be thor-
oughly analyzed to reduce safety risks, allow for easy
maintenance access, provide safe and environmentally
sensitive conveyance to the pond, and reduce the future
maintenance burden. The final plan is then submitted for
review and approval by the appropriate local and state
regulatory agencies.
4. Preparation of a Pondscaping Plan
This stage of the design process is critical but frequently
overlooked. The design team jointly prepares an aquatic
and terrestrial landscaping plan for the pond or wetland,
known as a pondscape. It specifies the trees, shrubs,
ground cover, and wetland plants that will be established
to meet specific functional objectives within different
moisture zones in and around the pond.
The pondscaping plan is more than a landscaping ma-
terials list, it also specifies necessary soil amendments,
planting techniques, maintenance schedules, reinforce-
ment plantings, and wildlife habitat elements needed to
establish a dense and diverse pondscape over several
growing seasons. Although landscape architects take
the lead in the development of the pondscape, other
members of the team can provide important contribu-
tions. For example, the engineer projects soil moisture
zones, the contractor provides practical guidance on
tree protection during construction and temporary stabi-
lization, and the environmental consultant provides input
on native wetland plants and propagation techniques.
5. Construction of the Pond
Appropriate designs only work when they are con-
structed properly. Therefore, it is essential to conduct a
field meeting with the entire construction crew prior to
construction. The design team outlines the purpose of
the project, the sequence of construction activities, and
walks through the no-disturbance limits. Short but regu-
lar meetings to inspect progress are helpful during the
construction process, especially to modify decisions in
the field. After construction is complete and the pond site
is stabilized, the engineer performs an as-built survey
for submission to local government authorities that veri-
fies that the pond was constructed in accordance with
the approved plans.
6. Establishment of the Pondscape
Establishing a functional pondscape requires frequent
adjustment of the original pondscaping plan. Initially, the
design team modifies the plan to account for actual
moisture conditions and water elevations that exist after
construction. The design team then reexamines the
pondscape after the first growing season to determine if
reinforcement plantings are needed.
7. Inspection and Operation of the Pond
The final stage of the process involves the final inspec-
tion of the facility, development of the maintenance prac-
tices and schedules, and the transfer of maintenance
responsibilities to the responsible party.
Resolving the Regulatory Status of
Stormwater Ponds
Although pond and wetland systems are attractive op-
tions for urban nonpoint source control, their regulatory
status has recently become very confused. This is due
to the fact the these systems fall under the scope of
three often conflicting sections of the Clean Water Act
Section 401 (water quality certification permits), Section
402 (stormwater National Pollutant Discharge Elimina-
tion System [NPDES] permits), and Section 404 (wet-
land permits). Confusion about these systems also
stems from a number of particular factors:
First, pond systems often acquire wetland charac-
teristics over time, whether by design or simply with age.
At some point, they may become delineated wetlands,
subject to the same protection and restrictions as natural
wetlands. If a stormwater pond system does evolve into
wetland status, then Section 404 wetland permits may
be required and all future maintenance activities con-
ducted on the stormwater pond system would likely
require a permit. Conversely, it also is possible that a
well-designed stormwater wetland would be eligible
for a partial mitigation "credit" when it "evolves" into
wetland status.
Second, most pond systems are located on waters of
the United States (i.e., intermittent or perennial streams
or drainage channels) and are thus subject to the Sec-
345
-------
tion 404 permit process, even when the system is not
located within a delineated wetland. Some regulators
have advocated that the prohibition against "instream
treatment" should apply to stormwater pond systems,
while others have required that an extensive alternatives
analysis be undertaken before a permit is issued. In the
former interpretation, the use of stormwater pond sys-
tems would be limited to off-line or pocket applications.
Under the latter interpretation, the design team might
have to demonstrate that all upland best management
practice (BMP) alternatives are exhausted before a
pond system can be constructed. While upland BMPs
are an alternative, they do not possess the performance
or longevity of pond and wetland systems and may not
be adequate to protect streams or meet pollutant reduc-
tion targets.
Third, construction of stormwater ponds and wetlands
within or adjacent to delineated natural wetlands can
radically alter the characteristics of that wetland, either
through excavation, fill, pooling, or inundation. In most
cases, construction of stormwater ponds in natural wet-
land areas is strongly discouraged. In other cases, how-
ever, it may actually be desirable to convert degraded
natural wetlands into stormwater wetlands. The condi-
tions, if any, where these conversions might take place
are the subject of considerable controversy. The influ-
ence of stormwater ponds on wetlands need not always
be negative, however. In many cases, stormwater ponds
can help protect downstream wetlands from degradation
caused by uncontrolled stormwater flows and construc-
tion-stage sediment deposition.
Fourth, stormwater ponds have a dual nature: They can
help to meet water quality standards in receiving waters,
while at the same time contributing to possible violations
of other standards. For example, ponds can help meet
sediment, turbidity, nutrient, and toxics limits. At the same
time, they may amplify the stream warming associated
with urban development and thus lead to violations of
temperature standards in some sensitive streams. This
creates a great dilemma for regulators that must perform
water quality certification on stormwater ponds.
The resolution of the uncertain and confusing regulatory
issues relating to stormwater ponds is critical if applica-
tion of this effective technology is to continue on a
widespread basis. The challenge for designers will be to
acknowledge and avoid the potential for negative envi-
ronmental impact, whereas the challenge for the regu-
latory community will be to recognize the benefits of
stormwater ponds and craft a regulatory policy that is
practical rather than merely legal. Otherwise, the fifth
member of the pond design team may have to be a
lawyer. Hopefully, a workable policy can be developed
in the near future that sets guidelines on the appropriate
use of this effective nonpoint source control technology.
References
1. Schueler, T. 1993. Performance of stormwater pond and wetland
systems. Submitted to ASCE International Symposium on Engi-
neering Hydrology, San Francisco, CA (July).
2. Schueler, T. 1992. Design of stormwater wetland systems. Wash-
ington, DC: Metropolitan Washington Council of Governments.
Additional Reading
1. Galli, J. 1992. Analysis of urban BMP performance and longevity
in Prince Georges County, Maryland. Prepared for Prince Georges
Watershed Protection Branch. Washington, DC: Metropolitan
Washington Council of Governments.
2. Schueler, T., and J. Galli. 1992. The environmental impacts of
stormwater ponds. In: Kumble, P., and T. Schueler, eds. Water-
shed restoration sourcebook. Washington, DC: Metropolitan
Washington Council of Governments, pp. 161-180.
3. Schueler, T., M. Heraty, and P. Kumble. 1992. A current assess-
ment of urban best management: Techniques for reducing non-
point source pollution in the coastal zone. Prepared for U.S. EPA
by Metropolitan Washington Council of Governments.
346
-------
Practical Aspects of Storm water Pond Design in Sensitive Areas
Richard A. Claytor, Jr.
Loiederman Associates, Inc., Frederick, Maryland
Abstract
This paper's purpose is to provoke thought in estab-
lishing some considerations and techniques for the de-
sign of stormwater management ponds in sensitive
areas, not to describe a step-by-step process for design-
ing stormwater management ponds. The reader should
have a basic understanding of the principles of small
pond design, urban hydrology, water quality control, and
best management practices.
First, practical design requires an inventory of the sen-
sitive resources that need protection and an estimate of
the project goals and potential environmental benefits.
The next step is to develop a concept plan, which initi-
ates the design process and ensures agency and public
involvement in early stages of the project. Several tech-
niques can be used to avoid or minimize negative im-
pacts on sensitive areas, which this paper groups into
techniques for either warm water or cool water environ-
ments. In addition, the paper covers three new theoreti-
cal techniques that combine warm water design
practices with cool water mitigation approaches. Main-
tenance and monitoring issues are also discussed. Cou-
pling a common sense approach with the need for
innovative thinking should be a primary goal, and de-
signers must factor into this challenge the goal of reach-
ing a consensus with different interest groups.
Goals and Expectations
Stormwater management ponds are often installed or con-
structed to fulfill regulations for the control of urban runoff.
Controlling urban runoff usually means providing some
kind of detention facility that controls the increased runoff
frequency and volume in developing areas.
Good, practical stormwater management requires an
assessment of what the pond needs to protect and an
estimate of how well pond is likely to work. This involves
conducting an inventory of existing natural and con-
structed features, which then becomes a basis for de-
sign considerations. For example, stormwater ponds
often need to be located in the lower portion of a site to
maximize the area and runoff draining toward them. This
can create a conflict with existing, sensitive natural fea-
tures, such as wetlands, seeps, springs, or even inter-
mittent or perennial streams.
A natural resources inventory, which is essential for
design, should at a minimum incorporate the following
features:
Topography
Wetlands (including springs and seeps)
Soils
Floodplains
Forest lands (vegetation)
Watercourses
Specimen trees
Steep slopes, rock outcroppings, etc.
Historical or archeological features
Habitat
After a reasonably detailed natural resources inventory
has been conducted, design should continue with an
analysis of the receiving stream or ground-water aqui-
fers. This may be very detailed and use various habitat
analyses or biological indicators, or it can be a general
overview. To pursue a sensitive design approach, how-
ever, establishing the type of aquatic resource fisheries
(cold water versus warm water) is important.
After establishing the natural resources inventory and
assessing what level of aquatic resource protection is
warranted, a concept plan should be developed.
Concept Plan Development
One of the most important elements in implementing a
successful stormwater management plan is the devel-
opment of a good concept plan. A concept plan allows
347
-------
various agencies and interest groups the opportunity to
offer input at a time when change is reasonably inex-
pensive. Later in a program, change becomes much
more difficult. Many resource protection agencies and
special interest groups have conflicting goals, which
should be resolved as much as possible in the early
stages of the concept plan process so that meaningful
projects ultimately become a reality.
One of the key elements of working in an environmentally
sensitive area is compromise, but ingenuity is equally
important. To advance technology and find different and
possibly more successful methods of stormwater man-
agement pond design, new techniques should be pro-
posed and implemented, even if unproven.
Techniques for Avoiding or Minimizing
Impacts to Sensitive Areas
Warm Water Environments
For warm water fisheries, where thermal impacts are not
a major consideration, wet ponds (permanent pools of
water) represent the most reliable and maintenance-
free option for stormwater runoff quality control (1). Sev-
eral techniques can enhance the pollutant removal
efficiency of wet ponds and simultaneously minimize the
impact that a large body of water has on surrounding
sensitive areas. Some of these techniques are:
Location of a pond "off-line" from active flowing
streams reduces the impact to existing aquatic environ-
ment and does not necessarily inhibit fish migration.
Diversion structures or "flow splitters" provide a tech-
nique for conveying both base flow and storm flow
away from sensitive areas (see Figure 1).
Pond grading techniques that provide storage vol-
umes direct impacts away from sensitive areas.
Pond grading techniques that give curvilinear geome-
try to the pond can increase flow lengths and de-
crease ineffective storage areas.
Pond grading techniques that use shallow aquatic
zones, peninsulas and/or islands, and low-lying areas
for riparian vegetation provide varied water regimes.
Incorporating vegetative practices into the design,
such as shallow marsh emergent wetlands, sub-
merged aquatic vegetation, and riparian fringe plant-
ings, can create additional wildlife habitats.
Figure 2 depicts a wet pond concept for a warm water
environment.
Cool Water Environments
For cool water fisheries, where thermal impacts are a
major consideration, a design must attempt to maximize
Diversion Pipe
to Wetlands
Large Storm
Overflow Weir'
Small Storm
Overflow Weir
Base-Flow Slot-
Figure 1. Diversion structure or "flow splitter."
pollutant removal efficiencies but also to reduce and/or
offset thermal impacts.
The following are some of the techniques that incorpo-
rate these goals:
The facility should avoid open bodies of water where
solar radiation would heat up the water column. Ex-
amples in descending order of preference would be
infiltration facilities, filtration facilities, dry extended
detention ponds, and shallow stormwater wetland
ponds (2).
The location and orientation of the facility should ac-
count for the hours of potential solar radiation, such
as a north/south dominant orientation.
Shading of the pool area by maximizing tree canopy
can minimize solar penetration.
Incorporating underdrain and toe drain ground-
water collection systems can provide an additional
source of cool water release, where available,
while implementing an earthen embankment safety
consideration.
Shading and covering a pond's outlet channel helps
prevent thermal impacts associated with water run-
ning over heated rocks.
Watershedwide landscaping, including shading of im-
pervious asphalt surfaces, helps reduce thermal
loading at the source.
Figure 3 depicts a dry pond concept for a cool water
environment.
New Theorized Techniques
New approaches may afford the opportunity to combine
the pollutant removal efficiencies of wet ponds with tem-
perature mitigation measures. Three approaches are to:
348
-------
Shrub/Scrub
Wetland Plantings
Terrestrial
Landscaping
Flow Splitter
Diversion Structure
Emergency
Spillway
Figure 2. Wet pond concept with diversion structure for warm water environment.
Incorporate "cooling tower" design practices into the
outlet structure of the spillway system (Figure 4) (3).
Investigate vegetative practices that cover the open
water surface of ponds to minimize solar radiation of
the water column (4).
Incorporate a ground-water siphon system into the
design of the release structures to siphon ground water
as the low flow release (Figure 5) (5).
Maintenance and Monitoring
An effective design cannot become a practical application
without a good implementation program, an effective
monitoring program, and a maintenance program that
keeps a facility functioning at its best. Many of the tech-
niques and considerations previously discussed are new
and may not meet expectations. These techniques require
short- and long-term monitoring to ensure that they are
meeting the expectations of the designer and agency.
In addition, many of the more innovative design ap-
proaches require periodic maintenance. It is not practi-
cal to assume that these approaches will function without
the necessary observations and periodic maintenance.
Some of the approaches (e.g., flow splitters) require
only periodic trash removal to keep them functioning as
designed, while others (e.g., filters and infiltration ba-
sins) require a more intensive maintenance program.
Conclusion
In sensitive areas, design approaches need to combine
innovative alternatives, common sense, and compro-
mise. Everyone agrees that our sensitive resources
349
-------
Flow Splitter
Diversion Structure
Open Water
Stilling Basin
Pond Located on North/
South Orientation
Check Dam
Infiltration or
Filtration Trench
Spillway
Terrestrial
Landscaping
Shading Sides
of Pond
Dry Extended
Detention Basin
Shrub/Scrub
Wetland
Plantings in
Pond Bottom
Shaded Outlet
Channel
Figure 3. Dry pond concept with diversion structure for cool water environment.
Emergency Spillway
Natural Convection
Warmed Air Out
Cool Air In
Splash Pieces (Saddles) to Maximize Exposure of Water Molecules to Cool Air
Figure 4. Combination atmospheric and natural draft cooling tower to cool water discharged from a wet pond system (3).
350
-------
10-yrWS_EL
2-yrWSEI """
WSEL = Water surface elevation
GWSEL = Ground-water surface elevation
Figure 5. Siphon thermal cooler concept for stormwater management ponds (5).
need special protection and require the utmost care if a
disturbance occurs. There is not agreement, however,
on the best approaches and on what resources are the
most important. Therefore, it is vital to document the
existing conditions carefully, prepare flexible concepts
and designs, and be prepared to revise plans and de-
sign approaches as new information and monitoring
results emerge. Practical aspects of stormwater pond
design will not remain static but will continue to change
as new technologies and techniques advance and older
considerations become obsolete.
Acknowledgments
The author would like to acknowledge the following
individuals who have played a key role in the develop-
ment of some of the techniques and considerations
contained herein: John Galli, Carter McCamy, Daniel
O'Leary, Mary Jo Garreis, Tom Schueler, Tim Schueler,
and Andrew Der.
References
1. Schueler, T. 1987. Controlling urban runoff: A practical manual for
planning and designing urban BMP. Metropolitan Washington
Council of Governments (July).
2. Galli, J. 1990. Thermal impacts associated with urbanization and
stormwater management best management practices. Metropoli-
tan Washington Council of Governments for the Sediment and
Stormwater Administration, Maryland Department of the Environ-
ment (December).
3. McCamy, C. 1992. Combination atmosphere and natural draft
cooling tower. EQR report (October 12). Environmental Quality
Resources, Inc., 1738 Elton Rd., Suite 310, Silver Spring, MD
20903.
4. McCamy, C. 1992. The use of water hyacinth to control thermal
loading in wet pond systems. EQR report (October 12). Environ-
mental Quality Resources, Inc., 1738 Elton Rd., Suite 310, Silver
Spring, MD 20903.
5. O'Leary, D. Syphon thermal cooler concept for stormwater man-
agement ponds. Water Management Administration, Maryland De-
partment of the Environment, 2500 Broening Highway, Baltimore,
MD 21224.
351
-------
Infiltration Practices: The Good, the Bad, and the Ugly
Eric H. Livingston
Florida Department of Environmental Regulation,
Tallahassee, Florida
Abstract
Of all the best management practices (BMPs) in the storm-
water treatment tool box, infiltration practices are the most
effective in removing stormwater pollutants and, equally
important, in reducing both stormwater volume and peak
discharge rate. This paper explains the concept of on-line
and off-line systems, and discusses factors that influence
their treatment effectiveness. Design guidelines for infil-
tration systems, including the importance of the BMP
treatment train approach, will be reviewed, focusing on
soil types, water table elevation, geology, vegetation,
and determination of infiltration rates. Construction con-
siderations will be reviewed. Because of their likelihood
for clogging, the importance of regular inspection and
maintenance programs is stressed.
Infiltration practices that the paper covers include road-
side swales, retention basins, landscape retention, ex-
filtration systems, infiltration trenches, and porous
pavement. For each type of system, information on
treatment effectiveness, design criteria, advantages,
and disadvantages is presented, along with discussion
of the good, the bad, and the ugly. The paper reviews
the effect of infiltration practices on ground-water quality
and presents recommendations to limit adverse im-
pacts. Special design guidelines for infiltration practices
in areas with karst geology, which is characterized by
sinkholes, will also be reviewed.
Introduction
To achieve the desired objectives of flood and water
quality protection, erosion control, improved aesthetics,
and recreation, a stormwater management system must
be an integral part of the site planning for every site.
Although the basic principles of stormwater manage-
ment remain the same, each individual site and each
specific project presents unique challenges, obstacles,
and opportunities. The many variations in climate, soils,
geology, ground water, topography, vegetation, and
planned land use require site-specific design. Each site
contains natural attributes that will influence the type
and configuration of the stormwater system.
The variety of features contained on a site suggest
which particular combination of best management prac-
tices (BMPs) can be successfully integrated into an
effective system. Whenever site conditions allow, the
stormwater management system should be designed to
achieve maximum onsite storage (and even reuse) of
stormwater 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 BMPs are the
individual cars. Generally, the more BMPs that are in-
corporated 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 stormwa-
ter system designers.
Infiltration practices are one of the few BMPs that can
help to ensure that all four stormwater characteristics
(the volume, rate, timing, and pollutant load) after devel-
opment closely approximate the conditions that occurred
before development. This is because infiltration prac-
tices help to maintain predevelopment site perviousness
and vegetative cover, thereby reducing stormwater vol-
ume and discharge rate, which further promotes infiltra-
tion and filtering of the runoff.
The benefits of infiltration include:
Reducing stormwater volume and peak runoff rate.
Recharging ground water, which helps to replenish
wetlands, creeks, rivers, lakes, and estuaries.
Augmenting base flow in streams, especially during
low flow times.
Aiding in the settling of pollutants.
Lowering the probability of downstream flooding,
stream erosion, and sedimentation.
Providing water for other beneficial uses.
352
-------
Another benefit of infiltration practices is their ability to
serve multiple uses because they are temporary storage
basins. Recreational areas (e.g., ballfields, tennis
courts, volleyball courts), greenbelt areas, neighbor-
hood parks, and even parking facilities provide excellent
settings for the temporary storage of stormwater. Such
areas are not usually in use during periods of precipita-
tion, and the ponding of stormwater for short durations
does not seriously impede their primary functions.
Determining Treatment Effectiveness
To design a BMP for water quality enhancement, a
pollutant reduction goal must first be established. Storm-
water treatment regulatory programs in Florida and
Delaware are based on a performance standard of re-
ducing the annual average total suspended solids (pol-
lutant) load by 80 percent for stormwater systems
discharging to waters classified as fishable and swim-
mable. In Florida, stormwater systems discharging to
potable supply waters, pristine waters, or highly polluted
waters may be required to remove up to 95 percent 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 (1). Design criteria for various
types of stormwater management systems that achieve
the desired performance standard (treatment efficiency)
are then adopted, thereby providing guidance to the
design community and making it relatively easy to obtain
a stormwater permit.
The average annual pollutant removal efficiency is calcu-
lated by considering the annual mass of pollutants avail-
able for discharge and the annual mass removed. The
primary removal mechanism for infiltration practices is
the volume of stormwater that is infiltrated, because this
eliminates the discharge of stormwater and its associ-
ated pollutants. As with any type of stormwater manage-
ment practice, its actual field efficiency depends on many
factors. For infiltration practices, these factors include:
Long-term precipitation characteristics such as mean
number of storms per year along with their intensity
and volume; average interevent time.
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.
"On-line" or "off-line" design.
Cumulatively, the above three factors determine the
minimum treatment volume and maximum storage re-
covery time.
The National Weather Service (within the National Oce-
anic and Atmospheric Administration) has measured
weather statistics at many locations around the country.
Long-term precipitation records, including information
such as day and duration of event, intensity, and vol-
ume, are available from either the federal government
or private vendors. Statistical analysis of these records
can develop probability frequencies for storm charac-
teristics such as the mean storm volume and the mean
interevent period between storms.
"First flush" describes the washing action that storm-
water has on accumulated pollutants in the watershed.
In the early stages of runoff, the land surfaces, espe-
cially impervious ones such as streets and parking
areas, are flushed clean by the stormwater. This flushing
creates a shock loading of pollutants. The occurrence
and prevalence of first flush, however, depends largely
on precipitation patterns. Studies in Florida have deter-
mined that for urban land uses there is a first flush for
many pollutants, especially particulates (2, 3). In areas
such as Oregon and Washington, however, where rain-
fall consists of low intensity, long-duration "events," the
first flush is not very prevalent. Where it exists, the
first-flush effect generally diminishes as the size of the
drainage basin increases and the amount of impervious
area decreases.
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 sys-
tem, 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 do provide both benefits.
Off-line practices are designed to divert the more polluted
stormwater first flush for water quality treatment, isolat-
ing it from the remaining stormwater that is managed for
flood control. 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
ground waters. Off-line retention practices are the most
effective for water quality enhancement of stormwater.
Because 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 1 is a sche-
matic of an off-line system, commonly referred to as a
"dual pond system," in which a smart weir directs the
first flush stormwater into the infiltration area until it is
filled, with the remaining runoff routed to the detention
facility for flood control.
Using the three factors above, design criteria have been
developed and implemented in Florida to achieve the
353
-------
Runoff From Site
Table 1. Cumulative Frequency Distributions on Efficiencies
per Storm Event as a Function of Storage Volume
(Area = 4.6 Ac, 85 percent Impervious, Tc = 20 min)
Retention Pond
for Water
Quality
Detention Pond
for Water Quality
__ Outlet Structure For
~ Pre-Peak Discharge
"Smart" Box Schematic
Figure 1. Schematic of an off-line system (4).
desired 80 or 95 percent treatment performance stand-
ard (5). The pollutant removal efficiency of an off-line
system depends on the annual volume of stormwater
that is diverted and infiltrated. For each storm, pollutant
removal efficiencies will vary from 100 percent for
storms producing less runoff than the diversion design
volume to lower efficiencies for much larger storms. If
the time between storms is less than the design intere-
vent period, then the design treatment volume will not
be available, and more runoff will not be captured and
treated. Wanielista (6) developed cumulative frequency
distributions for storm-related efficiencies using a simu-
lation model dependent on 20 years of rainfall data and
16 measured storm event runoff quantities and qualities.
The results shown in Table 1 are based on Florida
rainfall characteristics (90 percent of all annual rainfall
events are less than 2.54 cm) and a distinct first flush
(up to 90 percent of the pollution load carried in the first
2.54 cm of runoff). An off-line retention system designed
to accept at least the first 1.25 cm of runoff (or the
volume calculated by 1.25 times the percent impervious-
ness of the site) will remove more than 80 percent of the
average annual pollutant load.
A more recent investigation of the influence of long-term
rainfall characteristics on the efficiency of retention prac-
Average3
Efficiency
100
>96
>92
>88
>84
>80
>76
>72
>68
>64
Volume
0.25 (0.1)
35.4
42.5
46.0
47.8
50.4
65.6
61.1
66.4
72.6
82.3
of Storage,
0.64 (0.25)
66.4
74.3
77.9
81.4
90.3
92.9
96.3
97.3
98.2
100.0
centimeters
1.27 (0.50)
92.9
97.3
97.4
98.2
100.0
(inches)
2.54 (1.0)
99.0
100.0
1 Average efficiency is the average removal of BODs, suspended
solids, nitrogen, and phosphorus over a 20-year period. Average
number of rainfall events producing runoff per year is 116.
tices led to the development of diversion volume curves
for interevent dry periods of varying length (7). Figure 2
shows an example diversion volume curve for the Or-
lando 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,
which also include consideration of the probability that
a storm greater than the design storm will occur. The PIF
analysis looked at exceedance probabilities for storms
with a return period of 2, 3, 4, or 6 months, representing
a chance that the storm will exceed the design volume
six, four, three, or two times a year.
1.5 2 2.5
Diversion Volume
3.5
4-Hr I/E 24-Hr I/E 72-Hr I/E
Figure 2. Diversion volume curve for Orlando, Florida.
354
-------
Design of Infiltration (Retention) Practices
Infiltration practices also are commonly called retention
practices because they retain the runoff on site. They
are designed to infiltrate a design volume (treatment
volume) of stormwater, and the tool box includes on-line
and off-line percolation ponds and trenches, infiltration
areas, exfiltration systems, and vegetated swales. De-
sign factors that influence the treatment effectiveness
and feasibility of infiltration practices include choice of
on-line or off-line system, use of the BMP treatment train
concept, and soil type, geology, water table elevation,
topography, and vegetation.
Infiltration areas, especially off-line ones, can be incor-
porated easily into landscaping or open space areas of
a site. These can include natural or excavated grassed
depressions, recreational areas, and even parking lot
landscape islands. If site conditions prevent the exclu-
sive use of infiltration, then off-line retention areas
should be used as pretreatment practices in a stormwa-
ter 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. Parking lots with their land-
scape islands offer an excellent opportunity for the use
of this concept because even the infiltration of a quarter
inch of runoff will greatly reduce sediments, metals, and
oils and greases. Placing storm sewer inlets within re-
cessed parking lot landscape areas, raising the inlet a
few inches above the bottom, and using curb cuts to
allow runoff to enterthis area represent a highly effective
treatment train.
Siting, Design, and Planning
Considerations for Infiltration Practices
The suitability of a site for certain infiltration practices
depends on a careful evaluation of the site's natural
attributes. Proposed infiltration areas should be evalu-
ated for feasibility on any particular site or project by
examining the following.
Soils
Soils must have permeability rates that allow the di-
verted volume to infiltrate within 72 hours, or within 24
to 36 hours for infiltration areas that are planted with
grasses. Soil textures with minimum infiltration rates of
0.43 cm/hr or less are not suitable for infiltration prac-
tices (8). These unsuitable soils include soil textures that
have at least 30 percent clay content.
Infiltration Rates
One of the most difficult aspects of designing infiltration
practices is obtaining reliable information about the ac-
tual infiltration rate of the soil where the practice will be
constructed. Unfortunately, such information is not easily
obtainable. Avellaneda (9) conducted 20 hydrologic studies
of vegetated swales constructed on sandy soils with a
water table at least 1 ft below the bottom during dry
conditions. Infiltration rates were measured using labo-
ratory permeability tests, double-ring infiltrometers, and
field mass balance experiments. The field mass balance
method measured a minimum infiltration rate of 5 to
7.5 cm/hr. This measured rate was much less than lab
permeabilities, rates measured by double-ring infil-
trometer tests (12.5 to 51 cm/hr), or rates published in
the detailed soil survey. Recommendations for deter-
mining the infiltration rate for retention practices include
the following:
Because 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 per-
colated into the soil and not discharged downstream.
Onsite infiltration measurements must be taken at the
locations where retention practices will be located.
More importantly, because 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.
Infiltration rates should be determined by mass bal-
ance field tests if possible. These provide the most
realistic estimate of the percolation rate. If field tests
are not possible, then infiltrometer tests should be used,
with lab permeability tests a third option. In either of
these 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. The lowest rate should
be used, however, as should a safety factor of two.
Water Table
The seasonal high water table should be at least 1 m
beneath the bottom of the infiltration area to ensure that
stormwater pollutants are removed by the vegetation,
soil, and microbes before contacting the ground water.
When considering the ground-water elevation, it is im-
portant to remember that the retention area can cause
a mounding effect on the water table, thereby raising it
above the predevelopment level.
Geology
Bedrock should be at least 1 m beneath the bottom of the
infiltration area. In those parts of the country where lime-
stone is at or near the land surface, special precautions
must be taken when using infiltration practices. The
potential for ground-water contamination in such areas
is quite high, especially in "karst sensitive areas"
(KSAs), where sinkhole formation is common. In KSAs,
solution pipe sinkholes may form in the bottom of infil-
355
-------
tration areas, creating a direct conduit for stormwater
pollutants to enterthe ground water. Solution pipes often
open in the bottom of retention areas because the natu-
ral soil plug capping the solution pipe is thinned by
partial excavation to create the retention area and be-
cause the stormwater creates hydraulic pressure that
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:
Include several small offsite areas.
Use swale conveyances for pretreatment.
Be as shallow as possible.
Be vegetated with a permanent cover such as sod-
ded grasses.
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 percent to minimize the chance of
downstream water seepage from the subgrade. Sloping
sites often require extensive cut and fill operations. In-
filtration practices should not be sited on fill material
because fill areas are very susceptible to slope failure,
especially when the interface of the fill/natural soil be-
comes saturated.
Vegetation
To reduce the potential for stormwater pollutants to enter
ground waters and to help maintain the soil's capacity
to absorb water, infiltration practices should be vege-
tated with appropriate native vegetation, especially
grasses. This type of vegetation cannot tolerate long-
term inundation, however, so the retention area must be
capable of infiltrating all of its runoff within a relatively
short period (i.e., 24 to 36 hours).
Set Backs
Infiltration areas should be located at least 33 m from
any water supply well and at least 3.5 m downgradient
from any building foundations. Additionally, they should
be set back at least 15 m from onsite wastewater sys-
tems, especially drain fields.
Land-Use Restrictions
Certain infiltration practices can only be applied to par-
ticular land uses. Some sites are so small or intensively
developed that space is insufficient for practices that
require a large area (e.g., retention basin). Other prac-
tices (e.g., porous pavement) can only be used on sites
with parking lots and limited truck traffic.
Sediment Input
Infiltration practices must be protected from large loads
of sediment to prevent clogging and subsequent failure.
Although sediment loads drop sharply after construction
is complete, gradual clogging of infiltration practices can
still occur. Pretreatment practices such as swale con-
veyances or vegetated buffer strips can help to filter out
sediments and extend the life of retention practices.
Construction Considerations
To prevent clogging of infiltration areas, special precau-
tions must be taken during the entire construction phase
of a project. These are needed to prevent sedimentation
during construction, compaction of the soil, and sub-
sequent reduction in its infiltration capacity. Areas with
suitable characteristics that are selected for infiltration
use should be well marked during site surveying and
protected during construction. Heavy equipment, vehi-
cles, and sediment laden runoff should be kept out of
infiltration areas to prevent compaction and loss of infil-
tration capacity.
Before the development site is graded, the area
planned for use as infiltration areas should be well
marked during site surveying. Then, the area should
be roped off to prevent heavy equipment from com-
pacting the underlying soils.
Diversion berms should be placed around the pe-
rimeter of the infiltration area during all phases of
construction. Sediment and erosion control plans for
the site should be oriented to keep sediment and
runoff completely away from the area. Actual con-
struction of the infiltration practice should not begin
until after the site has been stabilized completely.
Infiltration areas should never be used as a tempo-
rary sediment basin during the construction phase. It
is somewhat common for infiltration areas, especially
basins, to be used as a sediment trap, with initial
excavation to within 2 ft of the final design elevation
of the basin floor. Sediment that accumulates during
the construction phase can then be removed when
the basin undergoes final excavation after the devel-
opment has been completed. Recent experience,
however, indicates that even with this type of con-
struction practice infiltration areas used as sediment
traps tend to fail.
Infiltration areas/basins should be excavated using
light earth-moving equipment with tracks or oversized
tires. Normal rubber tires should be avoided because
they compact the subsoil and reduce its infiltration capa-
bilities. For the same reason, the use of bulldozers or
front-end loaders should be avoided. Because some
356
-------
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.
The basin should be stabilized with vegetation within
a week after construction. Use of low maintenance,
rapid-germinating grasses such as fescues are rec-
ommended. The condition of the newly established
vegetation should be checked several times over the
first 2 months and any necessary remedial actions
taken (e.g., reseeding, fertilization, and irrigation).
Maintenance
All infiltration practices require regular and nonroutine
maintenance to maintain their ability to infiltrate storm-
water. The frequency and need for maintenance depends
primarily on the loading of particulates and the use of
pretreatment practices. Inspections should be conducted
on a regular basis after storm events, and maintenance
activities should be conducted whenever stormwater
remains in the practice beyond the designed time. Spe-
cific maintenance needs are discussed for each of the
different types of infiltration practices in the next section.
Discussion of Various Infiltration Practices
Infiltration Basins
An infiltration basin is made by constructing an em-
bankment or by excavating in or down to relatively per-
meable soils. The basin temporarily stores stormwater
until it infiltrates through the bottom and sides of the
system. The infiltration "basin" can actually be a landscape
depression within open spaces, even parking lot islands
or a recreational area such as a soccer field. Infiltration
areas generally serve drainage areas ranging from 2 to
20 hectares. Infiltration basins should be designed as
off-line systems but they can be on-line, especially if pre-
development stormwater volume is being maintained.
Advantages of infiltration basins are that they preserve
the natural water balance of a site, can serve larger
developments, and can be integrated into a site's land-
scaping and open spaces. Disadvantages of infiltration
basins can include their land area; fairly high rate of
failure due to unsuitable soils, poor construction, or lack
of maintenance; the need for frequent maintenance; and
possible nuisances such as odors, mosquitos, or soggy
ground (all signs of a failing system).
The function of infiltration basins can be improved if the
following design tips are followed:
Basin floor and sides: The rate and quantity of infil-
tration are enhanced by increasing the surface area
of the bottom. Large, relatively shallow areas are
preferable, especially in KSAs, so that the stormwater
spreads evenly over the entire surface area. There-
fore, it is very important that the bottom be evenly
graded with a zero slope. If the bottom is uneven,
these low spots will remain underwater for a longer
time and may become chronically wet as the floor
clogs and infiltration is reduced. Side slopes should
be no steeper than 3:1 to allow for vegetative stabi-
lization, easier mowing and access, and better public
safety.
Vegetation: The side slopes and bottoms of infiltra-
tion areas should be vegetated with a dense turf of
water-tolerant grass immediately after construction.
Not only does the vegetation stabilize these areas,
but it also helps to filter stormwater pollutants, re-
move dissolved nutrients and metals, enhance aes-
thetic qualities, reduce maintenance needs, and even
maintain or improve infiltration rates.
Reducing incoming water velocities: Inlets to an in-
filtration area should be stabilized to prevent inflowing
runoff velocities from reaching erosive levels and
scouring the bottom. Riprapping inlet channels or
pipe outfalls and using bubble-up inflow devices or
perimeter swale and berms can address this prob-
lem. Because the stormwater should spread evenly
over the entire infiltration area, riprap inlets should
terminate in a broad apron that serves as a crude
level spreader.
Construction requirements: Proper construction and
routine maintenance as discussed above are essen-
tial for successful infiltration basin implementation. In
a recent survey, approximately 40 percent of the in-
filtration basins had partially or totally clogged within
their first few years of operation (10). Many of the
systems failed almost immediately after construction
or never worked properly from the beginning.
Routine maintenance requirements: Infiltration areas
should be inspected following major storms, espe-
cially in the first few months after construction. If
stormwater remains in the system beyond the design
drawdown time (typically 24 to 36 hours if grassed,
48 to 72 hours if not grassed), either the infiltration
capacity was overestimated or maintenance is
needed. Factors responsible for clogging may include
upland erosion and sedimentation, low spots, exces-
sive compaction, or poor soils. Cleaning frequently
depends on whether the basin is vegetated or non-
vegetated and is a function of storage capacity, sedi-
ment and debris load, and land use. Litter, leaves,
brush, and other debris should be removed regularly,
perhaps during the mowing of vegetation. The buffer,
side slopes and bottom of the retention area should
be mowed as needed, with the grass clippings re-
moved. Eroded or barren areas should be immedi-
ately revegetated. Nonvegetated basins can be tilled
annually after accumulated sediments are removed.
Sediments should be removed only after the basin is
thoroughly dry, preferably to the point where the top
357
-------
layer begins to crack. To reduce soil compaction, only
light equipment should be used.
Nonroutine maintenance requirements: Over time,
the original infiltration capacity of the bottom will
gradually decline. Deep tilling every 5 to 10 years can
be used to break up clogged surface layers, followed
by regrading, leveling, and revegetation. If the origi-
nal infiltration rate was overestimated, underdrains
may be installed beneath the bottom, or perhaps the
system should be converted to a shallow marsh or
wet detention system.
Infiltration Trenches
An infiltration trench generally consists of a long, narrow
excavation, ranging from 1 to 3 m in depth, that is back-
filled with stone aggregate, allowing for the temporary
storage of the first-flush stormwater in the voids between
the aggregate material. Stored runoff then infiltrates into
the surrounding soil. To minimize clogging potential and
maximize treatment effectiveness, infiltration trenches
should always be designed as off-line systems. Infiltra-
tion trenches usually are designed to serve drainage
areas of 2 to 4 hectares and are especially appropriate
in urban areas where land costs are prohibitive. As with
any infiltration practice, the treatment train concept must
be employed to capture sediment before it enters the
trench to minimize and reduce clogging.
Advantages of infiltration trenches include ground-water
recharge, reduced stormwater volume, and the ability to
fit into perimeters or other underused areas of a devel-
opment, even beneath parking areas. Disadvantages
include potential clogging, especially if sediment is not
kept out during construction, the need for careful design
and construction, and maintenance.
Infiltration trenches can be located on the surface or below
the ground. Surface trenches receive sheet flow runoff
directly from adjacent areas after it has been filtered by
a grass buffer. Underground trenches can accept runoff
from storm sewers but require use of special pretreat-
ment inlets to prevent coarse sediment, soils, leaves,
and greases from clogging the stone reservoir.
Surface trenches typically are used in residential areas
where smaller loads of sediment and oil can be trapped
by grass filter strips that are at least 6 m wide. While
surface trenches may be more susceptible to sediment
accumulations, their accessibility makes them easier to
maintain. Surface trenches can be used in highway
medians, parking lots, and narrow landscape areas.
Underground trenches can be applied in many develop-
ment situations and are particularly suited to accept
concentrated runoff; however, pretreatment is essential.
Inlets to underground trenches should include trash racks,
catch basins, and baffles to reduce sediment, leaves,
debris, and oil and grease. Maintenance of underground
trenches can be very difficult and expensive, especially
if placed beneath parking areas or pavement.
The most commonly used underground trench is an
exfiltration system, in which the stormwater treatment
volume is diverted into an oversized perforated pipe
placed within an aggregate envelope. The first-flush storm-
water is stored in the pipe and exfiltrates out of the holes,
through the gravel and filter fabric, and into the surround-
ing soil. The city of Orlando, Florida, has installed exfil-
tration systems using perforated corrugated metal pipe
and slotted concrete pipe throughout the downtown area
to reduce stormwater pollution of its lakes.
Dry wells are used extensively in Maryland to store and
infiltrate runoff from rooftops. The downspout from the
roof gutter is extended into an underground trench,
which is constructed at least 3 m away from the building
foundation. Rooftop gutter screens are used to trap
particles, leaves, and other debris. Additional design
information on dry wells is available from the Maryland
Department of the Environment (11).
The following design and construction guidelines are
provided for infiltration trenches.
Infiltration Rates
The actual rate at which water leaves the infiltration
trench is determined by several factors. Whether infiltra-
tion primarily occurs through the trench bottom or sides
depends on the elevation of the water table and soil
properties. To prevent ground-water contamination,
trench bottoms should be at least 4 ft above the sea-
sonal high water table (remember to consider ground-
water mounding). This will also ensure infiltration
through the bottom. In addition to the infiltration rate of
the parent soil, the permeability of the surrounding filter
fabric (if used) is crucial and can become a limiting
factor. A recent investigation of exfiltration systems (12)
provides the following:
Permeability of the parent soil is not the limiting ex-
filtration rate.
The limiting exfiltration rate is set by the geotextile
filter fabric, not the soil.
A maximum rate of 1.27 cm/hr should be used, as-
suming infiltration through the sides and bottom.
A maximum rate of 2.54 cm/hr should be used if the
geotextile filter fabric is matched correctly to the soil
type and only the trench side areas are assumed to
exfiltrate.
Construction of Infiltration Trenches
Successful use of infiltration trenches requires thorough
site planning and evaluation and proper construction. In
addition to the construction recommendations for all
358
-------
infiltration practices discussed above, the construction
of infiltration trenches should also include the following:
Excellent erosion and sediment control should be main-
tained during construction to keep sediments away
from the trench. Allowing even an inch or two of soil
to get into the trench between the aggregate and the
fabric will almost ensure clogging. If constructed be-
fore the drainage area is entirely stabilized, then the
trench should be covered with heavy plastic to pre-
vent any inflow until stabilization is completed.
The trench should be excavated using a backhoe or
trencher equipped with tracks or oversized tires. Nor-
mal rubber tires should be avoided because they
compact the subsoil and may reduce infiltration ca-
pability. For the same reason, the use of bulldozers
or front-end loaders should be avoided. Excavated
material should be stored at least 3 m from the trench
to avoid backsliding and cave-ins.
Once the trench is excavated, the bottom and sides
should be lined with a geotextile filter fabric to prevent
upward piping of underlying soils. The fabric should
be placed flush with the sides and bottom, with a
generous overlap at the seams. Care should betaken
in selecting the proper kind of filter fa brie, as available
brands differ significantly in their permeability and
strength. The geotextile fabric must be handled care-
fully to prevent holes and tears that allow soil to get
into the trench. As an alternative, a 15-cm deep filter
of clean, washed sand may be substituted for filter
fabric on the bottom of the trench.
Clean, washed 2.5- to 7.5-cm stone aggregate
should be placed in the excavated reservoir in lifts
and lightly compacted with plate compactors to form
the coarse base. Unwashed stone has enough asso-
ciated sediment to pose a risk of clogging at the
soil/filter cloth interface. Where possible, the use of
limestone or bluestone aggregate should be avoided.
A simple observation well should be installed in every
trench. Wells can be made of secure foot plate, per-
forated polyvinyl chloride pipe, and locking cover. The
observation well is needed to monitor the performance
of the trench and is also useful in marking its location.
The drain time for a trench can be measured by placing
a graduated dipstick down the well immediately after
a storm and again 24, 48, and 72 hours later.
Postconstruction sediment control is critical. It is
therefore important that 1) sediment and erosion con-
trols be inspected to make sure they still work, 2)
vegetated buffer strips are established immediately,
preferably by sodding, and 3) if hydroseeding is used,
reinforced silt fences are placed between the buffer
and trench to prevent sediment entry before the buff-
er becomes fully established.
Maintenance of Infiltration Trenches
If properly constructed with pretreatment practices to
prevent heavy sediment loading, infiltration trenches
can provide stormwater benefits without tremendous
maintenance needs. Because trenches are "out of sight,
out of mind," getting property owners to maintain them
can be difficult. Accordingly, a public commitment for
regular inspection of privately owned trenches is essen-
tial, as is a legally binding maintenance agreement and
education of owners about the function and mainte-
nance needs of trenches.
Trenches should be inspected frequently within the first
few months of operation and regularly thereafter. In-
spections should be done after large storms to check for
water ponding, with water levels in the observation well
recorded over several days to check drawdown. Grass
buffer strips should maintain a dense, vigorous growth
of grass and receive regular mowing (with bagging of
grass clippings) as needed. Pretreatment devices
should be checked periodically and sediment removed
when the sediment reduces available capacity by more
than 10 percent.
Swales
Swales, or grassed waterways, are one of the oldest
stormwater BMPs, having been used along streets and
highways and by the farmer for many years. By defini-
tion, a swale is a shallow trench that:
Has side slopes flatter than 3 ft horizontal to 1 ft
vertical.
Contains contiguous areas of standing or flowing
water only following a rainfall.
Is planted with or has stabilized vegetation suitable
for soil stabilization, stormwater treatment, and nutri-
ent uptake.
Is designed to take into account the soil erodability,
soil percolation, slope, slope length, and drainage
areas so as to prevent erosion and reduce stormwa-
ter pollutants.
Traditionally, swales have been and are used primarily
for stormwater conveyance; as such, they are consid-
ered an on-line practice. The removal of stormwater
pollutants by swales can occur by either infiltration or
vegetative filtration and uptake. Investigations in Florida
(13, 14) 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. To achieve Florida's performance standards,
swales must be designed to infiltrate the runoff from a
3-yr/1-hr storm (about 7.5 cm) within 72 hours. Investi-
gations in Washington state (15, 16), however, indicate
that swales can also act as a biofilter, with removal of
particulate pollutants without infiltration of stormwater.
359
-------
Avellaneda (9) developed the following equation for a
triangular shaped swale to estimate the length of swale
necessary to infiltrate the design runoff volume:
L =
KQ5/8 S3/16
n3/8i
(1-1)
where:
L = swale length (m)
n = Mannings roughness coefficient
Q = average runoff flow rate (m3/sec)
i = infiltration rate (cm/hr)
S = longitudinal slope (m/m)
K = constant that is a function of side slope
(see Table 2)
For most residential, commercial, and highway projects,
the length of swales necessary to percolate the storm-
water needed to achieve the 80 percent 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 vege-
tative filtration. By using a raised stormsewer inlet, swales
can provide water quality enhancement via retention
and still serve as effective conveyances for flood protec-
tion. Swales can incorporate retention by using swale
blocks, small check dams, or elevated driveway cul-
verts to create storage, thereby reducing runoff velocity,
reducing erosion, and promoting infiltration.
Using the runoff from 7.5 cm of rainfall as a design
treatment volume, equations have been developed for
swale block designs to store and infiltrate the runoff (17).
Table 2. Constant (K) for Design Equation for Triangular
Shape
Z (Side Slope)
1 Vertical
Z Horizontal
1
2
3
4
5
6
7
8
9
10
K
(U.S. Units)
10,516
9,600
8,446
7,514
6,784
6,203
5,730
5,337
5,006
4,722
K
(SI Units)
75,552
68,971
60,680
53,984
48,740
44,565
41,167
38,344
35,966
33,925
The swale block volume can be calculated for a fixed
length of swale using:
Volume of runoff- volume infiltrated = swale block volume
Q (At)- QJ (At ) = swale block volume
Q (A t) - Ln3/8' (A t) = swale block volume (1-2)
[ K S/16 J
where
QI = average infiltration rate (m3/sec)
A t = runoff hydrograph time (sec)
Wanielista and Yousef (18) present the following example
problem using Equations 1 and 2 for designing a swale
with cross blocks to satisfy a specific water quality goal:
Given
n = 0.05
i = 7.5 cm/hr
S = 0.0279
z = 7
QI = 0.0023m3/sec for A t = 100 min
what swale length would be necessary to percolate
all the runoff?
Using Equation 1,
41,167 (0.0023)5/8 (0.0279)3/16
L =
(0.05)/8 7.5
= 193 meters.
If only 76 m is available, how much storage volume is
necessary?
Using Equation 2,
(0.0023)(60)(100)-
(76)(0.05)/8(7.5)
41,167(0.0279)3/16
60(100) = volume,
and the volume of storage is equal to 10.7 m3.
In highway designs for high-speed situations, safety
must be considered; thus, a maximum depth of water
equal to 0.5 m (about 1.5 ft) and flow line slopes on the
berms of 1 vertical/20 horizontal are recommended.
Along lower speed highways or in some residential/com-
mercial urban settings, steeper flow line berm slopes
(1/6) are acceptable.
The studies of swales in Washington state resulted in
the following recommendations to improve water quality
benefits (15):
360
-------
Maximum design velocity should not exceed 27 cm/sec.
A hydraulic residence time of at least 9 min is rec-
ommended for removal of about 80 percent of the
total suspended solids. Longer residence times will
provide higher removal effectiveness.
Swale width should be limited to 2 to 2.5 m unless
special measures are provided to ensure a level swale
bottom, uniform flow spreading, and management of
flows to prevent formation of low-flow channels.
Swale slopes should be between 2 and 4 percent.
Water depth should be limited to no greater than
one half the height of the grass, up to a maximum
of 7.5 cm.
Swale length will be a function of the hydraulic resi-
dence time, swale width, and stormwater volume
and velocity.
Porous 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 re-
viewed carefully to ensure that they are necessary (Is
paving really required in every case?) and that the num-
ber of spaces is related to actual traffic demands. After
these requirements have been reviewed and verified,
the use of porous pavement within a parking lot should
be examined. Porous pavement materials include po-
rous asphalt, porous concrete, turf blocks, and even
Geoweb covered with sod.
Overall, experiences with porous pavements have not
been very good. Porous 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. Field investigations
of porous concrete that has been in use for up to 15
years in Florida, however, indicate that these parking
lots can continue to infiltrate rainfall and runoff if they
were installed and maintained properly (19). Recom-
mendations to improve the utility of porous pavements
include the following:
Be sure that the installer is properly trained in the
design, mixing, installation, and finishing of the po-
rous pavement material. Both porous asphalt and
concrete must be mixed and installed much differ-
ently than regular asphalt or concrete.
Exemplary erosion and sediment control during con-
struction and complete site stabilization after con-
struction are essential to prevent clogging of the void
spaces within the porous pavement.
The porous pavement must receive regular, routine
vacuuming to remove accumulating solids. At times,
nonroutine maintenance may involve cleaning with
high-pressure water.
The entrance to any porous pavement area should
have a large sign warning those about to enter that
porous pavement is in use. Precautions should in-
clude prohibiting vehicles with large amounts of soil
on their tires.
Problems Associated With Infiltration
Practices
There have been several concerns regarding the use of
infiltration practices, including their propensity to fail,
their potential effects on ground-water quality, and their
need for maintenance.
Infiltration systems seem to have a very high rate of
failure. The author believes, however, that this high
failure rate is a reflection of improperly estimated infil-
tration rates and improper erosion and sediment control
during the construction process. A 1990 field survey of
stormwater infiltration facilities constructed in Maryland
replicated a 1986 field survey, thereby providing data on
the performance of infiltration practices after they have
been in operation for several years (20). Table 3 sum-
marizes the information from this project.
From Table 3 it can be seen that the overall condition
and functioning of infiltration systems declined over
time. In 1986, about two-thirds of all facilities were func-
tioning as designed, while in 1990 only about half were.
Only 42 percent of the facilities were functioning as
designed in both 1986 and 1990, while about 27 percent
were not functioning as designed in both years. About
24 percent of the systems were functioning in 1986 but
not in 1990, while only 7 percent of those not working in
1986 were working in 1990. Maintenance was needed
at more facilities in 1990 (66 percent) than in 1986
(45 percent). Additionally, many facilities (38 percent)
that needed maintenance in 1986 still needed mainte-
nance in 1990, while 32 percent of the facilities that
did not need maintenance in 1986 did need it in 1990.
Only 10 percent of the systems that needed mainte-
nance in 1986 did not need maintenance in 1990. These
data indicate that little effort is expended on maintain-
ing the operational capabilities of stormwater manage-
ment systems.
A second concern about infiltration practices is whether
they simply are transferring the stormwater pollution
problem from surface waters to ground waters. Harper
(14) has shown that stormwater pollutants, especially
heavy metals, quickly bind to soil particles, while vege-
tation is effective in filtering pollutants, thereby minimiz-
ing the risk of ground-water contamination. Ground
water beneath swales and retention areas located in
361
-------
Table 3. Comparison of the Operation of Maryland Infiltration Practices
Type of BMP
Basins
Trenches
Dry wells
Porous pavement
Swales
Totals
1986 Number
of Sites
63
94
30
14
6
207
1986 Number of
Sites Working
30 (48%)
75 (80%)
23 (77%)
7 (50%)
3 (50%)
138 (67%)
1990 Number
of Sites
48
88
25
13
3
177
1990 Number of
Sites Working
18(38%)
47 (53%)
1 8 (72%)
2 (15%)
2 (67%)
87 (49%)
highly sandy soils with low organic content and sparse
vegetative cover, however, did show elevated levels of
heavy metals down to depths of 20 ft (21).
References
1. Livingston, E.H. 1988. State perspectives on water quality crite-
ria. In: Design of urban runoff controls, an Engineering Founda-
tion conference. Potosi, MO.
2. Yousef, Y.A., M.P. Wanielista, and H.H. Harper. 1986. Fate of
pollutants in retention/detention ponds. In: Stormwater manage-
ment: An update. Publication 85-1. Orlando, FL: University of
Central Florida (July), pp. 259-275.
3. Miller, R.A. 1985. Percentage entrainment of constituent loads in
urban runoff, South Florida. U.S. Geological Survey Document
No. WRI 84-4329.
4. Livingston, E.H., and E.M. McCarron. 1992. Stormwater manage-
ment: A guide for Floridians. Fl. Tallahassee, FL: Department of
Environmental Regulation.
5. St. Johns River Water Management District. 1992. Rule 40C-42
and associated applicant's handbook. Palatka, FL.
6. Wanielista, M.P. 1977. Quality considerations in the design of hold-
ing ponds. Presented at the Stormwater Retention/Detention Basins
Seminar, University of Central Florida, Orlando, FL (August).
7. Wanielista, M.P, Y.A. Yousef, G.M. Harper, T.R. Lineback, and L.
Dansereau. 1991. Precipitation, interevent dry periods, and
reuse design curves for selected areas of Florida. Final report.
Contract WM-346. Submitted to the Florida Department of Envi-
ronmental Regulation, Tallahassee, FL.
8. Shaver, H.E. 1986. Infiltration as a Stormwater management com-
ponent. In: Proceedings of the Urban Runoff Technology Engineer-
ing Foundation Conference, Henniker, NH (June), pp. 270-280.
9. Avellaneda, E. 1985. Hydrologic design of swales. M.S. thesis.
University of Central Florida, Orlando, FL (December).
10. Sediment and Stormwater Division. 1986. Maintenance of storm-
water management structures: A departmental summary. Annapolis,
MD: Department of Natural Resources, Water Resources Admini-
stration (July).
11. Sediment and Stormwater Administration. 1984. Standards and
specifications for infiltration practices. Baltimore, MD: Maryland
Department of the Environment.
12. Wanielista, M.P, M.J. Gauthier, and D.L. Evans. 1991. Design
and performance of exfiltration systems. Final report. Contract
C3331. Submitted to the Florida Department of Transportation,
Tallahassee, FL.
13. Yousef, Y.A., M.P. Wanielista, H.H. Harper, D.B. Pearce, and
R.D. Tolbert. 1985. Best management practices: Removal of
highway contaminants by roadside swales. Final report. Contract
99700. Submitted to the Florida Department of Transportation,
Tallahassee, FL.
14. Harper, H.H. 1985. Fate of heavy metals from highway runoff in
Stormwater management systems. Ph.D. dissertation. University
of Central Florida, Orlando, FL.
15. Horner, R.R. 1988. Biofiltration systems for storm runoff water
quality control. Final report. Submitted to City of Seattle, Wash-
ington.
16. Water Pollution Control Department. 1992. Biofiltration swale per-
formance, recommendations, and design considerations. Publi-
cation 657. Seattle, WA.
17. Wanielista, M.P, Y.A. Yousef, L.M. Van DeGroaff, and S.H.
Rehmann-Koo. 1986. Best management practices: Enhanced
erosion and sediment control using swale blocks. Final report.
FL-ER-35-87. Submitted to the Florida Department of Transpor-
tation, Tallahassee, FL (September).
18. Wanielista, M.P, and Y.A. Yousef. 1986. Best management prac-
tices overview. In: Proceedings of the Urban Runoff Technology
Engineering Foundation Conference, Henniker, NH (June), pp.
314-322.
19. Florida Concrete Products Association. 1991. Pervious pavement
manual and field performance tests. Orlando, FL.
20. Lindsey, G., L. Roberts, and W Page. 1992. Inspection and
maintenance of infiltration facilities. J. Soil Water Conserv.
47(6):481-186.
21. Harper, H.H. 1988. Effects of Stormwater management systems
on ground water quality. Final report. Contract WM-190. Submit-
ted to the Florida Department of Environmental Regulation, Tal-
lahassee, FL.
362
-------
Stormwater Reuse: An Alternative Method of Infiltration
Marty Wanielista
University of Central Florida, Orlando, Florida
Abstract
Runoff water stored in a wet detention pond can be an
asset if it is used to recharge surficial aquifer levels. The
recharge can occur directly from the pond by infiltrating
the detained water, or the detained water can be irri-
gated over the watershed. Reuse in the watershed or
infiltration at the pond lessens the quantity of water
discharged, thus reducing the pollutant mass dis-
charged to surface waters. A benefit of irrigation is a
reduction in the use of potable water otherwise used
for irrigation.
A mass balance on pond storage volume using rainfall
data for select areas in the southeastern United States
was completed to determine the percentage of storm-
water runoff that can potentially be irrigated or infiltrated
for each area as a function of contributing area, runoff
coefficient, volume of temporary storage, and irrigation
rate. Design curves were developed that relate the effi-
ciency (E), or the percentage of runoff that is irrigated
on a yearly basis, to the volume of temporary storage
(V) in a reuse pond and the rate of irrigation (R). The
design curves, called REV curves, permit the selection
of a temporary storage volume and irrigation rate for a
given efficiency, runoff coefficient, and geographic area.
This paper contains example REV curves and presents
simplified uses of the results.
Introduction
The pollutants associated with stormwater and the volume
of stormwater discharges can result in significant impacts
to the natural and manufactured environments of any
watershed. As watersheds are made more impervious due
to paving and other construction activities, the volume of
runoff and pollutant mass discharged to surface waters
increases relative to predeveloped conditions.
Potential impacts from uncontrolled runoff are loss of
freshwater from an area where the rainfall occurred,
additional fresh water discharges to estuaries, increased
pollutant mass loadings, decreased river base flows,
reduced wetland areas, and an economic loss associ-
ated with the need to replace discharged freshwater with
potable or other waters.
Water policy in the state of Florida requires a perform-
ance standard for all stormwater management methods.
A stormwater pollutant annual average load reduction of
80 percent for discharges to most waters and of 95
percent for those discharging into outstanding Florida
waters (1) are required. Of the currently used stormwa-
ter management methods, off-line retention and chemi-
cal treatment can achieve the stated pollutant removal
efficiencies. Wet detention ponds that discharge to ad-
jacent surface waters, however, do not. If some of the
detained water can be used within the watershed and
not discharged to surface waters, the wet detention
ponds may also meet the standards.
A Stormwater Reuse Pond
A stormwater reuse pond is proposed to retain runoff
water within a watershed and to reduce the mass of
pollutants in the discharges to surface water bodies. The
difference between a wet detention pond and a reuse
pond is the operation of the temporary storage volume.
A wet detention pond is designed to discharge the
runoff water and possibly some ground water to adja-
cent surface waters, while a reuse pond is designed to
reuse a specific fraction of the runoff volume and not
discharge that fraction. In this paper, mathematical rela-
tionships are developed between the reuse volume
(temporary storage volume), the rate at which stormwa-
ter is reused, and the percentage of annual surface
runoff that is reused.
The traditional design of pond temporary storage vol-
ume for a wet detention pond has been based on the
consideration of water quality and uses a design storm.
The design storm, however, usually ignores the preced-
ing rainfall record and assumes that there is an antece-
dent dry period long enough to ensure that the pond is
at some control elevation. The usual assumption is a
zero temporary storage.
363
-------
To address the sensitivity of the temporary storage vol-
ume to interevent dry periods, long-term rainfall records
were used from 25 Florida and seven other southern
states' rainfall stations in a model that simulates the
behavior of a reuse pond overtime. A spreadsheet was
used to build a 15-year mass balance for a pond. After
each rainfall event, surface runoff and reuse volumes
were respectively added to and subtracted from the
previous pond storage volume. If the temporary storage
volume exceeded the available storage volume, dis-
charge occurred. If the temporary storage volume was
less than zero (the permanent pool volume was used for
reuse water), supplemental water was used to replenish
the pond and maintain the permanent pool. Both the rate
of reuse from the pond and the reuse volume were
varied. The reuse efficiency, defined as one minus the
total volume of surface discharge divided by the total
volume of runoff times 100, was calculated for each
combination.
Simulation of a Reuse Pond
To establish a relationship between the efficiency, the
reuse rate, and the reuse volume of a pond, a continu-
ous time model was used to simulate the dynamics of a
reuse pond. Continuous models are reported to be most
representative (2). The efficiency of the pond, or the
percentage of runoff that is reused, was calculated for
different reuse volumes and reuse rates. Charts for
different regions were produced using the local rainfall
records of these regions. The term "model" is used to
refer to the basic unchanged equation of the mass
balance in which different rainfall records were inserted
and reuse volumes and reuse rates were varied. "Simu-
lation" is used to refer to the complete calculations of the
model in which volume and rate were defined. There is
only one model, while many simulations were done.
Figure 1 depicts a cross section of a typical reuse pond.
The sediment storage volume lies at the bottom to re-
ceive settled matter. Above this is the permanent pool
volume, which provides a minimum residence time for
stormwater. The reuse volume (temporary storage vol-
ume) is the volume above the permanent pool and
below the flood control structure. The flood control vol-
ume would typically be above the reuse volume.
The reuse pond differs from a typical detention pond in
that instead of the temporary storage volume being
depleted by a surface water discharge device (such as
a bleed-down orifice in an outlet pipe), it is drawn down
by a reuse system and is thus called the reuse volume.
A reuse pond may deplete the pond volume below the
permanent pool boundary requiring a supplemental vol-
ume to maintain this volume. Adischarge structure is still
necessary for flood control. Common practice should be
used for the design of sediment storage, permanent
pool, and flood control volumes, and their elevations and
side slopes. This paper provides methodology and de-
sign criteria for the reuse volume only.
The water level of a typical reuse pond fluctuates during
a year. During and following a rainfall event, there is
runoff into the pond, and the water level rises to some
depth above the permanent pool. If this new water level
exceeds the level of the surface discharge control, dis-
charge will occur at some rate until the water level drops
back to the elevation of the control structure. The reuse
pond volume is incremented daily, removing an amount
of water for reuse. If the reuse volume is expended,
supplemental water, such as ground water, may be used
to maintain the permanent pool volume. This could
occur as seepage through the sides of the pond or by
mechanical pumping. This scenario was simulated by
creating a mass balance for pond operation.
Surface Water Discharge
Control Elevation >
Seasonal High Water Table
\
Reuse Control
Flood Control Volume* i \
Reuse Volume
Littoral Shelf*
T
Permanent Pool Volume
Shallow
Slope
Steep Slope
Sediment Storage
Volume
* Can be measured above permanent pool; however, some
regulatory agencies measure above the reuse volume.
** The reader should consult local water management districts
and other regulatory agencies to determine specific geometric
and littoral zone design requirements.
Figure 1. Schematic of a stormwater reuse pond.
364
-------
The Model
The model is based on the continuity equation
INPUTS-OUTPUTS = AS.
(Eq. 1)
If all potential water movements are considered, a complete
hydrologic balance may be expressed in volume units as
RE +G + P + F - R - D - ET = AS,
where
(Eq. 2)
RE = rainfall excess or runoff volume
G = supplemental water (ground water)
P = precipitation directly on the pond
F = water movement through the sides of the pond
R = reuse (infiltration)
D = discharge
ET = evapotranspiration
S = storage in pond
In Florida, the average evapotranspiration rate for a pond
is generally equal to the average precipitation on the pond
in a 1-year period (approximately 50 in.). Additionally,
evaporation data are only available in mean monthly rates
compared with the daily time step of the model, making
the estimate of evaporation potentially inaccurate.
These parameters were dropped from the mass bal-
ance. Also, because of its complexity, the flow of ground
water through the sides of the pond was assumed to
equal zero, and Equation 2 was further simplified to
RE + G - R - D = AS.
(Eq. 3)
For Florida modeling purposes, there were two inputs,
runoff and supplement, and two outputs, reuse and
discharge (Figure 2). Runoff was established from
known precipitation and watershed data. The reuse rate
was a controlled variable. Both supplemental water and
discharge were functions of the water level of the pond,
or the storage volume. Because ground-water move-
ment was assumed to equal zero, supplemental water
is considered as that which is pumped into the pond
mechanically. Supplement occurs at a rate necessary to
maintain the permanent pool; the maximum required
rate would equal that of reuse. Because potential stor-
age capacity is being constantly eliminated by supple-
ment, this may be considered as being conservative.
With the previous simplifications, the actual pond may
be simulated by the model.
The calculations for each simulation were done using
Quattro Pro, an electronic spreadsheet. The top and
bottom calculations and input data for one simulation
can be seen in Figure 3. The columns of the upper
portion of the simulation are the incremental registers of
G
RE
^^ \
PI A
, HET
R
D
Reuse Volume
Permanent Pool
R
E
7 +G-R-D = AS
Figure 2. Summary of mass balance of reuse pond, simplified
for Florida conditions.
the various parameters, which are labeled along the top.
Each of these variables is defined as follows:
EVENT
DATE
DRY
RAIN
RUNOFF
REUSE
DISCHARGE
Potential:
A distinct rainfall occurrence; for
computational purposes, each day of
a multiday rainstorm is considered a
separate event.
The date on which an event occurs.
The dry period separating rainfall events
(days); if events occur on consecutive
days there are no dry days. This value
is not used in the basic model but is
needed for the sensitivity analysis of
the discharge potential.
The amount of rainfall recorded dur-
ing each event (inches). This infor-
mation was taken directly from
National Oceanic and Atmospheric
Administration (NOAA) rainfall data.
The amount of runoff that enters the
pond during an event (inches).
The amount of water reused during
the day of an event and the dry days
following the previous event (inches);
the rate of reuse remains constant
during a single simulation.
The potential amount of discharge for
an event (inches); the amount that could,
if necessary, physically discharge during
the time since the previous event. This
was established as 2 in./day over the
equivalent impervious area (EIA).
365
-------
ORLANDO RAINFALL STATION (Hay 1974 - Dec. 1988) Volume = 3 in. Rate = 0.2 in/day
EVENT
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
1386
1387
DATE DRY
Days
04-May-74
05 -May- 74
06-May-74
07- May- 74
08- May- 74
12-May-74
14-May-74
15-May-74
16- May- 74
17- May- 74
23 -May- 74
27-May-74
01-Jun-74
02-Jun-74
03-Jun-74
10-Jun-74
11-Jun-74
14-Jun-74
15-Jun-74
17-Jun-74
24-Jun-74
25-Jun-74
26-Jun-74
27-Jun-74
28-Jun-74
30-Jun-74
01-Jul-74
02-Jul-74
23-Dec-88
28-Dec-88
Summation:
% Discharged =
% Reused =
Inputs:
Runoff:
Supplement:
Outputs:
Reuse :
Discharge:
0
0
0
3
1
0
0
0
5
3
4
0
0
6
0
2
0
1
6
0
0
0
0
1
0
0
0
4
RAIN
In.
0
0
0
0
0
0
0
0
0
0
0
1
0
0
2
0
0
0
0
0
1
3
1
.12
.77
.04
.33
.15
.11
.46
.07
.23
.35
.06
.19
.07
.05
.19
.18
.05
.54
.09
.95
.07
.47
.89
3.36
0
0
0
0
0
706
.17
.12
.88
.04
.05
.88
RUNOFF
In.
0.12
0.77
0.04
0.33
0.15
0.11
0.46
0.07
0.23
0.35
0.06
1.19
0.07
0.05
2.19
0.18
0.05
0.54
0.09
0.95
1.07
3.47
1.89
3.36
0.17
0.12
0.88
0.04
0.05
706.88
REUSE DISCHARGE SUPLMNT
In. Poten. Actual In.
0.
0.
0.
0.
0.
0.
0.
0.
0.
1.
0.
0.
0.
2
2
2
2
8
4
2
2
2
2
8
1
2
2
1.4
0.2
0.
0.
0.
1.
0.
0.
0.
0.
0.
0.
0.
1.
6
2
4
4
2
2
2
2
4
2
2
4
1
2
2
2
2
8
4
2
2
2
12
8
10
2
2
14
2
6
2
4
14
2
2
2
2
4
2
2
14
10
1070.40
Total Discharge/Total
1 -
Runoff
=
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1.14
1.69
3.16
0
0
0
0
0.08
0.00
-0.00
0.00
0.11
0.29
0.00
0.00
0.00
0.69
0.74
0.00
0.00
0.09
0.00
0.00
-0.00
0.00
0.00
0.20
0.00
0.00
-0.00
0.00
0.00
-0.00
0.00
1.36
0.95
NET
In.
0
0.00
0.57
0.41
0.54
0.00
0.00
0.26
0.13
0.16
0.00
0.00
0.19
0.06
0.00
0.79
0.77
0.22
0.56
0.25
0.00
0.87
4.14
4.69
6.16
2.77
2.69
3.37
0.00
0.00
75.72 439.24
Total Discharge/Total Runoff =
706.88
439.24
1146.12
1070.40
75.72
in.
in.
in.
in.
in.
-
Inputs
Outputs
Storage
10.71%
89.29%
1146.12 in.
-1146.12 in.
0.00 in.
1146.12 in.
Figure 3. Example of computer model using rainfall data from Orlando, Florida.
366
-------
Actual: The amount that does discharge
during an event (inches); depends on
the water level of the pond but is
restricted to the potential discharge.
SUPLMNT The amount of water needed between
events to maintain the permanent
pool volume (inches).
NET The amount of water above the
permanent pool recorded at the end of
each event (inches).
Every day in which a rainfall event takes place repre-
sents one line in the simulation. This is the fundamental
time step of the model. All inputs and outputs occur
during this 24-hour period. At the end of the period, the
net storage value of the pond is calculated. From this
value, decisions are made concerning discharge and
supplement. The process then repeats itself.
The 15-year totals for rain, runoff, reuse, actual dis-
charge, and supplement are calculated as shown in
Figure 3. From these values, the efficiency, or the per-
centage of runoff reused, can be determined for a par-
ticular simulation. The efficiency is equal to one minus
the volume of water that is discharged divided by the
volume of runoff times 100. The percent discharged, the
volume of water discharged divided by the volume of
runoff, is also calculated. The percent reused plus the
percent discharged equals 100.
At the bottom of Figure 3 is a summary of the mass
balance for the entire record. Both the inputs and out-
puts are listed and totaled. The difference between the
inputs and outputs, labeled "Storage," is compared with
the final value for NET. The values should be identical.
This is used primarily to check the calculations.
This single model was used to predict the behavior of a
reuse pond subjected to the rainfall record of 32 different
locations in the southeastern United States. Previously,
one location in Florida was reported (4). To simulate a
pond in a particular region, the rainfall record of that
region was inserted into the DATE and RAIN columns
of the model. The model was then lengthened or short-
ened to match the span of the rainfall record. Otherwise,
no changes were made to the model. By using one
model and varying only the rainfall record, the consis-
tency of the simulations was assured.
Length of Rainfall Record
An investigative question that arises when examining
the random behavior of rainfall is how large a record
must be to accurately represent the meteorological
characteristics of a region. In other words, how many
years of rainfall data must be used to estimate the
ultimate dynamics of the pond? Obviously, the greatest
accuracy can be obtained by using the most data. But
the incremental benefit of each additional unit of data
diminishes so that there is a point beyond which using
more is no longer reasonable. This is the limit for inves-
tigation.
Twenty-four individual simulations were run for the
Moore Haven and Tallahassee stations using, first, 1
year of rainfall data (1988) and then incrementally add-
ing the next previous year to the rainfall record. The
yearly efficiencies for several combinations of reuse
volumes and reuse rates were recorded. As expected
with only a few years of data, the average yearly effi-
ciencies fluctuated widely but then leveled out as more
years of data were added. As the size of the database
increased, each additional year had less impact. Be-
yond 15 years, there was very little change in the aver-
age annual estimate.
Volume Units
Runoff, discharge, reuse, supplement, and net storage
are volumes of water that are expressed in units of
inches. Volumes are commonly expressed as inches
over a defined area and, likewise, the parameters of this
model are based on a variable unit area that the user
defines. Rates are merely volumes delivered over a
period and thus can be expressed in the same manner.
This unit area is the ElAof the watershed or the product
of the runoff coefficient and the contributing watershed
area. The volumetric unit of inches on the EIA is a way
in which the results are generalized for any runoff coef-
ficient and contributing area. Once the EIA is known, the
values can be converted to more practical units using
simple conversions.
Model Output
The basic function of the model is to determine a rela-
tionship between the reuse rate, the reuse volume, and
the efficiency. This was done by varying the reuse rate
and the reuse volume, then calculating the efficiency.
Thus, a simulation was done for each combination of
reuse rate and reuse volume. The reuse volumes con-
sidered varied between 0.25 and 7.0 in. on the EIA. The
reuse rates varied between 0.04 and 0.30 in./day on an
area equivalent to the EIA. The respective efficiencies
are shown as fractions. The results are presented in
chart form as shown in Figure 4. The ultimate functional
product of the reuse pond model is the rate-efficiency-
volume (REV) chart. Wanielista et al. (5) presents the
REV charts for all of the 25 locations in Florida for which
accurate and long-term rainfall data were available. In-
dividual REV charts are specific to geographical regions
with similar meteorological characteristics.
367
-------
0.04
0.5
Orlando Rainfall Station
May 1974 - Dec. 1988
Mean Annual Rainfall = 48.2 in.
Figure 4. REV chart for Orlando, Florida.
Use of the REV Charts
1.5 2 2.5 3 3.5 4 4.5 5 5.5
Reuse Volume (inches on equivalent impervious area)
6.5 7
Examples of Direct Use
REV charts relate the reuse rate, the efficiency, and the
reuse volume of a pond. Recommended irrigation rates
for Florida are between 0.38 in./week in the winter to
2.25 in./week in the summer (6). Information concerning
any two of these three variables is necessary for the
determination of the third. The use of a REV chart
requires an understanding of the concept of the EIA. The
units of both the reuse rate and the reuse volume are
based on this area. A REV chart is specific for an area,
and the accuracy of the predictions are related to the
accuracy of the input data. The REV charts of this paper
have been placed in a computer program that reduces
the possibility of calculation errors (7).
The efficiency is defined as the average percentage of
runoff that is reused over a period, specifically 15 years.
A pond that discharges to surface waters 10 percent of
the runoff that flows into it must reuse the remaining and
so is 90 percent efficient. It may sometimes be desirable
to determine the efficiency of an existing pond. More
often it will be necessary to achieve a required efficiency
established by local regulations, thus making the effi-
ciency one of the known values. On every REV chart,
there is a curve for each of the following efficiency levels
(in percentage): 50, 60, 70, 80, 90, and 95.
Example 1
A watershed in Orlando must reuse 80 percent of the
annual runoff from a 10-acre impervious area. The pond
area is included in the impervious area. The maximum
reuse storage volume available for the pond is equal to
the runoff from a 3-in. rainfall event. At what rate must
the runoff be reused?
Because the entire watershed is impervious, the EIA is
equal to 10 acres. Because runoff equals rainfall on
impervious areas, the storage volume is equal to 3 in.
on the EIA. The reuse rate is a function of the efficiency
and the reuse volume:
R = f (E,V)
= f (80%, 3 in.)
= 0.152 in./day
By referring to the Orlando REV chart (Figure 4), the
necessary reuse rate is estimated at 0.152 in./day on
the EIA. The rate and volume can be expressed in other
units:
368
-------
V=3 in. x ElAx
10 ac
EIA
43,560 ft2 ft
= 30 ac-m. x x
109,000 ft3
ac
12 in.
From the REV chart for Tallahassee (Figure 5), the
required reuse volume is determined to be 3.5 in. on
the EIA:
V = f (E,R)
= f (90%, 0.26 in./day)
= 3.5 in.
and
day
EIA
H Co ac-'n- 43,560 ft2 ft
= 1.52 ; x ! x
= 5,520
day
ft3
day.
ac
12 in.
Again, the volume and rate can be expressed in other
units:
V=3.5in.xEIAx
4 ac
EIA
43,560 ft2 ft
= 14 ac-m. x x
= 50,800 ft3
ac
12 in.
Example 2
and
An apartment complex located in Tallahassee needs to
reuse 90 percent of the runoff from its parking lots. The
EIA is equal to the directly connected impervious area
and is 4 acres. The complex wants to use 0.26 in. of
water per day over the EIA. What must the reuse volume
be to maintain these conditions?
R = 0.260
-
day
EIA
nA ac-m.
1.04 x
EIA
43,560 ft2
= 3,780
day
ft3
day.
ac
12 in.
Percentage of Runoff Reused
0.04
0 0.5
1.5 2 2.5 3 3.5 4 4.5 5 5.5
Reuse Volume (inches on equivalent impervious area)
Tallahassee Rainfall Section
Jan. 1974-Dec. 1988
Mean Annual Rainfall = 64.3 in.
Figure 5. REV chart for Tallahassee, Florida.
369
-------
The previous examples illustrate the most simple appli-
cation: the watershed being impervious and the volume
and rate given in terms of the EIA. Much more complex
design problems, however, can be solved using the
same technique. The following steps can be used in any
design situation:
1. Select the appropriate chart.
2. Compute the EIA of the watershed (EIA = contribut-
ing area x effective C).
3. Determine known variables in terms of the EIA.
4. Reference the chart to obtain a solution.
5. Convert the answer to desired units.
Evaporation and Rainfall on Pond
One of the initial simplifications of the pond mass balance
was the assumption that the mean annual evaporation
from the pond is equal to the mean annual rainfall on the
pond. The evaporation totals in the Southeast may
range from 30 to over 60 in./yr. Precipitation rates range
from 37 in./yr in Key West to 64.5 in./yr in Tallahassee.
While evaporation and direct rainfall rates are based on
the size of the pond, all other model parameters were
based on the EIA. Therefore, a ratio was established
between the size of the pond and the EIA. Because
detention ponds usually require no more than 5 percent
of the total area of the watershed, depending on the
impervious area, a conservative estimate of pond area
to a completely impervious area was chosen as 1:10. As
an example, a 1-in. rainfall event, through direct precipi-
tation, would add 1 in. of rainfall to the pond or 0.10 in.
over the EIA.
Evaporation data were obtained from NOAA Clima-
tological Data publications for the years 1985 through
1989. Because the locations of climatological stations
match those of precipitation stations in only a few in-
stances, evaporation data from nearby stations were
used with selected model locations. Evaporation data
from Lisbon and Lake Alfred were introduced into the
models of Orlando and Parrish, respectively. The evapo-
ration data were available in monthly pan evaporation
totals. Fifteen years of records were used and converted
to surface water evaporation rates by multiplying by a
pan coefficient. The mean annual total evaporation for
the two locations is 56.46 in. for Lake Alfred and 41.07
in. for Lisbon.
The evaporation function was added to the models by
distributing evaporation depths in inches for each time
interval. The amount of evaporation for each interval is the
product of the number of days in that interval and mean
daily evaporation rates for the month. To ensure the as-
sumed distribution did not affect the total evaporation vol-
ume, the mean annual evaporation volumes for the 15-year
simulations were compared with the mean volumes ob-
tained from NOAA. The totals were almost identical.
To use the REV charts, rainfall on the pond must be
included in the calculation of the EIA. When the area of the
pond (approximated at 15 percent of the EIA) was added
to the EIA, the pond reuse volume increased, and for a
fixed reuse rate the average annual efficiency increased
by at least 2.5 percent. Because rainfall on the pond
reflects an impervious condition (all rainfall yields rainfall
excess), it must be added to the EIA while maintaining
consistent units (depth on an impervious area).
Recommendations
A mathematical mass balance model can be developed
to simulate the operation of a stormwater reuse pond.
This can be done for areas that have daily rainfall data
available for a significant period, about 15 years.
The reuse of stormwater within a watershed from which
it came should be encouraged and in some areas re-
quired. Reuse ponds can be designed to conserve water
within a watershed and to reduce the mass of pollutants
entering the surface waters.
The effective impervious area for a watershed should
include the area of the pond when using the REV
curves. The effective impervious area calculation is nec-
essary for the use of the REV curves. More than one
REV curve for a location is expressed in a figure called
a REV chart.
For an average annual pollutant mass removal of 80
percent in a wet detention pond, at least 50 percent of
the runoff volume should be reused when the REV
charts are used for design. For a 95 percent annual
pollutant mass removal, at least 90 percent of the runoff
volume should be reused. The reuse percentages as-
sume a wet detention pond will remove an average 60
percent of the incoming runoff pollution mass annually
before surface discharge, which may overestimate the
actual efficiency.
The reuse of stormwater is both an environmentally and
economically sound management practice. The current
common practice is to release stormwater to adjacent
surface waters from detention ponds using weirs and
orifices. Frequently, if not all the time, this detained
volume of water is greater than the volume of water
released from the land in its natural condition. Some
fraction of this detained water can be reused within the
watershed to 1) irrigate open areas, 2) recharge ground
water, 3) supplement water used for certain industrial
purposes, 4) enhance and create wetlands, and 5) sup-
ply water for agricultural users.
Currently, the most popular reuse method has been the
irrigation of relatively open spaces, for example, golf
courses, cemeteries, recreation areas, citrus groves,
370
-------
and common areas of apartment complexes. The pri-
mary reason for these reuse systems is economics.
Many irrigation systems use treated ground water. An
alternative to the use of ground water is detained storm-
water. Treated ground water cost about $1.00/thousand
gallons. A golf course of 100 acres using treated ground
water at a cost of $1.00/thousand gallons and irrigating
at 2 in./wk would pay almost $300,000/yr for the irriga-
tion. Using detained stormwater, the irrigation system
yearly cost could be less than $40,000.
In this paper, continuous modeling for reuse ponds was
completed and was based on a mass balance using
area-specific rainfall data to develop design criteria for
stormwater reuse ponds. The design procedure relates
pond temporary storage (reuse volume) to reuse rate
and a percent reuse of the runoff water and is expressed
as a REV curve. Also, mathematical equations for the
curves have been computer coded.
The REV curves can be used for various watershed
sizes or runoff coefficients. They may be used to deter-
mine the reuse rate, the reuse volume, or the efficiency
of a pond. Supplemental water needs in a hydrologic
balance also can be estimated. The REV charts pre-
sented in this paper could facilitate the rational planning
of stormwater reuse systems.
Acknowledgments
The outcomes of this paper resulted from a research
project funded by the Florida Department of Environ-
mental Regulation in Tallahassee, Florida. The authors
also appreciate the technical assistance given to them
by Eric Livingston and John Cox of that department.
References
1. Cox, J. 1991. State water policy: Summary of Chapters 17-40,
Florida Administrative Code. Tallahassee, FL: State Department
of Environmental Regulation.
2. James, W, and M. Robinson. 1982. Continuous models essential
for detention design. Proceedings of the Conference on Stormwa-
ter Detention Facilities. American Society of Civil Engineers.
3. Wanielista, M.P. 1990. Hydrology and water quantity control. New
York, NY: John Wiley and Sons. pp. 4-5.
4. Wanielista, M.P., YA. Yousef, and G.M. Harper. 1990. Precipitation
and interevent dry periods. Kyoto, Japan: International Association
of Water Pollution Research.
5. Wanielista, M.P, YA. Yousef, G.M. Harper, and L. Dansereau.
1991. Design curves for the reuse of stormwater. Tallahassee, FL:
Florida Department of Environmental Regulation.
6. Augustin, B.J. 1991. Watering your Florida lawn. Fact Sheet No.
OH-9. Gainesville, FL: Institute of Food and Agricultural Sciences.
7. Wanielista, M.P, and YA. Yousef. 1993. Stormwater management.
New York, NY: John Wiley and Sons.
371
-------
Use of Sand Filters as an Urban Stormwater Management Practice
Earl Shaver
State of Delaware, Department of Natural Resources and Environmental Control,
Dover, Delaware
Background
As our recognition of the need for stormwater control,
from both quantity and quality perspectives, has in-
creased, efforts to develop strategies and practices to
address stormwater runoff have emerged all over the
country. Many of these efforts have been developed on
a state or local level depending on the specific issues
that motivated program development.
The concerns over stormwater control and strategies for
dealing with stormwater are now international in scope.
Society as a whole needs to learn about what individuals
have already accomplished to allow for evolution of
control strategies and individual practices. Efforts under
way at the state level (in Delaware, Florida, Maryland,
South Carolina, and Washington) and at the municipal
level (in Austin, Texas; Washington, DC; and Alexandria,
Virginia) provide some hands-on knowledge regarding
the programs and types of stormwater control practices
that have been used successfully.
The intent of this paper is to discuss stormwater control
practices, in particular, filtration systems. Experience
with stormwater control ponds and infiltration systems
has led to considerable knowledge about these meth-
ods, but interest is increasing in the use of sand filters
in several locations around the country for stormwater
treatment. Use of these systems will expand as national
efforts addressing stormwater control are implemented.
Existing Efforts in the Use of Sand Filter
Systems
The first interesting point is the way that sand filter
systems have been used historically around the country.
These systems are being used for onsite and regional
control, as well as for water quality control only and for
both water quality and water quantity control.
Austin, Texas
The city of Austin has pioneered the use of sand filters
for stormwater treatment. Other areas have experimented
over the years with sand filters, but Austin has made a
long-term commitment to their use and evolution. The
design standards for sand filters have evolved based on
performance and maintenance considerations.
Sand filters are used on site and on a regional basis
(usually less that 50 acres of drainage), and the filters
are sized to accept and treat the first half-inch of storm-
water runoff from the contributing drainage area (1).
They are frequently used in conjunction with a stormwa-
ter detention basin, which provides for control of larger
storms from a water quantity perspective. Good water
quality data for the performance of these systems have
resulted, which indicates that sand filters can be very
effective at pollutant removal.
Washington, DC
Sand filter use is based on a design standard developed
by the Stormwater Management Branch of the Depart-
ment of Consumer and Regulatory Affairs. The sand
filter system design is based on whether water quantity
is a concern in addition to waterquality on a specific site.
Washington, DC, has a combined sewer system, and
sites that discharge into a combined sewer system must
design their sand filters to provide for peak control of the
15-year storm. If only water quality is an issue, a design
procedure is established based on the degree of site
imperviousness. For water quality control alone, storage
requirements are between 0.3 and 0.5 in. of runoff per
acre (2). The Stormwater Management Branch is initiat-
ing a monitoring program to determine the performance
of the sand filters.
State of Delaware
Delaware has developed a sand filter design system
based on the Austin design but that serves for water
372
-------
quality control only. It is intended for sites where storm-
water runoff, only from impervious areas, may drain to
the sand filter. The sand filter is designed to accept and
treat the first inch of stormwater runoff and is used as
either a "stand alone" practice or in conjunction with
another practice, such as an infiltration practice (3).
Where infiltration practices are used, the sand filter
provides pretreatment of the runoff to reduce premature
clogging of the infiltration practice. At this time, design
performance is not being monitored, but achieving an
80-percent reduction in suspended solids is considered
an acceptable practice as required under the statewide
stormwater management law.
Alexandria, Virginia
The city of Alexandria has developed a design manual
that supplements the northern Virginia BMP handbook (4).
The Alexandria supplement details the design require-
ments of "no net increase" in pollutant loading for new
development and a 10-percent reduction in pollutant load-
ing at site redevelopment locations. To achieve these
goals, phosphorus was accepted as a "keystone" pollutant
for design purposes. The Alexandria supplement provides
information on a number of different sand filter design
procedures and is probably the single best compilation
of information relating to design procedures developed
in areas such as Austin, Delaware, and Washington.
Other Areas and Efforts
The only other procedure that is more experimental
(although, in reality, they all still are) is the peat-sand filter
developed by the Washington Council of Governments.
This procedure is a variation of the traditional sand filter
design that uses peat as a medium for enhanced nutri-
ent reduction. The State of Washington has recently
completed a stormwater design manual that presents a
sand filter design based on the Austin system.
Discussion
Sand filters represent an emerging technology with sig-
nificant potential for evolution in coming years. The
procedure developed for the State of Delaware was
intended for use on small sites where overall site imper-
viousness was maximized. Examples of these sites
would be fast food restaurants, gas stations, or industrial
sites, where space for retrofitting is not readily available.
Another emphasized use for sand filters is as a pretreat-
ment system for stormwater infiltration practices. Infiltra-
tion practices are very susceptible to clogging by
particulates, and sand filters could provide an effective
means to reduce particulate loading and to block oil and
grease from entry into infiltration systems.
Sand filters are especially appropriate for highway sys-
tems where site conditions and right-of-ways limit the
types of feasible stormwater treatment practices. Sand
filter systems generally have lower maintenance needs
than infiltration practices have, so their use appeals to
highway officials if the costs can be made reasonable.
If the sand filter is moved to the edge of the parking
lot or roadway, where structural strength is not as
important, the system can be installed at significantly
lower cost. The City of Alexandria has developed a
variation of the Delaware approach where the sand
filter is behind curb openings. In addition, increasing
the head over the filter can increase the time between
required maintenance of the filter, thus lowering the
system's operation and maintenance costs. Consid-
eration should be given to placing stone over the
sand to prevent scour of the sand as water drops on
the filter, in addition to increasing the overall depth
of the sand to improve performance.
The design procedure developed for use in Delaware
is meant as guidance and can be modified or en-
hanced as needed depending on specific site condi-
tions. The practice as presented may be used in the
middle of a parking lot, where concrete and grate
strength are established, so that automobiles or
trucks could travel over the system. Consultants have
taken that design standard literally, which has made
construction costs extremely high.
Any one of these systems could be modified or im-
proved with proper engineering. Conversations have
started with different manufacturers to see if sand filter
units could be prefabricated which would reduce the
overall cost of installation. The use of sand filters will
dramatically increase if construction costs are reduced.
Conclusion
Sand filters have a strong potential for becoming an
effective tool for stormwater treatment, but engineering
expertise is necessary to improve performance and
cost. With proper maintenance and in conjunction with
other practices, sand filters can assist in water quality
protection. They also have potential in arid regions,
where more conventional practices such as wet ponds
are not feasible.
We live in an era where our desires and mandates for
clean water exceed our abilities to actually protect our
aquatic resources when structural controls are consid-
ered as the only method of stormwater control. The term
"treatment train" is certainly a concept that must be
expanded if resource protection is to be realized. Sand
filters are one car of the "treatment train," but the overall
train must include many different considerations. Ulti-
mately, land use must be a consideration in overall site
stormwater planning, and considerations of roadway
widths, curbing, and site compaction and utilization
must be flexible depending on individual site needs.
Why does a residential street have to be wide enough
373
-------
for a fire engine to turn around in? We need to question
basic planning assumptions with respect to resource
protection, and to evaluate whether a specific design
requirement is necessary in light of that requirement's
impact on our natural resources. Otherwise, we need to
recognize and accept the fact that a decline in quality
and productivity of our resources will occur.
References
1. City of Austin. 1988. Environmental criteria manual. Environmental
and Conservation Services Department, City of Austin, TX (June).
2. Truong, H.V. 1989. The sand filter water quality structure. Storm-
water Management Branch, Government of the District of Colum-
bia, Department of Consumer and Regulatory Affairs (May).
3. Shaver, E. 1991. Sand filter design for water quality treatment.
Proceedings of the Engineering Foundation Specialty Conference,
Crested Butte, CO.
4. City of Alexandria. 1992. Alexandria Supplement to the Northern
Virginia BMP Handbook. Department of Transportation and Envi-
ronmental Services, City of Alexandria, VA (February).
374
-------
Application of the Washington, DC, Sand Filter for
Urban Runoff Control
Hung V. Truong
DC Environmental Regulation Administration, Washington, DC
Mee S. Phua
University of DC, Washington, DC
Abstract
Conventional infiltration systems are frequently used for
water quality control of urban runoff. These types of
urban best management practices (BMPs), however,
may adversely affect ground-water quality through the
migration of pollutants into ground-water aquifers. Addi-
tionally, these BMPs may not be feasible in high-density
urban areas because of the large land areas required
for their installation.
To address these problems, this paper presents an alter-
native solution: to replace conventional infiltration BMPs
with the confined, underground sand filter water quality
(SFWQ) control structure. Over 70 of these structures
have been installed in Washington, DC, since 1988.
The Washington, DC, underground sand filter is a grav-
ity flow system consisting of a concrete structure with
three chambers. It is designed to provide quality control
for the first 1/2 in. of runoff. The first chamber performs
pretreatment of stormwater runoff by removing floating
organic material such as oil, grease, and tree leaves.
The second chamber is the filter chamber (process
chamber) and optimally contains a 3-ft filter layer. The
filter layer consists of gravel, clean sand, and geotextile
filter fabric. At the bottom of the filter is a subsurface
drainage system of polyvinyl chloride perforated pipes
in a gravel bed. The third chamber is a discharge cham-
ber that collects flow from the underdrain pipes.
The SFWQ structure may vary in size and shape. The
depth can range from 8 to 10 ft depending on the final
grading of the site.
Introduction
Urbanization resulting in surface- and ground-water con-
tamination is a serious and constant threat to water quality.
In turn, poor water quality is an undesirable economic
burden on taxpayers. Because of the extremely high
cost involved in restoring contaminated surface and
ground water, prevention seems to be the only economi-
cal course of action to protect natural water systems.
To regulate and provide protection for surface- and ground-
water systems, the federal government passed the Clean
Water Act. As part of this effort, the District of Columbia
enacted stormwater management regulations (DC Law
5-188, section 509-519) in January 1988. These regula-
tions require new developments and redevelopments to
control nonpoint source pollution transported from con-
struction sites by urban runoff, using best management
practices (BMPs) or best available technologies (BATs).
Infiltration devices are the most frequently used BMPs for
controlling stormwater runoff in urban areas. These con-
ventional BMPs have limitations, however, due to soil and
site-specific constraints. These BMPs may also adversely
affect ground water through the migration of pollutants
into ground-water aquifers. Additionally, conventional in-
filtration systems may not be feasible in an urban envi-
ronment because of the large land areas required for
their installation. In an effort to mitigate these problems,
an alternative design is outlined in this paper to replace
the conventional infiltration BMPs, where applicable.
This alternative system is called the confined sand filter
water quality (SFWQ) structure and is illustrated in Fig-
ure 1. The system uses multiple filter layers combined
with a moderate detention time to filter the suspended
pollutant particles and hydrocarbons from urban runoff.
A multiple-layer filter was chosen because it has proven
to be more effective than a single-layer filter design.
Background
Infiltration practices have been widely used to improve
the quality of urban stormwater runoff. Several limitations,
however, are associated with the use of conventional
375
-------
Access Manhole
Structural Concrete Vault
Designed for Load and Soil
Conditionsv
Access Manholes
Overflow Weir
6-ln. PVC Dewatering
Drain With Gate Valve
^~
Outflow to Storm Sewer
Clean/veil Chamber
6-ln. Perforated PVC Collector
in 8-ln. Gravel Bed (3 required)
2-Ft Sand Filter Between Geotextile
Filter Cloth Layers
Inspection Well/Cleanout Pipe With
Waterproof Cap (3 required)
First 1/2 In. of
Runoff (WQV) From
Flow Separator
Sediment Chamber With Water Seal
To Trap Hydrocarbons
Figure 1. DC three-dimensional sandfilter centerline cutaway (source: District of Columbia).
infiltration systems. According to several studies (1-3),
the practice of infiltration may have a negative impact
on ground-water quality. In addition, infiltration practices
are only recommended for sites with soil infiltration rates
higher than 0.27 in./hr and with a clay content of less
than 30 percent. Recently, a study by the Metropolitan
Washington Council of Government (MWCOG) shows
that over 50 percent of the infiltration trenches installed
in the Metropolitan Washington region either partially or
totally failed within the first 5 years of construction (4).
Research has also found that clogging may occur in
infiltration trenches and is also very common in other
infiltration systems. In surface systems, clogging is most
likely to occur near the top of the structure, between the
upper layer of stone and the protective layer of filter
fabric. For underground infiltration systems, clogging is
likely to occur at the bottom of the structure, at the filter
fabric, and at the soil interface.
Restoration of both surface and underground infiltration
systems is tedious and very costly, requiring the removal
of the vegetation layer, top soil, protective plastic layer,
stone aggregate, and filter fabrics. If the surface layer is
pavement or concrete, the rehabilitation effort becomes
even more difficult and expensive. Conventional infiltration
systems also require relatively large areas of land for their
installation; therefore, this family of BMPs is not feasible
due to the high cost of land in an urban environment.
Design Rationale
Whenever a liquid containing solids in suspension is
placed in a relatively quiescent state, solids having a
higher specific gravity than the liquid settle, while those
having a lower specific gravity rise. The design of the
SFWQ structure uses the one-dimensional "falling head
test" in Darcy's Law for calculating the head loss of fluid
flow through a multiple-layer filter medium to treat storm-
water runoff. The design uses various media layers with
different permeabilities to intercept pollutant particles as
fluid flows vertically through the filter layers. This princi-
ple can be used to accelerate the removal of pollutants
by increasing the residence times of stormwater runoff,
and to facilitate the filtering process in the filter chamber.
The SFWQ structure also utilizes Stoke's Law for termi-
nal falling velocities of individual particles in allowing
time for particles to settle out of stormwater runoff. The
average detention time of this system ranges from 6 to
8 hrforan optimal design consideration.
Functional and Physical Description
The SFWQ structure is a gravity flow system consisting
of three chambers. The facility may be precast or cast-
in-place. The first chamber (same as water quality inlet)
is a pretreatment facility removing any floating organic
material such as oil, grease, and tree leaves. The chamber
has a submerged weir leading to the second chamber
376
-------
(filter chamber) and may be designed with a flow splitter
or with a bypass weir if the system is for off-line storage,
as illustrated in Figure 2.
The second chamber contains 3 ft of filter material con-
sisting of gravel, geotextile fabric, and sand, and is
situated behind a 3-ft weir. At the bottom is a subsurface
drainage system consisting of a parallel polyvinyl chlo-
ride (PVC) pipe system in a gravel bed. A dewatering
valve is at the top of the filter layer for maintenance
purposes and for safety release in case of emergency.
It also has an overflow weir at the top to protect the
system from backing up when the storage volume is
exceeded, if the system is designed for on-line storage
(Figure 3).
Water enters the first chamber of the system by gravity
or by pumping. This chamber removes most of the
heavy solid particles, floatable trash, leaves, and hydro-
carbon material. A submerged weir (designed to mini-
mize the energy of incoming stormwater) conveys the
effluent to the second chamber. The effluent enters the
filter layer by overflowing the weir typically 3 ft above the
bottom of the structure. The water is filtered through
various filtering layers to remove suspended pollutant
particles. The filtered stormwater is then picked up by
the subsurface drainage system that empties it into the
third chamber. The third chamber also receives any
overflow from the second chamber for an on-line system
and overflow from the first chamber flow splitter for an
off-line system.
Applicability
The SFWQ structure is specifically designed for highly
urbanized areas where open space is not available. The
30-Ft Manhole 30-Ft Manhole
With Ladder With Ladder Bypass Pipe
\ i /
Bottom
Opening ~
|C[
:
f
I
v
/
1
. _ _ V
V
r
i
24-ln. Manhole
6-ln. PVC Dewatering
Drain With PVC
1 Gate Valve
... / ... V 3-6-ln. PVC Pipe
3VFtHVrgahM1n.PLAN Pejorated
30-Ft Manhole
Frame/Cover
Inflow
Pipe
30-Ft Manhole
. Frame/Cover
Filter Layer
SECTION
24-ln. Manhole
Frame/Cover
Bypass Pipe
6-ln. PVC Dewatering
Drain With PVC
| Gate Valve
6-ln. PVC Cleanout
Pipe With Cap
30-ln. Manhole
With Ladder
30-ln. Manhole
With Ladder
Bottom
Openin;
24-ln. Manhole
6-ln. PVC
Dewatering
Drain With PVC
Gate Valve
^MjSKSSOS^^&^^SSS^S^iS^SSSSSXSSR
Weir Wall pi AM 3-6-|n
3 Ft High Min.
PLAN
30-ln. Manhole 30-ln. Manhole
Frame/Cover Frame/Cover
i-ln. PVC Pipe
Perforated
24-ln. Manhole
Frame/Cover
Filter Layer
SECTION
Overflow
Weir
6-ln. PVC
Dewatering
Drain With PVC
Gate Valve
6-in. PVC Cleanout
Pipe With Cap
Figure 2. DC off-line underground sand filter (source: District
of Columbia).
Figure 3. DC on-line underground sand filter (source: District
of Columbia).
structure works best for impervious catchment areas of
1 acre or less. Multiple systems are recommended for
catchment areas greater than 1 acre.
Over 70 underground and surface sand filter structures
have been installed in Washington, DC, since 1988. In
fact, the structure has been adopted and incorporated
in the stormwater management programs of several
states and neighboring jurisdictions.
The structure may also be designed to provide deten-
tion, especially for on-line application when discharge
rates must be modified in accordance with local and
municipal regulations. Recommended areas where this
device may be used include:
Surface parking lots.
Underground parking lots or multilevel garages.
Parking apron, taxiway, and runway shoulders at
airports.
Emergency stopping and parking lanes and sidewalks.
Vehicle maintenance areas.
On-street parking aprons in residential areas.
Recreational vehicle camping area parking pads.
Private roads, easement service roads, and fire lanes.
Industrial storage yards and loading zones.
Driveways for residential and light commercial use.
Office complexes.
377
-------
Planning Considerations
Location
The SFWQ structure must be located in areas where it
is accessible for inspection and maintenance, as well as
to the vacuum trucks that are usually required to provide
maintenance.
Ground Water and Bedrock
The seasonally high ground-water table and bedrock
should be at least 2 to 4 ft below the footing of the
structure.
Size
The SFWQ structure may vary in size from a small-site
single installation to large or multiple facility installations.
Site topography and the presence of underground utili-
ties, however, may limit the size and depth of the sys-
tem. Use of other practices in combination with the
SFWQ structure may solve this problem.
Hydraulic Head
Because the SFWQ structure is a gravity flow system,
sufficient vertical clearance between the inverts of the
inflow and outflow pipes must be provided. When eleva-
tion is insufficient, a well pump may be used to dis-
charge the effluent from the third chamber into the
receiving drainage system.
Water Trap
In combined sewer areas, a water trap must be pro-
vided in the third chamber to prevent the backflow of
odorous gas.
Design Criteria
In designing the SFWQ structure, the nature of the area,
such as imperviousness, determines the control volume
of the sand filter chamber. Other recommended steps to
consider when designing a SFWQ structure are the
following:
Examine the site topographical conditions and select
possible outfalls from the existing drainage or sewer
map.
Review the final grading plans and determine the
maximum head available between the proposed in-
flow and outflow pipes.
Determine the total connected impervious area.
Select the design (first flush) runoff based on land
use characteristics. (Washington, DC, uses 0.5 in. for
surface parking lots, 0.3 in. for rooftops, and 0.4 in.
for other impervious surfaces.)
Estimate the storage volume and the release rate.
The storage volume and release rate depends on
local stormwater management regulations.
Select design storm(s). This should be based on the
storm frequencies selected by the stormwater man-
agement authorities.
Determine the size of the inflow, outflow, and emer-
gency release pipes. These should be sized to pass
the lowest selected storm frequency permitted by lo-
cal stormwater regulations. (Washington, DC, uses
15-yr, 5-min storms for postdevelopment runoff.)
Determine detention time. All SFWQ structures
should be designed to drain the design (first flush)
runoff from the filter chamber 5 to 24 hr after each
rainfall event.
Determine structural requirements. A licensed struc-
tural engineer should design the structure in accord-
ance with local building codes.
Provide sufficient headroom for maintenance. A mini-
mum head space of 5 ft above the filter is recom-
mended for maintenance of the structure. If 5 ft of
headroom is not available, a removable top should
be installed.
Design Procedures
Determine Design Invert Elevations
Determine the final surface elevation, invert in, invert
out, and bottom invert elevation of the structure (see
Figure 4):
D, = (Inv. in - Inv. out) + Hw + 1, (Eq. 1)
where
D, = total depth of structure (ft)
Inv. in = final invert elevation of inflow pipe (ft)
Inv. out = final invert elevation of outflow pipe (ft)
Hw = vertical height of overflow weir (ft)
1 = freeboard constant (ft)
Peak Discharge Calculation for Bypass Flow
Using the Rational Method:
QPk = CIA,
where
QPk = bypass peak flow (ft3/sec)
C = runoff coefficiency (dimensionless)
I = rainfall intensity (in./hr)
A = drainage area (ac)
(Eq. 2)
378
-------
Determination of Filter Area
L = Chamber Length W = Chamber Width
A1
A3
W
1
PLAN
Inflow ,
"<* "*TTvi-
h max
6-ln. PVCCIeanout j
Pipe With Cap M
H
t
L1-H
PVC Perforated Pipe
L1 = 1/3 L2
1
SECTION
Figure 4. Design guide for DC sandfilter (source: District of
Columbia).
Determine Area of Sand Filter
Use Figure 5 or the following equation:
Af = 50 + [la - 0.1 acres] x 167 ft3/ac, (Eq.3)
where
Af = sand filter area (ft2)
la = impervious area (ac)
Determine Storage Volume
Use the equation
Vw = (QiXla)-(FxTxAf), (Eq. 4)
where
Vw = volume storage needed (ft3)
QI = first flush runoff (in)
la = impervious area (ft2)
F = final infiltration rate for filter (ft/hr)
T = filling time (1 hr, based on empirical data)
Af = sand filter area (ft2)
Calculate Bottom Storage Volume in Second
Chamber
Use the equation
\/2b = Aixdx Vv,
200
, 150
TO
0)
< 100
<5
50
X
X
X
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Impervious Area (la) Acre
Af = 50 + (la - 0.1 acre) * 167 sq ft per acre
Figure 5. Filter area versus watershed imperviousness (source:
District of Columbia).
where
V2b = bottom volume of filter chamber (ft3)
Af = surface area of filter layer (ft2)
d = depth of filter layer (ft)
Vv = sum of void ratio for filter media
Calculate Bottom Storage Volume in First
Chamber
Use the equation
(Eq. 6)
where
V1b = bottom volume of first chamber (ft3)
A! = surface area of first chamber (ft2)
d = depth of filter layer (ft)
Note: Af/3 < A1 < Af/2 for optimum design condition.
Calculate Storage Volume in First and Second
Chambers
Use the equation
(Eq. 7)
(Eq. 5)
where
V1t + V2t = sum of top volume of first and
second chambers
379
-------
Vw = volume of water from Equation 4
V2b + V1b = sum of bottom volume of first
and second chambers
Determine Maximum Storage Depth for
On-Line System
Use the equation
D = [(V1t + V2t)/(A1+A2)] + d,
where
(Eq. 8)
D = maximum storage depth (ft)
V1t + V2t = sum of top volume of first and second
chambers
A! + A2 = sum of surface area of first and second
chambers
d = depth of filter layer (ft)
Note: D must be equal to or smaller than the difference
between the invert in and invert out from Equation 1.
Determine Size of Submerged and Overflow
Weirs
Submerged weir opening in first chamber:
A(h x I) = Qpk/C x (2 x g x hmax)05, (Eq. 9)
where
A(h x I) = area of weir opening (ft2)
QPk = bypass flow from Equation 2 (ft3/sec)
C = 0.6, weir coefficient
g = 32.2 ft2/sec
hmax = hydraulic head above the center line of
weir (ft)
h = weir height, minimum 1 ft
Overflow weir opening in second chamber:
H1-5 = Qpk/CL, (Eq. 9a)
where
H = height of weir opening (ft)
QPk = bypass flow (ft3/sec)
C = 3.33, weir coefficient
L = length of weir opening (ft)
Determine Flow Through Filter and Detention
Time After Storage Volume Fills Up
where
qf = flow through the filter (ft3/hr)
k = sand permeability (ft/hr)
Af = filter area
i = hydraulic gradient (Hmax/2 x filter depth)
Estimate the detention time:
Ts = Vw/q, (Eq. 11)
where
Ts = average dewatering time for SFWQ structures
(hr)
Vw = volume of first flush storage from Equation 3
(ft3)
q = average flow from Equation 10 (ft3/hr)
Develop Inflow and Outflow Hydrographs
Figure 6 is a typical illustration of inflow/outflow hy-
drographs for the SFWQ structure.
For inflow hydrograph, use Modified Rational Method
Hydrograph with:
T = TC
TR=1.67TC,
where
T = time to peak
Tc = time of
concentration
TR = recession period
For outflow hydrographs, use the following equations to
determine when flow occurs:
when
TcxQpk<2Vw, + (Eq. 12)
T = [2*TC2 - (2TC2 - 2Vw*Tc/Qpk)°-5].
Average flow through the filter:
qf = k x Af x i,
when
Tc x Qpk = 2VW, +
T = (0.5TC) + (Vw/Qpk)
when
Tc x Qpk > 2VW, +
(Eq. 10) T = [(2Vw*Tc)/Qpk]°-5.
(Eq. 13)
(Eq. 14)
380
-------
Washed Gravel
1 6 11 16 21 26 31 36 41 46 51 56 61
Time (min)
Discharge vs. Time
Series 1 -+- Series 2
(Outflow) (Inflow)
Figure 6. Typical inflow-outflow hydrograph (source: District of
Columbia).
Filter Layer Details
Figure 7 is a typical cross section of the filter chamber.
Upper Gravel Layer
The washed gravel or aggregate layer at the top of the
filter may be 1 to 3 in. thick and meet American Society
for Testing Materials (ASTM) standard specifications for
1-in. maximum diameter or DC #57 gravel.
Geotextile Fabrics
The filter fabric (geotextile fabric) beneath the top gravel
layer should be Enkadrain 9120 or equivalent with the
specifications shown in Table 1.
The filter cloth beneath the sand should meet the speci-
fications shown in Table 2.
The fabric roll should be cut with sufficient dimensions
to cover the entire wetted perimeter of the filter area with
a 6-in. minimum overlap.Sand Filter Layer
Sand Filter Layer
The sand filter layer should be 18 to 24 in. deep. ASTM
C33 Concrete Sand is recommended, but sand with
similar specifications may be used.
Table 1. Geotextile Fabric Specifications
Property Test Method Unit
Specification
Material
Unit weight
Flow rate
Puncture
thickness
Nonwoven
geotextile fabric
ASTM D-1 777
"Falling head test"
ASTM D-751
oz/yd2
gpm/ft2
Ib
in.
4.3 (min)
120 (min)
60 (min)
0.8 (min)
Enkadrain 9120
or Equivalent
Sand Filter
Layer
Geotextile
Fabric
6-ln. PVC Pipe Wrapped in
Geotextile Fabric
Figure 7. Cross section of filter compartment (source: District
of Columbia).
Bottom Gravel Layer
The bottom gravel layer surrounding the collector (per-
forated) pipes should be 1/2- to 2-in. diameter gravel
and provide at least 3 in. of cover over the tops of the
drainage pipes. No gravel is required under the pipes.
The gravel and the sand layer above must be separated
by a layer of geotextile fabric that meets the specifica-
tions listed above.
Underdrain Piping
The underdrain piping consists of three 6-in. pipes with
3/8-in. perforations and should be reinforced to withstand
the load of the overburden. All piping should be to schedule
40 polyvinyl chloride (PVC) or greater strength.
The minimum grade of piping shall be 1/8 in./ft or 1
percent slope. Access for cleaning all underdrain piping
is needed. Cleanouts for each pipe should extend to the
invert of overflow weir or maximum surface elevation of
the storage water.
Each pipe should be carefully wrapped with geotextile
fabric that meets the above specifications before place-
ment in the filter.
Table 2. Filter Cloth Specifications
Property Test Method
Unit Specification
Material
Unit weight
Filtration rate
Puncture strength
Mullen burst
strength
Tensile strength
Equivalent
opening size
Nonwoven
geotextile fabric
ASTM D-751
(Modified)
ASTM D-751
ASTM D-1 682
U.S. Standard Sieve
oz/yd2
in. /sec
Ib
psi
Ib
no.
8.00 (min)
0.08 (min)
125 (min)
400 (min)
300 (min)
80 (min)
381
-------
Construction Specifications
Maintenance Requirements
The SFWQ structure may be either cast-in-place or
precast. In Washington, DC, precast structures require
advanced approval. The approved erosion and sedi-
ment control plans should include the specific measures
to provide the protection of the filter system before the
final stabilization of the site.
Excavation and Installation
Excavation for SFWQ structure and connecting pipes
should include removal of all materials and objects en-
countered in excavation; disposal of excavated material
as specified in the approved erosion and sediment con-
trol plans; maintenance and subsequent removal of any
sheeting, shoring and bracing; dewatering and precau-
tions; and work necessary to prevent damage to adja-
cent properties resulting from this excavation. Access
manholes and steps to the filtration system should con-
form to local standards.
Leak Test
After completion of the SFWQ structure shell, a leak test
may be performed to verify watertightness before the
filter layers are installed.
Filter Materials
All filter materials in the second chamber should be
placed according to construction and materials stand-
ards and specifications, as specified on an approved
construction plan.
Completion and Site Stabilization
No runoff should be allowed to enter the sand filter
system before completion of all construction activities,
including revegetation and final site stabilization. Con-
struction runoff should be treated in separate sedimen-
tation basins and routed to bypass the filter system.
Should construction runoff enter the filter system prior
to final site stabilization, all contaminated materials must
be removed and replaced with new, clean filter materials
before a regulatory inspector approves its completion.
System Calibration and Verification
The water level in the filter chamber should be moni-
tored by the design engineer after the first storm event
before the project is certified as completed. If the dewa-
tering time of the filter chamber takes longer than 24 hr,
the top gravel layer and filter fabric underneath must be
replaced with a more rapid draining fabric and clean
gravel. The structure should then be checked again to
ensure a detention time that is less than 24 hr.
The SFWQ structure is designed to minimize mainte-
nance. It is subject to clogging, however, by sediment,
oil, grease, grit, and other debris. Actual performance
and service life of the structure is not available at this
time. Nevertheless, it is still very important to provide
general standard maintenance guidelines to maintain
adequate structure operation. The maintenance of the
system includes the following steps:
The water level in the filter chamber should be moni-
tored by the owner on a quarterly basis and after
every large storm for the first year after completion
of construction. A log of the results should be main-
tained, indicating the rate of dewatering after each
storm and the water depth for each observation.
Once the regulatory stormwater inspector indicates
that satisfactory performance of the structure has
been demonstrated, the monitoring schedule may be
reduced to an annual basis.
As with other pretreatment structures, the first cham-
ber must be pumped out semiannually. If the chamber
contains an oil skim, it should be removed by a firm
specializing in oil recovery and recycling. The remain-
ing material may then be removed by a vacuum pump
truck and disposed of in an approved landfill. After each
cleaning, refill the first chamber to a depth of 3 ft with
clean water to reestablish the water seal.
After approximately 3 to 5 yr, the upper layer of the
filter can be expected to become clogged with fine
silt. When the drawdown time for the filter exceeds
72 hr, the upper layer of gravel and geotextile fabric
must be removed and replaced with new, clean ma-
terials conforming to the original specifications.
Conclusion and Discussion
At the present time, the environmental and economic im-
pacts of the SFWQ structure have not been fully evaluated.
A long-term monitoring program is being implemented in
Washington, DC, to determine water quality benefits and
address long-term maintenance concerns. The results
from this monitoring effort will provide important informa-
tion on the removal efficiency of common urban pollut-
ants. In addition, the monitoring data will provide
information on actual headless in the system, which will
indicate the need for filter replacement.
Based on the results of the Austin, Texas, monitoring
program on its sand filter systems and on several years
of success in the application of the SFWQ structure in
Washington, DC, the feasibility of the SFWQ structure
has been demonstrated for use in an urban environ-
ment. The authors believe that the SFWQ structure may
be used as an alternative urban BMP for highly devel-
oped areas where other options are not available.
382
-------
In conclusion, the design presented here is an attempt
to provide an alternative solution to control nonpoint
source pollution from urban stormwater runoff. The ap-
plication of this system should be viewed with some
caution, as the structure has not been monitored for
optimal effectiveness.
When the SFWQ structure is used strictly as a gravity
flow system, one of its limitations is that it requires a
hydraulic head of at least 4 ft relative to the outflow pipe.
To minimize this problem, further study is needed to
evaluate the different thicknesses of the sand layers
(with thicknesses such as 18,12, and 6 in.) to determine
the relationship between the depth of sand layer and
pollutant removal efficiency.
Acknowledgments
Computer-aided design was performed by Renette
Dallas, DC Environmental Regulation Administration,
Washington, DC. The authors would also like to thank
Renette Dallas and Collin R. Burrell, also of the DC
Environmental Regulation Administration, for technical
assistance.
References
1. Washington Area NURP Project. 1983. Final contract report.
Manassas, VA: Occoquan Watershed Monitoring Lab.
2. U.S. EPA. 1983. Results of the nationwide urban runoff program,
Vol. I. Final report. Water Planning Division.
3. Nightingale, H.T. 1987. Water quality beneath urban runoff water
management basins. Water Resour. Res. 23(2):197-208.
4. Galli, J. 1992. Analysis of urban BMP performance and longevity
in Prince George's County, Maryland. Washington, DC: Metro-
politan Washington Council of Governments.
Additional Reading
1. Van Truong, H. 1989. The sand filter water quality structure.
Washington DC: Environmental Regulation Administration.
2. Van Truong, H. 1993. The DC sand filter water quality structure,
2nd version. Draft. Washington DC: Environmental Regulation
Administration.
3. Alexandria Department of Transportation and Environmental
Services. 1992. Alexandria supplement to the Northern Virginia
BMP handbook (adopted in February).
4. Chang, F.M., M.H. Watt, and H. Van Truong. 1986. Study of
erosion and sedimentation of selected small streams in the District
of Columbia. NTIS PB-86-246758. Washington, DC: WRRC.
5. Watt, H.M., J.V. O'Conor, and H. Van Truong. 1985. Ground-water
problem in the mid-Atlantic fall line cities. NTIS PB-85-225985/8H.
Washington, DC: WRRC.
6. Karikari, T.J., H. Van Truong, and M.K. Mitchell. 1988. DC storm-
water management guidebook. Washington, DC: Environmental
Control Division.
7. Truong, H.V. 1987. DC groundwater protection strategy. The Dis-
trict of Columbia Department of Consumer and Regulatory Affairs.
8. City of Austin. 1991. Design guidelines for water quality control
basins. Austin, TX: Public Works Department.
9. Department of Public Works. 1986. DC public works water and
sewer specifications and detail drawing. Washington, DC.
10. Das, B.M. 1990. Principle of geotechnical engineering, 2nd ed.
Boston, MA: PWS-KENT Publishing Company.
383
-------
Stormwater Measures for Bridges:
Coastal Nonpoint Source Management in South Carolina
H. Stephen Snyder
South Carolina Coastal Council,
Charleston, South Carolina
Abstract
Although stormwater runoff from bridges has a direct
pathway to estuaries, rivers, and lakes, little research
has been undertaken to directly measure the concentra-
tion of pollutants flushed from the bridge surface or the
impact of those pollutants on the receiving water body.
A general correlation can be made, however, from the
body of research available concerning runoff from roads
and streets in general and from the wider body of infor-
mation regarding urban runoff characteristics. The gen-
eral assumption is that runoff from highways (and
bridges) can negatively affect the water quality of receiv-
ing waters through the shock of acute loadings during
rainfall events and through long-term exposure and/or
accumulations of pollutants in sediments or marine or-
ganisms. Research does indicate a relationship be-
tween the average daily traffic volume and potential
water quality impacts. Concern is heightened where the
runoff has a direct, unobstructed pathway to the receiv-
ing waters and, even more so, where the receiving
waters are extremely sensitive, such as shellfish habitat.
This paper provides a brief overview of potential water
quality pollutants from highway and bridge runoff, then
focuses on management and control measures for run-
off from bridges. These include requirements of Section
6217 of the Coastal Zone Act Reauthorization Amend-
ments and stormwater management requirements for
bridges in the coastal zone of South Carolina. Included
is a case study of retrofitting a major bridge already
designed and under construction, which transverses
significant shellfish resources in coastal South Carolina.
Introduction
South Carolina's 187-mile coastline is only the facade
forsome 3,000 shoreline miles of estuaries, bays, rivers,
and creeks that intertwine among some 500,000 acres
of coastal marshes and wetlands. This immense coastal
system supports approximately 279,000 acres of estu-
arine shellfish-growing waters and thousands of acres
of other sensitive habitats. For people to live and work
in this environment, all of these coastal resources, riv-
ers, bays, marshes, and sensitive habitats must be
transversed in one form or another, most often by road-
ways and bridges. These roadways and bridges and
their associated uses can provide a direct source of
contaminants to our coastal waters and, as such, must
be managed to reduce or alleviate the potential impacts.
For coastal states, addressing pollution from bridges
may no longer be a choice. Section 6217 of the Coastal
Zone Act Reauthorization Amendments of 1990 requires
states with coastal zone programs to develop coastal
nonpoint source pollution programs. Such programs
must address pollution in the following areas: agricul-
ture, silviculture, hydrologic modifications, marinas, and
urban settings, the latter of which include roads and,
even more specifically, bridges.
A basic assumption contained herein is that the results
of studies on highways and their associated pollution po-
tential from runoff are also applicable to highway bridges.
Contaminants
A series of studies sponsored by the U.S. Department
of Transportation in the 1980s (1-3) confirms the pres-
ence and possible sources of a wide variety of contami-
nants that may be associated with roadways and
bridges. A basic listing is presented in Table 1. These
contaminants accumulate on roadway surfaces be-
tween major removal events, such as rainfalls and street
sweeping (which may be rare or nonexistent in nonur-
ban areas). The severity and order of magnitude of
these contaminants are site specific and variable, and
can depend on such factors as traffic characteristics,
highway or bridge design, maintenance activities, acci-
dental spills, surrounding land use, and climate.
384
-------
Table 1. Common Highway Runoff Constituents and Their Primary Sources (1)
Constituent Primary Sources
Participates
Nitrogen, phosphorus
Lead
Zinc
Iron
Copper
Cadmium
Chromium
Nickel
Manganese
Bromide
Cyanide
Chloride
Sulphate
Petroleum
Polychlorinated biphenyls
(PCBs), synthetic pesticides
Pathogenic bacteria
(indicators)
Rubber
Asbestos
Pavement wear, vehicles, atmosphere, highway maintenance
Atmosphere, roadside fertilizer application
Leaded gasoline (auto exhaust), tire wear (lead oxide filler material), lubricating oil and grease, bearing
wear
Tire wear (filler material), motor oil (stabilizing additive), grease
Auto body rust, steel highway structures (guardrails, bridges, etc.), moving engine parts
Metal plating, bearing and bushing wear; moving engine parts; brake lining wear; fungicides and
insecticides (roadside maintenance operations)
Tire wear (filler material), insecticides
Metal plating, moving engine parts, brake lining wear
Diesel fuel gasoline (exhaust) and lubricating oil, metal plating, bushing wear, brake lining wear, asphalt
paving
Moving engine parts
Auto exhaust
Anticake compound (ferric ferrocyanide, etc.) used to keep deicing salt granular
Deicing salts
Roadway beds, fuel, deicing salts
Spills, leaks, or blow-by of motor lubricants; antifreeze, and hydraulic fluids, asphalt surface leachate
Spraying of highway right-of-ways, background atmospheric deposition, PCB catalyst in tires
Soil, litter, bird droppings, trucks hauling livestock and stockyard waste
Tire wear
Clutch and brake lining wear
The studies have revealed some interesting results that
may influence management decisions. To elaborate on
one pollutant, tests (1) indicated that the pathogenic
bacteria indicators fecal coliform and fecal Streptococ-
cus were not consistently present on roadway systems
at any given time or place; their presence is most often
associated with nonspecific events, i.e., animal and bird
droppings, soil spills, and road kills. When present, how-
ever, the bacteria can remain viable for relatively long
periods in highway sweepings (up to 7 weeks) and up
to 13 days in stagnant storm sewer systems. As one
would expect, the tests showed that the coliform bacte-
ria were consistently lower when runoff was conveyed
through a grassy area, although none of the standard
nonpoint source management measures effectively kills
conforms and their associated microbes (2).
According to the U.S. Department of Transportation (1),
the major portion of priority pollution load in highway
runoff was attributed to metals (e.g., lead, zinc, and
copper), although a significant number of organic pol-
lutants were present in the highway environment.
Studies (4, 5) indicate that the magnitude of pollutants
associated with highway runoff is related to traffic vol-
ume. Research (2) tends to indicate that 30,000 average
daily traffic (ADT) is a general threshold for the potential
of impacts from highway runoff; however, several vari-
ables must be factored into this conclusion, including
sensitivity of receiving waters, distance to receiving wa-
ters, type of traffic, road or bridge design, and others.
The U.S. Department of Transportation (2) has drawn
the following conclusions from these studies and other
literature concerning highway runoff pollution potential:
Highway runoff does have the potential to adversely
affect the water quality and aquatic biota of receiving
waters.
The significance of these adverse effects is variable
by highway type and design, receiving water, and
runoff event.
Runoff from urban highways with high ADT volumes
may have a relatively high potential to cause adverse
effects.
Runoff from rural highways with low ADT volumes
has a relatively low potential to cause adverse effects.
385
-------
Basic Management Practices and
Processes
Of the variety of best management practices available
for nonpoint source pollution control, four basic manage-
ment measures are generally considered cost effective
for treatment of highway runoff based on effectiveness
for specific pollutants, relative capital costs, land re-
quirements, and operation and maintenance costs (2):
Vegetative controls
Wet detention basins
Infiltration basins
Wetlands
Pollution measures that were not considered effective
when used as a sole management tool were street
cleaning, catch basins, filtration devices for sediment
control, dry detention ponds, and porous pavements (2).
The first three methods were not effective in capturing
the fine sediments to which many pollutants attach
themselves, while the dry detention pond tended to
reflush the settled particles after each rainfall event.
Porous pavement is limited to low-volume traffic areas,
such as parking lots, because of current highway con-
struction standards.
All of the measures have in common several physical or
biochemical processes that occur to provide the neces-
sary control of pollutants: settling, filtering, adsorption,
bioassimilation, biodegradation, and volatilization or
evaporation. Table 2 lists the process associated with
each management measure as related to the general
type of pollutant control.
Management Measures for Bridges
Although bridges can be assumed to cause the same
types of water quality impacts as highways, and al-
though the techniques to manage those impacts are
fairly straightforward and generally well accepted, the
unique location of bridges presents some problems.
First, the runoff from the bridge must be intercepted from
seeking its natural pathway and routed back to high land
or another area suitable for treatment; secondly, land
areas for treatment are usually limited.
Collection and transportation are most easily solved in
the design of the bridge, although in coastal areas runoff
may have to be transported long distances with little
grade. The physical land requirements for the appropri-
ate treatment method, however, tend to be the most
limiting factors. Solutions are very site specific and must
be included in the earliest planning stages of the bridge.
Topography at the bridge/land junction is often the single
most important factor in considering the design of an
appropriate treatment method, although other factors,
such as high water tables, soil types, and adjacent land
use, also can be important in the design consideration
process. The design of the stormwater system should
not drive the design of the bridge, but neither should the
design of the bridge preclude the design of an effective
stormwater treatment system.
All of the traditional stormwater management methods
can be considered for treatment of runoff from bridges:
wet detention ponds, infiltration systems, grassed wa-
terways, and wetlands. These can be used even in
combination with less favorable methods, such as fre-
quent sweeping or catch basins, if the lack of good
alternatives so dictates. Other opportunities that may be
present in the area should also be considered, such as
nearby spoil disposal containment areas, preexisting
treatment systems for nearby development, or dis-
charge routing to less sensitive areas.
The U.S. Environmental Protection Agency (EPA) (6) lists
several general guidelines and management practices for
illustrative purposes, specifically for bridges, in the Section
6217 management measure guidance document:
Table 2. Principal Pollutant Fate Processes by Major Management Measures
Management Measures
Pollutant
Heavy metals
Toxic organics
Nutrients
Solids
Oil and grease
Biochemical oxygen
Vegetative Control
Filtering
Adsorption
Bioassimilation
Filtering
Adsorption
Biodegradation
Detention Basins
Adsorption, settling
Adsorption, settling,
volatilization
Bioassimilation
Settling
Adsorption, settling
Biodegradation
Infiltration Systems
Adsorption, filtration
Adsorption,
biodegradation
Absorption
Adsorption, settling
Adsorption
Biodegradation
Wetlands
Adsorption, settling
Adsorption, settling,
biodegradation,
volatilization
Bioassimilation
Adsorption
Adsorption, settling
Biodegradation
demand
Pathogens
NA
Settling
Filtration
NA
NA = information not available
386
-------
Coordinate design with the Federal Highway Admini-
stration (FHWA), U.S. Coast Guard, U.S. Army Corps
of Engineers, and other state and federal agencies
as appropriate.
Review National Environmental Policy Act require-
ments to ensure that environmental concerns are met.
Avoid highway locations requiring numerous river
crossings.
Direct pollutant loadings away from bridge decks by
diverting runoff waters to land for treatment.
Restrict the use of scupper drains on bridges less
than 400 ft in length and on bridges crossing very
sensitive ecosystems.
Site and design new bridges to avoid sensitive eco-
systems.
On bridges with scupper drains, provide equivalent
urban runoff treatment in terms of pollutant load re-
duction elsewhere on the project to compensate for
the loading discharged off the bridge.
Regardless of the "illustrative" nature of the above prac-
tices, EPA and the National Oceanic and Atmospheric
Administration (NOAA) expect the states to address
nonpoint pollution from bridges and to adopt enforce-
able policies by 1995 to manage the runoff or to docu-
ment why such runoff is not a problem.
South Carolina's Approach
In 1988, the South Carolina Coastal Council was faced
with the permitting of a new 2-mile bridge connecting
the mainland with a major developed barrier island
(see below) and crossing a major shellfish-producing
area. As an outcome of the permitting of this project,
the Coastal Council developed a set of guidelines to use
in conjunction with the South Carolina Department of
Highways and Public Transportation to allow all parties
to anticipate the design of stormwater controls in new
bridges. It is not unusual for bridges to be designed
well in advance of the permitting process, and the inclu-
sion of new design criteria can cause both new expenses
and a politically unpleasant situation. The guidelines
have been in use since 1989 and have been introduced
as regulations to the 1993 South Carolina General
Assembly. The regulations appear to meet the basic
intent of the EPA/NOAA Section 6217 guidance,
although this has yet to be determined. The basic regu-
lations are as follows.
Stormwater Management Requirements for
Bridge Runoff
The following are the criteria used to address stormwa-
ter management for bridges traversing saltwater and
critical areas.
No treatment is necessary for runoff from bridge sur-
faces spanning Class SA and Class SB tidal salt-
waters. (SA and SB waters are suitable for primary
and secondary contact recreation, crabbing, and fish-
ing. The two classes differ in their dissolved oxygen
[DO] limitations: SA waters must maintain daily aver-
ages of not less than 5.0 mg/L, and SB waters must
maintain DO levels not less than 4.0 mg/L.) This
runoff can be discharged through scupper drains di-
rectly into surface waters. The use of scupper drains,
however, should be limited as much as possible.
If the receiving water is classified as either outstand-
ing resource waters (ORW) or shellfish harvesting
waters (SFH), then the stormwater management re-
quirements shall be based on projected traffic vol-
umes and the presence of any nearby shellfish beds.
Table 3 lists the necessary treatment practices over
the different classes of receiving waters.
The ADT volume is based on the design carrying
capacity of the bridge.
Table 3. Requirements for Stormwater Management on
Bridges in the Coastal Zone, South Carolina
Water Quality Classification
ADT Volume
0-30,000 30,000
ORW (within 1,000 ft of shellfish beds) A A
ORW (not within 1,000 ft of shellfish beds) B B
SFH (within 1,000 ft of shellfish beds) B A
SFH (not within 1,000 ft of shellfish beds) B B
SFH (not within 1,000 ft of shellfish beds) B B
SA (exceptional) C C
SB (high quality) C C
A = The first 1-in. of runoff from the bridge surface must be collected
and routed to an appropriate stormwater management system or
routed so that maximum overland flow occurs, encouraging
exfiltration before reaching the receiving water body. Periodic
vacuuming of the bridge surface should be considered.
B = A stormwater management plan must be implemented that may
require the overtreatment of runoff from associated roadways to
compensate for the lack of direct treatment of runoff from the
bridge surface itself. Periodic vacuuming should be considered.
The use of scupper drains should be limited as much as possible.
The Isle of Palms Connector: A Case
Study in Retrofitting
The incorporation of a stormwater management system
into a bridge design usually can be done without any
great difficulty. Trying to incorporate a system into a
bridge already designed and ready for permitting, how-
ever, can be much more difficult. Such was the case with
the Isle of Palms Connector, an 11,500-ft, $30 million
bridge that was to provide alternate access to the Isle
387
-------
of Palms, a barrier island town just outside of Char-
leston, South Carolina. The bridge route called for the
crossing of some 9,000 ft of marsh, two major marsh
creeks, and the Intracoastal Waterway. Location and
environmental studies and basic bridge design were
completed in 1979, the same year the state's coastal
zone management program was authorized. Funding
limitations slowed the process until 1987, when federal
funds became available.
The proposed route for the Isle of Palms Connector
crossed over some of the state's most productive com-
mercial and recreational shellfish grounds. The live oys-
ter volume in Hamlin Creek and Swinton Creek alone
was surveyed by the South Carolina Wildlife and Marine
Resources Department at 32,000 bushels. Annual clam
production potential in the immediate area of the bridge
is estimated to be between 140,000 and 250,000 clams.
The bridge was originally designed with traditional meth-
ods of handling stormwater; water was drained directly
from the bridge through scuppers except at one pre-
viously identified sensitive area, where discharge was
eliminated. Because there were no objections to the
stormwater design in the original environmental impact
assessment, approved by the FHWA in 1986, the South
Carolina Department of Highways and Public Transpor-
tation was reluctant to make any changes. Relocation
of the bridge was not an option, nor, as it turned out, was
redesign of the bridge. The bridge was designed with
approximately 9,000 ft at 0.0 percent grade, with ele-
vated spans over the Intracoastal Waterway and one of
the creeks (Figure 1). The State Highway Department
estimated redesign to accommodate positive flows to
both ends of the bridge at $10 million, a one-third in-
crease in bridge cost (7).
The South Carolina Coastal Council, however, as pri-
mary permitting agency for the bridge, was sensitive to
public demand that the bridge must incorporate a storm-
water management system that met basic coastal
stormwater guidelines (8). After several meetings, which
included public input, the South Carolina Department of
Highways and Transportation agreed to work with the
Coastal Council in addressing stormwater within the
limits of two constraints: the bridge location could not
be changed, and the stormwater system must be adapt-
able to the existing bridge design. Once this decision
was reached, both agencies began a serious and coop-
erative effort in resolving the problem. It was immediately
apparent that the traditional methods of stormwater
treatment usually employed on high land must be ruled
out; other than pumping, which was explored and rejected
due to cost, there was no way to get the runoff back to
high ground fortreatment. Therefore, the study team threw
Overall Length of Bridge = 11,460 Ft
r 100
50
J.
r PVI El. 15.0
/ LVC = 600 Ft
PRCEI. 17.5 -i
0.00%
v
» +s.oo%y
0 7
C
| PVI El. 25.0 pvc EL 43 5 /r~%
co i / ~~ ,uuu ri Y ^^ ro
_»i./ ,- PRC El. 17.5 Y 8
I/ / / £
\ * J ooo°/ X Mean 1
J, p-J^ 0-°% ^sf High Water
I ovol Fl 9 ^T-*
,-PVI El. 108.5
/ LVC = 2,500 Ft
V -5.00%
A. El. 76.0
CT
g1 >v ,- PVT El. 43.5
10 \
>- -u- .^m
r- El. 10.5
M
Proposed Profile
Scale: Horiz. 1 In. = 1,000 Ft Vert. 1 In. = 40 Ft
Overall Length of Bridge = 11,460 Ft
r~ 100
- 50
- 0
-20
PVI El. 15.0
J LVC = 600 Ft
___
0
0
O
c
o
c
^
rPVI El. 108.5
y LVC = 2,500 Ft
+5.00%/Ty -5.00%
PVT El. 43.5 -v J El. 76.0 r F
\
\ ^~-
r El. 26.43 \ / |
/ ^L* ro
/ 4-n ^o^o/ ! - ro
^ ^ ' ' Mean £
' /
t-~N^ /
1 c |\
1 11 \
High Water «r .^ X
El. 10.5
I O
Closed Drainage System Profile
Scale: Horiz. 1 In. = 1,000 Ft Vert. 1 In. = 40 Ft
Figure 1. Proposed and closed drainage system profiles for Isle of Palms Connector.
388
-------
out the preconceived traditional approaches and fo-
cused on the basic tenents of stormwater management:
retention, settling, and pollutant removal. A variety of alter-
natives were identified, evaluated, and rejected for various
reasons. Among these alternatives were storage and re-
tention in gutters of several configurations along the shoul-
der of the bridge roadway and the design of an "in the
marsh" sand filtering system constructed in large cylinders.
What emerged from this process was the design of an
open-faced "runoff pan," 15 ft long by 32 in. wide, to be
bolted in place to catch the discharge from each scupper
drain (Figure 2). The pan, constructed of fiberglass, was
1 ft deep with a baffle overflow to prevent the discharge
of oil and grease. In addition to containing the first 3/4
in. of runoff, the pans were to be managed with a
vigorous maintenance program that would include
dry/wet vacuuming on a to-be-determined basis and
disposal of the residue in accordance with state hazard-
ous waste regulations. The estimated cost for the storm-
water management system, to include piping of runoff
from the vertical expansions of the bridge to high ground
and an adjacent spoil disposal area, was about 3.5
percent of the total bridge cost.
Accompanying this alternative was the commitment of
the State Highway Department and the Coastal Council
to develop a monitoring program to test the effective-
ness of this technique. The monitoring program was to
be implemented on completion of the bridge, estimated
for the fall of 1993. Background data was collected in
the summer and fall of 1993.
Both agencies, along with the concerned public, eagerly
await the results of the monitoring. If successful, the
runoff pan may provide one alternative for addressing
stormwater management on existing bridges crossing
sensitive waters.
Conclusion
Roadways and bridges are certainly not unique in their
potential contribution to lessened water quality. Virtually
all human activities on the land, on the water, and in the
air contribute to the problem. No one solution to correct
the problem exists; rather, the solution lies with the
incremental "micromanagement" of each specific activ-
ity that contributes to the problem.
References
1. U.S. Department of Transportation, Federal Highway Administra-
tion. 1984. Sources and migration of highway runoff pollutants, Vol.
III. Research report. Pub. No. FHWA/RD-84/059. McLean, VA.
2. U.S. Department of Transportation, Federal Highway Administration.
1988. Retention, detention, and overland flow for pollutant removal
from highway stormwater runoff: Interim guidelines for management
measures. Pub. No. FHWA/RD-87/056. McLean, VA.
3. U.S. Department of Transportation, Federal Highway Administra-
tion. 1988. Effects of highway runoff on receiving waters, Vol. III.
Resource document for environmental assessments. Pub. No.
FHWA/RD-84/064. McLean, VA.
6 In. on Center Drain @ 15 Ft
3-ln. x 8-ln. Outfall
(One per Unit)
1 Ft 1 In.
15 Ft Long Runoff Pan
(One per Drain)*
Support Bracket Spaced as Needed
"Shape and size of runoff pan may vary upon completion of final design.
Figure 2. Schematic "runoff pan" detail: proposed Isle of Palms Connector between U.S. 17-701 and 14th Avenue, Charleston County,
South Carolina.
389
-------
4. Dupuis, T.V., and N.P Kobriger. 1985. Effects of highway runoff on
receiving waters, Vol. IV. Procedural guidelines for environmental
assessments. Draft Report No. FHWA/RD-84/065. Washington, DC:
Federal Highway Administration.
5. Portele, G.J., B.W. Mar, R.R. Horner, and E.B. Welch. 1982. Ef-
fects of Seattle area highway stormwater runoff on aquatic biota.
WA-RD-39.11. Washington State Department of Transportation.
6. U.S. EPA. 1993. Guidance specifying management measures for
sources of nonpoint pollution in coastal waters. EPA/840/B-92/002.
Washington, DC.
7. South Carolina Coastal Council. 1989. Coastal Council Permit
CC-89-275. Charleston, SC: South Carolina Department of High-
ways and Public Transportation for the Isle of Palms Connector.
8. South Carolina Coastal Council. 1988. Stormwater management
guidelines. Charleston, SC.
Additional Reading
1. Gupta, M.K., R.W Agnew, D. Gruber, and W. Kreutzberger. 1981.
Constituents of highway runoff, Vol. IV Characteristics of highway
runoff from operating highways. Report No. FHWA/RD-81/045.
Washington, DC: Federal Highway Administration.
2. Maestri, B., F. Johnson, C.W Burch, and B.L. Dawson. 1985.
Management practices for mitigation of highway stormwater runoff
pollution, Vol. IV Executive summary. Report No. FHWA/RD-
85/004. Washington, DC: Federal Highway Administration.
390
-------
Controlling Pollutants in Runoff From Industrial Facilities
Kevin Weiss
Storm Water Section, NPDES Permits Division, U.S. Environmental Protection Agency,
Washington, DC
Abstract
Industrial facilities can be significant contributors of pol-
lutants to urban runoff. On November 16,1990, the U.S.
Environmental Protection Agency (EPA) published Na-
tional Pollutant Discharge Elimination System (NPDES)
permit application requirements for "stormwater dis-
charges associated with industrial activities." These
regulations provide a framework for reducing pollutants
in runoff from the industrial facilities addressed. EPA
subsequently developed a long-term strategy for issuing
NPDES permits for these discharges. As the initial step
in this strategy, the Agency issued general permits on
September 9, 1992, and September 25, 1992, for the
majority of stormwater discharges in states where EPA
issues NPDES permits. This paper provides an over-
view of major categories of sources that contribute pol-
lutants to runoff at industrial sites and describes
pollution prevention measures in EPAs NPDES general
permits.
Introduction
Pollutants in urban runoff depend in part on the nature
of land use. Several studies indicate that runoff from
industrial land uses is of relatively poorer water quality
than runoff from other general land uses (1-5). In addi-
tion, industrial sites can be significant sources of pol-
luted, uncontrolled nonstormwater to separate storm
sewers (6, 7).
Source of Pollutants to Industrial Runoff
The volume and quality of stormwater discharges asso-
ciated with industrial facilities depend on several factors,
including the industrial activities occurring at the facility,
the nature of precipitation, and surface imperviousness.
The sources of pollutants that can affect the quality of
stormwater from industrial facilities differ with the type
of operations and specific facility features. For example,
air emissions may be a significant source of pollutants
at some facilities, material storage operations at others,
and still other facilities may discharge stormwater asso-
ciated with industrial activity with relatively low levels of
pollutants.
Six classes of activities can be identified as major po-
tential sources of pollutants in stormwater discharges
associated with industrial activity (7-11):
Loading or unloading of dry bulk materials or liquids.
Outdoor storage of raw materials or products.
Outdoor process activities.
Dust or particulate generating processes.
Illicit connections or inappropriate management
practices.
Waste disposal practices.
The potential for pollution from many of these activities
may be influenced by the presence and use of toxic
chemicals.
Loading and unloading operations typically are per-
formed along facility access roads and railways and at
loading/unloading docks and terminals. These opera-
tions include pumping of liquids or gases from trucks or
rail cars to a storage facility or vice versa; pneumatic
transfer of dry chemicals to or from the loading or un-
loading vehicle; transfer by mechanical conveyor sys-
tems; and transfer of bags, boxes, drums, or other
containers from vehicles by forklift trucks or other mate-
rials handling equipment. Material spills or losses may
discharge directly to the storm drainage systems or may
accumulate in soils or on surfaces, to be washed away
during a storm or facility washdown.
Outdoor storage includes the storage of fuels, raw ma-
terials, byproducts, deicing chemicals, intermediates,
final products, and process residuals and wastes. Meth-
ods of material storage include use of storage contain-
ers (e.g., drums or tanks), platforms or pads, bins, silos,
boxes, and piles. Materials, containers, and material
391
-------
storage areas exposed to rainfall or runoff may contrib-
ute pollutants to stormwater when solid materials wash
off or materials dissolve into solution.
Other outdoor activities include certain types of manu-
facturing and commercial operations and land-disturb-
ing operations. Although many manufacturing activities
are performed indoors, some activities (e.g., equipment
and vehicle maintenance and cleaning, timber process-
ing, rock crushing, vehicle maintenance and cleaning,
and concrete mixing) typically occur outdoors. Process-
ing operations may result in liquid spillage and losses of
material solids to the drainage system or surrounding
surfaces, or creation of dusts or mists that can be de-
posited locally. Some outdoor industrial activities cause
substantial physical disturbance of land surfaces that
result in soil erosion by stormwater. For example, dis-
turbed land occurs in construction and mining. Disturbed
land may result in soil losses and other pollutant load-
ings associated with increased runoff rates. Facilities
whose major process activities are conducted indoors
may still apply chemicals such as herbicides, pesticides,
and fertilizer outdoors for a variety of purposes.
Dust or particulate generating processes include indus-
trial activities with stack emissions or process dusts that
settle on plant surfaces. Localized atmospheric deposi-
tion can be a particular concern with heavy manufactur-
ing industries. For example, monitoring of areas
surrounding smelting industries has shown much higher
levels of metals at sites nearest the smelter. Other in-
dustrial sites, such as mines, cement manufacturing
plants, and refractories, generate significant levels of
dusts.
Illicit connections or inappropriate management prac-
tices result in improper nonstormwater discharges to
storm sewer systems. Pollutants from nonstormwater
discharges to the storm sewer systems are caused
typically by a combination of improper connections,
spills, improper dumping, and improperly disposed of
rinse waters, cooling waters, or other process and sani-
tary wastewater. Often dischargers believe that the ab-
sence of visible solids in a discharge is equivalent to the
absence of pollution. Illicit connections are often asso-
ciated with floor drains that are connected to separate
storm sewers. Rinse waters used to clean or cool ob-
jects discharge to floor drains connected to separate
storm sewers. Large amounts of rinse waters that dis-
charge to floor drains may originate from industries
using regular washdown procedures; for example, bot-
tling plants use rinse waters for removing waste prod-
ucts, debris, and labels. Rinse waters can be used to
cool materials by dipping, washing, or spraying objects
with cool water; for example, rinse water is sometimes
sprayed over the final products of a metal plating facility
for cooling purposes. Condensate return lines of heat
exchangers often discharge to floor drains. Heat ex-
changers, particularly those used under stressed condi-
tions (e.g., exposure to corrosive fluids), such as in the
metal finishing and electroplating industry, may develop
pinhole leaks that result in contamination of condensate
by process wastes. These and other nonstormwater
discharges to storm sewers may be intentional, based
on the belief that the discharge does not contain pollut-
ants, or they may be inadvertent, if the operator is
unaware that a floor drain is connected to the storm
sewer.
Waste management practices include temporary stor-
age of waste materials and operations at landfills, waste
piles, and land application sites that involve land dis-
posal. Outdoor waste treatment operations also include
wastewater and solid waste treatment and disposal
processes, such as waste pumping, additions of treat-
ment chemicals, mixing, aeration, clarification, and sol-
ids dewatering.
Options for Control
Options for controlling pollutants in stormwater dis-
charges associated with industrial activity are discussed
below in terms of two major pollutant sources: 1) mate-
rials discharged to separate storm sewers via illicit con-
nections, improper dumping, and spills; and 2) pollutants
associated with runoff.
Nonstormwater Sources
As discussed above, nonstormwater discharges to
separate storm sewers come from a wide variety of
sources, including illicit connections, improper dumping,
spills, or leakage from storage tanks and transfer areas.
Measures to control spills and visible leakage can be
incorporated into the best management practices dis-
cussed below.
In many cases, operators of industrial facilities may be
unaware of illicit discharges or other nonvisible sources
of nonstormwater to a storm sewer. In such cases, the
key to controlling these discharges is to identify them.
Several methods for identifying the presence of non-
stormwater discharges are discussed below. (A more
complete discussion of methods to identify illicit connec-
tions can be found in U.S. EPA [6, 12]). A comprehen-
sive evaluation of the storm sewers at a facility often
should incorporate several of the following methods:
Evaluation of drainage map and inspections: Drain-
age maps should identify the key features of the
drainage system (i.e., each of the inlet and discharge
structures, the drainage area of each inlet structure,
storage and disposal units, and materials loading ar-
eas) that may be the source of an illicit discharge or
improper dumping. In addition, floor drains and other
water disposal inlets thought to be connected to the
sanitary sewer should be identified. A site inspection
392
-------
can be used to augment and verify map develop-
ment. These inspections, along with the use of the
drainage map, can be coordinated with other identi-
fication methods discussed below.
End-of-pipe screening: Discharge points or other ac-
cess points such as manhole covers can be in-
spected for the presence of dry weather discharges
and other signs of nonstormwater discharges. Dry
weather flows, material deposits, and stains are often
indicators of illicit connections. Dry weather flows can
be screened by a variety of methods. Inexpensive
onsite tests include measuring pH; observing for oil
sheens, scums, and discoloration of pipes and other
structures; and colorimetric detection for chlorine, de-
tergents, metals, and other parameters. In some
cases, it may be appropriate to collect samples for
more expensive analysis in a laboratory for fecal coli-
form, fecal Streptococcus, volatile organic carbon, or
other appropriate parameters.
Manhole and internal TV inspection: Inspection of
manholes and storm sewers, either physically or by
television, can be used to identify a potential entry
point for illicit connections. TV inspections are rela-
tively expensive and generally should be used only
after a storm sewer has been identified as having
illicit connections.
Dry weather testing: Where storm sewers do not
normally discharge during dry weather conditions,
water can be introduced into floor drains, toilets, and
other points where nonstormwater discharges are
collected. Storm drain outlets are then observed for
possible discharges.
Dye testing: Dry weather discharges from storm
sewers can occur for several legitimate reasons, in-
cluding ground-water infiltration or the presence of a
continuous discharge subject to a National Pollutant
Discharge Elimination System (NPDES) permit.
Where storm sewers do have a discharge during dry
weather conditions, dye testing for illicit connections
can be used. Dye testing involves introducing
fluorometric or other types of dyes into floor drains,
toilets, and other points where nonstormwater dis-
charges are collected. Storm drain outlets and man-
holes are then observed for possible discharges. Dye
testing can also be used to identify unknown sub-
merged outfalls to nearby receiving waters.
Water balance: Many sewage treatment plants require
that industrial discharges measure the volume of ef-
fluent discharged to the sanitary sewer system. Similarly,
the volume of water supplied to a facility is generally
measured. A significantly higher volume of water sup-
plied to the facility relative to that discharged to the
sanitary sewer and other consumptive uses may be
an indication of illicit connections. This method is
limited by the accuracy of the flow meters used.
Schematics: Where they exist, accurate piping sche-
matics can be inspected as a first step in evaluating
the integrity of the separate storm sewer system. The
use of schematics is limited because schematics usu-
ally reflect the design of the piping system and may
not reflect the actual configuration constructed. Sche-
matics should be updated or corrected based on ad-
ditional information found during inspections.
Smoke tests are sometimes listed in the literature as a
method for detecting illicit connections to separate storm
sewers. While smoke tests can be used to identify inflow
of stormwater to sanitary sewers, they can be much less
effective for identifying discharges of nonstormwater to
storm drains. This is because many nonstormwater
drainage locations have a sewer gas trap that blocks
smoke used in a test. Smoke tests can identify non-
stormwater discharges to storm drains if the piping for
the nonstormwater discharge has a vent or does not
have a sewer gas trap.
Options for Preventing Pollutants in
Stormwater
The following five categories describe options for reducing
pollutants in stormwater discharges from industrial plants:
Providing end-of-pipe treatment.
Implementing best management practices (BMPs) to
prevent pollution.
Diverting stormwater discharge to treatment plants.
Using traditional stormwater management practices.
Eliminating pollution sources/water reuse.
A comprehensive stormwater management program for
a given plant often includes controls from each of these
categories. Development of comprehensive control
strategies should be based on a consideration of plant
characteristics.
End-of-Pipe Treatment
At many types of industrial facilities, it may be appropri-
ate to collect and treat the runoff from targeted areas of
the facility. This approach was taken with the 10 indus-
trial categories with national effluent guideline limita-
tions for stormwater discharges: cement manufacturing
(40 CFR411), feedlots (40 CFR412), fertilizer manufac-
turing (40 CFR 418), petroleum refining (40 CFR 419),
phosphate manufacturing (40 CFR 422), steam electric
(40 CFR 423), coal mining (40 CFR 434), mineral mining
and processing (40 CFR 436), ore mining and dressing
(40 CFR 440), and asphalt emulsion (40 CFR 443).
393
-------
Best Management Practices
BMPs encompass a wide range of management proce-
dures, schedules of activities, prohibitions on practices,
and other management practices to prevent or reduce
the pollution of waters of the United States. BMPs also
include operating procedures, treatment requirements and
practices to control plant site runoff, and drainage from
raw materials storage, spills, or leaks. Requirements for
BMP-based pollution prevention plans generally appli-
cable to all industries are discussed in more detail in the
paper in the context of the U.S. Environmental Protec-
tion Agency's (EPA's) general permits for stormwater
discharges associated with industrial activity.
In addition to generic BMPs or pollution prevention plans,
industry- or activity-specific BMPs can be used. Table 1
provides a listing of industry-specific BMPs that the
Washington State Department of Ecology has developed.1
Diversion of Discharge to Treatment Plant
Where stormwater discharges contain significant amounts
of pollutants that can be removed by a wastewater or
sewage treatment plant, the stormwater discharge can
be diverted to a wastewater treatment plant or sanitary
sewage system. Such diversions must be coordinated
with the operators of the sewage treatment plant and the
collection system to avoid problems with either combined
sewer overflows (CSOs), basement flooding, or wet
weather operation of the treatment plant. Where CSO
discharges, flooding or plant operation problems can
result, and onsite storage followed by a controlled re-
lease during dry weather conditions may be considered.
Traditional Stormwater Management Practices
In some situations, traditional stormwater management
practices such as grass swales, catch basin design and
maintenance, infiltration devices, unlined onsite reten-
tion and detention basins, regional controls (offsite re-
tention or detention basins), and oil and grit separators
can be applied to an industrial setting. Care must be
taken, however, to evaluate the potential of many of
these traditional devices for ground-water contamina-
tion. Other types of controls, such as secondary contain-
ment systems, can be used to prevent catastrophic
events that can lead to surface or ground-water con-
tamination via traditional stormwater measures. In some
cases, it is appropriate to limit traditional stormwater
The document Best Management Practices for the Use and Storage
of Hazardous Materials (14) also provides examples of industry-
specific BMPs. The guidance addresses small mechanical repair
facilities, large mechanical repair facilities, dry cleaning facilities,
junkyards, photo processing facilities, print shops and silk screen
shops, machine shops and airport maintenance facilities, boat
manufacturing and repair facilities, concrete plants and mining fa-
cilities, agricultural facilities, paint manufacturers and distributors,
and plastics manufacturers.
management practices to those areas of the drainage
system that generate stormwater with relatively low lev-
els of pollutants (e.g., many rooftops, parking lots, etc.).
At facilities located in northern areas of the country,
snow removal activities may play an important role in a
stormwater management program.
Elimination of Pollution Sources/Water Reuse
In some cases, the elimination of a pollution source or
water reuse may be the most cost-effective way to
control pollutants in stormwater discharges associated
with industrial activity. Options for eliminating pollution
sources include reducing onsite air emissions affecting
runoff quality, changing chemicals used at the facility, and
modifying materials management practices such as
moving storage areas into buildings. Water reuse in-
volves collecting runoff and using it in a process or in some
mannerthat does not release the pollutants in the storm-
water to the environment. For example, many inorganic
wood preserving facilities use drip pad runoff to dilute
wood preserving fluids used in their processes. In some
cases, it may be less expensive to store and treat storm-
water for subpotable, industrial water supply purposes
than purchasing municipal potable water.
Clean Water Act Requirements
In 1972, the Clean Water Act (CWA) was amended to
provide that the discharge of any pollutants to waters of
the United States from a point source is unlawful, except
where the discharge is authorized by an NPDES permit.
The term "point source" is broadly defined to include "any
discernible, confined and discrete conveyance, includ-
ing but not limited to any pipe, ditch, [or] channel, . . .
from which pollutants are or may be discharged." Con-
gress has specifically exempted agricultural stormwater
discharges and return flows from irrigated agriculture
from the definition of point source.
Most court cases have supported a broad interpretation of
the term "point source" under the CWA. For example, the
holding in Sierra Club v. Abston Construction Co., Inc.,
620 F2d.41 (5th Cir., 1980) indicates that changing the
surface of land or establishing grading patterns on land
where the runoff from the site ultimately is discharged
to waters of the United States will result in a point
source:
A point source of pollution may be present where
[dischargers] design spoil piles from discarded over-
burden such that, during periods of precipitation,
erosion of spoil pile walls results in discharges into
a navigable body of water by means of ditches,
gullies and similar conveyances, even if the [dis-
chargers] have done nothing beyond the mere col-
lection of rock and other materials. . . . Nothing in
the Act relieves [dischargers] from liability simply
because the operators did not actually construct
394
-------
Table 1. Categories of Targeted Stormwater Controls Addressed in Puget Sound Guidance (13)
Category Targeted Stormwater Controls
Manufacturing facilities
Transportation and communication
Wholesale and retail businesses
Service businesses
Pubic agencies
Source controls
Cement
Chemical
Concrete products
Electrical products
Food products
Glass products
Industrial machinery and equipment, trucks and trailers, aircraft, parts and aerospace,
railroad equipment
Log storage and sorting yards, debarking
Metal products
Petroleum products
Printing and publishing
Rubber and plastic products
Ship and boat building and repair yards
Wood products
Wood treatment
Other manufacturing businesses
Airfields and aircraft maintenance
Fleet vehicle yards
Railroads
Private utility corridors
Warehouses and miniwarehouses
Other transportation and communication businesses
Gas stations
Recyclers and scrap yards
Restaurants/fast food
Retail general merchandise
Retail/Wholesale vehicle and equipment dealers
Retail/Wholesale nurseries and building materials
Retail/Wholesale chemicals and petroleum
Retail/Wholesale foods and beverages
Other retail/wholesale businesses
Animal care services
Commercial car and truck washes
Equipment repair
Laundries and other cleaning
Marinas and boat clubs
Golf and country clubs, golf courses, and parks
Miscellaneous services
Professional services
Vehicle maintenance and repair
Multifamily residences
Construction businesses
Public buildings and streets
Vehicle and equipment maintenance shops
Maintenance of open space areas
Maintenance of public Stormwater facilities
Maintenance of roadside vegetation and ditches
Maintenance of public utility corridors
Water and sewer districts and departments
Port districts
Fueling stations
Vehicle/Equipment washing and steam cleaning
Loading and unloading liquid materials
Liquid storage in aboveground tanks
Container storage of liquids, food wastes, and dangerous wastes
Outside storage of raw materials, byproducts, and finished products
Outside manufacturing activities
Emergency spill cleanup plans
Vegetation management/integrated pest management
Maintenance of storm drainage facilities
Locating illicit connections to storm drains
395
-------
those conveyances. . . . Conveyances of pollution
formed either as a result of natural erosion or by
material means, and which constitute a component
of a drainage system may fit the statutory definition
and thereby subject the operators to liability under
the Act.
Although the definition of point source is very broad,
before 1987 efforts under the NPDES program to control
water pollution focused on controlling pollutants in dis-
charges from publicly owned treatment works (POTWs)
and industrial process wastewaters. The major excep-
tions to this are the 10 effluent limitation guidelines that
EPA has issued for stormwater discharges: cement
manufacturing (40 CFR 411), feedlots (40 CFR 412),
fertilizer manufacturing (40 CFR 418), petroleum refin-
ing (40 CFR 419), phosphate manufacturing (40 CFR
422), steam electric (40 CFR 423), coal mining (40 CFR
434), mineral mining and processing (40 CFR 436), ore
mining and dressing (40 CFR 440), and asphalt emul-
sion (40 CFR 443).
As part of the Water Quality Act of 1987, Congress
added Section 402(p) to the CWA to require EPA to
develop a comprehensive, phased program for regu-
lated stormwater discharges under the NPDES program.
One of the first priorities under the stormwater program
was to develop NPDES requirements for stormwater
discharges associated with industrial activity.
On November 16,1990, EPA published the initial NPDES
regulations under Section 402(p) of the CWA (see 55
FR 47990). The November 16, 1990, regulations:
Defined the initial scope of the program by defining
the terms "stormwater discharge associated with in-
dustrial activity" and large and medium "municipal
separate storm sewer systems."
Established permit application requirements.
The regulatory definition of the term "stormwater dis-
charge associated with industrial activity" is provided at
40 CFR 122.26(b)(14) and addresses point source dis-
charges of stormwater from eleven major categories of
facilities. Table 2 summarizes these 11 major categories.
The NPDES regulations provided three options for sub-
mitting permit applications for stormwater discharges
associated with industrial activity: 1) individual applica-
tions, 2) group applications for groups of similar indus-
trial discharges, and 3) where an appropriate general
permit has been issued, submittal of a notice of intent
(NOI) to be covered by a general permit. The group
application option is no longer available; EPA received
over 1,100 group applications covering over 45,000 fa-
cilities. The Agency has organized these applications
into the 32 industrial sectors shown in Table 3 and
intends to develop guidance on issuing permits for the
32 industrial sectors.
Table 2. Summary of Classes of Industrial Facilities
Addressed by Regulatory Definition of "Stormwater
Discharge Associated With Industrial Activity"
Class Description
(i) Facilities subject to stormwater effluent limitations
guideline, new source performance standards, or toxic
pollutant effluent standards (see 40 CFR Subpart N)
(ii) Manufacturing facilities classified as Standard Industrial
Classification (SIC) 24 (except 2434), 26 (except 265
and 267), 28 (except 283), 29, 311, 32 (except 323),
33, 3441, and 373
(iii) Active and inactive mining operations classified as SIC
10-14
(iv) Hazardous waste treatment, storage, or disposal
facilities that are operating under interim status or a
permit under Subtitle C of RCRA
(v) Landfills, land application sites, and open dumps that
receive industrial wastes
(vi) Recycling facilities, including metal scrapyards, battery
reclaimers, salvage yards, and automobile junkyards
(vii) Steam electric power generating facilities
(viii) Transportation facilities classified as SIC 40, 41, 42
(except 4221-25), 43, 44, 45, and 5171, which have
vehicle maintenance shops, equipment cleaning
operations, or airport deicing operations
(ix) Sewage treatment plants with a design flow of 1.0
million gal/day or more or required to have an approved
pretreatment program
(x) Construction activities except operations that result in
the disturbance of less than 5 acres of total land area
and that are not part of a larger common plan of
development or sale
(xi) Facilities under SIC 20, 21, 22, 23, 2434, 25, 265, 267,
27, 283, 285, 30, 31 (except 311), 323, 34 (except
3441), 35, 36, 37 (except 373), 38, 39, and 4221-25
(and which are not otherwise included within categories
(i)-(x))
Table 3. Industrial Sectors Identified in NPDES Group
Application Process
Sector SIC Codes/Activities Represented
Number of
Facilities
1 SIC 24Lumber and Wood Products 2,640
2 SIC 26Paper and Allied Products 1,023
3 SIC 28Chemicals and Allied Products 1,498
4 SIC 29Petroleum Refining and 2,245
Related Industries
5 SIC 32Stone, Clay, Glass, and 4,786
Concrete Products
6 SIC 33Primary Metal Industries 730
7 SIC 10Metal Mining 188
8 SIC 12Coal Mining 495
9 SIC 13Oil and Gas Extraction 457
10 SIC 14Mining and Quarrying of 2,437
Nonmetallic Minerals
11 Hazardous Waste Treatment Storage 77
or Disposal Facilities
396
-------
Table 3. Industrial Sectors Identified in NPDES Group
Application Process (continued)
Sector SIC Codes/Activities Represented
Number of
Facilities
12 Industrial Landfills, Land Application 1,430
Sites, and Open Dumps
13 SIC 5015Used Motor Vehicle Parts 2,009
14 SIC 5093Scrap and Waste Materials 1,688
15 Steam Electric Power Generating 162
Facilities
16 SIC 40Railroad Transportation 1,024
17 SIC 41Local and Suburban Transit 13,089
and Interurban Highway Passenger
Transportation
SIC 42Motor Freight Transportation
SIC 43United States Postal Service
18 SIC 44Water Transportation 368
19 SIC 3731Ship Building and Repairing 498
SIC 3732Boat Building and Repairing
20 SIC 45Air Transportation 1,581
21 SIC 5171Petroleum Bulk Stations 131
and Terminals
22 Domestic Wastewater Treatment Plants 1,249
23 SIC 20Food and Kindred Products 2,608
SIC 21Tobacco Products
24 SIC 22Textile Mill Products 872
SIC 23Apparel and Other Finished
Products Made From Fabrics and
Similar Materials
25 SIC 25Furniture and Fixtures 339
26 SIC 27Printing, Publishing, and 65
Allied Industries
27 SIC 30Rubber and Miscellaneous 190
Plastic Products
28 SIC 31Leather and Leather Products 61
29 SIC 34Fabricated Metal Products 965
SIC 391Jewelry, Silverware, and
Plated Ware
30 SIC 35Industrial and Commercial 935
Machinery
SIC 37Transportation Equipment
31 SIC 36Electronic Components 14
SIC 357Computer and Office
Equipment
SIC 38Measuring, Analyzing, and
Control Instruments; Photographic and
Optical Goods, Watches, and Clocks
32 SICMiscellaneous Manufacturing 769
Industries
Long-Term Strategy
Many of the initial concerns regarding the NPDES
stormwater program focused on adapting the NPDES
permit program to effectively address the large number
of stormwater discharges associated with industrial ac-
tivity. In response to these concerns, EPA developed a
strategy for permitting stormwater discharges associ-
ated with industrial activity that will serve as a foundation
for future program development and technology trans-
fer. The strategy consists of two major components: a
tiered framework for developing permitting priorities and
a framework for the development of state stormwater
management plans.
Permitting Priorities
Underthe strategy, most stormwater permitting activities
are described in terms of the following four classes of
activities:
Tier IBaseline permitting: One or more general per-
mits will be developed initially to cover the majority
of stormwater discharges associated with industrial
activity.
Tier IIWatershed permitting: Facilities within water-
sheds shown to be adversely affected by stormwater
discharges associated with industrial activity will be
targeted for individual or watershed-specific general
permits.
Tier IIIIndustry-specific permitting: Specific industry
categories will be targeted for individual or industry-
specific general permits.
Tier IVFacility-specific permitting: A variety of fac-
tors will be used to target specific facilities for indi-
vidual permits.
These four classes of activities will be implemented over
time and will reflect priorities within given states. In most
states, Tier I activities will be the starting point. Initially,
the coverage of the baseline permits will be broad. As
priorities and risks within the state are evaluated, how-
ever, classes of stormwater discharges or individual
stormwater discharges will be identified for Tier II, III, or
IV permitting activities.
State Stormwater Management Programs
State stormwater management programs are to provide,
among other things, a description of NPDES permit
issuing activities for stormwater discharges associated
with industrial activity, including categories of industrial
activity that are being considered for industry-specific
general permits. These plans will assist EPA in develop-
ing technology transfer activities with other states,
evaluating states' progress in implementing stormwater
permitting activities, and identifying both successes and
difficulties with ongoing program implementation.
EPA's Baseline General Permits
Consistent with the long-term permit issuance strategy,
EPA published Tier I general permits, which potentially
could apply to the majority of stormwater discharges
associated with industrial activity located in 12 states on
397
-------
September 9, 1992, and September 25, 1992 (see 57
FR 41236 and 57 FR 44438). The 12 states where the
EPA general permits apply are Alaska, Arizona, Florida,
Idaho, Louisiana, Maine, Massachusetts, New Hamp-
shire, New Mexico, Oklahoma, South Dakota, and
Texas. Other states have authorized NPDES state pro-
grams, and the state issues NPDES permits instead of
EPA.
Consolidating many sources under a general permit
greatly reduces the administrative burden of issuing
permits for stormwater discharges associated with indus-
trial activity. Several advantages to this approach are:
Pollution prevention measures and/or BMPs are es-
tablished for discharges covered by the permit.
Facilities whose discharges are covered by the per-
mit are certain of their legal responsibilities and have
an opportunity to comply with the CWA.
EPA and authorized NPDES states will begin to col-
lect and review data on stormwater discharges from
priority industries, thereby supporting subsequent
permitting activities.
The public, including municipal operators of munici-
pal separate storm sewers, will have the opportunity
to review data and reports developed by industrial
permittees pursuant to NPDES requirements.
The baseline permits will provide a basis for coordi-
nating 1) requirements for stormwater discharges as-
sociated with industrial activity with 2) requirements
of municipal stormwater management programs in
permits for discharges from municipal separate storm
sewer systems.
The baseline permits will provide a basis for bringing
selected enforcement actions.
The baseline permit, along with state stormwater per-
mitting plans, will provide a focus for public comment
on draft permits and subsequent phases of the per-
mitting strategy for stormwater discharges.
The Agency believes that Tier I permits can establish the
appropriate balance between monitoring requirements
and implementable controls that will initiate facility-spe-
cific controls and provide sufficient data for compliance
monitoring and future program development.
Permit Requirements
The major requirements of EPAs Tier I stormwater gen-
eral permits are notification requirements, requirements
for stormwater pollution prevention plans, and special
requirements for selected facilities.
Notification Requirements
The general permits require the submittal of an NOI by
the discharger before the authorization of discharges. In
addition, operators of stormwater discharges that dis-
charge through a large or medium municipal separate
storm sewer system must, in addition to submitting an
NOI to the Director, submit a copy of the NOI to the
municipal operator of the system receiving the discharge.
Tailored Pollution Prevention Plan
Requirements
All facilities covered by EPAs general permits must
prepare and implement a stormwater pollution preven-
tion plan. These tailored requirements allow the imple-
mentation of site-specific measures that address
features, activities, or priorities for control associated
with the identified stormwater discharges. The approach
taken allows the flexibility to establish controls that can
appropriately address different sources of pollutants at
different facilities.
The pollution prevention approach adopted in the general
permits focuses on two major objectives: 1) to identify
sources of pollution potentially affecting the quality of
stormwater discharges from the facility, and 2) to de-
scribe and ensure implementation of practices to mini-
mize and control pollutants in stormwater discharges.
The stormwater pollution prevention plan requirements
in the general permits are intended to facilitate a proc-
ess whereby the operator of the industrial facility thor-
oughly evaluates potential pollution sources at the site
and selects and implements appropriate measures to
prevent or control the discharge of pollutants in storm-
water runoff. The process involves the following four
steps:
Formation of a team of qualified plant personnel re-
sponsible for preparing the plan and assisting the
plant manager in its implementation.
Assessment of potential stormwater pollution sources.
Selection and implementation of appropriate man-
agement practices and controls.
Periodic evaluation of the ability of the plan to prevent
stormwater pollution and comply with the terms and
conditions of this permit.
This process is shown in Figure 1. A complete descrip-
tion of this process can be found in U.S. EPA (15).
Pollution Prevention Team
As a first step in the process of developing and implement-
ing a stormwater pollution prevention plan, permittees
must identify a qualified individual or team of individuals
to be responsible for developing the plan and assisting
the facility or plant manager in its implementation. When
398
-------
Planning and Organization
Form pollution prevention team
Review other plans
1
Assessment Phase
Develop site map
Inventory exposed materials
Test for nonstormwater
Describe pollutant sources
BMP Identification Phase
Baseline BMPs
Select activity- and site-specific BMPs
Plan Review
and
Revision
Implementation Phase
Implement BMPs
Train employees
Evaluation/Monitoring
Conduct site inspections
BMP evaluations
Recordkeeping and reporting
Review and revise plan
Figure 1. Pollution prevention plan process.
selecting members of the team, the plant manager
should draw on the expertise of all relevant departments
within the plant to ensure that all aspects of plant opera-
tion are considered. The plan must clearly describe the
responsibilities of each team member as they relate to
specific components of the plan. In addition to enhanc-
ing the quality of communication between team members
and other personnel, clear delineation of responsibilities
will ensure that a specified individual or group of indi-
viduals addresses every aspect of the plan.
Description of Potential Pollution Sources
Each stormwater pollution prevention plan must de-
scribe activities, materials, and physical features of the
facility that may contribute significant amounts of pollut-
ants to stormwater runoff or, during periods of dry
weather, result in pollutant discharges through the sepa-
rate storm sewers or stormwater drainage systems. This
assessment of stormwater pollution risk will support
subsequent efforts to identify and set priorities for nec-
essary changes in materials, materials management
practices, or site features, as well as aid in the selection
of appropriate structural and nonstructural control tech-
niques. Plans must describe the site drainage, provide
an inventory of exposed materials, describe significant
spills and leaks that have occurred at the facility, and
include existing sampling data.
Each pollution prevention plan must include a certifica-
tion that discharges from the site have been tested or
evaluated for the presence of nonstormwater dis-
charges. The certification must describe possible signifi-
cant sources of nonstormwater, the results of any test
and/or evaluation conducted to detect such discharges,
the test method or evaluation criteria used, the dates on
which tests or evaluations were performed, and the
onsite drainage points directly observed during the test
or evaluation. Acceptable test or evaluation techniques
are discussed earlier in this paper.
The description of potential pollution sources culminates
in a narrative assessment of the risk potential that
sources of pollution pose to stormwater quality. This
assessment should clearly point to activities, materials,
and physical features of the facility that have a reason-
able potential to contribute significant amounts of pollut-
ants to stormwater. Any such activities, materials, or
features must be addressed by the measures and con-
trols subsequently described in the plan. In conducting
the assessment, the facility operator must consider
loading and unloading operations, outdoor storage ac-
tivities, outdoor manufacturing or processing activities,
significant dust or particulate generating processes, and
onsite waste disposal practices. The assessment
must list any significant pollution sources at the site
and identify the pollutant parameter or parameters (i.e.,
biochemical oxygen demand, suspended solids, etc.)
associated with each source.
Measures and Controls
Following completion of the source identification and
assessment phase, the permittee must evaluate, select,
and describe the pollution prevention measures, BMPs,
and other controls that the facility will implement. BMPs
include processes, procedures, schedules of activities,
prohibitions on practices, and other management prac-
tices that prevent or reduce the discharge of pollutants
in stormwater runoff.
The plan requirements emphasize the implementation
of pollution prevention measures that reduce possible
pollutant discharges at the source. Source reduction
measures include, among others, preventive mainte-
nance, chemical substitution, spill prevention, good
housekeeping, training, proper materials management,
material segregation or covering, water diversion, and
dust control. The remaining classes of BMPs, which
involve recycling or treatment of stormwater, allow the
reuse of stormwater or attempt to lower pollutant con-
centrations before discharge.
The pollution prevention plan must include a schedule
specifying the time or times during which each control
or practice will be implemented. In addition, the plan
should discuss ways in which the controls and practices
relate to one another and, when taken as a whole,
produce an integrated and consistent approach for pre-
venting or controlling potential stormwater contamina-
tion problems. The portion of the plan that describes the
399
-------
measures and controls must address the following mini-
mum components:
Good housekeeping: Good housekeeping involves
using common sense to identify ways to maintain a
clean and orderly facility and keep contaminants out
of separate storm sewers. It includes establishing
protocols to reduce the possibility of mishandling
chemicals or equipment, and training employees in
good housekeeping techniques.
Preventive maintenance: Permittees must develop a
preventive maintenance program that involves regu-
lar inspection and maintenance of stormwater man-
agement devices and other equipment and systems.
The program description should identify the devices,
equipment, and systems that will be inspected; provide
a schedule for inspections and tests; and address
appropriate adjustment, cleaning, repair, or replace-
ment of devices, equipment, and systems. For storm-
water management devices such as catch basins and
oil/water separators, the preventive maintenance pro-
gram should provide for periodic removal of debris to
ensure that the devices are operating efficiently.
Spill prevention and response procedures: Based on
an assessment of possible spill scenarios, permittees
must specify appropriate material handling proce-
dures, storage requirements, containment or diver-
sion equipment, and spill cleanup procedures that will
minimize the potential for spills and in the event of a
spill enable proper and timely response. Areas and
activities that typically pose a high risk for spills in-
clude loading and unloading areas, storage areas,
process activities, and waste disposal activities.
These activities and areas, and their accompanying
drainage points, must be described in the plan. For
a spill prevention and response program to be effec-
tive, employees should clearly understand the proper
procedures and requirements and have the equip-
ment necessary to respond to spills.
Inspections: Qualified facility personnel must be iden-
tified to inspect designated equipment and areas of
the facility at appropriate intervals specified in the
plan. A set of tracking or followup procedures must
be used to ensure that appropriate actions are taken
in response to the inspections.
Employee training: The pollution prevention plan must
describe a program for informing personnel at all
levels of responsibility of the components and goals
of the stormwater pollution prevention plan. Where ap-
propriate, contractor personnel also must be trained
in relevant aspects of stormwater pollution preven-
tion.
Recordkeeping and internal reporting procedures:
The pollution prevention plan must describe proce-
dures for developing and retaining records on the status
and effectiveness of plan implementation. At a mini-
mum, records must address spills, monitoring, and
inspection and maintenance activities. The plan also
must describe a system that enables timely reporting
of stormwater management-related information to ap-
propriate plant personnel.
Sediment and erosion control: The pollution preven-
tion plan must identify areas that, due to topography,
activities, soils, cover materials, or other factors, have
a high potential for significant soil erosion. The plan
must identify measures that will be implemented to
limit erosion in these areas.
Management of runoff: The plan must contain a nar-
rative evaluation of the appropriateness of traditional
stormwater management practices (i.e., practices other
than those that control pollutant sources) that divert,
infiltrate, reuse, or otherwise manage stormwater
runoff to reduce the discharge of pollutants. Appro-
priate measures may include, among others, vegeta-
tive swales, collection and reuse of stormwater, inlet
controls, snow management, infiltration devices, and
wet detention/retention basins.
Based on the results of the evaluation, the plan must
identify practices that the permittee determines to be
reasonable and appropriate for the facility. The plan also
should describe the particular pollutant source area or
activity to be controlled by each stormwater manage-
ment practice. Reasonable and appropriate practices
must be implemented and maintained according to the
provisions prescribed in the plan.
In selecting stormwater management measures, it is
important to consider the potential effects of each
method on otherwater resources, such as ground water.
Although stormwater pollution prevention plans primar-
ily focus on stormwater management, facilities must
also consider potential ground-water pollution problems
and take appropriate steps to avoid adversely affecting
ground-water quality. For example, if the water table is
unusually high in an area, an infiltration pond may con-
taminate a ground-water source unless special preven-
tive measures are taken. Under EPA's July 1991 Ground
Water Protection Strategy, states are encouraged to
develop comprehensive state ground-water protection
programs (CSGWPP). Efforts to control stormwater
should be compatible with state ground-water objectives
as reflected in CSGWPPs.
Comprehensive Site Compliance Evaluation
The stormwater pollution prevention plan must describe
the scope and content of comprehensive site inspections
that qualified personnel will conduct to 1) confirm the
accuracy of the description of potential pollution sources
contained in the plan, 2) determine the effectiveness of
the plan, and 3) assess compliance with the terms and
400
-------
conditions of the permit. The plan must indicate the
frequency of such evaluations, which in certain cases
must be at least once a year.
Material handling and storage areas and other potential
sources of pollution must be visually inspected for evi-
dence of actual or potential pollutant discharges to the
drainage system. Inspectors also must observe erosion
controls and structural stormwater management de-
vices to ensure that each is operating correctly. Equip-
ment needed to implement the pollution prevention plan,
such as that used during spill response activities, must
be inspected to confirm that it is in proper working order.
The results of each site inspection must be documented
in a report signed by an authorized company official.
Based on the results of each inspection, the description
of potential pollution sources and the measures and
controls in the plan must be revised as appropriate
within 2 weeks after each inspection.
Special Requirements for Selected Facilities
EPA's general permits also establish special require-
ments for selected classes of facilities. These include:
Sampling requirements: Targeted classes of facilities
are required to monitor their stormwater discharges for
specified parameters. Facilities that are a member of
a targeted class but that can certify that they do not
have materials or equipment exposed to precipitation
are not required to monitor. This is intended to pro-
vide facilities with an incentive to eliminate exposure
to precipitation.
EPCRA facilities: Certain facilities that are subject to
reporting requirements under Section 313 of the Emer-
gency Planning and Community Right-to-Know Act
(EPCRA) because they manufacture or use large
amounts of toxic chemicals are subject to special re-
quirements under the NPDES general permits. These
special requirements include provisions that are similar
to spill prevention, countermeasure, and control
(SPCC) plan requirements, and include provisions for
secondary containment or equivalent controls for liquid
storage areas. In addition, a professional engineer (PE)
must inspect the site, review the plan, and certify that
the stormwater pollution prevention plan has been pre-
pared in accordance with good engineering practices.
Salt piles: Salt piles must be enclosed or covered to
prevent exposure to precipitation.
Coal pile runoff: The permit establishes numeric ef-
fluent limitations for coal pile runoff.
Municipal Role in Implementation
The NPDES stormwater program establishes a permit
approach that envisions complementary, cooperative ef-
forts by the permit-issuing agency and municipal opera-
tors of large and medium municipal separate storm sewer
systems to develop programs that result in controls on
pollutants in stormwater discharges associated with in-
dustrial activity that discharge through municipal systems.
Under the complementary permit approach, stormwater
discharges associated with industrial activity that dis-
charge through large and medium municipal separate
storm sewer systems are required to obtain permit cov-
erage. Permits for these discharges will establish re-
quirements (such as pollution prevention requirements
or monitoring) for industrial operators. Any records, re-
ports, or information obtained by the NPDES permit-is-
suing authority as part of the permit implementation
process, including site-specific stormwater pollution pre-
vention programs that are developed pursuant to the
draft general permit, are available to municipalities. This
will assist municipalities in reviewing the adequacy of
such requirements and developing priorities among in-
dustrial stormwater sources. In addition, these permits
provide a basis for enforcement actions directly against
the owner or operator of stormwater discharges associ-
ated with industrial activity.
A second permit, issued to the operator of the large or
medium municipal separate storm sewer, establishes
the responsibilities of the municipal operators in control-
ling pollutants from stormwater associated with indus-
trial activity that discharges through their systems.
Municipal programs to reduce pollutants in industrial site
runoff specifically will address municipal responsibilities
in controlling pollutants from industrial facilities. In addi-
tion, programs to identify and control nonstormwater
discharges to municipal separate storm sewer systems
will in many cases focus on industrial areas because
these areas often have a significant potential for illicit
connections, spills, and improper dumping.
Municipal operators of these systems can assist
NPDES permit issuing authorities:
By identifying priority stormwater discharges associ-
ated with industrial activity to their systems.
In inspecting facilities and reviewing and evaluating
stormwater pollution prevention plans that industrial
facilities are required to develop under the draft gen-
eral permit.
In compliance efforts regarding stormwater discharges
associated with industrial activity to their municipal
systems.
A pilot program conducted by municipalities in the Santa
Clara Valley illustrates how a municipality can work with
an NPDES authority to control pollutants in stormwater
discharges associated with industrial activities. (A more
complete description of the pilot program and its findings
is provided in the Santa Clara Valley Nonpoint Source
Pollution Control Program [3]). One of the major goals
401
-------
of the prog ram was to reduce discharges to storm drains
of dry- and wet-weather heavy metals that result from
activities such as processing, storage, and maintenance
activities conducted at industrial sites. Components of
the program included the following:
Municipalities developed industrial inspection and il-
legal dumping/illicit connection programs to ensure
that activities focus on priority industries.
Monitoring requirements were established in the Cali-
fornia NPDES general permit for industries. Munici-
palities evaluated monitoring data collected by
priority industries.
The California NPDES general permit allowed for ex-
emption for industries from monitoring where the mu-
nicipality provides certification that the industry
pollution prevention plan is adequate.
Municipalities developed industry specific guidance.2
Municipalities implemented a "Clean Bay Business"
award program.
Market-based incentives were considered, such as
trading reductions from car pooling and telecommu-
nication programs for pretreatment requirements.
Key findings of the pilot programs identified the following
components needed for a successful program:
Hands-on field training conducted by an experienced
industrial inspector.
Classroom training on industrial stormwater require-
ments and on methods of communicating with facility
managers.
Classroom training on other related industrial regula-
tory programs (e.g., HAZMAT, pretreatment).
A reference manual on the regulations and local legal
authority.
Adequate legal authority to allow site access and
take progressive enforcement actions.
See California Storm Water Best Management Practice Handbook:
Industrial/Commercial(16), which addresses how to prepare a storm-
water pollution prevention plan and how to select BMPs. The guidance
also addresses source controls for nonstormwater discharges; vehicle
and equipment fueling; vehicle and equipment washing and steam
cleaning; vehicle and equipment maintenance and repair; outdoor
loading/unloading of materials; outdoor container storage of liquids;
outdoor process equipment operations and maintenance; outdoor
storage of raw materials, products, and byproducts; waste handling
and disposal; contaminated or erodible surface areas; building and
grounds maintenance; building repair; remodeling and construction;
and overwater activities. In addition, the guidance covers treatment
control BMPs and measuring BMP performance.
Prioritizing facilities based on existing information be-
fore conducting inspections.
Advance communications, in the form of a letter, to
industries before conducting the inspections.
A plan for followup actions, including enforcement,
where necessary.
References
1. Pitt, R. 1992. Stormwater, baseflow, and snowmelt pollutant con-
tributions from an industrial area. Presented at the 65th Annual
Conference, Water Environment Federation, New Orleans, LA
(September).
2. U.S. EPA. 1983. Results from the Nationwide Urban Runoff Pro-
gram, Vol. 1. Final report. NTIS PB84185552.
3. Santa Clara Valley Nonpoint Source Pollution Control Program.
1992. Source identification and control report. December 1.
4. Ontario Ministry of the Environment. 1986. Toronto area water-
shed management strategy study: Number River pilot watershed
project. June.
5. U.S. EPA, Region 5. 1990. Urban targeting and BMP selection:
An information and guidance manual for state nonpoint source
program staff engineers and managers. Region 5, Water Divi-
sion, Chicago, IL 60604 (November).
6. U.S. EPA. 1993. Investigation of inappropriate pollutant entries
into storm drainage systems: A user's guide. EPA/600/R-92/238
(January).
7. U.S. EPA. 1991. Federal Register 56:40948. August 16.
8. American Society of Civil Engineers. 1988. Design of urban runoff
quality controls. New York, NY: American Society of Civil
Engineers.
9. Torno/American Society of Civil Engineers. 1989. Urban storm-
water quality enhancement: Source control, retrofitting, and com-
bined sewer technology.
10. U.S. EPA. 1979. NPDES best management practices guidance
document.
11. U.S. EPA. 1991. Analysis of implementing permitting activities for
stormwater discharges associated with industrial activity.
12. U.S. EPA. 1990. Manual of practice: Identification of illicit connec-
tions. September.
13. Washington State Department of Ecology. 1992. Stormwater
management manual for the Puget Sound Basin, Vol. I. Minimum
technical requirements. February.
14. Alachua County Office of Environmental Protection. Best man-
agement practices for the use and storage of hazardous materi-
als. Gainesville, FL.
15. U.S. EPA. 1992. Stormwater management for industrial activities:
Developing pollution prevention plans and best management
practices.
16. California State Stormwater Task Force. 1992. California storm-
water best management practice handbook: Industrial/Commer-
cial. March.
402
-------
The Role of Education and Training in the Development of the Delaware
Sediment and Stormwater Management Program
Frank M. Piorko and H. Earl Shaver
Delaware Department of Natural Resources and Environmental Control, Dover, Delaware
On May 31,1990, the General Assembly of the State of
Delaware enacted new legislation on stormwater man-
agement and placed it within the revised framework of
the state's sediment control law to emphasize the inte-
gral relationship between the two programs. Governor
Castle signed the legislation into law at a public cere-
mony on June 15, 1990. The effective date of the regu-
lations was January 23, 1991. Program implementation
was initiated on July 1, 1991.
The role of education and training in the development
and implementation of Delaware's sediment and storm-
water program was recognized at the legislative onset.
The educational effort continued through the evolution,
development, and promulgation of the regulations and
remains an essential component of program strategy.
The sediment and stormwater regulations are specific
as to the training requirements and opportunities for
education that are to be provided for contractors, con-
struction review/inspection personnel, and plan design
professionals.
This paper discusses the education and training accom-
plishments to date, their value to successful program
inauguration, and specific training objectives being de-
veloped to meet the requirements of the new law and
regulations in Delaware.
Background
The State of Delaware has had an erosion and sediment
control program since 1978. That program was only
marginally successful due to budget and personnel
limitations. Environmentally oriented initiatives in other
states and within the federal government have since
provided an impetus for the Department of Natural Re-
sources and Environmental Control (DNREC) to attempt
program improvements with respect to sediment control
and stormwater management.
In 1989, DNREC representatives conducted onsite
reviews of the existing sediment control program to
document program effectiveness. It was readily appar-
ent that too few resources were devoted to a program
that lacked legislative and regulatory authority. The site
problems were recorded through slide documentation so
that a public education program could be developed that
clearly showed the need for program improvements.
At the same time, DNREC, in association with local
conservation districts, was considering the need for a
statewide stormwater management program that con-
sidered water quantity and water quality requirements.
Fortunately (or unfortunately, depending on the per-
spective), during the summer of 1989, Delaware had
several severe flooding events that reinforced the con-
cept that the state needed a stormwater management
program that would prevent existing problems from getting
worse.
Delaware does not have a strong environmental lobby
group to advocate the passage of new environmental
programs, so DNREC has developed a consensus-
style approach to get legislation and subsequent regu-
lations accepted by the legislative bodies and the
regulated community.
Legislative Process
As the legislation was developed, DNREC sponsored two
workshops at which the concept behind the proposed
legislation was discussed in a public forum accompanied
by slide presentations. The slide presentation focused
on problem identification, the proposed state program to
address the problems, and the degree to which, in the
opinion of DNREC, the sediment and stormwater pro-
gram was going to evolve. Individual meetings were
held with contractors' associations, engineering consult-
ants, land developers, and the general public.
In addition to those workshops and meetings, presenta-
tions were made to legislative committees in an informal
setting so that individual committee members would
have a basic understanding of the need for legislation.
403
-------
The proposed legislation passed through a state senate
committee and the full senate in only 2 days, with not
one negative vote. The passage of the legislation
through two committees in the state house of repre-
sentatives and the full house took approximately 11/2
months and again received no negative votes. The edu-
cational process prior to submission of the legislation
and during the legislative process was so successful
that not one affected group submitted comments that
were in opposition to the legislation. The legislation
passed through three committees and two houses unan-
imously. The legislation was signed into law by Governor
Castle in a public ceremony on June 15, 1990.
Regulatory Process
The legislation has several components that specifically
address education and training, but one component
critical to the process of regulation adoption was the
requirement in the law that the regulations were to be
developed with the assistance of a regulatory advisory
committee. Recognizing the need for program consen-
sus, DNREC placed the regulatory advisory committee
requirement within the legislation so that the affected
entities would participate in the regulatory process.
The regulatory advisory committee was composed of
representatives of 20 organizations representing such
groups as contractors, developers, consulting engineers,
utility companies, local governments, and conservation
districts. DNREC prepared drafts of the regulations prior
to meetings. Each section, subsection, paragraph, sen-
tence, and word that was proposed for the regulations
was subject to the scrutiny of the regulatory review
committee. Each member of the committee did not have
to approve all aspects of the regulations, but rather the
committee needed to substantially concur. Eight full
committee meetings were held, and through the meeting
process committee members could understand the ra-
tionale behind the various regulatory requirements. As
a result, the committee members substantially con-
curred on all aspects of the regulations. In fact, commit-
tee members tended to become advocates of the
regulations when they were published for public input.
In addition to the regulatory review committee process,
meetings were also held with any interested individual
or entity. Once the regulations were in a rough state of
completion, three public workshops were held around
the state to solicit input from a broader range of interests
than just those represented by the regulatory review
committee. The input received during this public review
process was limited, but the informal public process
prepared people for what was intended in the regula-
tions so that any significant opposition to any of the
requirements could be addressed before the formal
regulation adoption process.
On the basis of the input received from the workshops,
DNREC initiated formal regulation adoption procedures
with no major changes to the body of the regulations.
Announcements were placed in newspapers regarding
DNREC's intentions, and a formal public hearing was
held on January 16, 1991. Due to the consensus-build-
ing process, in which the regulated community partici-
pated in developing the regulations, not one adverse
comment was received during the public hearing proc-
ess. The entire public hearing took less than 15 minutes,
as there were no questions or comments due to public
awareness of the regulations' contents.
The entire process of legislative and regulatory develop-
ment and approval clearly demonstrates that a consensus-
building approach to environmental requirements may
be an effective means of obtaining the programmatic
infrastructure needed to implement an effective program.
In large part due to the strong involvement of the regu-
lated community, there is a significant effort in the law
and regulations regarding education and training of
contractors, inspectors, consultants, and the general
public. It is the position of the authors that environ-
mental programs can only be effective if the regulated
community is involved in program development and
evolution, recognizes the program need, and under-
stands and accepts their obligations under the regulatory
requirements. The individual educational and training ob-
ligations under the law and regulations are discussed as
they affect the overall sediment and stormwater pro-
gram.
Delaware Sediment and Stormwater
Contractor Certification Program
During the development of the Delaware Sediment and
Stormwater Regulations, a provision was made to pro-
vide for mandatory training and certification of individu-
als performing sediment and stormwater related
construction. Section 13 of the regulations states that
"After July 1, 1991, any applicant seeking sediment and
stormwater plan approval shall certify to the appropriate
plan approval agency that all responsible personnel
involved in the construction project will have a certificate
of attendance at a Departmental sponsored or approved
training course for the control of sediment and stormwa-
ter, before initiation of land-disturbing activity."
"Responsible personnel" means any foreman or super-
intendent who is in charge of onsite clearing and land-
disturbing activities for sediment and stormwater control
associated with a construction project.
"Land-disturbing activity" means a land change or con-
struction activity for residential, commercial, silvicultural,
industrial, and institutional land uses that may result in
soil erosion from water or wind or movement of sediments
or pollutants into state waters or onto lands in the state,
404
-------
or which may result in accelerated stormwater runoff
including, but not limited to, clearing, grading, excavat-
ing, transporting, and filling of land.
Contractor Certification Program Development
The development of the Contractor Certification Program
was part of a general sediment and stormwater educa-
tional package funded by a Section 205 (G) grant under
the Clean Water Act from the U.S. Environmental Pro-
tection Agency. Other tasks included a review of similar
programs throughout the mid-Atlantic region, contract-
ing for aerial photography of sites under construction,
preparation of a portable soils exhibit, and identifying
future training and educational needs. The grant tasks
were carried out jointly through a memorandum of
understanding between DNREC's Division of Water
Resources and the New Castle Conservation District.
A steering committee was formed in April 1990 and met
seven times over the course of the following 9 months.
The purpose of the committee was to provide input for
the development and implementation of the grant tasks.
It was determined that the certification program was to
use a slide presentation format since excellent docu-
mentation was already available and additional field
slides were easily obtained. In addition to the field slides
of sediment and stormwater construction practices, text
and technical slides needed preparation. A local com-
pany was contracted to produce this material.
The certification program was developed with a 31/2- to
4-hour time frame in mind. This would allow for morning
or afternoon sessions, even occasional evenings, as
necessary. Maryland has enjoyed success for many
years in their sediment control training program using a
similar format and time frame.
A 55-page narrative describing the slide presentation
was developed and made available to the audience upon
request. This was done to encourage attention to the slide
presentation rather than preoccupation with taking
notes. Finally, it was decided that participants should
receive a durable plastic laminate card with the state
logo and the individual's name and certification number
imprinted on it. This would give the participants a tangi-
ble item to associate with the completion of the program.
Contractor Certification Program
Implementation
By the end of January 1991, the program was ready to
be presented. Certain restrictions were placed upon class
size in order to communicate most effectively. Optimal
class size was 30 to 40 members. Limiting the class size
meant that the program would have to be presented
many times; therefore, by July 1, 1991, not all of the
contractors needing to complete the certification program
would have the opportunity to do so. The Sediment and
Stormwater Regulations provide for interim certification
if individuals notify DNREC of their intent to register for
the next available course.
The certification program was designed for presentation
in two ways. First, the conservation districts, counties,
and other agencies given the responsibility of certain
program elements would set up the programs in their
own jurisdictions, giving them a chance to meet with the
regulated community and explain local program require-
ments. Second, DNREC would present the program to
any regulated company, business or organization if they
could provide a suitable location and a minimum of 15
individuals to be trained. DNREC also provided training
for DNREC staff and several hundred Delaware Depart-
ment of Transportation inspectors, technical staff, and
engineers.
Throughout the first 6 months of presentations, we were
surprised and pleased not only with the response from
the contractors but also from the engineers, consultants,
and developers who wanted to attend the certification
program. All told, from February 1991 until July 1991,
DNREC presented the program on 37 occasions, certi-
fying over 1,100 individuals from 300 companies and
organizations.
As stated earlier, this was possible only with the assis-
tance from the three state conservation districts, county
governments, the Department of Transportation, and
organizations such as the Associated Builders and Con-
tractors and the Delaware Contractors Association. As
of January 1, 1993, almost 2,000 individuals have com-
pleted this training.
Initially, a program quiz was developed not so much to
grade the participants but to obtain feedback on the
retention of the material being provided. A program
evaluation was later substituted for the quiz so that we
could determine if any changes or improvements should
be made to the training program. A representative sam-
ple of 100 evaluations was compiled, the results of
which appear in Figure 1. Most notable is that 96 percent
of respondents would recommend this training (Ques-
tion 7), and 86 percent wished to continue in this training
(Question 8).
By continuing the Contractor Certification Program, not
only are the requirements of the Delaware Sediment
and Stormwater Regulations being met, but the knowl-
edge gained by the participants in this program is being
transferred to the field through proper construction prac-
tices.
Delaware Certified Construction Reviewer
Course
The Delaware Sediment and Stormwater Regulations
also provide for special site inspection or review require-
ments under certain site conditions. Section 12 of the
405
-------
Very Useful 53%
Met Expectations
83%
Less Than
Expected 5%
Exceeded
'Expectation 12%
Question #1
Did this course meet your expectations?
\ Not Very
Useful 1%
Somewhat
Useful 46%
Question #2
Was the course material useful?
Very
Interesting 62%
Excellent 78%
Not Very
Interesting 1%
Somewhat
Interesting 37%
Question #3
Was the course material interesting?
Fair 5%
Good 17%
Question #4
Was the speaker knowledgeable?
Excellent 52%
Fair 6%
Excellent 44%
Question #5
Rate the audio/visual materials.
Good 41
Poor 2%
Fair 5%
Question #6
Was the facility appropriate?
Yes 96%
If Material
/?Changes 4%
Question #7
Would you recommend this training?
Yes 86%
Not Sure 7%
No 7%
Question #8
Would you continue in this training?
Figure 1. Sediment and stormwater contractor certification program course evaluation.
regulations identifies these site conditions that allow
DNREC or the appropriate plan approval agency to
require that a certified construction reviewer be present
on site. Examples of site conditions that would warrant
this requirement would be a site in excess of 50 acres
of disturbed area or any site experiencing significant
sediment and stormwater problems. The owner or de-
veloper of the site in these cases would be responsible
for providing a certified construction reviewer for any or
all parts of the construction phase as deemed necessary
by the plan approval agency. The main responsibility of
these individuals is to ensure the adequacy of construc-
tion pursuant to the approved sediment and stormwater
management plan.
As with the Contractor Certification Program, DNREC
has the responsibility to provide training to certify these
construction reviewers. A formal Sediment and Stormwater
Management Certified Construction Reviewer Course
was developed in cooperation with Delaware Technical
and Community College. Course material was devel-
oped to instruct participants in basic hydrology and hy-
draulics, soils, vegetative establishment, construction
practices, plan preparation and implementation, inspec-
tion, enforcement, and maintenance. To instruct this
course, over 20 professionals in the area of sediment
and stormwater management were recruited, representing
government agencies, private industry, and the consult-
ing and engineering community.
406
-------
The course format was developed to be presented in
eight 31/^-hour weekly sessions. An examination was
developed and arrangements made with Delaware
Technical and Community College for Continuing Edu-
cation Credits to be issued.
We anticipated a lot of interest in this course offering, so
registration was limited to one individual per company
or organization. In addition to the private community, an
attempt was made to include at least one individual that
works for each agency responsible for delegation of
sediment and stormwater program elements. In all, 85
seats were quickly filled for this course. The second time
this course was offered, the class sessions were re-
duced to four all-day sessions. This seemed to suit the
class participants' schedule better.
One important measure of success is the evaluation
question that asked class participants to indicate
whether the course did not meet, met, or exceeded
expectations. The breakdown is as follows:
41 responses, or 74 percent of the class, stated that
the course met their expectations.
12 responses, or 22 percent of the class, stated that
the course exceeded their expectations.
2 responses, or 3.5 percent of the class, stated that
the course did not meet their expectations.
The success of this program is directly attributable to the
preparation of the speakers, the attentiveness of the
class, and the hard work of the Delaware Sediment and
Stormwater Program staff.
Stormwater Management Technical
Sessions
The engineering and design community in Delaware has
also indicated the need for DNREC to present more
design-oriented training in sediment and stormwater
management. To date, there have been several work-
shops in U.S. Department of Agriculture Soil Conserva-
tion Service TR-55 and TR-20 hydrologic analyses
sponsored by local conservation districts and enlisting
the assistance of the Soil Conservation Service.
DNREC recognizes the need to expand this basic train-
ing and make available more design-oriented training for
the consultant community.
Coinciding with the development and release of the
Delaware Stormwater Management Design Manual in
the summer of 1993, training classes were scheduled to
present this material in modules, as the manual was
developed. This training will help ensure that stormwater
management practices are designed to meet estab-
lished minimum criteria.
Summary
The education and training component of the Dela-
ware Sediment and Stormwater Management Pro-
gram is one of several areas of program development
that will continue to respond to the needs of the regu-
lated community. One obvious benefit in a small state
like Delaware is that the efforts of a regulatory agency
in providing education and training to the regulated
community are recognized and appreciated. As pre-
viously discussed, the Sediment and Stormwater
Management Program depends highly on interagency
cooperation and communication with the businesses
and industry involved. By maintaining education and
training objectives as a high priority, DNREC will in-
crease chances for program success.
407
-------
Development and Implementation of an Urban Nonpoint Pollution Educational
and Informational Program
Richard Badics
Washtenaw County Environmental Services Department,
Ann Arbor, Michigan
Abstract
Sampling, Abatement, Follow-up, Education, and Re-
sponse (SAFER) was formed by the Washtenaw
County's Environmental Interest Group on January 1,
1992. SAFER includes the county departments of Envi-
ronmental Coordination, Environmental Services, Drain
Commissioner, Planning, and Cooperative Extension,
as well as the Soil Conservation District, Huron River
Watershed Council, Ecology Center of Ann Arbor, and
the Southeast Regional Groundwater Education Center.
The purpose of SAFER is to "provide for coordination of
water protection programs through inter- and intra-
county agencies and group cooperation."
Education is a key element of SAFER. Four groups are
targeted for education by SAFER: government, busi-
ness and industry, community groups, and schools.
SAFER members develop their own specific educa-
tional programs and materials. Through SAFER, these
are coordinated to provide uniform and accurate infor-
mation to targeted segments of the community. This
avoids costly duplication of services.
To effectively deliver an educational program, the target
audience must first be determined, then an analysis of
existing educational programs must be made to build on
past successes. Through this process, an approach is
determined that is most likely to be successful. Prior to
beginning the educational program, the establishment
of an evaluation process is critical.
Overview of Washtenaw County's SAFER
Group
Sampling, Abatement, Follow-up, Education, and Re-
sponse (SAFER) was formed by Washtenaw County's
Environmental Issues Group on January 1, 1992. The
Environmental Issues Group consists of departments
within Washtenaw County government that indirectly
or directly manage the environment of Washtenaw
County. This provides the county with a coordinated
approach to addressing environmental issues. The En-
vironmental Issues Group is chaired by the Environ-
mental Coordination Office. Other member groups
within the Environmental Interest Group are the Sheriff's
Department, Environmental Services, Emergency Man-
agement, Planning, Public Works, Drain Commis-
sioner, and Cooperative Extension, as well as the
county's Health Officer. This group meets monthly to
discuss the status of county programming, pending
state and federal legislation, "hot" environmental topics
or issues, and strategic planning.
SAFER was formed as a work group of the Environ-
mental Issues Group "to provide for coordinative water
protection programs through inter- and intracounty
agencies and group cooperation." SAFER consists of
groups internal and external to Washtenaw County gov-
ernment that are involved in dealing with the county's
ground and surface water. SAFER includes the county
departments of Environmental Coordination, Environ-
mental Services, Drain Commissioner, Planning, Coop-
erative Extension, Soil Conservation District, Huron
River Watershed Council, and Ecology Center of Ann
Arbor, as well as the Southeast Regional Groundwater
Education Center (SER-GEM). During its first year of
operation in 1992, the group focused on categorizing
and compiling all current water quality programs and
their products. The 1992 SAFER Directory compiled
over 100 products addressing water quality issues
within the county.
Education is a key element of SAFER. Four target
groups for educational programs in SAFER are govern-
ment, business and industry, community groups, and
schools. The SAFER Educational Subcommittee in
1993 is compiling all educational programs and materi-
als on water quality related issues, similar to the 1992
SAFER Directory. Through SAFER, educational materi-
als are coordinated to provide current and accurate
408
-------
information to the community while avoiding costly du-
plication of services.
Urban Nonpoint Pollution Education
The development and implementation of a nonpoint
pollution educational and informational program is criti-
cal to a successful urban project. Public awareness of
urban nonpoint pollution is relatively low, and the media
tends to focus on health or environmental risks that are
easy to define, such as AIDS or hazardous waste is-
sues. Due to its nature, nonpoint pollution is harder to
pinpoint. Urban nonpoint pollution prevention requires a
long-term commitment to changing attitudes.
Urban nonpoint pollution can be directly attributed to
people. We all contribute to it. People are accustomed
to focusing on easier issues, where the blame can be
attributed to activities outside their control. An example
is auto safety. People are very concerned about vehicle
safety when a manufacturing error is the cause, such as
exploding gas tanks. These same people, however, are
not as focused on actions that they control, such as
wearing seat belts.
An environmental example is oil spills. A study by the
Michigan Department of Natural Resources (MDNR) in
1989 found that more oil is illegally released into the
environment in Michigan annually than was released in
the Valdez tanker incident. Getting people to buy into
the idea that they are a major part of the problem is a
critical step in gathering their support and cooperation.
Target A udience
Before an education information program can be devel-
oped, the target audience must be identified. A general
educational approach will not change the habits of a
wide range of target groups. Each targeted group must
be analyzed independently to understand its particular
needs and to develop specific actions it can take. Next,
the various media options must be explored.
A multimedia approach enhances the opportunities of
reaching larger segments within the target audience.
For example, handing out flyers at a garden show will
not reach several socioeconomic classes; a spot on a
local radio station may be more appropriate. Some com-
mon public outreach materials are fact sheets, pam-
phlets, radio, television, newspapers, magazines,
displays, models, posters, group presentations, and
one-on-one or community events.
Using existing resources in your educational program is
important. An educational program workshop for com-
posting in the community could also be a forum for
supplying information to the public on preventing urban
nonpoint pollution through the proper application of fer-
tilizers and use of environmentally friendly alternatives
to pesticides. By networking with existing programs in
the community, nonprofit programs will not compete for
and confuse the audience.
Educational Gaps
After analyzing current educational resources within the
community, identify audiences and approaches not cur-
rently used. All targeted groups need to receive your
message. Target groups in the community must "buy
into" their contribution to nonpoint pollution and their
ability to prevent or minimize it. Urban educational
programs must be innovative, well conceived, multi-
media, and coordinated with other educational pro-
grams in the community.
A large number of ongoing urban nonpoint education
programs exist in communities throughout the country.
These programs have been developed for various types
of audiences. Prior to implementing a program "from
scratch," review all ongoing programs. These can be
found in EPA "News Notes," as well as through profes-
sional groups, conferences, and environmental publica-
tions. Regional EPA offices are also a valuable resource
for finding suitable ongoing programs. Using existing
programs saves time and money.
Program Evaluation
An integral part of all educational programs is evalu-
ation. Valuable time and resources can be wasted if
information supplied to an audience is not effective.
When developing the evaluation mechanism for the
educational process, make sure the educational pro-
gram focus enhances the overall water quality objec-
tives. One way to evaluate the educational process is to
apply Bennett's Hierarchy of Evidence for Program
Evaluation. Bennett uses seven steps of evaluation. In
an inverted scale, these steps are:
1. Inputs of program resources that are used to make
the program work.
2. Activities which can include internal events, such as
planning, or external events involving an audience.
3. Involvement of the target audience in activities, fo-
cusing on hands-on type activities.
4. The target audience's view of the program.
5. KASA change, or the change in knowledge, atti-
tudes, skills, or aspirations of the audience.
6. Changes in behavior that result from the educa-
tional program.
7. End results that reflect the program's goals and
objectives.
Many techniques can be used to measure the seven
Bennett attributes. The basic who, what, where, and
409
-------
when questions are useful when establishing the spe-
cific evaluation technique.
Many books and guides can help in developing program
evaluation. Studying these before finalizing an evalu-
ation process is highly recommended. If there are time
constraints or expertise is not available for evaluation,
this component can be done by an outside party. The
key is to establish the evaluation mechanism before
implementing the educational program.
Huron River Pollution Abatement Program
Overview
The Huron River Pollution Abatement Project (HRPAP),
which encompassed the urbanized area of Washtenaw
County, was formed and implemented in 1986 by the
county's Drain Commissioner's Office in conjunction
with the Environmental Services Department. Public
education was a major objective of the project. The
educational program used by the HRPAP was designed
after reviewing earlier area pilot water quality programs
and their targeted community groups. The HRPAP fo-
cused on business, industry, community, and school
groups.
Business/Industry
The HRPAP conducted surveys and dye tests of facili-
ties located in the urbanized areas of Washtenaw
County. Staff interviewed facility owners and managers
on their particular businesses and gained critical infor-
mation about their operations. When a common need
was foundfor example, an owner unable to dispose of
a certain type of wastethe project staff worked with
the owner to resolve the problem. For example, many
facility operators with oil separators were not familiar
with separators and were unable to find a licensed
waste hauler to service them. The HRPAP developed a
maintenance guideline for the operators, contacted all
local waste haulers, and developed a list of haulers that
would service oil separators. This information was then
distributed to all facilities with oil separators.
Community and Civic Group Education
Over 200 educational presentations were made to the
community during the HRPAP's 6 years. The HRPAP
used various media to educate the community. One of
the most effective was the local press. Articles concern-
ing the HRPAP were published on an ongoing basis.
Press releases noted significant events and common
problems found within the community.
A second approach to outreach was through community
events. Examples are the Ann Arbor City Art Fair and
the Ypsilanti City Heritage Festival. These events attract
hundreds of thousands of people. Display booths and
pamphlets were developed for participating in these
events. This became a forum for discussing water qual-
ity related issues one-on-one with the public.
School Education
The HRPAP made its first school educational presenta-
tion to a third-grade class in 1988. Word of mouth led to
over 25 presentations per year in six local school dis-
tricts. HRPAP student interns with an educational back-
ground formulated lesson plans for different grade levels
on nonpoint pollution and related topics, such as the
water cycle and household hazardous waste.
In classrooms, educational programs concentrated on
hands-on activities. Two water quality models were built.
One electronic model, entitled "Pathways to Pollution,"
lights up various pollution pathways when the appropri-
ate button is pushed. A second model is a transparent
representation of a town showing the sanitary and storm
sewer systems. The students place a dye into catch
basins, floor drains, and toilets to observe the route the
water takes directly to the stream or the wastewater
treatment plant. This model has examples of both
proper and improper connections.
Conclusions
The majority of urban nonpoint pollution can be directly
attributed to the activities of people. Most people are not
aware of the impacts their routine activities at home and
at work have on water quality. Education is a key com-
ponent to improving urban water quality problems. Key
target audiences in the community need to be identified,
existing educational resources studied, educational pro-
gram gaps identified, and an evaluation process in-
cluded to measure a program's effectiveness.
The key to an educational program is to focus on prac-
tical activities that the target group can do to eliminate
water pollution. A long-term, sustained educational effort
leads to an increased awareness and respect for the
interdependence of all elements in the ecosystem and
for how individual activities affect them. This ultimately
leads to a sense of mutual responsibility and a long-term
commitment to continued environmentally sound actions.
Acknowledgments
The author would like to acknowledge the support and
help of Dr. Rebecca Head, Group Director, Environment
and Infrastructure; Janis Bobrin, Drain Commissioner;
Robert Blake, Director, Environmental Services; David
Dean; H. Leon Moore; Jeffry Krcmarik; and David Wil-
son, as well as other members of SAFER.
410
-------
Training for Use of New York's Guidelines
for Urban Erosion and Sediment Control
Donald W. Lake, Jr.
U.S. Department of Agriculture,
Soil Conservation Service, Syracuse, New York
Introduction
New York State still does not have a statewide erosion
and sediment control law. Unlike many of its neighboring
states, New York continues to leave the initiation of such
control to local units of government. Historically, coun-
ties, towns, and villages have enacted ordinances once
a significant environmental accident has occurred. Ju-
risdiction occurs at the local level, with planning boards
having approval authority to issue permits to develop.
Because each board is dealing with its local area, the
regulations and processes for gaining approval vary
from locale to locale.
Technical standards for controlling erosion and sedi-
ment were developed by the Soil Conservation Service
in March 1988 and issued as New York Guidelines for
Urban Erosion and Sediment Control. This document
provides design details and specifications for both tem-
porary and permanent management practices, as well
as resource-planning concepts. Known as the "Blue
Book," the document provides consistency in the tech-
nical approach to erosion and sediment control plans
for construction sites. It has been adopted by the New
York State Department of Environmental Conserva-
tion and the U.S. Army Corps of Engineers, Buffalo
District, as criteria for erosion and sediment control
plans. The New York State Department of Transporta-
tion has incorporated many of its details into its high-
way design manual.
In April 1992, the New York State Department of Envi-
ronmental Conservation (NYS-DEC), Division of Water,
published Reducing the Impacts of Stormwater Runoff
From New Development. This document establishes
performance standards for stormwater management
control in New York for projects requiring NYS-DEC
review. Standards were set for both water quantity and
water quality. Water quantity is addressed by requiring
no greater discharges from the site after development
than present before development for the 2-, 10-, and
100-year frequency storm events. Water quality is ad-
dressed by retaining the "first flush," which is defined as
the greater of one-half inch of runoff or runoff resulting
from a 1-year, 24-hour storm, from the land area for
which the infiltration rate has been changed.
These two documents finally provide guidance for ero-
sion and sediment control and stormwater manage-
ment for local units of governments as well as
regulatory agency staffs. Their use and application de-
pends on what the site's size and resource constraints
are and whether a local ordinance is in place. The local
approval process, in communities with such a regula-
tion, generally requires a formal review of the plan with
its erosion and sediment control and stormwater man-
agement component by either the town or village engi-
neer and a local soil and water conservation district staff
person or health department official. Unfortunately,
many of these individuals are unable to identify prob-
lems or lack the knowledge of design details to control
sediment from the site.
Once a developer begins operations in the field, the
building inspector, code enforcement officer, or health
department official is responsible for inspecting the site
for compliance to the approved plan as well as to
ensure that the contractor maintains the installed prac-
tices. These field inspectors require training in the con-
cepts of erosion and sediment control installation and
maintenance.
Clean Water Act Mandates
On October 1, 1992, stormwater regulations went into
effect under the Clean Water Act that require individuals,
agencies, and municipalities to apply for a National
Pollutant Discharge Elimination System (NPDES) permit
for stormwater discharges from a variety of activities.
New York State is a NPDES-delegated state, and the
411
-------
Department of Environmental Conservation is administer-
ing this program through their State Pollutant Discharge
Elimination System (SPDES) permit. One of the 11
categories covered in the regulations is construction
activity. Under this activity, any site where 5 or more
acres are disturbed must have an erosion and sediment
control plan and a stormwater management plan. The
5-acre size limit has been challenged as arbitrary, and
the size limit could be changed to 1 acre of disturbed
area. A developer needs to file a Notice of Intent at
least 48 hours before beginning operations to have
"coverage." This notice is filed with the U.S. Environ-
mental Protection Agency in Newington, Virginia. Under
the regulations, copies of the erosion and sediment
control plan and stormwater management plan are to be
kept on site. Copies of each are also sent to the munici-
pality that has jurisdiction. NYS-DEC does not want the
notices or plans sent to its offices; they will not be review-
ing or approving these plans. Who will? What will be the
local impacts?
As a result of this mandate, many New York counties,
towns, and villages will be receiving many erosion and
sediment control and stormwater management plans.
The majority of these units of government are still un-
aware of the requirements of the national program and
of what their role is or should be. There is a great need
for administrators, planners, and legislators to become
aware of the program and the process. Technical staff
need to learn the principles of planning, design, con-
struction, and inspection for erosion and sediment con-
trol and stormwater management systems.
Positive aspects of the NYS-DEC approach to the pro-
gram include the opportunity for local policy develop-
ment, provisions for local ordinances, and the formation
of interagency partnerships. Because NYS-DEC recog-
nizes that authority should rest at the local level, com-
munities have control over the quality of the natural
resources in their backyards. Of course this may require
additional staff or cooperation with other agencies to
assist with implementation.
Training Programs
Early efforts in erosion and sediment control began with
awareness seminars at the local level. The seminars
usually lasted 2 hours an evening for local officials
involved in the site review and approval process. Rec-
ognizing problems, learning the planning steps, and
becoming familiar with practices and guidelines were
the limit of these seminars.
The complexity of requirements and the technical needs
have increased dramatically due to recent mandates.
The Soil Conservation Service, in cooperation with
NYS-DEC and Syracuse University, has developed a
tiered educational program in erosion and sediment
control and stormwater management.
A1 -day seminar has been developed for planning board
members, environmental management council members,
legislators, and town boards, and has included legal
advisors, consulting engineers, and other agency per-
sonnel responsible for environmental analysis. This
agenda is included as Figure 1. This seminar stresses
site planning through a slide presentation that demon-
strates problems without control and shows practices
necessary to maintain resources on the site. Stormwater
management performance standards are reviewed in
accordance with NYS-DEC criteria. This seminar is re-
inforced with two specific site examples. Attendees are
asked to work in small design teams to design an ero-
sion and sediment control plan for the first site. These
same design teams are asked to critique the second
site, which already has an erosion and sediment control
plan. Thus, attendees go from designers to reviewers in
applying their knowledge of these principles.
A 2-day workshop has been developed for the technical
staffs of resource agencies, consulting engineers, local
governments, and others with technical review or design
responsibility (see Figure 2). This session begins with a
quick overview of the principles of erosion and sediment
control, then continues with a class exercise to design
an erosion and sediment control plan for a development
site while working in design teams of approximately four
individuals. The afternoon of the first day is spent at a
field site gathering specific resource information and
data to design a detailed erosion and sediment control
plan for the site. The design teams also compute and
compare peak discharges for the site for predevelop-
ment and postdevelopment conditions using Soil Con-
servation Service Technical Release 55, Urban
Hydrology for Small Watersheds (TR-55). The session
concludes with group presentations.
A 3-day short course with Syracuse University has been
developed to address the specific technical needs of
consulting engineers working with stormwater and ero-
sion control systems. This tuition-based course provides
for more indepth design of erosion and sediment control
practices using a field site. Sizing stormwater detention
basins is also required. In addition to the increased
technical emphasis, additional speakers from state and
local agencies provide a component on rules and regu-
lations. Syracuse University awards two continuing edu-
cation units for this course, which 57 people have
completed to date. The agenda is included as Figure 3.
Urban Erosion Control and Stormwater Design (CIE
600) stands as a fully accredited 3-hour graduate level
course in the Civil and Environmental Engineering De-
partment at Syracuse University. It was taught for the
first time in the 1992 fall semester and will be taught
again this September. It was developed as a hands-on
course that requires detailed designs for two projects,
using field trips and six additional site review projects.
412
-------
EROSION AND SEDIMENT CONTROL SEMINAR
AGENDA
8:30 AM Registration
9:00 AM Introduction and Course Overview
9:15 AM Developing an Erosion and Sediment Control Plan
Planning Considerations
Factors That Influence Erosion
Elements for a Sound Plan
Vegetative and Structural Components
Standards and Specifications
11:00 AM Site Example
Develop Conceptual Erosion and Sediment Control Plans
12:00 PM LUNCH (ON YOUR OWN)
1:00 PM Site Review
Critique an Erosion and Sediment Plan for a Specific Site
3:30 PM Wrap Up/Summary
4:30 PM Adjournment
Figure 1. Erosion and Sediment Control Seminar agenda.
413
-------
EROSION AND SEDIMENT CONTROL WORKSHOP
AGENDA
First Day
8:30 AM Registration
9:00 AM Introduction and Course Overview
9:15 AM Developing an Erosion and Sediment Control Plan
Planning Considerations
Factors That Influence Erosion
Elements for a Sound Plan
Vegetative and Structural Components
Standards and Specifications
11:00 AM Site Example
Develop Conceptual Erosion and Sediment Control Plans
12:00 PM LUNCH (ON YOUR OWN)
1:00 PM Design SessionSite-Specific Practices
Temporary Swale
Sediment Trap
Urban Runoff
2:30 PM Field ProblemDesign Teams
Gather Data
Develop Concepts in Field
4:30 PM Adjournment
Second Day
8:30 AM Complete Group Designs
10:00 AM Design Critiques
12:00 PM LUNCH (ON YOUR OWN)
1:00 PM Design Session
TR-55 Analysis for Structures
Rock Outlet Protection
Class Discussion
3:00 PM Wrap Up and Summary
3:45 PM Adjournment
Figure 2. Erosion and Sediment Control Workshop agenda.
414
-------
SYRACUSE UNIVERSITY
UNIVERSITY COLLEGE
EROSION AND SEDIMENT CONTROL
9:00 AM
10:00 AM
11:00 AM
12:00 PM
1:00 PM
2:15 PM
2:30 PM
4:30 PM
8:00 AM
9:30 AM
9:45 AM
11:30 AM
12:00 PM
3:00 PM
5:00 PM
8:00 AM
10:00 AM
10:15AM
11:45 AM
1:00 PM
SHORT COURSE AGENDA
April 28-30, 1992
First Day
Registration and Coffee
Introduction and Course Overview
Legislation, Ordinances, and Regulatory Review Process
Developing Your Stormwater Management Plan and Practices
Lunch
Urban Hydrology and Flow Routing
Break
Urban Hydrology and Flow Routing (continued)
Adjourn
Second Day
Developing Your Erosion Control Plan
Break
Erosion and Sediment Control Practice Standards
Lunch (En Route to Field Site)
Field Tour/Site Problems
Group Design Session
Adjourn
Third Day
Group Presentations and Critiques
Break
Group Presentations (continued)
Wrap UpAdjourn Short Course
Certified Professional Erosion Specialist Exam
Part II (Optional)
Dr. Stephan Nix
Mr. Robin Warrender
Mr. William Morton
Mr. Russell Nemecek
Mr. William Morton
Mr. Donald W Lake, Jr.
Mr. Donald W. Lake, Jr.
Mr. Donald W. Lake, Jr.
Mr. Donald W. Lake, Jr.
Dr. Stephan Nix
Mr. Donald W. Lake, Jr.
Figure 3. Erosion and Sediment Control short course agenda.
415
-------
In addition, the class participates in a town planning
board meeting. Syllabus topics (see Figure 4) include
manual and computer analyses of stormwater dis-
charges and lectures by a plant materials specialist, a
code enforcement officer, and governmental repre-
sentatives dealing with rules and regulations. Twelve
students enrolled in the first class, which was extremely
well received by both students and the people who
provided the example sites.
Summary
Over 2,600 people have received training through 76
different seminars, workshops, short courses, and the
graduate course since the training effort began in the fall
of 1988. These tiered training sessions have evolved
one after another based on needs at the local level.
Leaders in the NYS-DEC recognized that benefits are
local so training efforts should be local. This has led to
interagency cooperative agreements between the U.S.
Department of Agriculture, Soil Conservation Service, and
NYS-DEC to bring training directly to the communities.
There is no sign of these training requests letting up. An
average of 10 requests for the seminar sessions are
made at the local level during the year. In addition, the
proposed cooperative agreement for Fiscal Year 1994
between the Soil Conservation Service and NYS-DEC
calls for five 1-day seminars, four 2-day workshops, four
2-day TR-55 hydrology workshops, and two short
courses. The Syracuse University graduate course will
be taught again this fall. Future projects also include
workshops for New York State code enforcement offi-
cers, development of a field notebook for job superin-
tendents, and field application courses for equipment
operators. After all, equipment operators have the last
word in installation.
We have come a long way, but we can see that chal-
lenges are still ahead of us to educate public planners,
legislators, consultants, technical staff, and contractors
in the use of sound erosion and sediment control and
stormwater management practices to protect and en-
hance water quality and the environment.
416
-------
DEPARTMENT OF CIVIL AND ENVIRONMENTAL ENGINEERING
SYRACUSE UNIVERSITY
CIE 600
URBAN STORMWATER AND EROSION CONTROL DESIGN
FALL 1992
SCHEDULE:
INSTRUCTOR:
TEXT:
GRADING:
Monday/Wednesday
6:15-7:45 PM
Peck Hall, University College
Donald W. Lake, Jr., PE
State Conservation Engineer, USDA-SCS
SWCS, Empire Chapter, New York Guidelines for Urban Erosion and Sediment
Control, October 1991; Soil Conservation Service, Technical Release 55, Urban
Hydrology for Small Watersheds, June 1986; New York State Department of
Environmental Conservation, Reducing the Impacts of Stormwater Runoff From New
Development, April 1992.
Assignments: 40%
Mid-Term Exam: 30%
Final Exam: 30%
Course Content:
Week:
8/31
9/7
9/14
9/21
9/28
10/5
10/12
10/19
10/26
Topics:
Introduction to Urban Stormwater and Erosion
Control Design (1)*
Resource Planning and Stormwater Impacts (2)
Computing and Controlling Sediment and Runoff (2)
Stabilizing Soil, Vegetative and Biotech (2)
No lecture E&S Field Exercise
(10/3, 8:30-11:30 AM)
(turn in 10/7)
Urban Hydrology (2)
Urban Hydrology (1) and Site Exercise Critique (1)
NO CLASS HYDROLOGY PROJECT
Urban Hydrology Computer Program (1) and
MIDTERM
Reading
Ch. 1, NY Guide
and DEC Manual
Ch. 8, Appendix B,
NY Guide
Chs. 4 and 5,
NY Guide
NY Guide
SCS-TR-55
Tr-55
Instructor
Lake
Lake
Lake
Dickerson
Lake
Lake
Lake
Lake
Lake
Chapman
Lake
Figure 4. Urban Stormwater and Erosion Control Design course agenda.
417
-------
Week:
Topics:
Reading
Instructor
*Number
11/2
11/9
11/16
11/23
11/30
12/7
12/14
12/21
of lectures that week
Construction/Maintenance/Code Enforcement
Town Planning Board Assignment and
Stormwater Field Exercise
(11/14 9:00 AM)
Performance Standards for Stormwater
Management
Flow Routing (1)
Flow Routing (2)
Stormwater Basin Design (2)
Course Review
FINAL EXAM
NY Guide Proietta
Lake
Chs. 5 and 6, Warrender
DEC Manual Morton
Nix
Nix
DEC Manual Lake
Nix
Lake
Instructors
Donald W. Lake, Jr., PE, State Conservation Engineer, USDA-SCS
John Dickerson, Northeast Plant Materials Specialist, USDA-SCS
Dana Chapman, Asst. State Conservation Engineer, USDA-SCS
Robin Warrender, Chief, Nonpoint Source, Division of Water, NYS-DEC
William Morton, Resource Specialist, NYS Department of Environmental Conservation
Dr. Stephan Nix, Professor, Syracuse University, Civil and Environmental Departments
Figure 4. Urban Stormwater and Erosion Control Design course agenda (continued).
418
-------
Field Office Technical Guide:
Urban Standards and Specifications
Gary N. Parker
U.S. Department of Agriculture, Soil Conservation Service
Champaign, Illinois
Abstract
The Field Office Technical Guide is the primary tech-
nical reference for the Soil Conservation Service
(SCS). It presently contains general resource refer-
ences and soil and site information, and describes
conservation management systems, practice stand-
ards and specifications, and conservation effects. Al-
though SCS maintains offices and provides
assistance in all Illinois counties, the technical guide
does not contain any information specific to natural
resource use and management in urban areas. There-
fore, in June 1992 the SCS in Illinois entered into an
agreement with the Illinois Environmental Protection
Agency to develop technical information describing
best management practices (BMPs) for controlling
urban nonpoint source water pollution.
Currently in development, this information will include
40 BMP standards and accompanying construction
specifications, material specifications, and standard
drawings. It will also include estimates of pollutant re-
moval effectiveness and stormwater pollutant export, as
well as planning and design criteria. When complete,
this material will become part of the Field Office Techni-
cal Guide. The Illinois Environmental Protection Agency
will also use the information in a separate, stand-alone
technical manual. This material will be useful to plan-
ners, engineers, architects, and construction contrac-
tors, as well as to local government staff.
Background
The Soil Conservation Service (SCS), an agency of
the U.S. Department of Agriculture, is the major fed-
eral agency providing natural resource management
assistance on nonfederal land. Its primary responsi-
bility is to provide leadership and expertise in manag-
ing natural resources in nonurban areas. Currently,
SCS maintains a network of field offices in nearly
every county in the country, providing local citizens with
direct access to a wide range of technical specialists.
These specialists include engineers, soil scientists, bi-
ologists, agronomists, and natural resource planners.
The technical material and expertise that has been de-
veloped to support SCS activities largely pertains to
agricultural or rural settings. For example, the seed
mixtures that most SCS specifications call for are those
appropriate for agricultural areas and not necessarily for
parks, recreation sites, or lawns. In addition, design
criteria for waterways and diversions assume an agri-
cultural land use context.
Despite this rural, nonurban emphasis within the
agency, SCS maintains a field staff in urban and urban-
izing areas. In Illinois, this urban staff serves over one-
half the state's population. This urban presence has
enabled SCS to develop some urban expertise. For
instance, SCS TR-55 hydrology modeling techniques
are widely used to estimate runoff from urban areas.
Moreover, the PL-566 watershed projects constructed in
the Chicago suburbs have given the agency some ex-
pertise in urban construction site issues. The SCS, how-
ever, has not provided any systematic technical support
to its field staff on natural resource management issues
in an urban setting. It has instead relied on the ability of
its staff to adapt the provided information from a rural to
an urban environment.
To become more effective in addressing key natural
resource issues in urbanizing areas, the SCS in Illinois
has initiated several activities:
It is actively participating in a coalition of state and
federal agencies to prepare a strategy for coordinat-
ing agency activities in northeastern Illinois.
It is reviewing and clarifying its policy relative to pro-
viding assistance in nonagricultural areas.
419
-------
It is expanding the technical information its staff uses
when providing assistance to decision-makers in ur-
ban areas.
The third initiative listed is the subject of this paper. In
June 1992, SCS entered into an agreement with the
Illinois Environmental Protection Agency to prepare a
set of standards and specifications describing BMPs for
controlling urban nonpoint source water pollution. In
addition, the SCS will provide estimates on the range of
pollutant removal effectiveness and criteria for planning
runoff management. The agency will incorporate all this
material into its Field Office Technical Guide.
Field Office Technical Guide
The Field Office Technical Guide is the primary technical
reference for the SCS. It contains technical information
about conservation of soil, water, air, plant, and animal
resources. The guide is designed for use by technically
trained people who are assisting landowners and users,
land managers, government officials, and other deci-
sion-makers to plan, apply, and maintain appropriate
conservation practices. The technical guide also is a
major reference for those addressing top-priority re-
source goals identified by the National Program for Soil
and Water Conservation. These goals are to reduce the
damage caused by excessive erosion and to protect
water from nonpoint source pollutants. The technical
guide identifies sediment, nutrients, animal waste, pes-
ticides, and salinity as nonpoint source pollutants.
The Field Office Technical Guide contains five sections:
The "General Resource References" section lists ref-
erences, cost data, maps, climate data, cultural re-
sources information, threatened and endangered
species, and pertinent state/local laws, ordinances,
and regulations.
The "Soil and Site Information" section describes the
soil survey of the local area. It contains soil descrip-
tions and interpretations that can be used to make
decisions about land use and management. This sec-
tion identifies soil characteristics that limit or affect
land use and management, and rates soils according
to limitations, capability, or potential.
The section on "Conservation Management Systems"
provides information for developing resource man-
agement systems to prevent or treat problems asso-
ciated with soil, water, air, and related plant and
animal resources. This section includes quality crite-
ria that describe the level of resource protection that
decision-makers should try to achieve to meet re-
source quality goals.
The "Practice Standards and Specifications" section
alphabetically lists conservation practices used by
the field office, followed by practice standards and
specifications. It may also include references and
documentation requirements for the individual prac-
tices. Practice standards establish the minimum level
of acceptable quality for planning, designing, install-
ing, operating, and maintaining conservation prac-
tices. Practice specifications describe the technical
details and workmanship required to install the prac-
tice, as well as the quality and extent of materials
used in the practice.
The last section, "Conservation Effects," contains in-
formation describing the economic and environ-
mental effects of implementing particular practices
and systems. The purpose of this section is to provide
decision-makers with a way to evaluate the extent to
which various alternatives can meet their goals.
As stated previously, this guide is the primary technical
reference for SCS staff, particularly those at the field
level. The guide is also useful to Soil and Water Conser-
vation District staff, and to consultants and staff of state,
county, and municipal governments. To expand its use-
fulness, however, the SCS urban field staff in Illinois
have recommended that the guide include information
that is directly relevant to natural resource management
in an urban environment and is user friendly to urban
clients. The material now being developed will attempt
to meet that need.
New Material for the Field Office Technical
Guide
The new material will supplement and expand the exist-
ing material in the guide's fourth section, Practice Stand-
ards and Specifications. The SCS will modify or develop
40 BMPs that deal specifically with urban natural re-
source management.
Each BMP standard will follow a uniform format:
"Definition": describes what the practice is.
"Purpose": explains what the intended effect of the
practice is, that is, why this practice is used.
"Conditions Where the Practice Applies": describes
the types of sites where the practice would be appro-
priate; this section also describes limiting factors
such as slope percent, maximum drainage areas,
and maximum flow velocities.
"Criteria": describes, in general terms, material and
construction requirements and usually provides ref-
erences to specific material and/or construction
specifications.
"Considerations": offers general information regard-
ing factors to consider when deciding on the appro-
priateness of a particular practice; in some cases,
this section is a brief, narrative, nontechnical sum-
mary of the "Conditions" section.
420
-------
"Plan and Specification Requirements": describes the
nature and extent of the information the contractor
needs to build the practice; it lists the requirements
of the plans and specifications needed to install a
practice.
"Operation and Maintenance Requirements": de-
scribes the needed operation and maintenance ac-
tions and suggests the frequency with which they
should be performed.
The revised fourth section of the technical guide will also
include all the material specifications and constructions
referenced in the practice standards, as well as a series
of standard drawings for the practices. The standards
and specifications will be available on computer disk.
The standard drawings, which will be developed using
a CAD system, also will be available on disk. This will
allow engineers and consultants to access the material
in preparing construction plans and specifications.
In addition to the SCS incorporating the new material
into the Illinois Field Office Technical Guide, the Illinois
Environmental Protection Agency plans to issue a
stand-alone technical manual of those standards for use
by consultants, state agencies, and local governments.
The project is scheduled for completion in December
1994.
421
-------
Storm water Outreach at the Federal Level:
Challenges and Successes
Kimberly O. Hankins
Office of Wastewater Enforcement and Compliance,
Office of Water, U.S. Environmental Protection Agency, Washington, DC
Background
Stormwater regulations brought a distinctly different com-
munity into the realm of U.S. Environmental Protection
Agency (EPA) regulation. Many members of this com-
munity have never before been regulated by an environ-
mental program. The regulated community now includes
all major cities and unincorporated areas with populations
of 100,000 or more, as well as a very large, diverse
group of industries. The most important factor influenc-
ing success with the stormwater regulations is educa-
tion. By educating all parties concerned with the
program, the community can begin to practice all that
EPA is learning about how to provide a cleaner, safer
environment.
The principal elements of an outreach program are com-
munication and education, with a focus on influencing
how people and organizations act. Given this, the Na-
tional Pollutant Discharge Elimination System (NPDES)
stormwater outreach program at the national level
should, among other things:
Disseminate information and educate people about
the effects of receiving water pollution from diffuse
sources, such as the loss of recreational activities.
Promote positive environmental results, including the
reduction of pollutant loadings into receiving waters.
Theoretically, accomplishing these goals should elicit a
successful outreach program at any level. In fact, success
is much more elusive. Of course, many outreach programs
implement this theory very effectively. At the federal
level, however, EPA has 16 different customers reflect-
ing 10 EPA regions, 50 states, thousands of municipali-
ties, and hundreds of thousands of facilities, trade
associations, and professional groups. Moreover, when
factoring in to this multitude Congress, EPAs own man-
agement, and scarce resources, a successful outreach
program becomes a tremendously complex and costly
endeavor.
At the federal level, it is crucial to provide as much
information as possible to as many people as possible.
Therein lies the biggest challenge in outreach at the
federal level. This paper presents some of the chal-
lenges in developing an outreach strategy for the storm-
water program at the federal level. It also describes
some of the projects EPAs Office of Water has under
way, some of which have worked very well and some of
which have not. In addition, the paper discusses what
the future holds for the stormwater outreach program.
Challenges of Developing a Stormwater
Outreach Strategy
For its first year or so, the strategy of the stormwater
outreach program consisted of a hotline, which ad-
dressed most needs, and speaking engagements,
which filled in the gaps.
Almost immediately after the NPDES stormwater pro-
gram was born, several years ago, the stormwater hot-
line was established. Since its inception, the hotline has
received over 90,000 calls. The hotline staff answers
questions, distributes documents, and handles registra-
tion for EPA workshops and seminars.
The other important element of the early stages of the
stormwater program was speaking engagements and
workshops. These continue to be one of the best ways
to get "the word out" correctly. Regulated communities
need to know exactly how the stormwater program af-
fects them. For example, the program held 12 work-
shops between 1990 and 1991 to explain the November
10, 1990, regulations.
As the stormwater program matured, it became appar-
ent that the community needed a more substantial out-
reach strategy. The hotline staff quickly found it difficult
422
-------
to refer all policy interpretation calls to EPA stormwater
staff. At that time, the staff at Headquarters was very
small and the regions were overburdened.
Consequently, the Headquarters stormwater staff ex-
panded, and one of its first tasks was to develop an
outreach plan. The first step was to identify the plan's
customers, which turned out to be just about everyone.
Primary customers are the regions and states. Of
course, there are 11 categories of regulated industries
and over 200 municipalities in Phase I alone. The list of
customers continues to grow when the general public,
elected officials, professional associations, trade
groups, and consultants all are factored in. These
groups require a different level of understanding of
stormwater regulations. This presented a major chal-
lenge because the staff needed to examine each docu-
ment and ensure that it satisfied the needs of more than
one group of customers.
This early outreach strategy assumed knowledge of
what the customers wanted. The assumption, however,
was wrong. There was one crucial step in strategy de-
velopment that the stormwater staff neglected to com-
plete: ask the customers. Because of their enormous
number, however, asking them all was impossible.
Some customers, of course, in addition to the regulated
community, are the states and regions, who are trying
desperately to run their own stormwater programs.
These customers were finally asked about the outreach
plan at the 1992 Stormwater Coordinator's Conference
in Atlanta, Georgia. The stormwater staff reviewed what
they had been doing to date, and customers offered
helpful suggestions on what to do next. Customers also
participated in a session specifically targeted at design-
ing the stormwater workshops held in April 1993 in
Annapolis, Maryland, so as to ensure customer input.
During this meeting, it became apparent that many
states and regions were duplicating work unnecessarily,
that is, developing something that another state had
already developed. This was very frustrating for all those
involved. Some kind of clearinghouse or electronic com-
munications system was desperately needed. Re-
search, however, had already shown that it could cost
from $750,000 to $1 million to set up such a system.
This cost prevented Headquarters from accomplishing
this effort on its own. Therefore, it asked the states to
help by directing their 104(b)(3) grant funds to this effort.
This seemed the only way to accomplish the goal
quickly and effectively. Although this sounded like it
would work, it has not. There is quite a bit of reluctance
to use that money for this task. Therefore, stormwater
personnel have begun to look for other avenues.
The challenges multiply when budget constraints are
considered. One of the biggest problems involves print-
ing a developed document. The printing budget at Head-
quarters has taken some very serious cuts. Despite
attempts to solve this problem, difficulties continue. For
instance, Headquarters has tried to distribute items
electronically, but this can cause more problems than it
solves. Budget cutbacks have seriously hampered
plans to develop more public education materials than
are currently available.
Of course, nearly everyone has been hit very hard by
budget problems. Some states and counties have of-
fered very creative ideas about getting the "most bang
for your buck!" This issue has shed new light on the
problem of getting out as much information as possible.
These are just some of the challenges stormwater staff
have faced in putting together an outreach strategy. The
next section describes some current outreach projects.
Current and Developing Outreach
Activities
Research
A primary task has been to research existing outreach
activities. Much information on these activities exists,
and both researchers and audiences find this an ongo-
ing educational process. Research efforts include:
Research on outreach activities
Audience: Headquarters management, regions
Research on videos
Audience: Headquarters management
Research on clearinghouses
Audience: Headquarters management, regions
Current research on existing outreach activities exam-
ines their successes and failures. Hopefully, this effort
will help target materials and practices that can be ex-
panded to a national level. While outreach videos have
had difficulty with funding, the staff is researching what
is out there, again, in case it finds something that works
well and can be expanded to a national level. Finally,
research on clearinghouses began before stormwater
staff heard from the regions and states. The staff tried
to learn of available clearinghouses to examine the
possibility of their use or adaptation.
Outreach Strategy
The strategy is expected to be presented in a dynamic
document. Its audience is Headquarters management
and the regions. Hopefully, the document will provide an
adaptable framework for designing and completing out-
reach projects within an assigned time frame.
Fact Sheet Development
Because the stormwater program involves so many is-
sues and firestorms, staff often produce fact sheets to
clear up confusion. Past fact sheets have focused on:
423
-------
The Transportation Act's effect on the stormwater
program.
The Ninth Circuit Court decision that affected munici-
palities.
The Municipal Part II guidance document.
Phase II progress and results of public meetings.
Question and Answer Document
The audience for this document is the regions and in-
dustries via trade associations. The first volume was
developed based on questions from the hotline. The
staff compiled over 50 commonly asked questions and
answers into one document, which has been distributed
through the hotline.
The second volume covers more complex interpreta-
tions of the regulations, including questions on sam-
pling, group applications, and the Ninth Circuit Court
decisions. Again, distribution will probably proceed
through the hotline.
Stormwater Workshops
In fiscal year (FY) 1991, the stormwater staff at Head-
quarters conducted 12 workshops on the basics of the
stormwater program. The workshop audience consisted
of regions, states, and the regulated community. The ob-
jective was to inform as many people as possible about
the requirements of the November 16,1990, rule. Atten-
dance was in the thousands. The effort was successful.
In FY 1992, the stormwater staff presented workshops
and spoke to over 4,000 people. These workshops fo-
cused on the requirements of the general permit and the
development of pollution prevention plans. In addition,
workshops for municipalities covered the requirements
of the Part 2 municipal application. All these workshops
were well received and also considered successful.
The FY 1993 workshops presented by Headquarters
focused on developing pollution prevention plans. The
staff developed a workshop series with the first day
targeted to reach state and EPA regional repre-
sentatives. This day is a train-the-trainer session to
teach the audience how to lead a workshop on pollution
prevention for industry. The second day is designed for
the industrial regulated community and focuses on in-
dustrial and construction pollution prevention plan de-
velopment. This day should include case studies and
interactive exercises.
These workshops mark the first effort by the stormwater
program to conduct workshops of this kind. The hope
was to meet the objectives identified by the regions and
states at the 1992 Stormwater Coordinator's Confer-
ence in Atlanta. Due to budget problems, Headquarters
was limited to the number of workshops it could conduct
in each region. The goal was, however, for state and
regional staff to be able to present the workshops on
their own. Each state was to receive a set of slides and
speaking materials for its own use.
Municipal Support Division/Permits Division
Pamphlet on Stormwater
The audience for this publication is Headquarters, the
regions, and the general public. This project has expe-
rienced difficulties getting started due to contractual
problems. It is, however, now moving ahead toward
completion. The pamphlet is predominantly aimed at
members of the general public who have little or no
knowledge of the stormwater program.
Updated Stormwater Overview
This document addresses general information needs.
Its audience consists of Headquarters, the regions, and
the general public. The Overview reviews who the
stormwater program covers, what their application op-
tions are, and what the deadlines are associated with
those applications. As the program grows and changes,
the Overview is updated. Distribution is currently
through the stormwater hotline.
Raindrop Report (Status of the Stormwater
Program)
This document is targeted to Headquarters, the regions,
and the general public. It supplies a brief update on current
activities in the stormwater program and features rele-
vant information from recent Federal Registers. In addi-
tion, it describes outreach activities and provides specfics
on applications submitted and general permits.
Articles for Newsletters
Stormwater staff are developing articles by request for
publication in various journals and newsletters. They are
trying to establish a regular submittal effort to some
publications, such as the Nonpoint Source News Notes,
which is published by the Headquarters nonpoint source
program to supplement the bulletin board.
General Permit Effectiveness Study
The purpose of this effort is to determine the effective-
ness of the general permit approach in implementing
Phase I. The evaluation assesses, among other things,
the rate of compliance, the level of awareness, and the
quality of pollution prevention plans being developed.
This effort also is identifying obstacles that prohibit the
general permit from being as effective as possible.
Monthly Conference Calls
As of March 1993, Headquarters had completed 15
regularly scheduled conference calls with stormwater
424
-------
regional coordinators. These meetings have proven
very successful, and they should continue.
Storm water A wards
These awards recognize municipalities and industries
that demonstrate a commitment to protecting and im-
proving the quality of the nation's waters through out-
standing implementation of innovative and
cost-effective stormwater control programs and pro-
jects. In 1991, the winner for a stormwater control pro-
gram or project by a municipality was Murray City, Utah.
In 1992, the city of Orlando, Florida, won, and Prince
George's County, Maryland, took second place. Nomi-
nations are sought from the 10 EPA regions.
National Stormwater Coordinator's
Conference
This annual event is indispensable for planning and
feedback from the states and regions. The meeting is
designed for regional and state stormwater coordina-
tors, as well as for Headquarters staff.
Continuous Speaking Engagements
Stormwater staff receive requests to speak to groups
twice a week on average. While they are not always able
to fill some requests because of a limited travel budget,
the staff respond to as many as possible. In FY 1992,
staff participated in about two dozen talks or seminars,
not including the workshops.
Phase II Outreach Meetings
The Phase II Outreach Meetings are a series of meet-
ings designed to include individuals that may be affected
by the Phase II regulations in the development of those
regulations. As of this writing, four meetings have been
held (two in Washington, one in Dallas, and one in Chicago)
to involve as many people as possible.
Information and Education Catalog
Another important project is the management and peri-
odic update of the Information and Education Catalog,
which was distributed at the National Urban Runoff Man-
agement Conference. The author and Tom Davenport
manage this project. Everyone concerned should
have a copy of this excellent document. Management
plans to expand the manual to include stormwater
information. In addition to putting out several calls for
information, the conference registration packet in-
cluded a form to fill out if individuals wanted this
catalog to include a particular document. Manage-
ment believes this document will help in the tremen-
dous demand for technology transfer in the stormwater
and nonpoint source programs. This, of course, is a top
priority that customers have requested.
Electronic Sources
Linking to other clearinghouses and bulletin boards
should improve communications. The nonpoint source
program at Headquarters has been extremely helpful by
placing information and announcements on its elec-
tronic bulletin board and in the Nonpoint Source News
Notes publication. This has proven to be a good way to
meet customer needs.
Further Considerations
Education is becoming one of the most important as-
pects of the stormwater program as people learn about
the regulation and how it affects their day-to-day lives.
Industries as part of their pollution prevention plans are
developing training and education programs for their
own employees. Cities are training their employees in
sampling techniques and safety procedures as well as
developing excellent public education programs. Tre-
mendous efforts involving stormwater education are be-
ing undertaken. Stormwater Headquarters needs to
know about the successful programs to help the lesser
programs learn.
As this program moves forward, each success in educat-
ing those affected by the stormwater program, including
the general public, leads to greater accomplishments.
As these successes continue to build, more people will
understand the intent and effects of protecting and
cleaning up the waters of our nation. It is a cycle in which
we all play a major role.
425
-------
Training for Construction Site Erosion Control and Stormwater Facility Inspection
Richard Horner
University of Washington, Seattle, Washington
Abstract
Probably the leading reason that stormwater manage-
ment programs fail in effectively protecting water re-
sources is the lack of followup to ensure that permit
conditions are met, approved designs are properly in-
stalled, and temporary and permanent management
practices and facilities are maintained. Avoiding this
downfall requires obtaining the legal authority for and
then instituting a coordinated program extending from
the first submission of permit applications through con-
struction and all phases of site operation. This program
should have components covering the construction
phase as well as permanent practices and facilities.
While somewhat different elements are appropriate for
the two components, they share the common precepts
of sound underlying planning; competent plan review;
and effective inspection, maintenance, and enforce-
ment. The University of Washington's Center for Urban
Water Resources Management and Office of Engineer-
ing Continuing Education have developed and are offer-
ing courses to train personnel responsible for various
aspects of the suggested program. This paper empha-
sizes the training for site inspectors. For construction-
site inspectors, it covers the role of the erosion and
sediment control (ESC) plan, the applicability of many
ESC practices, key points to check when inspecting
them, and how to deal with various circumstances that
can arise during inspections. For permanent drainage
system inspectors, the paper covers both the initial con-
struction and continuing operation of facilities and offers
guidance on key inspection points and such issues as
safety, tracking maintenance, and waste handling.
Introduction
Effective stormwater management requires successful
execution of steps at all phases of a project. These
phases and the accompanying management steps in-
clude analysis of potential problems in the planning
stage, quality design of programs and practices to pro-
tect aquatic resources as the project takes shape, com-
petent review of plans at the permit application point,
proper implementation of approved plans during con-
struction, and correct operation and practices at facilities
after their installation. All phases of the process need
improvement through a better basis in knowledge and
greater skills in application. Probably the weakest areas
and the leading causes of program failures and environ-
mental damage are implementation during construction
and long-term operations.
Redressing this weakness will require widespread de-
velopment of comprehensive and aggressive programs
of inspection during the construction of developments
and their stormwater management systems, followed by
ongoing inspection of operating systems to ensure suf-
ficient maintenance for continuing adequate perform-
ance. The diffusion of development and tradition of local
land-use control prevalent in most of the United States
will necessitate local acquisition of the legal authority,
where it does not now exist, to institute these programs.
As is already occurring in some places, it is likely that
larger units of government will become involved in set-
ting standards for these programs. The U.S. Environ-
mental Protection Agency's National Pollutant Discharge
Elimination System (NPDES) program is presently ex-
tending authority over programs in the largest cities and
counties and at sites of construction larger than 5 acres
and involving industrial activity. Still, the details and the
responsibility for conducting the programs will very likely
rest with local governments.
The concern of this discussion is the development and
execution of local programs to upgrade significantly the
quality of followup to increase the probability that ap-
proved stormwater management plans are effective.
The scope of the programs envisioned would extend
from the point of permit issuance through construction
and all the years of site operation to follow project
completion. The programs might be considered to have
distinct components, covering, for example, erosion and
sediment control (ESC) inspection at construction sites,
inspection of the construction of storm runoff quantity
426
-------
and quality control facilities, and the periodic inspection
and maintenance of operating facilities. However they
are structured, these programs should embrace some
common principles. They should be the logical exten-
sion of and ultimate implementation vehicle for the fore-
going phases of planning, design, and plan review.
Further, they should be conceived and conducted as
essential elements of a successful program, deserving
of the needed funding, staffing, support by administra-
tors and public officials, training of personnel, and en-
forcement authority.
This discussion covers aspects of program development
and especially emphasizes training for site inspectors.
For these purposes it divides the overall program into
two components. One covers construction site ESC
programs. The second covers permanent drainage
practices and facilities, both their inspection at construc-
tion and followup inspection and maintenance. In both
cases, the paper recommends program structures and
discusses some key program elements. It then offers
specific examples of inspection checks to perform in the
field. The goal of the paper is to give the reader a basis
for beginning program design and undertaking the key
element of training the staff who will be charged with its
performance.
The discussion was derived from two courses devel-
oped and offered by the University of Washington's
Center for Urban Water Resources Management and
Office of Engineering Continuing Education. The course
coverage is organized in the same manner as this pres-
entation, and course manuals are available for ESC
inspector training (1) and permanent drainage system
inspector training (2). Important contributions to the ma-
terial presented in these courses and in this discussion
have been made by local governments and state agen-
cies in the Puget Sound area of Washington state that
have been working actively to improve stormwater man-
agement through good followup.
Construction Site ESC Inspection
Programs
Program Development
Program Elements
The following elements are recommended fora compre-
hensive construction site ESC program:
ESC planning
A plan review process
Contractor education
An inspection and enforcement process
The subsections to follow cover two of these program
elements in detail, ESC planning and inspection and
enforcement. The latter discussion is then extended in
the following section to examples of inspection guide-
lines for common practices.
ESC Planning
ESC planning is an absolute prerequisite for an effective
program. A careful site analysis should produce a stand-
alone plan (i.e., a plan devoted exclusively to this aspect
of the project) developed with the same thoroughness
and care as any other plan in the overall construction
set. It is intended for use by the plan reviewer, the
construction superintendent and other contractor per-
sonnel, and the construction site inspector. This sub-
section outlines the ESC planning process from
beginning to end and concludes with an example of a
complete plan.
In approaching an ESC plan, the planner must:
Understand the erosion process, so that it can be
controlled.
Know the site and the construction plan, so that both
potential problems and solutions will be apparent.
Understand the various ways that erosion can be
prevented or that eroded sediments can be caught.
The erosion process is first reviewed for the lessons it
can offer ESC planning. Erosion has been understood
for thousands of years, as is attested by the extensive
evidence of terraced farmingsome continuing today
in steep terrain in ancient cultures. Figure 1 illustrates
the types of erosion and its nature. Soils can be loos-
ened and set in motion initially by the impact of falling
raindrops. Erosion progresses, although gradually, as
runoff flows in a sheet over a bare surface and exerts
shear stress, which is a function of velocity, on soil
particles. The rate of erosion increases when flow con-
centrates and increases in velocity. Channels formed by
these flows are known as rills. When rills join and form
highly concentrated, rapidly flowing channels, the rate
increases still further, a stage termed gully erosion. Ero-
sion can progress still further to mass wasting when a
whole area loses stability.
Several factors involving site soils, vegetation, and to-
pography influence the erosion process. Soil erodability
is greater in the case of silts and fine sands than clays
or soils with a substantial gravel content. Relatively high
organic content also offers cohesiveness that resists
erosion. Clays tend to produce a larger volume of runoff,
however, because of their relatively poor permeability,
which exerts more erosive stress on soil. Vegetative
cover offers a number of important advantages, includ-
ing reducing raindrop impact, slowing runoff velocity,
helping to absorb water, and holding soil in place. In
regard to topography, both slope gradient and length
tend to increase velocity and the resulting frictional
427
-------
Raindrop Erosion
Sheet Erosion
Rill and Gully Erosion
Stream and Channel Erosion
Stream Flow
Figure 1. Soil erosion processes (3).
shear stress. Erosion hazards relative to slope gradient
and length are listed in Table 1.
Acquiring the familiarity with the site and proposed con-
struction necessary to proceed with the ESC plan in-
volves data collection and analysis. Site data should be
collected in regard to:
Soils
Vegetation
Topography
Ground-water table
Neighboring water bodies
Adjacent properties
Drainage routes and patterns (define subbasins)
Potential areas of serious erosion problems
Existing development, utilities, and dump sites
The following construction plan information should be
cataloged at the outset of planning:
Grading (location, amount)
Topographic changes
Clearing and grading limits
Drainage changes
Table 1. Soil Erodability Relative to Slope Gradient and
Length
Erosion Hazard
Low
Moderate
High
Slope Gradient
0-7%
7-15%
>15%
Maximum Length
300 ft
150 ft
75ft
Materials to be used and locations of use and storage
Access points
ESC planning should proceed with reference to certain
basic principles, as follows:
First consider all means of preventing erosion; only
consider trapping sediments from unavoidable ero-
sion. Prevention has the potential to be more effec-
tive in resource protection than later treatment and
less costly.
Phase construction and post clearing limits to main-
tain as much natural vegetation as possible and for
as long as possible.
Plan construction to fit the site; use terrain advanta-
geously and avoid critical areas.
Cluster buildings and other developed features, and
minimize their impact on impervious area.
Plan for control of erosion subbasin by subbasin.
Minimize extent and duration of vegetation removal
(especially during wet season) and soil disturbance.
Stabilize and protect disturbed areas as soon as possible.
Use natural drainage features, existing vegetation,
and materials found on the site.
Minimize slope length and gradient to control runoff
velocities.
Divert offsite runoff away from disturbed areas.
Retain any released sediment within the construction
area and reduce tracking off site.
Have a thorough maintenance and followup program.
Take measures to control potential pollution from con-
struction materials (e.g., paving materials, petroleum
428
-------
products, other vehicle fluids, fertilizers, pesticides,
grinding and sanding debris, wastes).
An ESC plan consists of a narrative and site plans.
Points that should be covered by the narrative include
1) a project description, 2) a description of existing
and modified site conditions, 3) descriptions of ESC best
management practices (BMPs), 4) descriptions of
BMPs for pollutants other than sediments, 5) plans for
permanent stabilization, 6) calculations, and 7) provisions
for inspection and maintenance. Site plans are maps
and engineering plans illustrating and specifying the
project's location, existing and modified site conditions,
and BMPs. The set of site plans should include 1) a data
collection worksheet (principally showing topography,
soils, and vegetation), 2) a data analysis worksheet
(mainly indicating drainage subbasins and primary
drainage courses), 3) a site plan development work-
sheet (showing existing and finished contours, roadways,
and permanent stormwater facilities), 4) the ESC plan
(showing BMP locations), and 5) diagrams of repre-
sentative BMPs, as appropriate. The ESC plan (item 4
in the set) is the key element for implementing the plan.
BMPs are usually specified on this plan using a system
of symbols, which are defined in a legend.
Inspection and Enforcement
The most important general needs of an inspection and
enforcement program are a staff dedicated to the func-
tion, specific staff training, and administrative support.
These needs are best provided for by a dedicated reve-
nue source, such as a stormwater utility assessment.
The staff should not have unrelated and distracting
duties such as inspection of other facets of construc-
tion. Initial training should offer needed background in,
for instance, legal and regulatory requirements, water
quality, hydrology, soils, and vegetation. Subsequent
training should provide detailed coverage of BMP re-
quirements, such as discussed in the following section.
Strong support from administrators is essential for a staff
undertaking a relatively new function that might be un-
popular in terms of economic interests.
Beyond these basic needs are some specific issues to
clarify during program development for incorporation as
formal program elements. Recommendations on the is-
sues presented in this paper are drawn from experience
in the Puget Sound region, especially in King County
and the cities of Bellevue and Redmond. One of these
issues is the response to a situation in which measures
in an approved ESC plan proved inadequate. Strong
permit review should normally limit these instances, but
unforeseen circumstances can still arise. Inflexible adher-
ence to an ESC plan can be self-defeating when meas-
ures prove to be inadequate for whatever reason; thus,
the jurisdiction should retain the authority to require
additional measures if needed. This option should be
noted in a statement on each ESC plan.
A second issue is how field change orders will be han-
dled. The policy should call for careful but expeditious
consideration of requests for plan changes, generally
after consultation with plan review personnel. Finally at
issue is the granting of variances from code require-
ments. Conditions on granting variances should be strict
and specific, such as:
The expected result should be at least comparable
to the outcome expected to be achieved with the
approved method.
Sufficient background information and justification
should be presented for adequate assessment of the
alternative.
The ability should be retained with the variance to
meet objectives of safety, function, appearance, en-
vironmental protection, and maintainability based on
sound engineering judgment.
The variance should be in the public interest.
Enforcement authority must be obtained and the system
of enforcement defined and made clear to the regulated
parties. A system successfully used by the city of
Bellevue has a sequence of three steps, as follows:
A verbal warning, with a deadline for correction.
A correction notice (with specifications of correc-
tions), a deadline, and a warning about the conse-
quences of noncompliance.
A stop-work order, with a warning about the conse-
quences of noncompliance.
ESC Practices and Their Inspection
Categories of Practices
The numerous ESC practices in use can be categorized
in various ways. The most basic division is between
erosion control practices, which prevent or minimize
erosion, and sediment control practices, which attempt
to capture soil released through erosion. Within each of
these broad groupings are several categories that rep-
resent general strategies for achieving either erosion
control or sediment control. In addition to sediments,
construction sites can generate many other pollutants,
such as petroleum products, solvents, paints, sanding
dusts, pesticides, and fertilizers. It is most efficient to
manage those materials along with sediments and to
inspect the management practices for them simultane-
ously with ESC inspection. Therefore, these practices
represent another basic division.
Following is the breakdown of ESC practices used by
Reinelt (1), with the number of individual practices in
429
-------
each category. The 29 practices represented are by no
means the only ones, but they are the most widely
recognized and used. Twenty-two of the 29 (all but the
sediment trapping techniques) are preventive and are
thus generally the most cost-effective options; however,
the straw bale and filter fabric fences and sedimentation
ponds among the trapping techniques are most com-
monly used practices.
1. Erosion control
1.1. Natural vegetative covertwo practices
1.2. Temporary coverthree practices
1.3. Permanent vegetation establishmenttwo
practices
1.4. Stabilized construction entrance and roads
three practices
1.5. Runoff controleight practices
2. Sediment trapping techniquesseven practices
3. Management of other construction site pollutants
four practices
The following passages provide inspection checklists for
example practices, generally the most common, in each
category and subcategory. The checklists are divided
into checks to be made when the practice is imple-
mented and checks to be made on each followup visit
to determine the need for maintenance or replacement
of the ESC materials. Many of the points are illustrated
in diagrams that accompany the checklists.
While much of an inspector's work is performed in the
field, it is often advisable or even absolutely necessary
to do some background work in the office before going
out to inspect an installation. This work mainly consists
of consulting the ESC plan to determine the specifica-
tions. The plan should be retained on the construction
site should the inspector or construction personnel need
to refer to it.
1. Erosion control
1.1. Natural vegetative cover
1.1.1. Phasing construction
Phasing construction is a practice in
which clearing operations are per-
formed in stages to take advantage of
cover that exists on site before con-
struction.
Installation checks:
1. Are areas that will not be cleared
set off with plainly visible clearing-
limit fencing?
2. Is plainly visible flagging placed at
the drip line of trees to be pro-
tected (see Figure 2)?
3. Are fills and cuts near protected
trees treated as shown in Figure
2?
4. Is final vegetation established as
soon as portions of the site can be
made ready?
Maintenance checks:
1. Do fencing and flagging need re-
pair or replacement for personnel
to see it clearly?
2. Do exposed or injured roots of
protected trees need covering or
dressing?
1.2. Temporary cover
Temporary cover practices recognize that por-
tions of most construction sites remain un-
worked for months, during which time very
large amounts of erosion can occur unless
these areas are stabilized. Stabilization can be
achieved with temporary seeding or various
kinds of slope coverings, or both. Slope cov-
erings include both mulches and commercial
mats and blankets. It is often necessary to
apply temporary cover to different areas sev-
eral times during construction.
Mulches, mats, and blankets can serve sev-
eral purposes in erosion control: covering the
slope temporarily to prevent erosion by rain-
drop impact and the friction of runoff, holding
water to encourage grass growth, protecting
grass seedlings from heat, and enriching the
soil. Straw, hay, wood fiber, wood chips, and
other natural organic materials can serve as
mulches. Inspection guidelines for straw and
wood fiber are given below as examples. Mats
and blankets are manufactured from both
natural and synthetic materials. Guidelines are
given for several varieties.
1.2.1. Temporary seeding
Installation checks:
1. Is the soil stabilized within the pe-
riod specified by regulation?
(This period varies from place to
place, depending on climate pat-
terns. In the Puget Sound area of
Washington, which receives most
of its rainfall in the winter, the
specified periods are within 2
430
-------
Surface
Cleared
Roots compete
for water with
lawns and shrubs.
Individual Plants
Water-seeking roots
may clog tile lines.
Potential Problems
Drain Tiles
"Vertical
Tiles"
Mixture of
Peat Moss or
Leaf Mold
and Soil
Figure 2. Guidelines for preserving natural vegetation (3).
days during the months October
to April and within 7 days during
the months May to September.)
2. If used without slope covering
practices, is temporary seeding
limited to slopes of less than 10
percent and 100 ft in length? If the
slope exceeds either limit, is a
mulch or mat slope covering used?
3. Has the seedbed been prepared
with at least 2 to 4 in. of tilled
topsoil?
4. Is fertilizer use limited as much as
possible; if used, is it applied in
amounts no greater than the
needs of the grass for the prevail-
ing soil conditions?
431
-------
5. Is mulch applied for protection if
seeding occurs when tempera-
tures can be high or runoff is likely
to occur before the grass is well
established?
6. Is irrigation provided if planted
when rainfall might be insufficient
for good establishment?
Maintenance checks:
1. Is it necessary to irrigate and/or
reseed?
2. Is maintenance fertilizer needed?
1.2.2. Straw mulch
Straw mulch can be used without
seeding or, for better erosion control,
with seeding.
Installation checks:
1. Is the straw spread generally a
minimum of 2 in. deep (corre-
sponds to 2 to 3 tons per acre)
and greater on very steep slopes,
adjacent to sensitive areas, and
where concentrated flow passes
over the slope?
2. Is the mulch anchored as needed
by crimping, disking, rolling, or
punching into soil or by moisten-
ing, tackifying, or netting?
Maintenance checks:
1. Is replacement needed as a result
of blowing away or decomposition
over time?
2. Is there any fire hazard requiring
moistening?
1.2.3. Wood fiber mulch
Wood fiber mulch should only be used
with seeding and generally should be
used with a soil bonding agent.
Installation checks:
1. Is the mulch used with seeding
and a soil bonding agent? Were
the bonding agent distributor's
application guidelines followed?
2. Has the wood fiber been applied
to cover the soil completely, allow-
ing no bare soil to show through
(corresponds to about 1 ton per
acre and is adequate for most cir-
cumstances)? Are there any spe-
cial circumstances, such as
seeding during hot weather, when
the amount should be increased
by about 50 percent?
Maintenance checks:
1. Is replacement needed as a re-
sult of loss over time?
1.2.4. Excelsior
Excelsior is a product made of fine
wood shavings that assume a more-
or-less helical form. As a conse-
quence of this form, excelsior does
not lie in close contact with the soil
and allows runoff to drain beneath it
and cause erosion. Therefore, it
should be used only with seeding,
where it is very useful in holding mois-
ture and providing protection from di-
rect sun in hot periods. Suppliers
generally market several grades for
sheet and channelized flow and differ-
ent velocities.
Installation checks:
1. Is the excelsior used only with
seeding?
2. Was an appropriate material se-
lected according to manufac-
turer's recommendations and
then placed and stapled as rec-
ommended by the manufacturer?
3. On slopes, was it placed 3 ft over
the crest or in an anchor ditch?
4. In ditches, was it placed in the
direction of water flow with any
seams offset 6 in. from the ditch
centerline?
Maintenance checks:
1. Is replacement needed as a result
of damage or loss over time?
1.2.5. Mats and blankets
Examples of materials produced in a
mat or blanket form for erosion control
are jute, woven straw, and synthetics.
Mats can be used without seeding, or
with seeding for better erosion control.
As with excelsior, suppliers generally
market several grades for sheet and
432
-------
channelized flow and different veloci-
ties.
Installation checks:
1. Was an appropriate material se-
lected according to manufac-
turer's recommendations and
then placed and stapled as rec-
ommended by the manufacturer?
2. Was it placed in the direction of
water flow, in full contact with the
soil but not tightly stretched?
Maintenance checks:
1. Is replacement needed as a result
of damage or loss over time?
1.3. Permanent vegetation establishment
Permanent vegetation should be established
as soon as possible after all construction is
completed in each segment of the site. Grass
can be established by seeding or sodding.
Seeding is generally preferred because of the
lower cost and greater flexibility in selecting
grass species. Sod is often available only in
limited varieties, which may not be the most
suitable for erosion control and other purposes
unless grown to order. In some cases,
overseeding with preferred species is recom-
mended in the spring, when grass must be
established with sod in the winter. Species
should be selected based on local climatologi-
cal and soil conditions, with reference to re-
gional guidance documents, and, when
necessary, in consultation with regional ex-
perts.
1.3.1. Permanent seeding
Installation checks:
1. Has the seedbed been prepared
by loosening with a plow if sub-
soils are highly compacted,
spreading 2 to 6 in. of topsoil, and
lightly rolling?
2. Is fertilizer use limited as much as
possible; if used, is it applied in
amounts no greater than the needs
of the grass for the prevailing soil
conditions?
3. Is mulch applied for protection if
seeding occurs when tempera-
tures can be high or runoff is likely
to occur before the grass is well
established?
4. Is irrigation provided if planted
when rainfall might be insufficient
for good establishment?
Maintenance checks:
1. Is it nencessary to water, reseed, or
add fertilizer?
1.3.2. Sodding
Installation checks:
1. Is the sod placed from the lowest
area and perpendicular to water
flow?
2. Are sod strips wedged tightly to-
gether and joints staggered at
least 12 in.?
3. Is the sod stapled if on a steep
slope?
Maintenance checks:
1. Is overseeding needed, either to
repair damage or to install a pre-
ferred grass species?
1.4. Stabilized construction entrance and roads
The entrance is the most important access
route to stabilize, since it is the last point at
which tracking sediment off site can be
stopped. If equipment travels extensively on
unstabilized roads on the site, a tire and vehi-
cle undercarriage wash near the entrance will
be needed. Perform washing on crushed rock.
Wash water will require treatment in a sedi-
ment pond or trap.
1.4.1. Stabilized construction entrance (see
Figure 3)
4-to8-ln. Quarry Spalls
Provide Full Width of
Ingress/Egress Area
Figure 3. Stabilized construction entrance (from Washington
Department of Ecology, 1992).
433
-------
Installation checks:
1. Is the entrance constructed with
quarry spalls 4 to 8 in. in size and
at least 12 in. thick?
2. Is the stabilized entrance sized cor-
rectly for the site?
3. If the entrance sits on a slope, is
a filter fabric fence in place down-
gradient?
Maintenance checks:
1. Is the entrance clogged with sedi-
ments, requiring top dressing the
pad with clean 2-in. rock?
2. Is it necessary to clean up any
sediments carried from the site
onto the street?
1.5. Runoff control
Runoff control represents various practices
designed to keep water from coming in contact
with bare soil or controlling its velocity if it
does. Included are drains for surface and sub-
surface water, dikes and swales placed across
slopes to interrupt runoff, and roughness cre-
ated on the surface to reduce velocity. Exam-
ple guidelines presented below are for a pipe
slope drain and surface roughening.
1.5.1. Pipe slope drain (see Figure 4)
A temporary pipe slope drain is an
effective technique for preventing ero-
sion on a slope caused by runoff from
a higher elevation. Upslope runoff
needs to be collected and directed
into the drain effectively and then dis-
charged in a controlled way to prevent
erosion at the bottom of the slope.
Installation checks:
1. Are no more than 10 acres
drained into a single pipe slope
drain?
2. Was a minimum 6-in. metal toe
plate placed at the entrance to
prevent undercutting?
3. Is runoff directed into the pipe with
interceptor dikes at least 1 ft
higher at all points than the top of
the pipe?
4. Is there a slope toward the pipe
on a grade of at least 3 percent at
the inlet?
5. If the pipe is 12 in. in diameter or
larger, was a flared entrance sec-
tion installed and connected se-
curely to the drain with water-tight
connecting bands?
Discharge into a stabilized
watercourse or a sediment
trapping device or onto a
stabilized area
Earth Dike
Corrugated Metal
or CPEP Pipe
Slope = 2:1
H = D + 12in.
Elbows
Corrugated Metal
or CPEP Pipex
6D
Slope 3% or
Steeper
v
6-ln. Min.
Riprap per Table III-2.6
Depth of apron shall be
equal to pipe diameter.
Entrance Section Cutoff Wall
K
x Diameter D (for pipe >12 in.)
4 Ft Min. at Less
Than 1% Slope
Figure 4. Pipe slope drain details (3).
434
-------
Undisturbed Area
Tread Grooves of Track
Perpendicular to Slope
Direction
Undisturbed Vegetation
Diversion
Dozer Treads
Create Grooves
Perpendicular to
Slope Direction
Figure 5. Examples of surface
equipment (3).
roughening using heavy
6. Was the soil thoroughly com-
pacted at the entrance and under-
neath the pipe?
7. Were gasketed, water-tight fittings
placed between pipe sections,
were the sections securely fas-
tened, and was the drain an-
chored to the soil?
8. Was the area below the outlet sta-
bilized with a riprap apron?
9. If the drainage can carry sedi-
ment, is it treated in a sediment
pond or trap?
Maintenance checks:
1. Is undercutting or bypassing oc-
curring at the inlet, requiring rein-
forcing of the headwall with
compacted earth or sandbags?
2. Is erosion occurring at the outlet,
necessitating rebuilding the
apron?
1.5.2. Surface roughening (see Figure 5)
A roughened surface is an easy and
inexpensive way to reduce runoff ve-
locity, encourage the growth of vege-
tation, increase runoff infiltration, and
trap some sediment. It is not effective
enough to use alone but can reduce
the load on sediment trapping instal-
lations downstream. Roughening is
best used on slopes steeper than 3
horizontal to 1 vertical that do not re-
quire mowing. There are several
methods of roughening a surface, all
of which involve forming horizontal
depressions with equipment. Methods
include tracking perpendicular to the
slope direction, driving treaded equip-
ment along the slope direction to get
grooves perpendicular to the slope, or
tilling (preferred because it avoids
compaction). On steeper slopes
(steeper than 2 horizontal to 1 vertical)
a stair-step pattern should be formed.
Installation checks:
1. Have all exposed slopes steeper
than 3 horizontal to 1 vertical been
roughened, with 40- to 50-in. stair-
step patterns formed on slopes
steeper than 2 horizontal to 1 ver-
tical?
2. Was the soil scarified if it was
heavily compacted by the rough-
ening?
3. Was the area seeded as quickly
as possible?
Maintenance checks:
1. Have rills appeared that should be
regraded and reseeded?
2. Sediment trapping techniques
Trapping sediments once they are released requires
slowing the transport velocity sufficiently for soil
particles to settle (i.e., reducing the velocity below
the settling velocity of the particles). Soil particles
range over several orders of magnitude in size, from
the small clays to the large sands. Settling velocity
is approximately related to the square of the particle
diameter; thus, halving the diameter approximately
quadruples the time needed for settlement. There-
fore, as particles decrease in size, they become
435
-------
increasingly difficult to remove from a runoff stream.
This fact is largely why preventive techniques are
more cost effective than sediment trapping practices
and are strongly preferred.
The two basic types of sediment trapping tech-
niques in use are sediment barriers and settling
ponds. Sediment barriers include the commonly
used filter fabric and straw bale fences as well as
brush fences and barriers constructed of gravel.
Both types trap sediments in the same way, by
ponding water. Although that mechanism is more
obvious in the case of ponds than of barriers, prac-
tices of the latter type actually provide only a mini-
mum of filtering capability and primarily slow the
flow of water long enough for some particles to
settle. Thus, they can only trap relatively large par-
ticles, generally the larger silts and sands. The trap-
ping ability of settling ponds depends on their size.
While they can theoretically be made large enough
to trap any size particle, practical sizes generally
limit efficient removal to the medium silts and larger.
2.1. Sediment barriers
Several principles apply to the various types
of sediment barriers. Maximizing a sediment
barrier's ponding volume maximizes the
amount of sediment trapped. Therefore, the
barriers should be placed away from the im-
mediate toe of slopes in order to increase the
area for ponding. It is very important that sedi-
ment barriers be aligned on the contour, not up
and down slopes. This alignment places them
at a right angle to flow paths and also in-
creases ponding volume. Slopes draining to
sediment barriers generally should not be
more than 100 ft long. Sediment barriers must
be trenched in and staked to hold up under the
pressure of the wall of water they will dam.
Finally, sediment barriers do not provide effec-
tive sediment removal from concentrated
flows. While straw bales are sometimes used
in ditches, rock check dams are really a better
alternative for decreasing velocity in channels.
2.1.1. Filter fabric fence (see Figure 6)
Installation checks:
1. Are filter fabric fences used only
in the following applications:
Maximum of 1 acre served by a
single fence?
Maximum 1:1 slope gradient and
100-ft slope length?
Sheet flow situation (never in con-
centrated flow)?
Filter fabric material in
continuous rolls; use
staples or wire rings
to attach fabric to wire
5 Ft0 In.
n /
Wire mesh
support fence
for slit film fabrics
n
2 Ft 0 In.
' "
2 Ft 6 In
_
* i
'"*"' 'Bury bottom of filter material
in 8- by 12-in. trench N.
6 Ft Max.
, 2- by 2-in. wood posts, standard or
better or equivalent
'
F,i.j^
Wire mesh support fence
for slit film fabrics -
Filter Fabric
Materia
Provide washed gravel backfill
or compacted native soil as
directed by local government
Bury bottom of filter material-
in 8-by 12-in. trench
2- by 2-in. wood posts, standard
or better or equivalent
2 Ft 0 In
12ln.
3 In. Min
5 FtOln.
Figure 6. Filter fabric fence detail (3).
2. Is the fence aligned to slope con-
tours as well as possible?
3. Is the fence installed so that its
height above the soil is no more
than 3 ft?
4. Are posts 2 x 4 in. wood or 1.33
Ib/ft steel, or the equivalent?
5. Are posts buried 2.5 ft deep
whenever possible and spaced
no more than 6 ft apart?
6. Is fabric attached on the upslope
side with staples (at least 1 in.), tie
wires, or hog rings?
7. Is the end of the fabric buried in a
trench sized as shown in Figure 6
and backfilled on both the upslope
and downslope sides (as shown)?
8. Is splicing avoided if possible? If
impossible, is splicing done only
at posts and overlapped at least 6
in.?
9. Nonwoven and woven monofila-
ment materials have the best prop-
erties for silt fencing. If a woven
slit-film fabric is used, is wire
mesh reinforcing (14-gauge rein-
436
-------
forcing wire mesh with openings
no larger than 6 in.) placed on the
upslope side and fastened the
same as the fabric?
Maintenance checks:
1. Is it necessary to restake, reat-
tach, or replace the fence to main-
tain all of the above conditions?
2. Is sediment removal needed (be-
fore it reaches 1/3 the height of
the fence)?
2.1.2. Straw bale fence (see Figure 7)
Straw bale fences tend to swell when
they get wet and require frequent
maintenance. They are not highly rec-
ommended but could be more effec-
tive if used according to the following
guidelines.
Installation checks:
1. Are straw bale fences used only
in the following applications:
Maximum of 1/4 acre served per
100 ft offence length?
Maximum 2:1 slope gradient and
100-ft slope length?
2. Is the fence aligned to slope con-
tours as well as possible?
3. Are the bales bound with wire,
preferably, or string placed
around the sides of the bale, par-
allel to the ground?
1. Excavate the trench.
2. Place and stake straw bales.
3. Wedge loose straw between bales. 4. Backfill and compact the excavated soil.
CONSTRUCTION OF A STRAW BALE BARRIER
Points A should be higher than point B.
PROPER PLACEMENT OF STRAW BALE BARRIER IN DRAINAGE WAY
Figure 7. Proper installation of straw bale fences (3).
437
-------
4. Are the bales installed in a 4-in.
trench, as shown in Figure 7, and
backfilled with 4 in. of soil on the
upslope side?
5. Are the bales forced together as
tightly as possible and anchored
with at least two stakes or pieces
of rebar per bale driven toward
the previous bale and flush with
the top of the bale?
6. Are gaps wedged with straw, and
is straw spread on the upslope
side?
7. Are straw bale fences used in
channels with concentrated flow
only when velocities are low and
placed as shown in Figure 7 (per-
pendicular to flow and extending
at least one bale length above the
mid-channel bale)?
Maintenance checks:
1. Is it necessary to replace the
fence to maintain all of the above
conditions?
2. Is sediment removal needed (be-
fore it reaches 1/2 the height of
the fence)?
2.2. Settling ponds
Settling ponds have several advantages. They
can function through all construction phases
and have relatively low maintenance require-
ments. They can also be located to intercept
runoff both before and after the onsite drain-
age system is developed.
The three types of settling ponds in use differ
only in their outlet structure. The term sedi-
ment basin is used to describe a settling pond
with a pipe outlet that generally serves a drain-
age area of 3 to 10 acres. A sediment trap is
a settling pond with a stable spillway outlet and
a smaller service area. The third type is a
permanent water quantity control pond put in
temporary service during construction; such a
pond is designed to drain completely between
storms in permanent service. This operating
mode is not appropriate for ESC application,
however, because the residence time is too
short for good particle trapping and settled
material becomes resuspended during drain-
ing. Therefore, a temporary riser outlet needs
to be installed for use during construction.
A key point in the design and construction of a
settling pond is to avoid short-circuiting by the
water. Short-circuiting can cut the actual resi-
dence time far below the theoretical value and
harm performance. Ways of avoiding it are to
divide the pond into two or more cells, locate
the inlet and outlet far apart, and install baffling
to increase the flow path.
2.2.1. Sediment basin (see Figure 8)
Installation checks:
1. Is the bottom graded to be as
level as possible?
2. Is the pond no deeper than 7 ft
with 1 ft of freeboard?
3. Are side slopes no steeper than 3
horizontal to 1 vertical?
4. Does the pond have an emer-
gency spillway that is 1 ft deep,
with a width two to three times the
number of acres served by the
pond, and lined with 2 to 4 in. of
rocks?
5. Does the pond discharge through
a riser pipe having at least two
1-in. diameter orifices at the top of
the sediment storage zone?
6. Are inlet and outlet areas pro-
tected from erosion with riprap?
7. Is baffling installed if the length-to-
width ratio is less than 6 or if the
entrance velocity is high?
8. A good feature to prevent short-
circuiting of flow is a two-celled
pond, preferably with cells divided
by sandbags or a rock berm and
connected by a riser pipe similar
to that used for the outlet. A less
preferred arrangement is dividing
the pond with a filter fabric fence.
Is this feature installed if specified
in the design?
9. Is the pond fenced if it presents
any safety hazard to children?
Maintenance checks:
1. Is sediment removal needed (be-
fore 1.5 ft accumulates)?
2. Are any outlet orifices clogged
and in need of cleaning?
438
-------
Pond Length >3x Pond Width
Filter Fabric
Fence
Inflow
Perforated Drain Pipe* in
Gravel-Filled Trench
Outlet Pipe
Riser Pipe* With
Weighted Base
"Sediment dewatering may be accomplished with perforated pipe in trench as shown or with a perforated
riser pipe covered with filter fabric and a gravel "cone." A control structure may also be required; see Conditions
Where Practice Applies.
1-Ft Spillway Depth
1-Ft Freeboard
Riser Pipe, Open at Top
Provide a Rebar Trash Rack on Riser
___/ Pipes >18 In.
6 Ft Min.
Emergency Overflow
Spillway Crest
(Principal Spillway)
Dewatering Outlets
Max. 4 In.
Sediment Storage
3-Ft Maximum Depth
Perforated Drain Pipe in
Gravel-Filled Trench for
Silt Dewatering; Trench
Wrapped With Filter
Fabric Full Length
Weighted Base To
Prevent Flotation
Energy-Dissipating
Rock
Figure 8. Typical sediment basin (3).
3. Are any embankments damaged
and in need of compaction or re-
building?
4. Has riprap or spillway lining mate-
rial been lost and need to be re-
placed?
5. Are there signs of excessive
drainage to the pond, requiring re-
routing or pond enlargement?
6. Are there signs of excessive
sediment loading to the pond, re-
quiring stabilization of the drain-
age area?
3. Management of other construction site pollutants
Construction sites can create pollution problems
over and above erosion and sediments through pav-
ing operations, handling and storage of various ma-
terials, spills, and waste handling. Inspectors should
also be aware of the potential for runoff contamina-
tion from these sources and inspect the site accord-
ing to the following guidelines.
3.1. Handling cement and concrete
Inspection checks:
1. Do concrete trucks have a designated
washout area with a sediment trap?
439
-------
2. Is exposed-aggregate driveway wash
water drained toward a collection point at
the side or into a sediment trap, where it
cannot get into a street drainage system?
3.2. Material storage and handling
Inspection checks:
1. Are weather-resistant enclosures used for
the storage and handling of materials,
such as paints, coatings, wood preserv-
atives, pesticides, fuels, lubricants, and
solvents, and for potentially polluting
wastes?
2. Are there designated and clearly commu-
nicated procedures for handling materials
and wastes and washing containers?
3. Is a chemical inventory maintained, in-
cluding Material Safety Data Sheets?
4. Are containers and enclosures inspected
periodically for leakage, indicating the
need for maintenance?
3.3. Spill containment
Inspection checks:
1. Has a spill control plan been developed,
and have supplies been obtained to imple-
ment it? Does the plan include:
Who to notify if a spill occurs?
Specific instructions for different prod-
ucts?
Who is in charge?
Spill containment procedures?
Easy to find and use spill cleanup kits?
How a spill will be prevented from getting
into a drainage system (e.g., valving,
diversion, absorption)?
A disposal plan?
A worker education program?
3.4. Waste management
Inspection checks:
1. Have waste reduction practices been insti-
tuted (e.g., reusing solvents, substituting for
toxic products, minimizing quantities of
materials used)?
2. Have recycling practices been instituted
(e.g., waste separation for recycling, pur-
chasing recycled materials)?
3. Are hazardous and nonhazardous wastes
separated and each disposed of properly
and promptly?
4. Has an employee education program on
waste management been established?
Inspection Programs for Permanent
Drainage Practices and Facilities
Program Development
Program Elements
The following elements are recommended fora compre-
hensive inspection program for permanent drainage
practices and facilities:
Stormwater management planning
Plan review process
Construction inspection and enforcement process
Followup inspection and long-term maintenance process
The stormwater management planning step ensures
that each site considered for a permit receives compre-
hensive analysis. The extensive considerations in this
portion of the recommended program are beyond the
scope of this discussion. The third element refers to
inspection of the stormwater management facilities
themselves when they are built to determine whether
installation has been consistent with the approved
plans. The final element seeks to ensure that facilities
continue to operate properly. The next subsection cov-
ers programmatic aspects of the followup inspection and
long-term maintenance process. The discussion is then
extended in the following section to examples of inspec-
tion guidelines for common practices and facilities.
Followup Inspection and Long-Term Maintenance
Process
Recommended features for a followup inspection and
maintenance program are:
An ordinance designating public authority and public
and private responsibilities.
A tracking system.
An inspection schedule.
A maintenance schedule.
A safety program.
A citizen response program.
A detailing of proper waste disposal practices.
A maintenance contractor education program.
440
-------
The discussion below elaborates on several of these
features, drawing principally on experience in King
County, Bellevue, Olympia, and elsewhere in the Puget
Sound region of Washington. The examples in the sec-
tion that follows this discussion present guidance on
establishing schedules for common facilities and the
specific checks to be made during inspection visits.
Public Versus Private Responsibilities. Whereas in-
spection is usually a public function, the question of
responsibility often arises with respect to the upkeep of
privately owned facilities. One model involves estab-
lishing a multiyear bonding period, during which the
developer has all responsibility. Often after this period
and a demonstration of effective operation, the govern-
ment agency responsible for stormwater management
then takes over operation and maintenance. A second
model calls for leaving maintenance as a private func-
tion (performed by a commercial property owner or
homeowners'association), with inspection by the public
agency. In this approach, the government assumes the
responsibility and assesses costs if the private party
does not meet its responsibility. Effective application of
this strategy requires that private maintenance contrac-
tors competently perform the needed work. The frequent
lack of qualified contractors requires government agen-
cies to consider training and certifying them.
Tracking System. King County, Washington, offers a
useful model for a tracking system to organize long-term
inspections and maintenance. The King County approach
uses a computerized information system. Each inspector
is assigned an inventory of facilities to inspect and spec-
ify maintenance and is given a laptop computer to use
in the field. The information system contains an identifi-
cation number for each facility, its type (e.g., wet pond,
infiltration basin), location, any special needs, and data
on previous experiences. At the conclusion of each visit,
the inspector enters a maintenance needs assessment
in the computer database. The computer then generates
a maintenance work order.
Safety. Safety is a major consideration because of po-
tentially harmful air quality in below-ground spaces, cor-
roded supports, traffic, falling objects, sharp edges,
poisonous plants and insects, and lifting. The safety
portion of an inspection and maintenance program
should include:
Testing instruments for harmful atmospheres (explo-
sive, containing hydrogen sulfide, lacking in oxygen);
a tester should be capable of checking all potential
conditions of concern, and all enclosed spaces
should be tested before an inspector enters.
Ventilating equipment.
Checking for structural soundness before entering a
manhole.
Traffic warning devices.
Ladders, safety harnesses, and hard hats.
Removing poisonous plants and threatening insect
nests.
Adequate personnel.
Safety training.
Waste Handling. Major maintenance on large facilities
should be scheduled when the least runoff is expected.
It is often a good idea to use ESC-type installations such
as filter fabric fences, sandbags, grassed drainage ar-
eas, and revegetation to prevent escape of sediments
during maintenance.
Although the vactor truck is the maintenance work-
horse, a problem concerns mixing waste that may be
relatively clean with very dirty waste. A solution, but an
expensive one, is to have "clean" and "dirty" trucks.
Another issue concerns disposal of both solids and
separated "decant" water picked up by vactor trucks.
The best solution for decant water is to discharge it to a
special decant station that has sediment and oil separa-
tion equipment, before the water is discharged to a
sanitary sewer. Few facilities currently operate this way,
and most vactor waste is discharged directly to a sani-
tary sewer. This practice can result in pollutants entering
surface waters because of inadequate treatment at the
municipal wastewater plant. It can also deliver toxic
materials that can upset biological processes at the
treatment plant. Guidelines are needed but generally do
not exist for disposing of solids. The best programs now
send them to a lined municipal landfill, unless they fail
a "looks bad and smells bad" test, in which case they
are treated as hazardous waste.
Permanent Drainage Practices and Facilities
and Their Inspection
Categories of Practices and Facilities
Following is the breakdown of practices used by Reinelt
(2), with the number of individual practices in each category:
1. Stormwater devicesthree practices
2. Detention facilitieseight practices
3. Infiltration facilitiesfive practices
4. Biofiltersthree practices
The 19 practices represented include some variations
on common devices, depending on their intended func-
tion, as specified by the Stormwater Management Man-
ual for the Puget Sound Basin (3). For example,
detention facilities include "wet ponds," which have a
quantity control function, and "water quality wet ponds,"
which are treatment devices.
441
-------
The following passages provide inspection checklists for
example practices and facilities, generally the most
common, in each category. The practices and facilities
themselves are described only very briefly in this sec-
tion. For detailed descriptions, consult a stormwater
management manual or textbook. The checklists are
divided into checks to make when the practice or facility
is first installed and checks to be made on each followup
visit to determine the need for maintenance. Many of the
points are illustrated in diagrams that accompany the
checklists. Also presented for a number of practices are
tables of maintenance standards. These tables have
been developed overtime in the Puget Sound area, and
several jurisdictions have contributed to them.
While much of an inspector's work is performed in the field,
it is often advisable or even absolutely necessary to do
some background work in the office before going out to
inspect an installation. This work mainly consists of con-
sulting the design plans to determine the specifications.
Too infrequent inspection and maintenance is one of the
main reasons for poor performance by stormwater facili-
ties. The frequency of followup inspections should be
determined based on the type of device and the circum-
stances where it is installed. An inspection and mainte-
nance plan should be developed before an installation
goes into service. As a general rule, surface facilities
should undergo a drive-by inspection at least monthly
and after any rain totaling 0.5 in. or more in 24 hr.
1. Stormwater devices
This group includes devices used for collection and
conveyance of stormwater, as well as special-purpose
facilities. Within the category are catch basins, pipes
and culverts, and oil/water separators. Inspection
guidelines are given for oil/water separators as a
complete example. Tables of maintenance standards
are included for the other types of facilities.
1.1. Oil-Water separators
Figure 9 illustrates the three basic types of
oil-water separators. The spill control unit's
purpose is to catch small spills; it is not capa-
ble of separating dispersed oil. The American
Petroleum Institute (API) separator is a baffled
tank that can separate "free" (unemulsified) oil
but requires a relatively large volume for effec-
tiveness. The coalescing plate (CP) separator
can separate free oil in a much smaller volume
because of the large surface area provided for
oil collection by the corrugated plate pack. The
following guidelines generally apply to all
types, except as noted.
Installation checks
1. Is the type appropriate for the service?
2. Is the unit sized and installed as specified
in the plans?
3. Are adequate removable covers provided
for observation and maintenance?
4. Is runoff excluded from roofs and other
areas unlikely to contain oil?
5. Is any pump in use placed downstream to
prevent mechanical emulsification?
6. Is detergent use avoided upstream to pre-
vent chemical emulsification?
7. For API and CP separators, is a forebay
provided sized at 20 ft2 of surface area per
10,000 ft2 of drainage area?
8. For API and CP separators, is an afterbay
provided for placement of absorbents?
9. For the CP separator, are the plates no
more than 3/4 in. apart and at 45 to 60
degrees from horizontal?
Maintenance checks:
1. Is weekly inspection performed by the
owner?
2. Are oil and any solids removed frequently
enough (at least just before the main run-
off period and then after the first major
runoff event)?
3. Are absorbents replaced as needed, but at
least at the beginning and end of the main
runoff season?
4. Is the effluent shutoff valve operational for
closure during cleaning?
5. Are waste oil and solids disposed of as
specified by regulations?
6. Is any standing water that is removed dis-
charged to the sanitary sewer and then
replaced with clean water?
1.2. Pipes and culverts
Refer to Table 2 for a summary of maintenance
standards for conveyance facilities.
1.3. Catch basins
Catch basins are routinely placed between the
drain inlets in streets and parking lots and the
conveyances that transport water away to
settle large solids. Refer to Table 3 fora sum-
mary of maintenance standards.
2. Detention facilities
Detention facilities include ponds that are designed
and operated either to drain within hours after a
442
-------
Table 2. Maintenance Standards for Pipes and Culverts
Conditions When
Defect Maintenance Needed Maintenance Results
Sediment Accumulated sediment Pipe cleaned of all
and debris that exceeds 20% of the sediment and debris.
diameter of the pipe.
Vegetation Vegetation that reduces All vegetation
free movement of water removed so water
through pipes. flows freely through
pipes.
Damage Protective coating is Pipe repaired or
damaged; rust is causing replaced.
more than 50% of
deterioration to any part
of pipe.
Any dent that decreases Pipe repaired or
the end area of pipe by replaced.
more than 20%.
Debris Trash or debris that is Barrier clear to
barriers plugging more than 20% receive capacity flow.
of the openings in the
barrier.
Damaged/ Bars are bent out of Bars in place with no
Missing bars shape more than 3 in. bend >3/4 in.
Bars are missing or Bars in place
entire barrier is missing. according to design.
Bars are loose and rust Repair or replace
is causing 50% barrier to design
deterioration to any part standards.
of barrier.
storm (dry ponds), to drain within a day or two
(extended-detention dry ponds), or to retain a per-
manent or semipermanent pool (wet ponds). These
ponds can have water quantity control objectives, or
water quality control objectives, or both, although
dry ponds offer few water quality benefits. Detention
facilities also include below-ground concrete vaults
and storage pipes, the latter sometimes referred to
as tanks. These devices serve primarily quantity con-
trol purposes, although if they have relatively long
water residence times they can collect some solids.
Other facilities sometimes included in this category
are parking lot and rooftop storage. Constructed wet-
lands can be placed in either this group or with
biofilters. Inspection guidelines are given for wet ponds
as a complete example. A table of maintenance
standards is included for vaults and tanks as well.
2.1. Wet ponds
Figure 10 illustrates a typical wet pond. A wet
pond has a "dead storage" permanent or
semipermanent pool and a "live storage" zone
that fills during runoff events and then drains
fairly quickly. Its design basis differs depend-
ing on its purpose (quantity control or quality
control, or both), but the checks made when it
is installed and later while it is operating are
Table 3. Maintenance Standards for Catch
Defect
Trash and
debris
(including
sediment)
Structural
damage to
frame or top
slab
Cracks in
basin walls
or bottom
Settlement/
Misalignment
Fire hazard
Vegetation
Pollution
Conditions When
Maintenance Needed
Trash or debris of more
than 1/2 ft3 located in
front of the catch basin
opening or is blocking
capacity of basin by >10%.
Trash or debris in the
basin that exceeds 1/3 to
1/2 the depth from the
bottom of the basin to the
invert of the lowest pipe
into or out of the basin.
Trash or debris in any inlet
or outlet pipe blocking
more than 1/3 of the height.
Dead animals or debris that
could generate odors that
would cause complaints or
dangerous gases.
Deposits of garbage
exceeding 1 ft3 in volume.
Corner of frame extends
more than 3/4 in. past
curb face into the street
(if applicable).
Top slab has holes larger
than 2 in.2 or cracks
wider than 1/4 in. (intent
is to make sure all
material runs in to basin).
Frame not sitting flush on
top slab (i.e., separation
of >3/4 in. of the frame
from top of slab).
Cracks wider than 1/2 in.
and longer than 3 ft, any
evidence of soil particles
entering catch basin
through cracks, or
structure is unsound.
Cracks wider than 1/2 in.
and longer than 1 ft at the
joint of any inlet/outlet
pipe or any evidence of
soil particles entering
catch basin through crack.
Basin has settled more than
1 in. or has rotated more
than 2 in. out of alignment.
Presence of chemicals
such as natural gas, oil,
and gasoline.
Vegetation growing across
and blocking >10% of
basin.
Vegetation (or roots)
growing in inlet/outlet pipe
joints that is >6 in. tall
and <6 in. apart.
Nonflammable chemicals
of >1/2 ft3 per 3 ft of
basin length.
Basins
Maintenance
Results
No trash or
debris located
immediately in
front of catch
basin opening.
No trash or
HphriQ in f*3tf*h
UCUI lo Ml Isdllsl 1
basin
Inlet and outlet
pipes free of
trash or debris.
No dead animals
or vegetation
present.
No garbage in
catch basin.
Frame is even
with curb.
Top slab is free
of holes and
cracks.
Frame is sitting
flush on top of
slab.
Basin replaced
or repaired to
design standards.
No cracks more
than 1/4 in. wide
at joint of
inlet/outlet pipe.
Basin replaced
or repaired to
design standard.
No flammable
chemicals
present.
No vegetation
blocking opening
to basin.
No vegetation or
root growth
present.
No pollution
present other
than surface film.
443
-------
Water Outlet
Separator Vault
Oil
Outlet
Water inlet
from streets,
parking lots,
or other
catch basins
Grease and
oil float on
retained water
Watertight
Cleanout Gate
Water
Inlet
Flow
Baffle
CPS Separator
Separator
Vault
SC-Type Separator
Clear
Well
Oil
Retention
Baffle
Oil
Separation
Compartment
Flow
Distribution
Baffle
Water
Outlet
Water
Inlet
\
Inspection and
Sampling Tee
Grit/Sludge
Removal Baffle
API Separator
Figure 9. Types of oil/water separators (3).
444
-------
Maintenance Control MH
w/Shear Gate/Valve on
Both Outlet Pipes
Bypass Pipe - Activate During
' Maintenance Operations Only
Control Structure
PLAN
No Scale
Aquatic Bench
5FtWide(Min.)x6ln. Deep
SECTION
No Scale
3- to 6-Ft-Deep
Permanent Pool
(Dead Storage)
Figure 10. Typical wet pond (3).
generally the same, with the few exceptions
noted.
Installation checks:
1. Does construction comply with local re-
quirements for earthwork, concrete, other
masonry, reinforcing steel, pipe, water
gates, metalwork, and woodwork?
2. Are all dimensions as specified in the ap-
proved plan?
3. Are interior side slopes no steeper than 3
horizontal to 1 vertical and exterior side
slopes no steeper than 2:1?
4. Is the bottom level?
5. Are the spillways (between cells, if any,
and the emergency outlet spillway) sized
and reinforced as specified in the ap-
proved plan?
6. Is a drain provided that can drain the dead
storage zone within 4 hr if necessary?
7. Are inlet and outlet areas stabilized as
necessary to avoid erosion?
8. Are safety concerns addressed, for exam-
ple, with such features as a shallow bench
completely around the edge of the pond,
barrier plantings to discourage approach
by children, and/or fencing (should not be
necessary if sloped as recommended and
other safety features are provided)?
9. For a water quality pond, is the effective
length-to-width ratio at least 3:1 minimum,
5:1 preferably; are the inlet and outlet
separated to the greatest width possible?
Maintenance checks:
1. Has a maintenance plan and schedule
been developed?
2. Refer to Table 4 for specific checks and
maintenance standards (these standards
apply to other types of ponds as well).
445
-------
Table 4. Maintenance Standards for Detention Facilities
Defect Conditions When Maintenance Needed
Maintenance Results
Trash and debris
Poisonous vegetation
Pollution
Unmowed grass/
ground cover
Rodent holes
Insects
Tree growth
Erosion of pond side
slopes
Sediment
accumulation in
forebay/pond
Dike settling
Rocks missing from
overflow spillway
Inadequate spillway
Missing, broken, or
damaged fencing
Erosion under
fencing
Missing or damaged
gates
Blocked or damaged
access roads
Any trash or debris that exceeds 1 ft3/ 1,000 ft2. There should be no
evidence of dumping.
Presence of any poisonous vegetation that constitutes a hazard to
maintenance personnel or to the public (e.g., poison oak, stinging
nettles, devil's club).
One gallon or more of oil, gas, or contaminants, or any amount that
could 1) cause damage to plant, animal, or aquatic life, 2) constitute
a fire hazard, 3) be flushed downstream during storms, or 4)
contaminate ground water.
In residential areas, mowing is needed when the cover exceeds 18
in. in height. Otherwise match facility cover with adjacent ground
cover and terrain as long as there is no decrease in facility function.
Any evidence of rodent holes if facility is acting as a dam or berm, or
any evidence of water piping through dam or berm via rodent holes.
When insects such as wasps or hornets interfere with maintenance
activities.
Tree growth does not allow maintenance access or interfere with
maintenance activity. If trees are not interfering with access, leave
trees alone.
Eroded damage >2 in. deep where cause of damage is still present
or where there is potential for continued erosion.
Accumulated sediment 10% of the design forebay/pond depth, or
every 3 yr.
Any part of dike that has settled >4 in.
Only one layer of rock above native soil in an area of 5 ft2 or greater,
or any exposed soil.
Emergency overflow or spillway not large enough to handle flows
from large storm events.
Any defect in fencing that permits easy entrance to the pond.
Damaged fencing including posts out of plumb by >6 in., top rails
bent >6 in., missing or loose tension wire, missing or sagging barbed
wire, missing or bent extension arms.
Fencing parts that have a rusting or scaling condition that is affecting
structural adequacy.
Opening in fencing that allows passage of an 8-in. diameter ball.
Erosion >4 in. deep and 12 to 18 in. wide, permitting an opening
under fence.
Missing or damaged gate, locking device, or hinges.
Gate is out of plumb >6 in. and out of design alignment >1 ft.
Missing stretcher bar, bands, or ties.
Debris that could damage vehicle tires.
Obstructions that reduce clearance above road surface to <14 ft
(e.g., tree branches, wires).
Any obstructions restricting access to a 10- to 12-ft width for a
distance of >12 ft, or any point restricting access to a width of <10 ft.
Any road settlement, potholes, mushy spots, or ruts that prevent or
hinder maintenance access.
Weeds or brush on or near road surface that hinder access, or are
>6 in. tall and <6 in. apart within a 400 ft2 area.
Erosion within 1 ft of the roadway >8 in. wide and 6 in. deep.
Trash and debris cleared from site.
No evidence of poisonous vegetation.
Coordinate with health department.
No contaminents present other than
surface film. Coordinate with local
health department.
Grass/ground cover should be mowed
to 2 in. Maintain dense cover on
slopes and in bottom of dry ponds.
Rodents destroyed and dam or berm
repaired. Coordinate with local health
department.
Insects destroyed or removed from
site. Coordinate with people who
remove wasps for antivenom protection.
Trees do not hinder maintenance
activities.
Slopes stabilized with appropriate
erosion control BMPs (e.g., seeding,
mats, riprap).
Sediment cleaned out to design depth.
Reseed if necessary for erosion control.
Dike is rebuilt to design elevation.
Rock replaced to design standard.
Increase capacity of spillway to current
design standards.
Fencing repaired to prevent entrance.
Repair fencing and barbed wire to
design standards
Structurally adequate posts or parts
with protective coating.
No opening in fence.
No opening under fence >4 in.
Gates, locking devices, and hinges
repaired.
Gate is aligned and vertical.
Stretcher bar, bands, and ties in place.
Roadway free of debris.
Roadway clear overhead to 14 ft.
Obstructions moved to allow at least a
12-ft access route.
Road surface repaired and smooth.
Weeds and brush on or near road
surface cut to 2 in.
Shoulder and road free of erosion.
446
-------
2.2. Vaults and tanks
Referto Table 5 for a summary of maintenance
standards for closed detention systems.
3. Infiltration facilities
Infiltration facilities discharge most of the entering
water to the ground. They include surface basins
and trenches, below-ground perforated pipes, roof
drain systems, and porous pavements. Inspection
guidelines are given for infiltration basins as a com-
plete example. A table of maintenance standards is
included for infiltration trenches as well.
3.1. Infiltration basins (see Figure 11 for a typical
basin)
Installation checks:
1. Does construction comply with local re-
quirements for earthwork, concrete, other
masonry, reinforcing steel, pipe, water
gates, metalwork, and woodwork?
2. Are all dimensions as specified in the ap-
proved plan?
3. Does the timing of basin construction
avoid the entrance of any runoff containing
sediment from elsewhere on the site?
4. Is the basin preceded by a pretreatment
device (e.g., presettling basin or biofilter)
to prevent failure caused by siltation?
5. Is the basin at least 50 ft from any slope
greater than 15 percent and at least 100 ft
upslope and 20 ft downslope of any build-
ing?
6. Is the outlet orifice design consistent with
the infiltration capacity on which the facility
is based (e.g., to avoid the collection of
more water than can infiltrate in 48 hr)?
7. Are the spillways (between cells, if any,
and the emergency outlet spillway) sized
and reinforced as specified in the ap-
proved plan?
8. Are all disturbed areas stabilized to pre-
vent erosion?
9. After final grading, has the bed been
deeply tilled to provide a well-aerated,
highly porous surface texture?
Maintenance checks:
1. Has a maintenance plan and schedule
been developed?
2. Refer to Table 6 for specific checks and
maintenance standards.
Table 5. Maintenance Standards for Closed Detention Systems
Defect Conditions When Maintenance Needed
Maintenance Results
Plugged air vents
Debris and sediment
in storage area.
Cracks in joints
between tank/pipe
sections
Problems with
manhole cover
Ladder rungs of
manhole unsafe
Catch basins
Half of the end area of a vent is blocked at any point with
debris and sediment.
Accumulated sediment depth is >10% of the diameter of the
storage area for 1/2 the length of storage vault or any point
exceeds 15% of the diameter. Example: 72-in. storage tank
would require cleaning when sediment reaches a depth of 7 in.
for more than 1/2 the tank length.
Any crack allowing material to be transported into the facility.
Cover is missing or only partially in place. Any open manhole
requires maintenance.
Locking mechanism cannot be opened by one maintenance
person with proper tools. Bolts into frame have <1/2 in. of
thread (may not apply to self-locking lids).
Cover difficult to remove by one maintenance person applying
80 Ib of lift.
Local government safety officer or maintenance person judges
that ladder is unsafe due to missing rungs, misalignment, rust,
or cracks.
See Table 3.
Vents free of debris and sediment.
All sediment and debris removed from
storage area.
All joints between tanks or pipe
sections are sealed.
Manhole is closed and secured.
Mechanism is repaired or replaced so
it functions properly.
Cover can be removed and reinstalled
by one maintenance person.
Ladder meets design standards and
allows for maintenance access.
See Table 3.
447
-------
Inflow Pipe
Tract/Easement Lines as Required
Equipment Maintenance ^j
Pad - 25 Ft Min. j: Connecting
' Outflow/Overflow Pipe/Control
Structure as Required
6 Ft H < 3 Ft
15R H >3 Ft(Typ.) I
Vegetated
Basin Floor
Pretreatment
BMP
SECTION A-A
nts
Note: Detail is schematic representation only. Actual configuration will vary depending
on specific site constraints and applicable design criteria.
Figure 11. Typical infiltration basin (3).
3. In addition, is tilling necessary to restore
infiltration capacity (regular annual tilling is
recommended)?
3.2. Infiltration trenches
Referto Table 7 for a summary of maintenance
standards for infiltration trenches.
4. Biofilters
The term "biofilter" applies to vegetated land treat-
ment systems. Biofilters can be in the form of vege-
tated swales, in which water flows at some
measurable depth or in a thin sheet across broad
surface areas, sometimes called "filter strips." Con-
structed wetlands are also sometimes put in this
category. The guidelines given below generally per-
tain to swales and filter strips, although some excep-
tions are noted. Inspection of constructed wetlands
should be conducted with reference to both these
guidelines and those given above for wet ponds.
4.1. Biofiltration swales and filter strips
Installation checks:
1. Are the dimensions and plantings as
specified in the approved plan?
2. Is the vegetation cover dense and uni-
form?
3. If the biofilter is a swale, is it parabolic or
trapezoidal in shape, with side slopes no
steeper than 3 horizontal to 1 vertical?
4. Is the biofilter placed relative to buildings
and trees in such a way that no portion will
448
-------
Table 6. Maintenance Standards for Infiltration Basins
Defect Condition When Maintenance Needed
Maintenance Results
Sediment buildup in
system
Poor facility drainage
(more than 48 hr)
Sediment trapping area
No sediment trapping
facility
Soil texture test indicates facility is not functioning as
designed.
Soil texture test indicates facility is not functioning as
designed.
Sediment and debris fill >10% of sediment mapping facility
or sump.
Stormwater enters infiltration area without pretreatment.
Sediment is removed and/or facility is
cleaned so that system works
according to design. A forebay or
presettling basin is installed to reduce
sediment transport to facility.
Additional volume added through
excavation to provide needed storage.
Soil aerated and rototilled to improve
drainage.
Sediment trapping facility or sump
cleaned of accumulated sediment.
Trapping facility (presettling basin,
detention pond, biofilter) is added
before infiltration facility.
Table 7. Maintenance Standards for Infiltration Trenches
Defect Condition When Maintenance Needed
Maintenance Results
Sediment and debris
buildup in trench
Observation well
Water percolates up
from trench
Filter fabric exposed
By visual inspection, little or no water flows through
the trench during large storms.
Observation well buried, covered, or inaccessible.
Trench water or water with dye percolating to surface.
Filter fabric is exposed or damaged.
Debris blocking infiltration trench is removed.
Gravel in infiltration trench is replaced or cleaned.
The observation well/cap is accessible to the
inspector for opening and inspection.
Gravel and filter fabric in infiltration trench is
replaced or cleaned. Trench functions according to
design standards.
Filter fabric is replaced or repaired and covered
with proper backfill material.
be shaded throughout the day and possi-
bly experience poor plant growth?
5. If the longitudinal slope is less than 2 per-
cent or if the watertable can reach the root
zone of vegetation, is water-resistant
vegetation planted to survive a standing
water condition or is an underdrain system
installed to assist drainage (note: under-
drains may not be practical with a large
filter strip)?
6. If the longitudinal slope is in the range of
4 to 6 percent, are check dams provided
approximately every 50 to 100 ft to reduce
velocity (note: check dams may not be
practical on a larger filter strip)?
7. If the slope on which a swale is installed
exceeds 6 percent, does it traverse the
slope in such a way that no reach slopes
more than 4 percent, or 6 percent with
check dams?
8. Is the lateral slope entirely uniform to
avoid any tendency for the flow to chan-
nelize?
9. Is flow introduced in such a way that en-
trance velocity is dissipated quickly, flow is
distributed uniformly, and erosion is
avoided (e.g., by using a riprap pad or
some means of level spreading)?
10. Was construction-phase runoff excluded
or was the biofilter reestablished after con-
struction, and are upslope areas stabilized
to avoid erosion into the biofilter?
11. Is a bypass in place for flows larger than
the flow rate for which the biofilter is de-
signed to provide runoff treatment, or is
the facility sufficiently large to pass at least
the 100-yr, 24-hr storm without eroding (a
bypass is preferred to maintain the treat-
ment function and prevent resuspension
of settled material)?
Maintenance checks:
1. Has a maintenance plan and schedule
been developed?
2. Refer to Table 8 for specific checks and
maintenance standards.
449
-------
Table 8. Maintenance Standards for Biofilters
Defect Conditions When Maintenance Needed
Maintenance Results
Trash and debris
Sediment buildup
Poor vegetation cover
Erosion damage to slopes
Conversion to use incompatible
with water quality control
Poor drainage
Dumping of yard wastes. Accumulation of
nondegradable materials.
Accumulation >20% of design depth.
Vegetation sparse and/or weedy. Overgrown
with woody vegetation.
Erosion >2 in. deep where cause still present
or potential exists for continued erosion.
Filled, planted appropriately, or blocked.
Water stands in swale.
Remove degradable wastes and compost. Recycle
other waste when possible.
Cleaned or flushed to match design. Vegetation
restored as necessary.
Aerate soil and plant. Remove woody growth and
replace.
Find cause and eliminate. Stabilize with appropriate
erosion controls (e.g., seeding, mat, mulch).
Discuss with nearby property owners and specify
corrections to be made.
Determine cause. If water table is high, consider
rebuilding with liner or underdrain. If slope <1%, use
underdrain.
References
1. Reinelt, I.E. 1991. Construction site erosion and sediment control
inspector training manual. Seattle, WA: Engineering Continuing
Education, University of Washington.
2. Reinelt, I.E. 1992. Inspection and maintenance of permanent
stormwater management facilities: Training manual. Seattle, WA:
Engineering Continuing Education, University of Washington.
3. Washington Department of Ecology. 1992. Stormwater manage-
ment manual for the Puget Sound Basin. Olympia, WA: Wash-
ington Department of Ecology.
450
------- |