&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 management—the
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 area—whether 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 systems—natural 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
wetlands—an incentive to the developer to use, not
destroy,  the  wetland—and it  revitalizes ditched  and
drained wetlands by providing water.

Wetlands may be  viewed as nature's kidneys—they
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 hydroperiod—the duration that water stays  at
various levels—must 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 Program—A 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  document—the
foundation of the entire planning process—with 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  unknown—to 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 1978—September 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 growth—up to 900 people per day—could
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
  document—the  foundation of  the  entire planning
  process—with  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 Plan—the
  development and adoption of a capital plan and
  budget—was 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 Plan—the 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
metals—zinc, lead, and copper—from 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 fees—where
    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  problem—it  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 years—especially
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
force—the  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-
  riculture—Soil 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 light—they 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-
                                                       ties—and the DelMarVa Peninsula generally—fall 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 *****••***«»*** JuQJD™JL/ 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  stream—nearly  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 Inventory—1988  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 years—by
Congress under Section 118(c)(3) of the Water Quality
Act of 1987 and the Great Lakes Critical Program Act of
1990—to 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 models—Storm
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 Application—Lake
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 Study—Humber 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 Quality—Impact 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 Wisconsin—Madi-
                             son (November).

                         14. U.S. Soil Conservation Service.  1986. Urban hydrology for small
                             watersheds. U.S. Department of Agriculture Tech. Release No.
                             55—revised (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 techniques—espe-
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 examination—BODS,
ammonia, and iron—based 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 applications—from simple to sophis-
ticated—depending 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 process—a 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 site—the 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
basis—the 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 pollutants—especially sediment, nitrogen, and
phosphorus—from 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 need—water supply—is 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 capacity—for control of metals.

• High exchangeable aluminum and/or iron—for 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 zones—deep  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 small—research  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 fluctuations—this 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 condition—look 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 trap—and
when necessary remove—sediments 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 pollutants—the 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 Alternatives—Physical, 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 24—Lumber and Wood Products        2,640

 2       SIC 26—Paper and Allied Products          1,023

 3       SIC 28—Chemicals and Allied Products      1,498

 4       SIC 29—Petroleum Refining and            2,245
         Related Industries

 5       SIC 32—Stone, Clay, Glass, and            4,786
         Concrete Products

 6       SIC 33—Primary Metal Industries            730

 7       SIC 10—Metal Mining                     188

 8       SIC 12—Coal Mining                      495

 9       SIC 13—Oil and Gas Extraction              457

10       SIC 14—Mining 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 5015—Used Motor Vehicle Parts        2,009

14       SIC 5093—Scrap and Waste Materials       1,688

15       Steam Electric Power Generating            162
         Facilities

16       SIC 40—Railroad Transportation            1,024

17       SIC 41—Local and Suburban Transit        13,089
         and Interurban Highway Passenger
         Transportation
         SIC 42—Motor Freight Transportation
         SIC 43—United States Postal Service

18       SIC 44—Water Transportation               368

19       SIC 3731—Ship Building and Repairing       498
         SIC 3732—Boat Building and Repairing

20       SIC 45—Air Transportation                1,581

21       SIC 5171—Petroleum Bulk Stations          131
         and Terminals

22       Domestic Wastewater Treatment Plants      1,249

23       SIC 20—Food and Kindred Products        2,608
         SIC 21—Tobacco Products

24       SIC 22—Textile Mill Products               872
         SIC 23—Apparel and Other Finished
         Products Made From Fabrics and
         Similar Materials

25       SIC 25—Furniture and Fixtures              339

26       SIC 27—Printing, Publishing, and            65
         Allied Industries

27       SIC 30—Rubber and Miscellaneous          190
         Plastic Products

28       SIC 31—Leather and Leather Products       61

29       SIC 34—Fabricated Metal Products          965
         SIC 391—Jewelry, Silverware, and
         Plated Ware

30       SIC 35—Industrial and Commercial          935
         Machinery
         SIC 37—Transportation Equipment

31       SIC 36—Electronic Components             14
         SIC 357—Computer  and Office
         Equipment
         SIC 38—Measuring,  Analyzing, and
         Control  Instruments;  Photographic and
         Optical Goods, Watches, and Clocks

32       SIC—Miscellaneous  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 I—Baseline permitting: One or more general per-
  mits will be developed initially to cover the majority
  of stormwater discharges associated with industrial
  activity.

•  Tier II—Watershed 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 III—Industry-specific permitting: Specific industry
  categories will be targeted for individual or industry-
  specific general permits.

•  Tier IV—Facility-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 found—for example, an owner unable to dispose of
a certain type  of waste—the  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 Session—Site-Specific Practices
                 —Temporary Swale
                 —Sediment Trap
                 —Urban Runoff

2:30 PM          Field Problem—Design 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 Up—Adjourn 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 farming—some 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 cover—two practices

    1.2.  Temporary cover—three practices

    1.3.  Permanent  vegetation  establishment—two
         practices

    1.4.  Stabilized construction entrance and roads—
         three practices

    1.5.  Runoff control—eight practices

2.   Sediment trapping techniques—seven 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 devices—three practices

2.   Detention facilities—eight practices

3.   Infiltration facilities—five practices

4.   Biofilters—three 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

-------