Developing Reasonable Assurance:
A Guide to Performing Model-Based
Analysis to Support Municipal Stormwater
Program Planning
Submitted to:
. . U.S. EPA Region 9
75 Hawthorne St.
W, San Francisco, CA 94602
% <£
Prepared by:
Paradigm Environmental
9320 Chesapeake Dr.
PARADIGM Suitel00
San Diego, CA 92123
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
DISCLAIMER
This is not a U.S. Environmental Protection Agency (EPA) guidance document; nor does it
represent official EPA policy. This document does not substitute for the Clean Water Act (CWA) or
associated implementing regulations. Approaches and practices identified in this document are not
binding; other approaches consistent with the CWA and associated implementing regulations may
also be available. This document is intended to be consistent with but does not modify existing EPA
policy or guidance. This guide may be revised in the future to account for developments in this fast-
changing field.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use. Although a reasonable effort has been made to assure that the information
presented is correct, some of the approaches described in this guide are relatively new and
undergoing additional testing, refinement, and review. Therefore, the author and the U.S.
Environmental Protection Agency are not responsible and assume no liability whatsoever for any
results or any use made of the results obtained from these programs and approaches, nor for any
damages or litigation that result from the use of these programs and approaches for any purpose.
ACKNOWLEDGEMENTS
We would like to acknowledge the extensive technical and editorial contributions to this guide by
David Smith from EPA Region 9. We also thank the following people for their technical reviews of
and/or technical contributions to this guide: Dr. Nicole Beck (2NDNATURE), Dr. Dino
Marshalonis (EPA Region 10), Dr. Thomas Mumley (San Francisco Regional Water Quality
Control Board), Renee Purdy (Los Angeles Regional Water Quality Control Board), Dominic
Roques (Central Coast Regional Water Quality Control Board), Mark Vorhees (EPA Region 1), and
Dr. Jing Wu (San Francisco Estuary Institute). However, we remain solely responsible for the
content of this guide.
EXECUTIVE SUMMARY
This document is designed to assist municipal stormwater program managers, watershed
stakeholders, consultants, and permitting authorities in understanding, selecting, and using model-
based approaches to support development of rigorous, comprehensive municipal stormwater
program management plans. Over the past 5 years, many municipal stormwater NPDES permits
and associated local programs have shifted their planning focus to use robust analytical modeling
tools to identify the specific stormwater management strategies and practices that will be necessary
over the long term to attain specified water quality protection requirements. This general approach,
based on what has been termed "reasonable assurance analysis" (RAA), has been developed as an
alternative to traditional municipal stormwater permitting approaches that relied upon
implementation of programmatic minimum stormwater management efforts and an iterative
approach to stormwater control development that were often not grounded in rigorous analytical
frameworks. This document is based on evaluation of several recent permits and local programs that
are implementing this new RAA approach and is intended to provide a structured approach to
selecting among alternative analytical tools and efficiently using the selected tools to support
development of long-term stormwater management programs that will comply with NPDES permit
requirements. The document is organized as follows:
February 2017
i
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
1. Purpose of a Reasonable Assurance Analysis
Why have alternative approaches based on RAA been implemented? This section briefly reviews
the evolution of municipal stormwater program planning to address new water quality management
challenges and concerns about the efficacy of traditional program approaches. The section
summarizes the role of the RAA in the stormwater management planning process and the basic
elements for completing the RAA.
2. Emerging MS4 Permit Requirements
This section discusses changes in NPDES permits to authorize use of RAA-based program planning
as an alternative to traditional permits that rely solely on qualitative, iterative program
improvements or imposition of outcome-based water quality limitations. This discussion
summarizes alternative methods used by permitting authorities to develop prescriptive permitting
requirements concerning the RAA, and is intended to educate permitting authorities and permittees
about the pros and cons of establishing more prescriptive RAA expectations through NPDES
permits and associated implementation guidance.
3. Factors to Consider in Selecting RAA Methods
This section discusses key regulatory, planning, technical/analytical, and practical considerations
decision-makers should evaluate in selecting among a range of available RAA modeling and
planning methods. The intent of this section is to assist local programs in selecting analytical
methods that are practicable and will fully address their long-term permit compliance and program
planning and implementation needs.
4. Performing a Reasonable Assurance Analysis
This section presents a general framework for performing the RAA and discusses key considerations
that should be addressed in designing and implementing the RAA approach. Key elements in the
framework include:
1. Designation of the Area Addressed by the RAA
2. Characterization of Existing Conditions and Practices
3. Determination of Stormwater Control Needs and Improvement Goals
4. Demonstration that Proposed Stormwater Controls and Management Actions Will Attain
Goals
5. Documentation of Results to Inform Implementation, Tracking, and Evaluation of Success
5. Transitioning from Planning to Implementation
This section discusses how RAA results inform development of long-term stormwater management
plans and asset management systems that guide long-term program operations and implementation
of new infrastructure projects. The section also describes how RAA results and associated long-term
plans can be used to support development of program financing plans to ensure sufficient capital and
O&M resources are available to fund the program.
Appendix A provides seven in-depth case studies that illustrate a wide range of permitting
requirements and potential RAA technical approaches and applications in use by cities and states
around the country. The case studies demonstrate that different RAA approaches may be
appropriate given the differences in "on the ground" stormwater management needs, the state of
evolution of local stormwater programs, and the regulatory and planning frameworks present in
different areas.
ii
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
While this guide will not substitute for careful, site-specific consideration of local circumstances,
capabilities, and needs for stormwater management planning and analysis, it should assist local
decision makers and regulatory authorities in better understanding the utility of RAA approaches,
the key factors to consider in selecting among different RAA approaches, the practical experiences of
several entities that have developed and implemented RAA-based stormwater management plans,
and the potential applications of RAA-based planning for long- term asset management and
financial planning.
February 2017
iii
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
Contents
Disclaimer i
Acknowledgements i
Executive Summary i
1 Purpose of a Reasonable Assurance Analysis 1
1.1 Role of a Reasonable Assurance Analysis in Watershed or Stormwater Management
Planning 2
1.2 Elements of a Reasonable Assurance Analysis 6
2 Emerging MS4 Permit Requirements 8
3 Selecting an ANalytical Method for Performing the Reasonable Assurance Analysis 11
3.1 Factors to Consider in Selecting an Analytical Method 11
3.2 Size Matters when Selecting an Analytical Method 15
4 Performing the Reasonable Assurance Analysis 17
4.1 Element 1: Designation of Area Addressed by Analysis 17
4.2 Element 2: Characterization of Existing Conditions 20
4.2. i Analysis of Monitoring Data 20
4.2,2 Modeling Existing Conditions 21
4.3 Element 3: Determination of Stormwater Improvement Goals 24
4.4 Element 4: Demonstration that Management Actions Will Result in Attainment of Goals
27
4.4, i Quantification of Benefits of Nonstructural BMPs 28
4.4.2 Quantification of Benefits of Structural BMPs 29
4.5 Element 5: Documentation of Results that Inform Implementation and Tracking 33
5 Transitioning from Planning to Implementation 35
5.1 Implementation Tracking and Adaptive Management 36
5.2 Stormwater Program Planning Frameworks that Support Successful Implementation 36
6 References 38
Figures
Figure 1-1. Steps in the Watershed Planning and Implementation Process (USEPA 2008) 4
Figure 1-2. Role of the RAA in the Watershed or Stormwater Planning Process 5
Figure 3-1. Considerations for Selecting an Analytical Method for the RAA 16
Figure 4-1. Los Penasquitos Watershed Pollutant Discharge Responsibilities (LP WQIP Responsible
Agencies 2015) 19
Figure 4-2 Example analysis sequence for selecting priority water quality conditions (LP WMA
2015). 21
Figure 4-3. Example Process for Model Calibration to Minimize Propagation of Uncertainty 23
Figure 4-4 Example Assessment of Relative Sediment Loading Los Penasquitos (LP WQIP
Responsible Agencies 2015) 24
iv
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
Figure 4-5 Example Illustration for how Exceedance Volume is Derived for Metals (ULAR WMG
2016) 26
Figure 4-6. Identifying the Limiting Pollutant for the RAA Critical Condition using EVs (ULAR
WMG 2016) 27
Figure 4-7. WRIA 9 Study Area BMP Storage in Watershed Inches for Full Stormwater
Management in 2040 (King County 2014a) 30
Figure 4-8. Example Cost-Optimization for Two Jurisdictions in the Upper Los Angeles River
EWMP (ULAR WMG 2016) 32
Figure 4-9. Example Scheduling of BMP Implementation Strategy to Meet TMDL Milestones
(ULAR WMG 2016) 34
Tables
Table 4-3. Summary of Identified Critical Conditions for Example RAAs (Appendix A) 25
February 2017
v
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
vi February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
1 PURPOSE OF A REASONABLE ASSURANCE ANALYSIS
Stormwater runoff from urban areas is often a major contributor of pollutant loadings and can result
in impairments of beneficial uses of waterbodies. Throughout the US, thousands of Total
Maximum Daily Loads (TMDLs) have been developed that include an analysis of the impairments
and an identification and quantification of pollutant sources, including assignment of wasteload
allocations to urban runoff. In some cases, the TMDLs include wasteload allocations that have been
assigned to sources regulated by National Pollutant Discharge Elimination System (NPDES)
permits that address polluted runoff from industrial areas, highways, and Municipal Separate Storm
Sewer Systems (MS4s). Since the development of those TMDLs, many NPDES permits have
incorporated wasteload allocations as Water Quality Based Effluent Limits (WQBELs) to provide
regulation and oversight of the implementation of best management practices (BMPs) to reduce
pollutant loads. WQBELs have been expressed either as numeric limits or as BMP-based
requirements in-lieu of numeric limits. However, many MS4 and other stormwater permits do not
require the development of sufficiently detailed control plans or robust analytical evidence showing
that proposed control plans will adequately reduce pollutant loadings to meet WQBELs and
associated TMDL wasteload allocations. As a result, it has been difficult to demonstrate that
implementation of stormwater permit provisions results in pollution controls that are sufficient to
protect and restore water quality in many urban areas. To address this issue, EPA (2014) has
suggested that increased understanding of BMP performance should be reflected in proper
demonstration and supporting rationale showing that implementation of BMPs will likely result in
the attainment of WQBELs and associated water quality standards and TMDL wasteload
allocations, and including milestones or other mechanisms where needed to ensure that the progress
of implementing BMPs can be tracked and permit compliance evaluated.
A new generation of MS4 permits throughout the U.S. includes specific requirements for a
quantitative analysis to provide reasonable assurance that pollutant load reductions or reduced
stormwater impacts will be achieved through the implementation of detailed stormwater or
watershed management programs. Often called a Reasonable Assurance Analysis (RAA), the
process typically employs the use of computer modeling or other quantitative techniques to
demonstrate that a combination of specified BMPs or other control strategies will likely reduce
pollutant loads or other stormwater impacts (e.g., peak flows) as necessary to result in achievement
of WQBELs or TMDL wasteload allocations expressed as WQBELs within compliance schedules
established by NPDES permits.
Although relatively new to MS4 permits, the concept of the RAA has been integral to the watershed
planning process for several years. In the early 1990s, during the early stages of development of
TMDLs that emphasized stormwater as a pollutant source, models available at the time were
thoroughly evaluated regarding their ability to simulate urban runoff (USEPA 1991) and address the
quantitative needs of a TMDL source analysis and calculation of allowable loads (USEPA 1992).
During this period, EPA and many states also recognized that models can assist in the watershed
planning process, including targeting watersheds for management, developing goals and objectives,
defining solutions, developing plans for management implementation, simulating storage and
treatment effects of alternative management options, providing input to cost-benefit analyses, and
tracking progress toward achieving goals (USEPA 1991 and 1997). As the process for developing
watershed plans continued to evolve, including the collection of many lessons learned on effective
plans, EPA sought to develop guidance to states, territories, tribes, local governments, watershed
organizations, and the public regarding technical tools and sources of information for developing
watershed-based plans. In 2008, EPA released the Handbook for Developing Watershed Plans to
1
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
Restore and Protect Our Waters (Watershed Plan Handbook) to provide information on developing
and implementing watershed management plans to restore or protect water quality. The Watershed
Plan Handbook provides guidance on the selection and application of quantitative approaches to
characterize point and nonpoint sources and pollutant loads and predict load reductions associated
with planned management activities. In order to quantify anticipated pollutant reductions resulting
from management strategies, the Watershed Plan Handbook provides an overview of a range of
quantitative approaches from literature values and spreadsheet tools to more sophisticated modeling
approaches (USEPA 2008).
With the incorporation of new requirements in MS4 permits that include development of watershed
or stormwater management programs that incorporate RAAs, lessons learned from the watershed
planning process and early RAA implementation efforts can inform and assist communities
beginning the process. The purpose of this document is to assist with the selection and application of
technical frameworks for performing RAAs that are best suited to meet MS4 permit requirements
and watershed and stormwater planning needs. There is no "one-size-fits-all" approach for
performing RAAs, and opportunities always exist for tailoring approaches to meet local needs,
furthering the research and development of currently used tools or models, or developing new
methods. Therefore, this guide is not meant to define the full range of possible approaches to be used
for RAAs, but rather assist in the selection of an approach that best fits local needs while fulfilling
expectations of the MS4 permit, permitting authorities, and watershed stakeholders.
1.1 Role of a Reasonable Assurance Analysis in Watershed or
Stormwater Management Planning
The required components of a watershed or stormwater management plan vary depending on the
MS4 permit or, where available, local regulatory guidance. Often, the planning approach includes
components that are similar to traditional watershed or TMDL implementation planning processes.
For example, the Watershed Plan Handbook outlines multiple steps in the development and
implementation of a watershed management plan (Figure 1-1). For the Watershed Plan Handbook
steps outlined in Figure 1-1, "Characterization and Analysis Tools" are similar to approaches used
to perform an RAA, as they support the characterization of the watershed (Step 1) and the setting of
goals and identification of management solutions (Step 2) (USEPA 2008).
In California, similar guidance for TMDL implementation planning was developed by the California
State Water Resources Control Board (SWRCB). In 2005, the SWRCB released A Process for
Addressing Impaired Waters in California, which outlines the following steps for designing a TMDL
implementation plan (SWRCB 2005):
1. Identify Current Activities: Considers management actions (e.g., NPDES requirements)
that are already initiated in the watershed, and serves to establish the starting point for
identifying further actions required to improve water quality.
2. Identify Common Interests and Overlapping Objectives: Assesses other multiple benefits
(e.g., flood protection, water supply) that can be considered in the planning process.
3. Engage Stakeholders: Involves stakeholders in the planning process and selection of
management activities.
4. Identify Opportunities for Management Practices: Identifies viable opportunities for
management practices and considers source type, impairment type, and load reduction
required. Evaluates suitable locations for management practices.
5. Consider Alternatives and Cost: Analyzes multiple implementation scenarios and
associated costs for selection of the most cost-effective plan. This step can include an analysis
2
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
of the effectiveness of each scenario at meeting water quality standards and associated load
reductions, as well as the cost of each scenario, for identifying the cost-effective
implementation plan.
Steps 4-5 above are similar to an RAA in that a quantitative analysis of load reductions or water
quality improvement is performed to evaluate the expected effectiveness of alternative management
actions. The SWRCB also notes that "these analyses can be used to link the proposed management
actions with the desired load reductions, and determine whether the proposed management actions
will be sufficient to meet water quality objectives (e.g., through TMDL allocations)."
The above examples are for traditional watershed and TMDL implementation planning process and
the role of RAAs. However, watershed or stormwater management plans addressing requirements of
a MS4 permit should always consider components specifically described in that permit or associated
regulatory guidance. Although typically similar to traditional watershed or TMDL implementation
planning approaches, many MS4 permits have specific requirements for various planning activities
including TMDL implementation.
In summary, the RAA is only one component of an overall process for effective watershed or
stormwater management planning. The RAA can inform the entire planning process in terms of
evaluating combinations of potential management actions and their effectiveness in attaining
necessary pollutant reductions to meet TMDL wasteload allocations, WQBELs, or other water
quality targets. Figure 1-2 provides an overview of the RAA (blue) and its interaction with typical
processes included in the development of watershed or stormwater plans (gray).
February 2017
3
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
1. Build Partnerships
Identify key stakeholders
Identify issues of concern
Set preliminary goals
Develop indicators
Conduct public outreach
Characterize the Watershed
Gather existing data and create a watershed inventory
Identify data gaps and collect additional data if needed
Analyze data
Identify causes and sources of pollution that need to be controlled
Estimate pollutant loads
3. Finalize Goals and Identify Solutions
Set overall goals and management objectives
Develop indicators/targets
Determine load reductions needed
Identify critical areas
Develop management measures to achieve goals
4. Design an Implementation Program
Develop implementation schedule
Develop interim mdestones to track implementation of management measures
Develop criteria to measure progress toward meeting watershed goals
Develop monitonng component
Develop information/education component
Develop evaluation process
Identify technical and financial assistance needed to implement plan
Assign responsibility for reviewing and revising the plan
Characterization and
Analysis Tools
Jr
GIS
)r
Statistical packages
Monitoring
Load calculations
J*
Model selection tools
jr
Models
Databases
(environmental and
social tools)
5. Implement Watershed Plan
Implement management strategies
Conduct monitonng
Conduct information/education activities
6. Measure Progress and Make Adjustments
Review and evaluate information
Share results
Prepare annual work plans
Report back to stakeholders and others
Make adjustments to program
Watershed Plan
Document
Figure 1-1. Steps in the Watershed Planning and Implementation Process (USEPA2008).
4
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
input from Reasonable
Stormwater/Watershed _ ...
Planning Process Assurance Analysis
Assess Permitting
Responsibility
MS4 permit
Non-permitted areas
Areas addressed by
other NPDES permits
Analyze Monitoring Data
Stormwater and
receiving water
Assess when and where
numeric targets are
exceeded
~ 2
Designate Area for Analysis
Watershed boundaries
Jurisdictional boundaries
MS4 permitted area
Characterize Existin
Conditions
Stormwater flows and
pollutants conc./loads
Incorporate existing
mgt. practices
I d e ntify_Nu merle Ta [gets
TMDL wasteload
allocations
WQBELs
Water Quality Targets
Opportunities
Nonstructural or source
control measures
Structural BMPs {e.g.,
green infrastructure)
* 4
Implementation Support
Determine Stormwater
Improvement Goals
Compare existing
conditions with numeric
targets
* Reduce pollutant
loads/conc. or flows
Demonstrate M
Actions will Attain Goals
Models/analytical tools
Pollutant/flow
reduction over time
Document Results
Demonstrate reasonable
assurance
Inform implementation
Support tracking
Output to
Stormwater/Watershed
Planning Process
Inform Met. Actions
Select effective mgt.
actions
Develop conceptual
design assumptions
Stakeholder Engagement
¦ Provide assurance
that management
actions will result in
attainment of goals
Complete Watershed
or Stormwater
Management Plan
Additional Planning Efforts
Stormwater program
enhancements
Capital improvement planning or
asset mgt.
Funding investigations
Adaptive Management
Tracking of implementation over time
Assessment of progress towards
attainment of goals
Modifications to plan to take
advantage of lessons learned
Figure 1-2. Role of the RAA in the Watershed or Stormwater Planning Process.
February 2017
5
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
1,2 Elements of a Reasonable Assurance Analysis
What constitutes reasonable assurance? From a regulatory perspective, reasonable assurance can be
interpreted as the demonstration that the implementation of a watershed or stormwater management
plan will, in combination with operation of existing system assets and programs, result in sufficient
pollutant reductions or reduced stormwater impacts over time to meet TMDL wasteload allocations,
WQBELs, or other targets specified in the MS4 permit or identified in the plan (USEPA 2014).
From the perspective of a stakeholder in the watershed who is focused on the improvement of water
quality or restoration of a beneficial use of a waterbody, reasonable assurance is often interpreted as
a demonstration, as well as commitment, that specific management practices are identified with
sufficient detail, and implemented on a schedule to ensure that necessary improvements in the
receiving waters will occur. From the perspective of the MS4 permittee, reasonable assurance can be
interpreted as a detailed analysis of the TMDL wasteload allocations and associated permit
limitations themselves. RAA may also assist in evaluating the financial resources needed to meet
pollutant reductions based on schedules identified in the permit, and in preparing associated capital
improvement plans. Proper attention to each viewpoint is especially important in the selection and
application of an RAA to avoid many pitfalls that can jeopardize regulatory or stakeholder
acceptance of a plan or its usefulness to guide future implementation efforts. To help avoid these
pitfalls, the following basic elements of an RAA have been identified (corresponding to numbered
components of the RAA depicted in Figure 1-2).
Element 1: Designation \ dressed by Analysis
As the RAA associated with a stormwater management plan is developed in the context of the MS4
permit, the area where it is applied is typically (but not always) specific to urban areas within one or
more municipal jurisdictions addressed by that permit. This may exclude areas that are regulated by
other NPDES permits, including those issued to industrial dischargers, transportation agencies, or
other MS4 permits. However, RAAs that address all the areas within the watershed are more likely
to successfully target controls needed by regulated MS4 permittees as well as other entities or sources
not addressed by the MS4 permit. Section 4.1 provides a discussion of the considerations for
designating urban areas addressed by a watershed or stormwater management plan and associated
RAA.
Element 2: Characterization of Existir v' nditions
Critical to the RAA is careful characterization of stormwater pollutant loads or flows under existing
conditions. This understanding serves as the foundation of the RAA and identifies the starting point
for planning management actions. Where a TMDL is established, characterization of existing
conditions may be documented within the TMDL and can be cited by the RAA. However, some
TMDLs do not include a detailed characterization of existing conditions and specification of load
reductions needed to meet wasteload allocations. For other TMDLs that include a characterization
of existing conditions, the RAA may require revisiting of calculation methods to account for factors
or assumptions not addressed by the TMDL. The characterization of existing conditions should also
consider all stormwater management practices and system assets currently in place or implemented
at a specified point in time (e.g., date of approval of MS4 permit or TMDL). Section 4.2 provides a
discussion of methods used to characterize existing conditions and factors to consider in selecting
the appropriate approach for a RAA.
6
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
Element 3: Determination of Stormwater Improvement Goals
Based on the existing conditions characterized above, and in combination with water quality targets
established by TMDLs or other assessments, the MS4 permit, and/or the watershed or stormwater
management plan, goals can be determined for addressing stormwater-caused impairments. These
goals can be expressed in various forms, but usually include specified reductions in stormwater
pollutant loads or concentrations, volumes, or peak flows. Section 4.3 outlines typical methods used
to establish stormwater improvement goals for RAAs.
Element 4: Evaluation of Management Alternatives and Demonstration
that Management Actions will Result in Attainment of Goals
As part of the watershed or stormwater management planning process, several opportunities for
management actions can be identified. These management actions may include a combination of
programmatic activities (e.g., street sweeping, inspection and enforcement procedures, or source
controls), low impact development (LID) practices incorporated within new development, re-
development, or retrofit of urban areas, or municipal capital improvement projects that provide
stormwater capture or treatment (e.g., regional treatment or infiltration facility, green streets). The
goal of the RAA is not to identify these opportunities, as that is part of the overall watershed or
stormwater management planning process (see Section 4.4). The RAA helps identify combinations
of actions and practices that together will achieve desired water quality results. The RAA provides a
quantitative evaluation of the stormwater pollutant loads or concentrations, volumes, or peak flows
reduced via alternative management scenarios to ensure that the selected management scenario will
result in attainment of stormwater improvement goals established in Element 3. RAA methods
provide a critical framework for comparing stormwater management alternatives, including different
mixes of structural and non-structural practices and different options for distributing stormwater
management practices and facilities throughout the planning areas. By applying the selected RAA
framework to support the broader planning process, program managers can rigorously compare
implementation alternatives and their ability to ensure attainment of stormwater management goals
and requirements.
The RAA is not a one-time concept. Stormwater program managers should anticipate that the
modeling approach selected to perform the RAA will need to be used in the future in coordination
with other program tracking and assessment tools to support adjustment of program implementation
efforts over time. Stormwater program implementation opportunities and constraints change over
time, and programs can expect to learn critical information from early implementation efforts about
BMP performance and effectiveness of non-structural approaches. RAA models can and should be
adjusted over time to take advantage of such new information. Moreover, as additional data and
information are collected over time, it is important to update RAA modeling tools themselves to
reduce modeling uncertainty and improve model performance. For example, program managers
can benefit from uncertainty analysis conducted as part of an initial RAA modeling effort to identify
data needed to reduce important sources of model uncertainty over time. The key point is that all
RAA efforts and associated stormwater programs entail iterative adjustment over time. This means
that stormwater managers should assume an RAA is not a one-time analysis but is instead a critical
ongoing element supporting long term implementation and adjustment of stormwater program
actions. However, a commitment to iterative adjustment does not by itself provide reasonable
assurance. Program managers will need to ensure that the RAA method selected at the outset
provides sufficient analytical power to support detailed stormwater program planning and ensure
that the program plan is likely sufficient to meet regulatory requirements, including TMDL
allocations. A commitment to long term, iterative adjustment of the stormwater program andthe
February 2017
7
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
RAA methods used to support planning and evaluation will be key to long-term program
effectiveness.
Element 5: Documentation of Results that Inform Implementation and
Tracking
Documentation of RAA results is critical to ensuring their success in fulfilling regulatory
requirements to demonstrate that the selected stormwater management plan will result in attainment
of TMDL, water quality, and other management objectives. The RAA analytical framework also
can be used to inform and track future implementation efforts, evaluate effectiveness of implemented
projects and practices, and measure progress in attaining regulatory goals. This documentation can
serve various purposes (see Section 4.5), including:
Providing reasonable assurance to stakeholders and regulators that the plan will lead to
effective implementation;
Providing information that can support next steps for stormwater program enhancements,
capital improvement planning, and investigation of funding options; and
Highlighting the quantitative results that support the adaptive management process, tracking
of implementation over time, and/or assessment of progress towards attainment of
stormwater improvement goals and requirements.
2 EMERGING M34 PERMIT REQUIREMENTS
In recent years, MS4 permits have been issued that include methods for incorporating TMDLs and
associated wasteload allocations, as well as requirements for integrated watershed or stormwater
management planning and RAAs, to demonstrate how TMDL allocations and other water quality
based requirements will be met. Incorporation of TMDLs into MS4 permits presents numerous
challenges including:
Addressing a large number of discharge points;
Accounting for highly variable flows;
Assessing a wide array of pollutants-of-concern;
Evaluating the co-mingling among multiple municipal jurisdictions and areas addressed by
other stormwater permits;
Considering multiple types of available BMPs;
Incorporating limited data regarding BMP effectiveness;
Considering limited sources of capital and O&M funding dedicated to stormwater
management.
MS4 permits vary in terms of how they express water quality-based requirements. Some permits
incorporate effluent or receiving water quality or loading requirements, while other permits
incorporate more BMP-based approaches in lieu of water quality-based requirements. In either
situation, RAA methods are critical to evaluating and establishing needed connections between
selected stormwater management practices and intended water quality goals. Many recent permits
offer alternative compliance pathways among which permittees can select.
To address the complexities of municipal stormwater management and the needs for regulatory
oversight and compliance determination, many recent MS4 permits have included "BMP-based"
permit requirements that require development of a watershed or stormwater management plan and a
demonstration of the management actions needed over time to meet TMDL wasteload allocations
8
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
or WQBELs. In requiring a robust analysis to demonstrate the ability of the stormwater
management plan to meet water quality requirements within specified timeframes, this approach
differs from many traditional MS4 permits that only require iterative implementation and
adjustment of stormwater plans with the general goal of meeting water quality goals in the future.
There are many potential advantages to BMP-based permit requirements supported by RAAs (as
opposed to numeric WQBELs) including:
Requirements to implement BMPs may facilitate projects being placed "in the ground" in
the near-term;
A schedule of specific BMPs (with type, location, installation schedule, and expected
performance) could be easier to monitor and enforce than end-of-pipe or receiving water
limitations;
Requirements often viewed by municipal agencies as more tangible and controllable than
water quality outcome-based requirements; and
Requirements to develop a well-justified list and schedule of BMPs can increase a
municipality's ability to obtain funding for stormwater quality projects.
The analysis may assist in defining "maximum extent practicable" for purposes of iterative
program development.
Perhaps the biggest challenge with incorporating BMP-based permit requirements is development of
a comprehensive watershed or stormwater management plan that is quantitative, reliable, and
enforceable, yet flexible. A 2014 memo from EPA Office of Wastewater Management (OWM) and
Office of Wetlands, Oceans and Watersheds (OWOW) directors to regional Watershed Division
directors describes EPA's expectations for BMP-based effluent limits, including the following
excerpt:
"The permitting authority's decision as to how to express the WQBEL(s), either as
numeric effluent limitations or as BMPs, with clear, specific, and measurable
elements, should be based on an analysis of the specific facts and circumstances
surrounding the permit, and/or the underlying wasteload allocation, including the
nature of the stormwater discharge, available data, modeling results, and other
relevant information. As discussed in the 2002 memorandum, the permit's
administrative record needs to provide an adequate demonstration that, where a
BMP-based approach to permit limitations is selected, the BMPs required by the
permit will be sufficient to implement applicable wasteload allocations. Permits
should also include milestones or other mechanisms where needed to ensure that the
progress of implementing BMPs can be tracked. Improved knowledge of BMP
effectiveness gained since 2002 should be reflected in the demonstration and
supporting rationale that implementation of the BMPs will attain water quality
standards and be consistent with wasteload allocations."
By definition, the RAA provides the necessary analysis to support BMP-based compliance
mechanisms, and the demonstration that BMPs identified within a stormwater management plan
will be sufficient to result in attainment of TMDL wasteload allocations. Approaches to BMP-based
compliance and TMDL implementation have been incorporated within MS4 permits in various
forms. EPA Region 9 consulted with other EPA regions across the U.S. to identify MS4 permits that
include requirements for RAAs in order to highlight a range of geographic areas and approaches.
Appendix A summarizes notable MS4 permits that include approaches to stormwater management
planning and the incorporation of RAAs. For most of these MS4 permits, a case study is provided
that outlines methods used by municipalities to perform RAAs and be responsive to the permit
requirements. Each of these case studies include discussions regarding how the basic elements of an
February 2017
9
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
RAA (Section 1.2) were addressed. Appendix A includes the following example MS4 permits, which
are referenced throughout this document.
1. Los Angeles County Phase IMS4 Permit (A. 1)
2. Washington Phase IMS4 Permit (A.2)
3. San Diego Region Phase IMS4 Permit (A. 3)
4. Central Coast California Phase I and Phase IIMS4 Permit (A.4)
5. San Francisco Bay Area Regional MS4 Permit (A.5)
6. Virginia Phase IMS4 Permits Addressing the Chesapeake Bay TMDL (A. 6)
7. Massachusetts General Phase IIMS4 Permits (A. 7)
For each of the example MS4 permits, computer modeling or other quantitative approaches served
as a critical component of RAAs. However, the role of the RAA differs depending on the
prescriptiveness of the permit and/or the degree to which previous modeling was performed to
inform permit requirements or separate guidelines developed by the regulators. For example,
previous modeling efforts and analyses performed by EPA and the Virginia Department of
Environmental Quality (VDEQ) to address the Chesapeake Bay TMDL resulted in model output
and quantitative approaches that were included within guidelines and reduced the need for further
modeling. EPA Region 1 used a similar approach to established quantitative procedures for RAAs in
the Massachusetts General Phase IIMS4 Permits that are built upon previous modeling efforts, with
the release of additional tools to support municipal efforts to follow these procedures and reduce the
need for further modeling. For other California MS4 permits (Los Angeles and San Diego),
modeling was previously used to support development of TMDLs, however, the RAAs typically
require additional modeling or tools to provide assurance that these TMDLs or other water quality
goals are met. For King County, no previous model was available, so the RAA required the
development of new modeling approaches to address requirements of the Washington Phase IMS4
Permit. In summary, there are two main categories of approaches to use models to support RAAs,
including:
Deterministic Approach: The MS4 permit may include requirements for RAAs and/or
additional guidance, but the research and development associated with the modeling of
pollutant loads and BMP performance is performed during the RAA following permit
issuance. Some permits include associated guidance in terms of performance criteria or
acceptable models to be used for the RAA. Examples of the Deterministic Approach include
the Los Angeles County Phase IMS4 Permit, San Diego Region Phase IMS4 Permit,
Washington Phase IMS4 Permit, San Francisco Bay Area Regional MS4 Permit and
Central Coast California Phase I and Phase IIMS4 Permit (Appendix A).
Prescriptive Approach: The MS4 permit provides detailed procedures for performing the
RAA that are built upon output from past modeling and analysis efforts, typically used to
support development and implementation of TMDLs. In this way, the research and
development of modeling approaches and assumptions for pollutant loading and BMP
performance are performed up front, and provide a recipe for municipalities to follow using
simple processes to perform an RAA. Examples include Virginia Phase IMS4 Permits
Addressing the Chesapeake Bay TMDL, the Massachusetts General Phase IIMS4 Permits,
and the Central Coast California Phase I and Phase IIMS4 Permit (Appendix A).
Regardless of the approach used, at some point either before or after the permit requirements are in
place, modeling plays an integral role in the RAA process. The Prescriptive Approach to an RAA is
defined by specific permit language or associated guidance provided by the permitting agency. As
such, little or no other guidance is needed to inform a municipality on that approach. As
10
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Deterministic Approaches typically lack specific guidance in terms of selecting and performing
modeling and presenting results of the RAA, the following Sections 3 and 4 provide a preliminary
guide for evaluating and selecting among alternative analytical options. However, guidance
provided within MS4 permits or associated documentation will always govern local regulator
expectations of Deterministic Approaches to RAAs. For Example, the LARWQCB (2014) released
detailed guidance for performing Deterministic Approaches for RAAs to address the Los Angeles
County MS4 Permit (Appendix A.l), which includes local considerations and references specific
approaches to model selection and application for that region.
3 SELECTING AN ANALYTICAL METHOD FOR
PERFORMING THE REASONABLE ASSURANCE
ANALYSIS
Several key factors should be considered in selecting the most applicable and appropriate analytical
framework for performing a RAA that best meets the local needs for a stormwater management
plan. For those unfamiliar with models or other technical approaches, the number of options and
considerations can be challenging to navigate. This section provides a general outline of factors to
consider in the selection of an analytical method for performing the RAA.
3.1 Factors to Consider in Selecting an Analytical Method
When selecting an appropriate analytical method for the RAA, the following factors should be
addressed to ensure that the method meets the needs of the municipality. These factors include (1)
regulatory and planning needs, (2) analytical capabilities, and (3) practical considerations.
The analytical method must be able to address specific regulatory requirements
defined by the MS4 permit while meeting the stormwater and financial planning
needs of the municipality. This category is probably the most important to the
municipality, as it addresses the basic issues that are key to the overall purpose
of the RAA. Factors that may be key to assessing Regulatory/Planning Needs
can include:
Permit Requirements - The MS4 permit must be reviewed to identify specific basic
requirements of the analytical method. For instance, the permit may require that the RAA
address specific pollutants and applicable TMDLs. The permit may also define specific critical
wet or dry conditions that must be considered, or guide how critical hydrologic conditions that
informed TMDL development should be addressed during permit implementation. Some
permits may also include specific reporting requirements based on the output from the RAA,
including degree to which individual BMPs or combinations of BMPs contribute to pollutant
reductions. Permits may also vary in the degree of assurance and accuracy required to obtain
plan approval by the permitting authority. These permit requirements can guide the
municipality in evaluating specific analytical methods designed to address them. If the permit is
unclear regarding specific expectations of the RAA, the municipality should consult with the
regulating authority to obtain appropriate guidance and ensure that an effective analytical
method is selected.
February 2017
11
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
Planning Process Requirements - As discussed in Section 1.1 and depicted in Figure 1-2, the
RAA is a component of the overall stormwater management planning process. As the overall
planning process is strategized, the role of the RAA should be specifically considered by the
municipality in terms of how it should inform the planning process. For example, as the RAA
can provide the ability to quantify pollutant reductions and other multiple benefits of individual
BMPs, the municipality may choose to capitalize on this capability to compare the benefits of
alternative BMP implementation strategies. Also, depending on the analytical method, various
types of information will need to be compiled early in the planning process to provide sufficient
assumptions or input to a model. Data and information limitations may practically limit the
range of RAA methods that can be pursued in a specific situation. To support future
implementation of the plan, a municipality may desire to have the RAA produce specific
information to guide future planning efforts. For instance, a municipality may select an
analytical method with capability to conceptually size or select specific structural BMPs, with
output from the RAA that can inform future design or cost estimation efforts (linked to the
below factors). A municipality may also want to keep the analytical method simple, and focus
on development of a method that can support the tracking of BMP implementation and
associated pollutant reductions over time. To ensure that the analytical method meets the overall
planning requirements of a municipality, the approach to stormwater management planning and
implementation should be strategized prior to selection of the approach to the RAA. The
conceptual plan for the planning process can then set the expectations of the RAA method to be
sure that the proper tools are available to effectively support the stormwater managementplan.
BMP Types, Siting Opportunities, and Constraints - Prior to selecting an RAA method, a
municipality should consider the types of BMP opportunities that will be assessed. Analytical
methods for performing RAAs can differ considerably in terms of capability to represent
different BMPs and management actions.
Financial Planning Requirements - Looking beyond the stormwater management plan, a
municipality may choose to consider an analytical approach that supports cost estimation,
conceptual BMP design, or other capabilities that can inform financial planning. For example,
some analytical approaches consider sizes of structural BMPs and other attributes (e.g.,
hydraulic designs, underdrains) to estimate pollutant loads or stormwater captured. This same
information can be later used for estimating capital and O&M costs. Other RAA approaches
provide added capability to incorporate mathematical cost functions, which can be used directly
to estimate capital costs, O&M costs, etc. Selection of a simple RAA method may be sufficient
to meet regulatory requirements or needs of the stormwater management plan, but additional
investment may be needed later to develop necessary tools to support capital improvement
planning, asset management, or other strategic planning efforts needed to secure funding for
implementation. In summary, certain analytical approaches provide added capability to support
implementation planning, and there can be cost-savings associated with selecting an RAA
method that can support these tasks at a later time.
Multi-Purpose Planning Needs - For a stormwater management plan to be successful, it is often
beneficial if the plan considers multiple benefits of related project opportunities. For example, a
structural BMP project may provide infiltration and potential recharge of groundwater supplies,
reduced flooding of urban areas, or reduced hydromodifcation or erosion of downstream
creek/river channels. Different RAA methods vary in their capacities to assist evaluation of
multi-purpose project approaches. Consideration of multiple purposes and benefits can widen
funding opportunities and maximize returns on capital investments in stormwater projects.
12
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
The analytical method must provide sufficient capability and precision to: (1)
accurately characterize existing stormwater and pollutant loading conditions,
and (2) demonstrate reasonable assurance that planned BMPs will meet their
intended purpose at reducing stormwater and pollutant loads to meet
applicable targets. Some MS4 permits may include specific guidance on
expectations of the analytical capabilities of an RAA method. The following
factors can be considered in assessing Analytical Capabilities:
Spatial/Temporal Precision - The analytical method should provide the capability to assess
spatial and temporal variability with adequate precision that is consistent with the planning or
regulatory goals. For example, if the goal is to meet an allowable annual pollutant load, then the
approach should provide sufficient capability to estimate pollutant loads for that annual period.
If the goal is to meet a pollutant load or concentration target for a specific size storm, then the
approach should provide capability to provide hourly or daily estimates of storm volumes,
concentrations, or pollutant loads. The management plan may require assessment of pollutant
loads at multiple locations throughout the area of analysis, which may vary depending on site-
specific pollutant sources, rainfall variability, or other land characteristics that influence
pollutant transport.
Hydrologic Precision - Hydrology is the driving force for pollutant transport and forms the
foundation of the stormwater management plan. Analytical approaches often vary considerably
regarding representation of hydrologic processes, which typically include methods for estimating
rainfall/runoff, infiltration, evapotranspiration, subsurface flow, among others. More
sophisticated methods typically provide greater precision in representing multiple hydrologic
processes. However, this does not mean that the most sophisticated methods are needed to meet
the requirements of the RAA or stormwater planning goals. The hydrologic precision of a
selected analytical approach should be in agreement with planning and regulatory goals, and
sufficient to represent hydrology to a degree of accuracy that will provide reasonable assurance
to the permitting authority, while ensuring confidence in terms of setting management goals that
may have significant cost implications. Special attention should be placed on selection of an
analytical approach that provides documentation of hydrologic precision or procedures to apply
the method and measure precision. For example, models can be used to represent hydrology
and predict stormflows over extended periods, which can be compared with measured flows to
establish statistical representation of the precision of the model.
Pollutant Representation and Assessment - The RAA analytical approach should have
sufficient capability to represent the pollutants to be addressed by the stormwater management
plan, including key processes related to their sources and transport. This may require
assumptions for pollutant loads that are associated with land use, imperviousness, or other
characteristics in the drainage area to provided assessment of the spatial variability of pollutant
loads and inform selection and placement of BMPs. The method should also consider adequate
capability to represent processes that drive or influence pollutant transport, including relevant
hydrologic and/or sediment transport processes. Proper representation of hydrology and
pollutant transport is essential to understanding the pollutant loads delivered to specific locations
for BMPs, or discharged from the storm drain system to receiving waters where applicable water
quality targets apply.
BMP Process Precision - The analytical approach should provide BMP process simulation that
is consistent with the capabilities for Spatial/Temporal Precision, Hydrologic Precision, and
February 2017
13
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Pollutant Representation and Assessment discussed above. For example, if the Hydrologic
Precision results in selection of a method that estimates peak hourly stormflows, then the BMP
Process Precision should also provide the ability to estimate reductions of pollutant loads during
that same period. Similar to hydrology discussed above, analytical approaches vary significantly
in the degree to which BMP processes are represented. Some methods rely on literature
assumptions for BMP effectiveness that can be used for assigning percent reductions or effluent
concentrations for pollutants. Other methods attempt to represent the processes that contribute
to these reductions, including BMP processes associated with infiltration, filtration, settling,
evaporation, and others. Typically, the more processes represented, the more modeling
assumptions will need to be documented based on local data or literature. However,
representation of these multiple processes can assist in determining details for BMP sizing and
design that can provide greater precision from a planning perspective. As a result, selection of
the appropriate analytical approach should consider the Regulatory/Planning Needs, and ensure
that the selected method will provide adequate BMP Process Precision to support planning
decisions.
Multi-Benefit Assessment and Optimization - A stormwater management plan may identify
multiple opportunities for BMPs, with different combinations of BMP sizing and placement
representing alternative implementation scenarios. If desired, an analytical approach can have
the capability to provide relative comparison of these implementation scenarios and aid in the
selection of an ideal BMP implementation plan that meets multiple planning goals. This relative
comparison can also consider cost-effectiveness and other multi-benefit planning objectives (e.g.,
groundwater recharge, flood protection, habitat). Some analytical approaches provide
comparison of multiple implementation scenarios and direct comparison of pollutant reductions
and costs to support optimization of the plan and selection of the most cost-effective
combination of BMPs to meet targets. Certain RAA methods can represent characteristics that
provide a measure for a subset of benefits, ranging from the quantity of water infiltrated into the
ground to aquifers of interest, to the reduction of peak flows that impact downstream areas and
associated flood risk. Some approaches can also consider climate change scenarios and reduced
impacts resulting from stormwater projects. Often, separate approaches outside of the RAA are
used to estimate other multiple benefits, with varying metrics used to quantify these benefits.
There are multiple Practical Considerations that can influence the selection
of the analytical rigor of a RAA method. These considerations are based on
the perspective of the municipality and their ability to invest in the RAA
approach, including such factors as data needs, costs for performing the
RAA, and ability of municipal staff to apply the approach or need to rely on
contractor support. The following are examples of Practical Considerations
for selecting an RAA method:
Data Needs/Availability - Appropriate selection of an analytical approach is highly dependent
on the available data needed to develop that approach. All analytical approaches rely heavily on
available GIS datasets to represent land characteristics. At minimum, these approaches require
information on land use, impervious area, and topography. More sophisticated approaches may
require additional information related to soils, vegetative cover, storm drain networks, among
other datasets. Rainfall measurement data are typically used as the primary input for hydrologic
representation. Other flow measurement data can be used to assess the Hydrologic Precision of
the methodology. Although many of these datasets used to develop a RAA method are typically
available from federal (e.g., USGS, NOAA, USD A) or state agencies, some datasets must rely
on more site-specific information (e.g., water quality monitoring data). Typically, the more
14
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
sophisticated the analytical method, the more data is required to support its application.
However, in many cases these sophisticated approaches can be applied based on simplified
assumptions, and once more data are collected in the future, these assumptions can be further
investigated and tailored. As a result, selection of an analytical approach should consider the
data needed for application, the availability of these data, and future considerations for data
collection/development.
Costs for Performing the RAA - Generally, the more complex the technical approach for the
RAA, the more costly it is to perform. However, there are trade-offs for selecting an approach
that should be considered during the selection of a RAA method. For instance, although a
simpler approach may be sufficient to meet MS4 permit requirements or a permitting authority's
expectations for the RAA at the outset, such methods may provide insufficient information to
guide later implementation and financial planning. As a result, a municipality that invests in a
simpler approach may find themselves needing to invest later in development of tools and
information to guide cost-effective BMP implementation and financial planning. Also, some
more rigorous approaches can be applied in a simplistic way that results in lower costs for the
RAA. Later, these analytical frameworks can be revised or expanded to provide increased
functionality to support implementation planning, preventing the need to start over and waste
previous investments in the tools. As previously discussed, there is a need for iterative
adjustment of any RAA approach over time to take advantage of lessons learned, new data, and
adaptive management. In summary, when selecting an analytical method, a municipality should
consider not only the initial costs for performing the RAA, but also future costs for continued
maintenance and updates of the system over time to support ongoing planning needs as well as
the adaptability of the system. This is especially true for large municipalities with complex
systems and multiple BMP projects needed. However, for smaller municipalities with fewer
BMP projects needed, simpler approaches may be sufficient for both the RAA and future
implementation efforts.
User Skill/Training Needs - With RAA methods ranging from simple to complex, there is a
corresponding range of skills needed to develop and use these systems. For example, if a simpler
RAA approach (e.g., one based on a spreadsheet analysis) is selected, a wider range of user skills
can be found to conduct the analyses, as long as there is a good understanding of the
assumptions used in the approach. A more intensive analysis that relies on models would require
modeling background to apply and operate. Often, municipal staff do not possess modeling
expertise and require training to be able to develop or run such systems. Many of these
municipalities rely on contractors to support model development efforts, although training is
often provided to either provide municipal staff a basic understanding of the model processes, or
to prepare staff for operation of the models. In summary, consideration must be given to whether
in-house user skills are adequate for the selected approach or other options for applying the
approach are available.
3.2 Size Matters when Selecting an Analytical Method
Typically, the size of a municipality influences the weight that each factor above has in selecting an
analytical method for performing the RAA. In this context, size corresponds to the amount of urban
area within the area of analysis. More urban area is also often associated with more impacts to
receiving waters, resulting in increasing number of impairments to beneficial uses of those waters
and TMDLs. From the perspective of the stormwater management plan addressing MS4 permit
requirements, more urban area and water quality impairments or TMDLs addressed by the plan can
result in increasing BMPs and higher costs for implementation. As a result, the sophistication of the
February 2017
15
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
RAA approach may increase with the size of the municipality to provide reasonable assurance that
the stormwater management plan will address water quality goals while informing responsible, cost-
effective decision-making for the municipality in terms of BMP selection. Given this variability,
there is no one-size-fits all approach to selecting an analytical method for performing an RAA.
Instead, each municipality must weigh the categories of factors (Figure 3-1) and make decisions as to
what factors are most important to their stormwater management plan. The following are some
general considerations that depend on size of the municipality. Each consideration includes a Venn
diagram to illustrate the overlap of factors (Regulatory Drivers, Analytical Capabilities, Practical
Considerations) that can influence the selection of an analytical approach, with the size of the dark
grey area labeled "RAA" providing an indicator of the robustness to be anticipated.
Planning Process Requirements
Financial Planning
Requirements
Spatial/Temporal Precision
Hydrologic Precision
Pollutant Assessment
and Representation
BMP Process Precision
User Skill/Training Needs
Permit Requirements
Multi-Purpose Planning Needs
BMP Types, Siting
Opportunities, and
Constraints
Data Needs/Availability
Costs
Multi-Benefits Assessment and Optimization
Figure 3-1. Considerations for Selecting an Analytical Method for the RAA
Regulatory/Planning
Needs
Large Municipality or Complex Setting - As
mentioned, large municipalities typically have a
complex setting to be addressed by the RAA, including
a large amount of urban area, multiple pollutants of
concern, and a complicated storm drain system that
may include various diversions/impoundments or
comingling with separate NPDES permitted systems.
Larger municipalities may also have an MS4 permit
that requires a robust RAA demonstration with
commitments for BMP planning and tracking. Given
the high potential BMP implementation costs, there
can be increased burden to demonstrate certainty to the
permitting authority, elected officials, or public that BMPs are sufficient at addressing planning
goals and to support funding of implementation. As a result, multiple factors within each
category in Section 3.1 may be important to the selection of the analytical method. Typically, the
more factors that apply, the more sophisticated the approach selected.
Analytical
Capabilitii
r «
16
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Medium Municipality or Moderate Setting - A medium
municipal jurisdiction can have a less urban area than the
large municipality described above, with fewer pollutants of
concern or applicable TMDLs. The M S4 permit may require
an RAA, but provide greater flexibility in rigor, planning
commitments, and tracking of BMP implementation.
However, implementation costs can still range from
moderate to high, requiring some sophistication to justify
financial commitments to BMPs. In considering the
categories of factors in Section 3.1, some factors may be
considered more important than others, and all factors may
not be addressed by the selected RAA. This can result in selection of an analytical method that is
less sophisticated than a method selected by a larger municipality as described above.
Small Municipalities or Simple Setting - For small
communities or areas with no specific pollutants of concern
or TMDLs, hydrology or pollutant loading may be highly
predictable or understood based on monitoring data. MS4
permits for these municipalities may not specifically require
separate RAA tools and/or require little long-term tracking
of BMP implementation, as a solid stormwater management
plan may be in place that is funded and demonstrating
success at preventing water quality impairments. As a result,
modest implementation costs are expected as a result of
future modifications of the stormwater management plan. If
an RAA is pursued by the municipality to inform BMP
implementation efforts, there may be less emphasis on several factors associated with categories
in Section 3.1. In this situation, a simple analytical method may be determined sufficient to
support planning efforts.
4 PERFORMING THE REASONABLE ASSURANCE
ANALYSIS
This section provides a discussion of the processes generally followed to support each of the key
Elements of an RAA outlined in Section 1.2. This RAA organizational framework can be applied in
the design and implementation of RAA requirements both by permitting authorities developing new
permit requirements and permittees as they develop specific RAA methods to address permit
requirements.
4.1 Element 1: Designation of Area Addressed by Analysis
While traditional approaches to watershed plans tend to use a holistic approach that considers all
point and nonpoint sources that are hydrologically connected (USEPA 2008), the permit-driven
approach aims to isolate, quantify, and manage pollutant sources that originate from within theMS4
permit boundary. In some cases, there may be more than one municipal jurisdiction that is
addressed by a permit that collectively drain and comingle within a receiving water. Furthermore,
areas addressed by separate NPDES permits, federal land, or state-owned land subject toother
February 2017
17
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
management that fall within the delineated hydrologic boundaries should also be considered and, in
some circumstances, removed from the designated planning area.
There is no single, standardized approach for designating an area to be addressed by an RAA. There
are many approaches and available datasets that can support Element 1, which typically include
processes for analyzing areal imagery or GIS datasets that represent land use, municipal
jurisdictional boundaries, storm drain collection systems, or impervious area. For some MS4 permits
that include a Prescriptive Approach to performing RAAs, specific guidance is provided on the
process to be followed for designating the MS4 jurisdiction to be addressed by the RAA. As
discussed in Appendix A.6, the Virginia Phase 1 MS4 permits prescribe how different land uses and
jurisdictions are to be evaluated and the Massachusetts General Phase IIMS4 permit provides
permittees a choice of whether to evaluate the entire areas within municipal boundaries or only the
urbanized areas within those boundaries.
For MS4 permits that utilize a Deterministic Approach to performing an RAA, little or no guidance
was provided in the example permits summarized in Section 2. This presented some challenges for
permittees in terms of establishing methods for designating areas for analysis that were defensible to
the permitting authority and stakeholders. It is important for permitting authorities to work with
permittees as the geographical scope of RAA analysis is selected to ensure the approach selected will
enable robust analysis of stormwater management alternatives and plans for which individual
permittees are responsible. As summarized in Appendix A.3, the San Diego Region Phase IMS4
Permit and Los Angeles County Phase IMS4 Permit provided no guidance on performing this task,
and some WQBELs associated with TMDL wasteload allocations in the permit were vague in terms
of specific urban areas or spatial extent of storm sewer systems for which they apply. The permittees
had discussions with the permitting authorities on this issue throughout the planning process and
reached varying agreement on methods to address Element 1. Figure 4-1 shows a map of MS4 and
non-MS4 permitted areas for an example watershed in the San Diego Region.
18
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
:igure 4-1. Los Penasquitos Watershed Pollutant Discharge Responsibilities (LP WQIP Responsible
Agencies 2015)
Another key consideration in the designation of permitted areas to be addressed in the analysis is the
hydrologic connectivity to receiving waters with associated WQBELs and/or TMDL wasteload
allocations that apply to permittees. In the MS4 permits that utilized a Prescriptive Approach,
specific guidance is provided in terms of the applicable watershed boundaries and the permittees
located within those boundaries. In performing a Deterministic Approach to an RAA, the
delineation of watershed boundaries and segmentation of smaller subwatersheds is typically
performed during the model development process, as the model utilizes this segmentation for model
configuration. This segmentation must account for both municipal jurisdictional boundaries and
hydrologic connectivity so that pollutant loading can be quantified within each jurisdiction (Element
2) and responsibility can be determined in terms of individual permittee pollutant reductions
(Element 3) and associated management actions to meet these reductions (Element 4). As
summarized in Appendix A.l, the Upper Los Angeles EWMP provides an example of how
jurisdictional and hydrologic boundaries were considered during segmentation of the planning area
for the RAA.
In summary, various approaches can be used for designation of the permitted area addressed by the
RAA. These approaches are highly dependent on preferences of permitting agencies and the
availability of GIS data to support the analysis. If a permit does not include guidance on methods to
be used to address Element 1, the permittee should consult with the permitting authority to reach
agreement on methods to be used. The following are key considerations for addressing Element 1:
If multiple municipal jurisdictions are addressed by the RAA, the analysis should be capable of
distinguishing among jurisdictions in terms of relative contributions of flow and pollutant loads.
If areas not subject to municipal jurisdiction are included, their flows and loads should be
distinguishable.
February 2017
19
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
The area of analysis should make sense in terms of hydrologic function and connectivity, and for
some approaches flows and loads may require routing through the system.
4.2 Element 2: Characterization of Existing Conditions
For the RAA, management objectives and actions are determined relative to an existing condition
that reflects the current understanding of stormwater flows, pollutant loads, and critical conditions to
be managed. Under a Prescriptive Approach with an established TMDL, existing conditions and
associated critical conditions may already be documented within the TMDL and can be cited by the
RAA. Nevertheless, the RAA may still require evaluation of certain components to account for
factors or assumptions not addressed in the TMDL. For example, while a TMDL and associated
wasteload allocations are calculated for a particular point in a drainage network, contributing flows
and loads may need to be further allocated to a scale that is compatible with associated jurisdictional
boundaries and management actions. Using a Deterministic Approach, the research and
development associated with modeling pollutant loads and BMP performance is performed during
the RAA. Sometimes there is a need to explicitly represent the impacts of existing BMPs as part of
the baseline condition so that it is possible to back-calculate their benefits. Other times, refinements
to the RAA model are needed to address management practices at a different point in time, such as
the date of approval of the MS4 permit or TMDL, or even a future date that includes committed or
planned management activities. The objective of such a study is to identify additional practices
(above and beyond planned actions) that are required to achieve management targets.
4.2.1 Analysis of Monitoring Data
During the early stages, the RAA process includes data assessment and trends analysis to build an
understanding of the applicability and limitations of available information used to characterize
existing conditions in the watershed. The aim of this effort is to establish causal linkages and
prioritize water quality conditions for management (e.g. limiting pollutant loading or flow
conditions). Figure 4-2 presents an example analysis sequence utilized in Los Penasquitos
Watershed Management Area WQIP (Appendix A.3) to identify and prioritize water quality
conditions.
20
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Priority and Highest Priority Water Quality Condition
Selection Process
Step 4 Determine Highest Priority
Water Quality Conditions
Step 3 Determine Priority Water Quality
Conditions
Step 2 Determine Potential Receiving Water Impacts
from MS4 Discharges
Step 1 Determine Receiving Water Conditions
Figure 4-2 Example analysis sequence for selecting priority water quality conditions (LP WMA2015).
Data analysis first begins with determining the receiving water conditions, with the goal of assessing
the relative impact and contribution of MS4 discharges to those conditions. Part of this involves
determining natural and anthropogenic impacts to the receiving water conditions. The analysis of
monitoring data can result in a thorough understanding of the pollutant sources addressed by the
RAA and their impacts on receiving water quality. This analysis can also inform the selection and
application of a model used to represent existing conditions. For instance, once the pollutant sources
and impacts are well-understood, an appropriate model can be selected that provides the best
representation of the processes associated with pollutant generation or transport. Depending on the
selected model, the monitoring data can also be used as direct model input (e.g., assignment of
stormwater pollutant concentrations). Most models used to support RAAs utilize monitoring data
for model calibration and validation (Section 4.2.2). As a result, a thorough understanding of
available monitoring data that characterizes stormwater runoff and water quality is key informing
the selection and application of a model that best represents the system.
4.2.2 Modeling Existing Conditions
Key to Element 2, Characterization of Existing Conditions, is the demonstration that a model
provides reasonably accurate representation of existing managed conditions of stormwater runoff,
pollutant concentrations, and/or pollutant loads. This representation of existing conditions becomes
the foundation for evaluation of management strategies that seek to improve conditions, through the
reduction of pollutant loads and/or capture of stormwater. Therefore, careful thought must go into
the selection and application of an analytical framework to represent the existing conditions.
Selection of the appropriate analytical method, including a model, can be based on several factors as
discussed in Section 3. EPA's Guidance for Quality Assurance Project Plans for Modeling outlines a
general three-step process that considers both the model selection and application process.
Throughout this three-step process, the following procedures are recommended for evaluating model
February 2017
21
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
performance, which lend to the demonstration of accurate representation of the existing conditions
(USEPA 2002):
Define Model Performance Evaluation Criteria - Step 1 includes procedures for defining
quality objectives, desired model performance criteria, and documentation of needs for model
output. This process addresses questions such as the following:
1. How accurately and precisely does the model need to predict a given quantity at the
application site or in the process step of interest in order to satisfy regulatory or scientific
objectives?
2. What are appropriate criteria for making a determination of whether the model estimate is
accurate and precise enough (e.g., based on past general expertise combined with site or
process specific knowledge)?
3. How would these criteria be used to determine whether model outputs achieve the needed
quality?
Often, model performance criteria or quality objectives are based on typical metrics reported in
literature, or can be specified in local guidance prepared by regulators. For example, Lumb et al.
(1994) outlines a number of numeric criteria that can be used to assess the performance of HSPF
models. To support RAAs performed in the Los Angeles Region, the Los Angeles Regional
Water Quality Control Board (2014) developed guidance that specified model performance
criteria to be used, as discussed in further detail in Appendix A.l. The selected numeric criteria
are used to measure model precision and accuracy through comparison of model output with
monitoring data (e.g., measure flow or pollutant concentration) or other benchmarks that
characterize expected hydrology or pollutant loading (e.g., typical land loadings in lbs/year).
Therefore, the selected criteria will need to be commensurate with the output available from the
model. For instance, a model that predicts annual loads will not provide comparison to
performance criteria associated with peak storm flows. However, a model that predicts hourly
flows or pollutant concentrations can be compared with performance criteria for a single storm
volume or peak flow event, or can be summed across longer time periods for comparison to
coarser criteria focused on annual volumes or pollutant loads. Selection of appropriate
performance criteria for an RAA should therefore consider both the expectations of permitting
authorities as well as capability of the selected model.
Model Calibration and Validation - Assessment and reporting of model performance is often
achieved through the model calibration and validation process. Model calibration and validation
is the method of adjusting rates and constants that represent physical, chemical, or biological
processes, while confirming those adjustments to produce a robust predictor of the system
modeled. The model calibration process is a step-wise procedure that starts with quality
assurance of model input (e.g., weather data), and continues with calibration of model
parameters that drive simulation of hydrology, transport, and water quality. Careful attention is
used in each step of the process to ensure that model uncertainty is not propagated to latter steps,
as many model processes are dependent on other calibrated processes. For instance, hydrology
calibration is one of the first steps in calibrating a model, and if not performed thoroughly,
uncertainty in hydrology simulations could impact calibrations for sediment transport, water
quality, etc. Figure 4-3 is a schematic describing a process for model calibration that aims to
minimize the propagation of uncertainty. Once a model is calibrated, model predictions are then
compared to an independent dataset for validation. This independent dataset may be monitoring
data collected at another location or during a different period than that used for calibration. If
the model is not validated, then the model calibration is revisited with an emphasis on the
adjustment of parameters that are hypothesized to result in the lack of validation. Throughout
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
each step of this iterative calibration and validation process, appropriate statistical and sensitivity
analysis should be performed to support assessment of model uncertainty. Through comparison
with the model performance criteria, the RAA should document the model uncertainty to
demonstrate that the model reasonably predicts existing conditions of the system. Additional
sensitivity analysis can be used to further investigate the degree of uncertainty of key modeling
assumptions that highly impact modeling results. Such an analysis can inform future monitoring
efforts and the collection of data that can be used to improve the model over time through an
iterative process.
Model Calibration
Weather Data
Land Hydrology
Stream Transport
Water Quality
Ensure quantity ~ quality of
identify influential land use
Include special hydraulic
1. Sediment «,
spatial/temporal cow-age
^1
features and factors. Define
features of stream routing
I. Associated Pollutants
of forcing data that drrvc
parameter groups. Check
network (i.e. point sources,
3. Other Pollutants
watershed hydrology.
base flow, runoff, seasonal
reservoirs, diversions).
4, Fate and Transport
It
it
it
it
Parallel Objective:
Figure 4-3. Example Process for Model Calibration to Minimize Propagation of Uncertainty.
Documentation of Results - The model performance is typically evaluated through the review
of documentation of the calibration and validation and comparison to model performance
criteria. There are various methods to present model comparisons to monitoring data and
performance criteria, ranging from time-series plots to tabulated results of statistical analyses.
Appendix A.2 includes example documentation of model calibration for hydrology (Stormwater
Retrofit Plan for Water Resources Inventory Area 9) and water quality (Upper Los Angeles
River EWMP).
Once the model is fully calibrated, the model is typically used to assess pollutant loads that can vary
by source category or spatially. This assessment can provide early indication of locations where
management actions should be emphasized in the stormwater management plan. Figure 4-4 shows
an example assessment of the spatial variability of sediment loading predicted by the model
supporting an RAA for the Los Penasquitos watershed management area (Appendix A. 3).
February 2017
23
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
POWAY
SOLANA
BEACH
SAN DIEGO
COUNTY
KiLoim
[Penasqultos]
* Creek
Los,
Pcnasquitos
Lagoon
SAN
OIEGO
. SAN DIEGO
COUNTY
Carroll
Canyon
Creek
Sol 411J
Heath
- Highway
Stream or Rrver
~ SubsvatorsheO
Los Pena&qultos Watershed
JuriMbcton Boundary
Wet-Weather Score (Sediment)
lOowt
I oil 4
5 (High)
:igure 4-4 Example Assessment of Relative Sediment Loading Los Penasquitos (LP WQIP Responsible
Agencies 2015)
4.3 Element 3: Determination of Stormwater Improvement Goals
Once the model of existing conditions is complete, it can be used to assess a range of flow conditions
and pollutant loading conditions that can be compared with relevant TMDL wasteload allocations,
WQBELs, or other hydrologic or water quality targets to help specify necessary stormwater
improvement goals. This modeling analysis is performed without inclusion of BMPs in the model,
and is instead a direct comparison of model scenarios representing the existing condition (Element
2) and target conditions. Key to this analysis is the simulation of a specific critical hydrologic
condition that ensures protection of beneficial uses under a managed scenario. Typically, a critical
condition is identified in the stormwater management planning process or designated by the MS4
permit or applicable TMDLs. The existing condition model can be used to simulate the critical
hydrologic condition to provide a baseline for evaluating the necessary reductions of flow or
pollutant loads to meet hydrologic or water quality targets. Critical conditions may be expressed at
different spatial and temporal scales, with diverse methodologies applied for both identifying and
managing the critical conditions. Table 4-1 summarizes critical conditions for the example RAAs
summarized in Appendix A. For many developed watersheds, water quality impairments tend to be
the direct result of an altered hydrologic condition. Some municipal agencies have developed or
adopted evaluation metrics that directly address impaired hydrologic conditions, while others focus
more 011 water quality targets. The methodologies vary considerably for the examples RAAs in
Table 4-1 and Appendix A, and there is no one-size-fits all approach to determining stormwater
improvement goals. Examples of methodologies that address hydrologic and water quality targets
are provided in the following sections.
24
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Table 4-1. Summary of Identified Critical Conditions for Example RAAs (Appendix A).1
RAA Case Study
RAA
Summary of Critical Condition and Hydrologic or Water
Quality Target Addressed
Approach
Flow
Regime
Constituent
Methodology
Upper Los Angeles
River EWMP
Deterministic
Wet
Metals
Manage Exceedance Volume for the
Limiting Pollutant
Bacteria
Retain Critical Storm
Dry
Bacteria
Eliminate or Retain Non-Stormwater
Runoff
Santa Monica Bay
Jurisdictional Group
2 and 3 EWMP
Deterministic
Wet
Metals
Stochastic Evaluation of Critical Storms
Bacteria
Manage Critical Storm
Dry
Bacteria
Monitoring
Stormwater Retrofit
Plan for Water
Resources Inventory
Area 9, King County,
WA
Deterministic
Wet
Peak Flow
Restore Flow Duration Curve and B-IBI1
Wet
Sediment
Achieve Water Quality Indicators
Los Penasquitos
Watershed
Management Area
WQIP and CLRP
Deterministic
Wet
Sediment
Reduce Annual Sediment Loads for
Average Year
Bacteria
Reduce Frequency of Exceedance of
Water Quality Targets
Dry
Eliminate or Retain Non-Stormwater
Runoff
Both
Protect Beach Contact Recreation Use
Optimal Stormwater
Management Plan
Alternatives: A
Demonstration
Project in Three
Upper Charles River
Communities
Deterministic
Wet
Volume and
Sediment
Reduce Annual Loads
Arlington County
Chesapeake Bay
TMDL Action Plan
Prescriptive
Wet
Nitrogen
Manage Algal Blooms that Create Low
Dissolved Oxygen for Aquatic Life
Phosphorus
Sediment
Restore Degraded Aquatic Habitats
Optimal Stormwater
Management Plan
Alternatives: A
Demonstration
Project in Three
Upper Charles River
Communities
Prescriptive
Wet
Phosphorus
Reduce Annual Phosphorus Load
1 Benthic Index of Biotic Integrity
As shown in Table 4-1, methods for assessment of critical conditions and hydrologic or water quality
targets can vary considerably. In some cases, reduction of peak flows or a critical storm volume can
1 For the example San Francisco Bay Regional MS4 Permit in Appendix A.5, although local analytical
methods have been developed to support RAAs, no completed RAAs were available at the time of this review
for inclusion of methods in the table addressing specific hydrologic or water quality targets.
February 2017
25
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
be demonstrated to be protective of biological or water quality targets. For these cases, reduction of
the critical flow or volume can be proven through the RAA to provide sufficient reductions of
impacts to biological conditions or pollutants to meet water quality targets. The King County
Stormwater Retrofit Plan for Water Resources Inventory Area 9 (Appendix A.2) provides an
example methodology used to address critical conditions to protect biological conditions. For the
example RAA summarized in Appendix A. 1 for the Upper Los Angeles River Enhanced Watershed
Management Plan, the MS4 permit required that pollutants be in compliance with water quality
targets for 90 percent of all wet-weather conditions over a recent 10-year period (wet weather
conditions were defined as the highest 10 percent of instream flows). Stormwater improvement goals
for the RAA were identified using a three-step process. First, modeled flow, pollutant
concentrations, and water quality standards were plotted together over a ten-year period. Second,
Exceedance Volumes (EV) were calculated for each wet-weather event as the sum of flows during
times when the modeled concentrations exceeded the water quality standard (Figure 4-5). Finally,
the critical condition was identified as the 90th percentile wet-weather EV. Expressing water quality
conditions on a volumetric basis allowed for normalized comparison among pollutants. The
pollutant with the highest EV was identified as the limiting pollutant for the critical condition
(Figure 4-6). Where multiple pollutants are being managed, the RAA can target the limiting
pollutant because it can be demonstrated that controlling it addresses the other pollutants of concern.
The limiting pollutant will therefore drive the selection of necessary BMPs to capture its associated
EV, and the BMPs will be sufficient for capturing the EV associated with all other relevant
pollutants.
Metals
(e.g. Cu, Zn, Pb)
24-hour Period
Figure 4-5 Example illustration for how Exceedance Volume is Derived for Metals (ULAR WMG 2016).
A Concentration
Metals
RWL
26
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
1.0
0.9
i
0.8
u
u
c
E
0.7
jj
!c
:>
|
0.6
0)
Q
o
0.5
c
c
ro
3
x>
0
0.4
0)
m
OJ
TJ
u
X
111
i
0.3
a
X
0.2
9i
0.1
0.0
0.20
01
E
3 0.15
0.10
3
01
u
c
CO
| 0.05
u
X
0.00
/ j
-
E. coli (Day 11 after HFS)
Zinc
lead
90th Percentile
80% 85% 90% 95%
Percentile
(10/1/2001-9/30/2011)
100%
0%
10%
20%
30%
70%
80%
90%
Figure
40% 50% 60%
Percentile
(10/1/2001-9/30/2011)
4-6. Identifying the Limiting Pollutant for the RAA Critical Condition using EVs (ULAR WMG 2016).
100%
Municipalities should always consult with the permitting agency on the selection of appropriate
critical conditions and hydrologic or water quality targets to be addressed by an RAA, which may
also be specified in the MS4 permit, relevant TMDLs, or local guidance. Where TMDLs have not
been developed that quantify wasteload allocations for the MS4 permit, methods for determining
stormwater improvement goals can have similarities to methods for calculating TMDLs and
associated allocations. Therefore, there are multiple guidance documents that municipalities can
reference to develop methods to address Element 3, including but not limited to the following:
Handbook for Developing Watershed Plans to Restore and Protect Our Waters (XJSEPA 2008)
Protocol for Developing Nutrient TMDLs (USEPA 1999a)
Protocol for Developing Pathogen TMDLs (USEPA 2001)
Protocol for Developing Sediment TMDLs (USEPA 1999b)
PCB TMDL Handbook (USEPA 2011a)
TMDLs with Stormwater Sources: A Summary of 17 TMDLs (USEPA 2007)
Helpful Practices for Addressing Point Sources and Implementing TMDLs in NPDES Permits
(USEPA 2015)
4.4 Element 4: Demonstration that Management Actions Will Result
in Attainment of Goals
Element 4 involves the quantification of the benefits of BMPs in terms of demonstrating capability to
attain stormwater improvement goals and requirements identified in Element 3. This progress can be
demonstrated in various ways, and typically include estimates of reduction of storm flows, pollutant
loads, or pollutant concentrations over time resulting from an array of proposed management
actions. Analytical methods for quantifying these reductions can range from simple calculations of
BMP effectiveness to more robust mechanistic computer modeling. It is important that the selected
February 2017
27
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
method have the capability to delineate necessary flow, loading, or concentration reductions at
geographical scales and timesteps specified in applicable permits and associated TMDLs and/or
water quality standards. Appendix A includes several example RAAs that vary in the
methodologies used.
As shown in Figure 1-2, separate efforts of the stormwater management plan typically identify
opportunities for nonstructural or structural BMPs. Nonstructural BMPs may include such practices
as street sweeping, pollutant source control, water conservation efforts that impact urban lawn
irrigation, public education, etc. Structural BMPs may include distributed stormwater capture
projects such as green infrastructure or Low Impact Development (LID), or regional stormwater
capture facilities that treat or infiltrate stormwater runoff from a larger drainage area. An example of
a regional stormwater capture facility could be an underground infiltration gallery placed in a public
park, which receives stormwater that is conveyed from the storm sewer system. A stormwater
management plan may identify a combination of opportunities for nonstructural and structural
BMPs (including both distributed and regional stormwater capture). The following sections discuss
model considerations for nonstructural and structural BMPs.
j ' | Quantification of Benefits of Nonstructural BMPs
Nonstructural BMPs are typically implemented across the area of RAA analysis and can vary
spatially in terms of density or intensity. Numerous nonstructural BMPs are often implemented in
combination as a stormwater management program. As a result, it can be challenging to estimate the
effectiveness of individual or specific combinations of nonstructural BMPs. The following are
methods typically used to represent the effectiveness of nonstructural BMPs in terms of
demonstrating stormwater or pollutant load reductions to support RAAs:
Assumptions for Nonstructural BMP Effectiveness based on Literature or Studies - Several
studies have been undertaken throughout the U.S. that attempt to quantify the effectiveness of
individual nonstructural BMPs, which have been documented by state/federal agencies,
municipalities, and academia. These studies vary in the type of BMPs assessed, pollutants
analyzed, and the methods used to estimate benefits. Typical methods for measuring
effectiveness rely on monitoring of stormwater and water quality, which can be costly and
challenging given the characteristics of nonstructural BMPs summarized above. As a result,
many RAAs formulate assumptions for nonstructural BMP effectiveness based on results
reported in past studies or literature. For example, the Massachusetts General Phase IIMS4
Permits (Appendix A. 7) provides methods for estimating nutrient load reductions associated
with enhanced street sweeping programs, catch basin cleaning, and organic waste and leaf litter
collection programs that are based on assumptions from various literature sources. To address
the Los Angeles County Phase IMS4 Permit (Appendix A.l), EWMPs made assumptions for
effectiveness of nonstructural BMP programs as a whole, representing a combination of
practices. The burden of proof for these EWMPs was based on an extensive list of new
nonstructural programs that were proposed, with pollutant reductions to be verified through
extensive monitoring programs also reported in the plans.
Modeled BMP Effectiveness - Some of the models typically used for RAAs have the capability
to represent a subset of nonstructural BMPs (e.g., street sweeping) to estimate pollutant
reductions. However, these models rely heavily on literature or local monitoring studies to base
modeling assumptions. For the example RAA supporting the Los Penasquitos Watershed
Management Area WQIP and CLRP (Appendix A.3), modeling was performed to quantify
watershed-wide pollutant reductions resulting from multiple nonstructural practices, including
28
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
street sweeping, catch basin cleaning, and a rain barrel incentive program, downspout
disconnection incentive program, and irrigation runoff reduction program for residential and
commercial areas. However, modeling assumptions for this RAA relied heavily on multiple
special studies previously performed by the City of San Diego to quantify the benefits of these
BMPs. Without substantial information to base such modeling assumptions, many
municipalities are limited to the literature-based assumptions above.
! i Quantification of Benefits of Structural BMPs
Separate from the RAA (Figure 1-2), stormwater management planning efforts should identify
opportunities for the placement of structural BMPs, which may include characterization of the BMP
drainage area, site conditions, and proposed BMP type for individual projects. Stormwater
management planning efforts can also identify design considerations of BMPs that can support
modeling efforts. Appendix A.7 shows an example typical BMP design used to support RAAs
addressing the Massachusetts General Phase IIMS4 Permits. The siting of structural BMP
opportunities can be performed through GIS analysis that considers multiple characteristics of the
site (e.g., soil infiltration rate, available space, slope) and associated BMP drainage areas (e.g., size,
imperviousness) that influence the likelihood that a BMP will be selected, constructed, and provide
meaningful performance. The robustness of this analysis is often dictated by the availability of local
GIS datasets. Some RAA modeling tools include capabilities to site BMP opportunities, which may
be considered in lieu of or in combination with GIS analysis techniques. For example, GreenPlan-IT
(Appendix A. 5) includes a GIS-based Site Locator Tool that combines the physical properties of
different types of green infrastructure with watershed GIS information to identify and rank potential
locations for green infrastructure implementation.
Once BMP opportunities and design considerations are identified by stormwater management
planning efforts, this information is then used as input to RAA analytical methods to quantify
stormwater flow or pollutant reductions expected to occur following BMP implementation. This will
require representation of the routing network for BMP locations, their associated drainage areas, and
placement within the area addressed by the RAA to provide assessment of area-wide benefits. The
methodology will also likely require characterization of the BMP drainage areas to provide
assessment of storm flows and pollutant loads to the individual BMPs. For a stormwater
management plan that includes a handful of BMPs, the explicit representation of the individual
BMPs may be straightforward for model representation. However, for plans that address large
watersheds that require representation of hundreds of BMPs, including many distributed BMPs
within small drainage areas, model representation can present some challenges. To address model
representation of a large number of BMP opportunities, assumptions can be made to reduce the
complexity of the problem and prevent the need to individually represent each distributed BMP.
Appendix A.2 and A.7 include example structural BMP routing networks used to model large
numbers of distributed BMP opportunities for RAAs supporting the King County Stormwater
Retrofit Plan for WRIA 9 and the Massachusetts General Phase IIMS4 Permits, respectively.
The methods for representing BMP processes and associated stormwater and pollutant reductions
vary considerably based on the analytical approach selected. As a result, documentation of the
various models should be consulted to more fully understand the considerations for modeling BMPs.
Once the BMPs are configured within a model, the model can then be used to test alternative BMP
scenarios and the necessary reductions of storm flows or pollutants to meet stormwater improvement
goals from Element 3. This requires model simulation of the same critical condition used in Element
3 to provide direct comparison to stormwater or pollutant reduction targets. Consistent with the
stormwater improvement goals established in Element 3, the model output for the BMP scenarios
February 2017
29
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
may consider reductions of stormwater peak flows or volumes, pollutant concentrations or loads, or
other metrics that coincide with water quality targets. Example RAA results for the King County
Stormwater Retrofit Plan for WRIA 9 are shown in Appendix A.2, illustrating how BMP model
scenarios can be used to demonstrate attainment of water quality and hydrology targets. Model
results can also be presented spatially to communicate where and to what degree BMPs will be
implemented. Figure 4-7 shows the King County RAA results with BMP volumes presented
spatially for the 2040 future land use/development scenario with full stormwater management. This
map provides an indication of the level of BMP implementation needed throughout the watershed to
meet the stormwater improvement goals.
Figure 4-7. WRIA 9 Study Area BMP Storage in Watershed inches2for Full Stormwater Management in 2040
(King County 2014a).
Puget Sound
Storage (watershed-inch)
_ 0 1-08
08-15
15-22
2^2 - 3 0
30-37
The analytical method may also need to show improvements to stormwater to meet progressive
goals over time. This is typical of stormwater management plans that address TMDL or MS4 permit
requirements for phased pollutant reductions over time to meet WQBELs or wasteload allocations.
As a result, the methodology will require simulation of multiple scenarios with increasing levels of
BMPs, and estimation of resulting stormwater flow or pollutant load reduction. Appendix A.3
2 Watershed inches are calculated as the amount of BMP volume needed divided by the area of developed land
use in the study area.
30
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
shows results of an RAA performed for the Los Penasquitos WQIP that demonstrates the increasing
and diverse types of BMPs planned over future years to meet phased sediment load reduction targets
defined by WQBELs and the TMDL wasteload allocation.
For stormwater management plans that include cooperative efforts of multiple municipalities, the
MS4 permit may require that separate BMP implementation plans, stormwater improvement goals,
and RAAs be performed for each individual participant. This was the case for plans developed in the
Los Angeles and San Diego Regions to address their respective MS4 permits, as discussed in
Appendix A.l and A.3, respectively. For the example Los Penasquitos WQIP in the San Diego
Region, four municipalities and the California Department of Transportation participated in the
plan. As a result, separate BMP implementation plans and RAAs needed to be reported for each
participant. This required separate modeling efforts to evaluate BMPs and associated pollutant
reductions for each individual jurisdiction.
For municipalities with a large jurisdictional area and many opportunities or alternatives for
structural BMP implementation, certain RAA analytical methods can provide additional capability
to assess numerous scenarios representing alternative BMP implementation strategies. As discussed
in Section 3.1, these approaches can provide automated comparison of multiple implementation
scenarios and direct comparison of pollutant reductions and costs to support optimization of the
plan and selection of the most cost-effective combination of BMPs to meet targets. Although not
central to the basic requirements of the RAA to address MS4 permit provisions, this capability can
have an important role in decision-making and strategizing of BMPs to be incorporated within a
stormwater management plan. Figure 4-8 shows an example cost optimization performed for two
jurisdictions in the Upper Los Angeles River EWMP (Appendix A.l). Separate cost optimizations
were performed for each permittee participating in the EWMP. The following example MS4 permits
in Appendix A provide additional summary of RAA analytical approaches that include methods for
BMP cost optimization:
Los Angeles County Phase IMS4 Permit
San Diego Region Phase IMS4 Permit
Washington Phase IMS4 Permit
Massachusetts General Phase IIMS4 Permits
San Francisco Bay Area Regional MS4 Permit
February 2017
31
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
120
QJ
u_
n
s-
G
OJ
Q.
a.
CO
"!5
k_
3
100
80
60
re 40
20
i Regional BMPs (private)
I Regional BMPs (Medium)
Regional BMPs (High)
Regional BMPs (Very High)
¦ Green Streets
HID (residential)
I LID (public)
ILID (existing/planned)
lUD (ordinance)
-Implementation Cost
Pasadena (Arroyo Seco)
Target: 39%
r (Prior to Allocation of Regional BMP
Capacity to Upstream Jurisdictions)
250
c
o
200 1
150
o
O
c
O
100
c
o
5
_oj
a
E
50 Z
re
OJ
>¦
o
PM
80
~ 70
ai
u
re
60
u 50
re
Q.
re
u
CL
5
00
40
30
u
3
20
10
Zinc Load Reduction (limiting pollutant)
i Regional BMPs (private)
HIRegional BMPs (Medium)
Regional BMPs (High)
Regional BMPs (Very High)
¦ Green Streets
hi LID (residential)
HI LID (public)
hi LID (existing/planned)
H LID (ordinance)
Implementation Cost
Uninc. LA County (Arroyo Seco)
Target: 39%
(Prior to Allocation of Regional BMP
Capacity to Upstream Jurisdictions)
180
160
c
o
140
120
100
e
t/v
o
u
c
o
80 £
¦
60 «
40 -i
20
re
¦
o
fM
Zinc Load Reduction (limiting pollutant)
Figure 4-8. Example Cost-Optimization for Two Jurisdictions in the Upper Los Angeles River EWMP (ULAR
WMG 2016)3
3 This example shows the set of optimized BMP solutions for two ULAR EWMP jurisdictions that drain to
Arroyo Seco. Each optimization curve represents over 1 million BMP scenarios that were evaluated for cost-
effectiveness.
32
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
4.5 Element 5: Documentation of Results that Inform Implementation
and Tracking
Apart from the RAA, a stormwater management plan can document information for a BMP
implementation strategy that can inform future implementation efforts and the tracking of
implementation progress (Figure 1-2). This can include maps of BMP locations, design information,
and other details developed through the planning process. The RAA can provide specific results that
will strengthen the documentation of the BMP implementation strategy, including key information
used to link BMP effectiveness to reductions in stormwater or pollutant loads. Depending on the
RAA analytical approach used, information on BMPs incorporated within the RAA can be reported
in varying methods and levels of detail. As discussed in Section 3, proper attention should be placed
on selection of an RAA analytical approach that provides necessary information to meet reporting
requirements of the MS4 permit. If the permit does not provide sufficient guidance on reporting and
documentation of the BMP implementation strategy associated with the RAA, the permitting
authority should be consulted to gain an understanding of expectations.
For Prescriptive Approaches for RAAs, expectations are typically clear in the MS4 Permit or
associated guidance in terms of the detail and format for documenting the RAA results based on
methods and tools developed by the permitting authority. Appendix A includes examples of
Prescriptive Approaches for three MS4 permits that summarize different methods to report
outcomes of the RAA.
For MS4 Permits that require Deterministic Approaches to RAAs, varying guidance is provided by
permitting authorities in terms of documenting results of the RAA. Of these approaches summarized
in Appendix A, the Los Angeles County Phase IMS4 Permit and guidance prepared by the
LARWQCB (2014) provided the most specificity in terms of expectations and detail to be
documented for the RAA. Appendix A.l includes example documentation of the RAA results for
the Upper Los Angeles River EWMP, which demonstrates how RAA optimization results (Section
4.4.2) can be used to inform the documented BMP implementation strategy. The RAA results
reported for the Upper Los Angeles River EWMP provide details on the BMPs to be implemented
within specific subwatersheds and municipal jurisdictions to cost-effectively meet TMDL
compliance targets. These results are documented through the reporting of stormwater volumes
managed (through treatment or capture) by the different categories of BMPs in separate
subwatersheds and municipal jurisdictions to meet these targets over time. The EWMP also reports
the capacity of storage within the BMPs (BMP capacity) to provide the needed stormwater volumes
managed over time, which can guide implementation and planning of the BMPs (Figure4-9).
Similar RAA results were reported for the King County Stormwater Retrofit Plan for WRIA 9, as
shown spatially in Figure 4-7. Through the adaptive management process, future opportunities may
be identified that can change the combination of BMP capacities or locations, but the ultimate
jurisdictional or watershed goals for stormwater volume capture or pollutant load reduction remain
valid to guide implementation and tracking of implementation efforts.
February 2017
33
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
San Gabriel
hinal EWMP Compliance by 2037
Rio Hondo
¦ Regional BMPs (p'ivate)
¦ Regional BMPs (Medium)
Regional BMPs (H
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
methods can be applied to estimate each project's reduction of stormwater volume or pollutant
loads, and the collective reductions associated with a suite of BMPs can be added and compared to
scheduled goals identified in the stormwater management plan or established by the MS4 permit or
TMDLs. For example, to meet requirements of the Los Angeles County Phase IMS4 Permit,
RAAs performed throughout the region included documentation of results with quantitative metrics
used to provide reasonable assurance that the BMP implementation strategies will result in
attainment of pollutant reductions. As summarized above for the Upper Los Angeles River EWMP,
many of these EWMPs used volumes managed as the measurable metric. As a requirement of the
MS4 permit, each permittee and EWMP group reports annually to the permitting authority their
progress in terms of BMP implementation, which can be measured based on these metrics. As
summarized in Appendix A. 7, a similar process is used to track implementation of Phosphorus
Control Plans addressing the Massachusetts General Phase IIMS4 Permits, which requires periodic
performance evaluations to quantitatively assess the progress and effectiveness of BMP
implementation strategies. Section 5.1 summarizes additional considerations for tracking of
implementation.
The quantitative tracking of BMP implementation can also support effective adaptive management
and continued improvement and refinement of BMP implementation strategies to ensure successful
attainment of planning goals. Traditional assessments that have supported the tracking of BMP
implementation or water quality improvement have relied on water quality monitoring combined
with an inventory of BMPs that have been implemented or other qualitative/quantitative
information for program effectiveness assessment. These traditional approaches can be combined
with quantitative tracking of BMP implementation, based on metrics established by the RAA, to
continuously evaluate the trajectory of the BMP implementation strategy to assess if refinements or
changes are required. This can inform cost-effective BMP planning and periodic updates of the
stormwater management plan over time that capitalize on lessons learned.
In summary, beyond meeting the requirements of the MS4 Permit, RAAs can provide important
information that can further support BMP project planning and tracking. Depending on the
analytical method used, information produced by the RAA can also inform future decisions
regarding selection of cost-effective BMPs for implementation that maximize benefits, conceptual
design or sizing of individual structural BMP projects, or development of project costs estimates to
support capital improvement planning or investigation of funding opportunities.
5 TRANSITIONING FROM PLANNING TO
IMPLEMENTATION
As municipalities transition from the development of stormwater management plans to the
implementation of associated BMP implementation strategies, municipalities will face new
challenges associated with the realities of implementation and demonstration of continued success of
the plan. Section 4.5 summarized potential needs for RAA documentation in terms of quantitative
metrics that can support future tracking of BMP implementation and inform the adaptive
management process for continued improvement of the plan over time. The RAA or analytical
approach can support additional efforts associated with implementation, including capital
improvement planning, individual project planning, engineering, construction, and operation and
maintenance (O&M), financial planning, and integrated water resources planning. As discussed in
Section 3, the needs of these future efforts can be considered in the selection and development of the
RAA analytical method to provide necessary information or technical capability to potentially serve
as a decision-framework or assist in the next phase of implementation of the stormwater
February 2017
35
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
management plan. The following sections briefly summarize considerations for these future planning
efforts.
5.1 Implementation Tracking and Adaptive Management
Section 4.5 discussed the potential need to provide quantitative metrics and other information to
support the tracking of implementation of the stormwater management plan. Quantitative tracking
of BMP implementation can support MS4 permit reporting requirements or the adaptive
management process that guides continued revisions or updates of the stormwater management plan
over time and adjustments of the RAA approach. Each municipality should review the reporting
requirements in their MS4 permit to understand the future needs, which may require regularly
scheduled reporting of implementation progress or evaluation of the performance of the plan,
including specific methods or quantitative procedures to be used. For the example, during
development of the RAA for the Upper Los Angeles River EWMP (Appendix A.l), the permittees
understood the quantitative burden of proof that would be required by both the RAA and the
following annual reports required by the MS4 permit to document progress towards attaining goals
and implementing BMPs. To meet these requirements, quantitative goals were established in the
form of volumes managed with BMPs to meet pollutant reduction goals (Section 4.5). To support
implementation tracking, the Los Angeles County Flood Control District has initiated the
development of the Watershed Reporting Adaptive Management and Planning System (WRAMPS),
a public domain, web-based system that will allow each individual permittee in the county to enter
BMP information (e.g., BMP type and size, characteristics of drainage area, infiltration rate), and
the tool will calculate stormwater volumes managed that are in line with methods used to calculate
quantitative goals in the RAA. Each permittee can input all BMPs implemented within their
jurisdiction, and WRAMPS will compile information to support development of annual reports and
assessment of progress. As the EWMP is updated in the future as part of the adaptive management
process, the necessary data will already be compiled for BMPs that have been implemented,
allowing and iterative process for streamlined updates and adjustments to the RAA to represent the
new managed condition or improve model predictions. As a holistic, planning, tracking and
reporting system, the iterative approach currently employed in the Central Coast Region (Appendix
A.4) is also purposely designed to meet the short and long term needs of managing, tracking, and
accounting of the implementation of structural and non-structural BMPs to demonstrate water
quality improvement progress and annual regulatory requirements.
5.2 Stormwater Program Planning Frameworks that Support
Successful Implementation
Although a stormwater management plan designed to meet MS4 permit requirements may be
robust, these plans typically lack much of the information needed to support necessary critical
decisions regarding funding for implementation. Funding for stormwater management is often in
competition with other municipal services that have comprehensive capital improvement plans
and/or asset management plans that provide specificity for individual project engineering needs and
detailed justification of funding needs. Although a cost estimate may be included as part of the
stormwater management plan, these estimates are often coarse and lack the burden of proof needed
by decision-makers and elected officials to justify the investment. If funding for the stormwater
management plan requires additional revenue or appropriation of general funds, a detailed plan is
often needed to support these funding decisions, including specific information on the design,
construction, O&M, and all other costs associated with each individual project. Although
development of the stormwater management plan addressing the MS4 permit may not include this
information, the planning process can consider the future needs for capital improvement planning,
36
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
financial planning, or asset management to provide an analytical method that can support these
efforts. The following are examples of capital improvement planning and asset management projects
that have been linked to stormwater management planning efforts:
City of San Diego Watershed Asset Management Plan and Masterplans - Parallel to efforts
preparing WQIPs to address the San Diego Region Phase IMS4 Permit (Appendix A.3), the
City of San Diego (2013) developed the Watershed Asset Management Plan to document the
current state of assets, project long-term rehabilitation and replacement requirements, and
serve as a roadmap for actions that address flood risk management and water quality.
Development of the Watershed Asset Management Plan has been closely coordinated with
the WQIP to comprehensively evaluate costs and funding strategies to meet future goals.
Upon completion of WQIPs addressing San Diego watersheds, the City initiated
development of processes for developing masterplans for each watershed that provide
improved information to support planning and engineering efforts, coupling stormwater
projects with other capital improvement projects planned by the City to capture efficiencies
in spending. Models used to support the RAAs have been used as analytical tools supporting
both the WAMP and watershed masterplanning.
City of Los Angeles Capital Improvement Planning and One Water LA - The City of Los
Angeles (2015) has begun capital improvement planning efforts in the development of a
Stormwater and Green Infrastructure 5-Year Capital Improvement Plan that combines
projects across five separate watersheds with separate stormwater management plans,
estimates costs for multiple project phases from design through construction, couples BMP
projects with other currently planned storm drain projects, and prioritizes the most cost-
effective projects for the first five years of the EWMP implementation schedule. Currently,
the City is developing the One Water LA Plan as a long-term, integrated framework for
managing the City's watersheds, water resources, and water facilities. The One Water LA
Plan will incoporate projects included in the EWMPs with a goal of integrating project
planning with other multiple planning objectives addressing a range of benefits, from
integrated management of water resources to improvement of watershed health. Models
used to support the RAAs have been used to support both planning efforts above.
Whether the audience are city managers, elected officials, voters, or state or federal agencies
soliciting grant opportunities, individuals who make funding decisions want to know what they are
receiving for the investment. Stormwater managers have the burden of proof to make this case,
which often require comprehensive planning efforts similar to the examples above. Stormwater
management that focuses on water quality has traditionally been out-competed by water supply,
wastewater, flood control, and other municipal services in terms of educating audiences on the
importance of investments and the benefits received. Recent efforts have focused more on integrated
watershed or regional planning efforts that emphasize management of all water as a collective
resource to collaboratively identify water management solutions that concurrently achieve social,
environmental, and economic objectives. The One Water LA Plan is a local example of an
integrated planning approach that serves the dual purpose of responsibly managing local water
resources in Los Angeles, while providing a forum for educating the public on the importance of all
sustainable water management, including stormwater and water quality.
State and federal agencies have also recognized the importance of integrated water management,
and have encouraged collaborative efforts with local agencies to develop watershed plans that
identify multi-benefit projects that pool funding resources. For example, the State of California
(2014) released the California Water Action Plan, which identifies the challenges of sustainably
managing water resources, and goals and actions needed to collaboratively plan and identify
February 2017
37
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
management solutions. To specifically address integrated stormwater management, the SWRCB
(2015) established guidance on the development of a Stormwater Resource Plan as a condition of
receiving funds for stormwater projects from any bond measure approved by voters. The Stormwater
Resource Plan is a watershed-based approach for integrated stormwater management that considers
waters quality and other aspects of aquatic resource protection, flood control, water supply, habitat
conservation, among others. The SWRCB guidelines for Stormwater Resource Plans also require
methods for analysis of water quality projects based simulation of proposed watershed-based
outcomes using modeling, calculations, pollutant mass balances, and/or other methods of analysis.
Such an analysis is equivalent to the RAA supporting the stormwater management plan addressing
an MS4 permit. Therefore, RAAs can serve the greater purpose of fulfilling requirements of state or
regional stormwater resource planning efforts that can support grant funding of proposed projects.
In summary, the RAA can serve as an analytical tool supporting a range of engineering, asset
management, and financial planning activities beyond the stormwater management plan. Linking
the RAA with other water management, economic, and financial planning tools, the resulting
evolving stormwater program planning framework can support quantitative assessment of the costs
and benefits of stormwater projects to inform long-term planning objectives, as well as coupling of
stormwater projects with other water resource project opportunities to capitalize on multiple project
benefits and improve funding opportunities.
6 REFERENCES
Arlington County. 2015. Arlington County Chesapeake Bay TMDL Action Plan. Arlington County,
Arlington, VA.
Bicknell, B. R., , J. C. Imhoff, A. S. Donigian, R. C. Johanson. 1997. Hydrological Simulation Program
- FORTRAN (HSPF), User's Manual For Release 11. EPA - 600/R-97/080. U.S.
Environmental Protection Agency, Athens, GA.
City of San Diego. 2013. Watershed Asset Management Plan. City of San Diego Transportation and
Storm Water Department, San Diego, CA.
City of Los Angeles. 2015. City of Los Angeles Stormwater and Green Infrastructure 5-Year Capital
Improvement Plan. City of Los Angeles, Bureau of Sanitation, Watershed Protection Division,
Los Angeles, CA.
Commonwealth ofVirginia. 2010. Chesapeake Bay TMDL Phase I Watershed Implementation Plan.
Commonwealth ofVirginia, Office of the Governor, Richmond, VA.
Commonwealth ofVirginia. 2012. Chesapeake Bay TMDL Phase II Watershed Implementation Plan.
Commonwealth ofVirginia, Office of the Governor, Richmond, VA.
DeGasperi, C.L., Berge, H. B., Whiting, K. R., Burkey, J. J., Cassin, J. L. and Fuerstenberg, R. R.
2009. Linking Hydrologic Alteration to Biological Impairment in Urbanizing Streams of the
Puget Lowland, Washington, USA. Journal of the American Water Resources Association
(JAWRA), 45: 512-533. Doi: 10.1111/j. 1752-1688.2009.00306.x.
Geosyntec Consultants. 2008. A User's Guide for the Structural BMP Prioritization and Analysis Tool
(SBPAT vl.0). Prepared for Heal the Bay, City of Los Angeles, and Los Angeles County
Department of Public Works by Geosyntec Consultants, Los Angeles, CA.
38
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
Geosyntec Consultants. 2012. A User's Guide for the Structural BMP Prioritization and Analysis Tool
(OCTA-SBPATvl.O). Prepared for Orange County Transportation Authority by Geosyntec
Consultants, Huntington Beach and Los Angeles, CA.
Horner, R.R. 2013. Development of a Stormwater Retrofit Plan for Water Resources Inventory Area 9: Flow
and Water Quality Indicators and Targets. Prepared for King County, Water and Land
Resources Division, Seattle, WA.
Huber, W.C., andR.E. Dickinson. 1988. Storm Water Management Model Version 4, User's Manual.
EPA/600/3-88/001a (NTIS PB88-236641/AS), U.S. Environmental Protection Agency,
Athens, GA.
King County. 2013. Watershed Model Development for Water Resources Inventory Area (WRIA) 9
Stormwater Retrofit Planning Project. King County Department of Natural Resources and
Parks, Water and Land Resources Division. Seattle, WA.
King County. 2014a. Development of a Stormwater Retrofit Plan for Water Resources Inventory Area 9:
SUSTAIN Modeling Report. King County Department of Natural Resources and Parks, Water
and Land Resources Division. Seattle, WA.
King County. 2014b. Development of a Stormwater Retrofit Plan for Water Resources Inventory Area 9:
Comprehensive Needs and Cost Assessment and Extrapolation to Puget Sound. King County
Department of Natural Resources and Parks, Water and Land Resources Division. Seattle,
WA.
LACDPW (Los Angeles County Department of Public Works). 2010a. Los Angeles County Watershed
Model Configuration and CalibrationPart I: Hydrology. Prepared for LACDPW, Watershed
Management Division, Los Angeles County, CA, by Tetra Tech, Pasadena, CA.
LACDPW (Los Angeles County Department of Public Works). 2010b. Los Angeles County Watershed
Model Configuration and CalibrationPart II: Water Quality. Prepared for LACDPW,
Watershed Management Division, Los Angeles County, CA, by Tetra Tech, Pasadena, CA.
LACDPW (Los Angeles County Department of Public Works). 2011. Phase IIReport: Development of
the Framework for Watershed-Scale Optimization Modeling. Prepared for LACDPW, Watershed
Management Division, Los Angeles County, CA, by Tetra Tech, Pasadena, CA.
LARWQCB (Los Angeles Regional Water Quality Control Board). 2002. Santa Monica Bay Beaches
Wet-Weather Bacteria TMDL. LARWQCB, Los Angeles, CA.
LARWQCB (Los Angeles Regional Water Quality Control Board) and USEPA (U.S.
Environmental Protection Agency). 2005a. Total Maximum Daily Load for Metals inBallona
Creek. LARWQCB, Los Angeles, CA, and USEPA Region 9, San Francisco, CA.
LARWQCB (Los Angeles Regional Water Quality Control Board) and USEPA (U.S.
Environmental Protection Agency). 2005b. Total Maximum Daily Load for Metals - Los Angeles
February 2017
39
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
River and Tributaries. LARWQCB, Los Angeles, CA, and USEPA Region 9, San Francisco,
CA.
LARWQCB (Los Angeles Regional Water Quality Control Board). 2006. Total Maximum Daily
Loads for Bacterial Indicator Densities in Ballona Creek, Ballona Estuary, & Sepulveda Channel.
LARWQCB, Los Angeles, CA.
LARWQCB (Los Angeles Regional Water Quality Control Board). 2010. Los Angeles River Watershed
Bacteria Total Maximum Daily Load. LARWQCB, Los Angeles, CA.
LARWQCB (Los Angeles Regional Water Quality Control Board) and USEPA (U.S.
Environmental Protection Agency). 2011. Dominguez Channel and Greater Los Angeles and Long
Beach Harbor Waters Toxics Total Maximum Daily Loads. LARWQCB, Los Angeles, CA, and
USEPA Region 9, San Francisco, CA.
LARWQCB (Los Angeles Regional Water Quality Control Board) 2012. NPDES Permit No.
CAS004001 Waste Discharge Requirements for Municipal Separate Storm Sewer System (MS4)
Discharges within the Coastal Watersheds of Los Angeles County, except those Discharges Originating
from the City of Long Beach MS4. Order No. R4-2012-0175. LARWQCB, Los Angeles,
CA.
LARWQCB (Los Angeles Regional Water Quality Control Board). 2014. Guidelines for Conducting
Reasonable Assurance Analysis in a Watershed Managmeent Program, Including an Enhanced
Watershed Management Program. LARWQCB, Los Angeles, CA.
LP WMA Responsible Agencies (Los Penasquitos Watershed Management Area Responsible
Agencies). 2015. Los Penasquitos Watershed Management Area Water Quality Improvement Plan
and Comprehensive Load Reduction Plan. Prepared for the LP WMA Responsible Agencies by
Amec Foster Wheeler Environment & Infrastructure, Inc., San Diego, CA.
Lumb, A.M., R.B. McCammon, and J.L. Kittle. User's Manualfor an Expert System (HSPEXP) for
Calibration of the Hydrologic Simulation Program -FORTRAN. U.S. Geological Survey, Reston,
VA.
MassDEP (Massachusetts Department of Environmental Protection). 2008. Structural BMP
Specifications for the Massachusetts Stormwater Handbook. Volume 2, Chapter 2. Massachusetts
Department of Environmental Protection, Worcester, MA.
MassDEP (Massachusetts Department of Environmental Protection) and USEPA (U.S.
Environmental Protection Agency). 2007. Total Maximum Daily Load for Nutrients in the Lower
Charles River Basin, Massachusetts - CN 301.0. Prepared for MassDEP, Worcester, MA, and
USEPA Region 1, Boston, MA by Charles River Watershed Association, Weston, MA, and
Numeric Environmental Services, Inc., Beverly Farms, MA.
MassDEP (Massachusetts Department of Environmental Protection) and USEPA (U.S.
Environmental Protection Agency). 2011. Total Maximum Daily Load for Nutrients in the
Upper/Middle Charles River, Massachusetts - CN272.0. MassDEP, Worcester, MA, and
USEPA Region 1, Boston, MA.
40
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
Prince George's County. 2005a. BMP/LID Decision Support System for Watershed-Based Stormwater
Management: Users Guide. Prepared for Prince George's County, Department of
Environmental Resources, MD, by Tetra Tech, Inc., Fairfax, VA.
Prince George's County. 2005b. BMP/LID Decision Support System for Watershed-Based Stormwater
Management: Step-by-Step Application Guide. Prepared for Prince George's County,
Department of Environmental Resources, MD, by Tetra Tech, Inc., Fairfax, VA.
Prince George's County. 2007. Prince George's County BMPDSS Calibration/Validation using Field
Monitoring Data. Prepared for Prince George's County, Department of Environmental
Resources, MD, by Tetra Tech, Inc., Fairfax, VA.
Riverson, J., K. Alvi, J. Zhen, R. Murphy. 2014. SUSTAIN Application User's Guide for EPA Region
10. USEPA - Region 10, Office of Water and Watersheds, Seattle, WA.
Shen, J., A. Parker, and J. Riverson. 2004. A New Approach for a Windows-based Watershed
Modeling System Based on a Database-supporting Architecture. Environmental Modeling and
Software, July 2004.
SDRWQCB (San Diego Regional Water Quality Control Board). 2010. Revised Total Maximum Daily
Loads for Indicator Bacteria Project I- Twenty Beaches and Creeks in the San Diego Region
(Including Tecolote Creek). SDRWQCB, San Diego, CA.
SDRWQCB (San Diego Regional Water Quality Control Board). 2012. Total Maximum Daily Load
for Sedimentation in Los Penasquitos Lagoon. SDRWQCB, San Diego, CA.
SDRWQCB (San Diego Regional Water Quality Control Board). 2015. NPDES Permit No.
CAS0109266 - National Pollutant Discharge Elimination System (NPDES) Permit and Waste
Discharge Requirements for Discharges from the Municipal Separate Storm Sewer Systems (MS4s)
Draining the Watersheds within the San Diego Region. Order No. R9-2013-001 as amended by
Order Nos. R9-2015-0001 and R9-2015-0100. SDRWQCB, San Diego, CA.
SMB JG2/JG3 EWMP Group (Santa Monica Bay Jurisdictional Group 2 and 3 Enhanced
Watershed Management Program Group). 2016. Santa Monica Bay Jurisdictional Group 2 and
3 Enhanced Watershed Management Program. Prepared for the SMB EWMP Group by MWH,
Pasadena, CA.
State of California. 2014. California Water Action Plan. California Natural Resources Agency,
California Department of Food and Agriculture, and the California Environmental
Protection Agency, Sacramento, CA.
Sutherland, R.C. 1995. Methodology for Estimating the Effective Impervious Area of Urban
Watersheds. Watershed Protection Techniques, Vol. 2, No. 1.
SWRCB (California State Water Resources Control Board). 2005. A Process for Addressing Impaired
Waters in Calfifomia. S.B. 469 TMDL Guidance. SWRCB, Sacramento, CA.
SWRCB (California State Water Resources Control Board). 2015. Storm Water Resource Plan
Guidelines. SWRCB, Sacramento, CA.
February 2017
41
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
Tetra Tech and USEPA (U.S. Environmental Protection Agency). 2002. The Loading Simulation
Program in C++ (LSPC) Watershed Modeling System - User's Manual. Tetra Tech, Fairfax, VA,
and U.S. Environmental Protection Agency, Washington, DC.
ULAR WMG (Upper Los Angeles River Watershed Management Group). 2016. Enhanced
Watershed Management Program (EWMP) for the Upper Los Angeles River. Prepared for the
ULAR WMG by Black and Veatch, Los Angeles, CA.
USEPA (U.S. Environmental Protection Agency). 1991. Modeling of Nonpoint Source Water Quality in
Urban and Non-urban Areas. EPA/600/3-91/039. USEPA, Office of Research and
Development, Washington D.C.
USEPA (U.S. Environmental Protection Agency). 1992. Compendium of Watershed-Scale Models for
TMDL Development. EPA841-R-92-002. USEPA, Office of Water, Washington D.C.
USEPA (U.S. Environmental Protection Agency). 1997. Compendium of Tools for Watershed
Assessment and TMDL Development. EPA841-B-97-006. USEPA, Office of Water, Washington
D.C.
USEPA (U.S. Environmental Protection Agency). 1999a. Protocol for Developing Nutrient TMDLs.
EPA841-B-99-007. USEPA, Office of Water, Washington D.C. November 1999.
USEPA (U.S. Environmental Protection Agency). 1999b. Protocol for Developing Sedimnet TMDLs.
EPA841-B-99-004. USEPA, Office of Water, Washington D.C. November 1999.
USEPA (United States Environmental Protection Agency). 2000. BASINS Technical Note 6:
Estimating Hydrology and Hydraulic Parameters for HSPF. EPA-823-R00-012. USEPA, Office of
Water, Washington D.C. July 2000.
USEPA (United States Environmental Protection Agency). 2001. Protocol for Developing Pathogen
TMDLs. EPA-841-R-00-002. USEPA, Office of Water. January 2001.
USEPA (United States Environmental Protection Agency). 2002. Guidance for Quality Assurance
Project Plans for Modeling - EPA QA/G-5M. EPA/240/R-02/007. United States
Environmental Protection Agency. Office of Environmental Information. December 2002.
USEPA (U.S. Environmental Protection Agency). 2003. Fact Sheet: Loading Simulation Program in
C++. USEPA, Watershed and Water Quality Modeling Technical Support Center, Athens,
GA.
USEPA (United States Environmental Protection Agency). 2006. BASINS Technical Note 8: Sediment
Parameter and Calibration Guidance for HSPF. United States Environmental Protection
Agency. Office of Water. January 2006.
USEPA (United States Environmental Protection Agency). 2007. Total Maximum Daily Loads with
Stormwater Sources: A Summary of 17 TMDLs. USEPA, Office of Wetlands, Oceans and
Watersheds, Washington, D.C. July 2007.
USEPA (U.S. Environmental Protection Agency). 2008. Handbook for Developing Watershed Plans to
Restore and Protect Our Waters. EPA 841-B-08-002. USEPA, Office of Water, Washington,
DC.
42
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
USEPA (U.S. Environmental Protection Agency). 2009. SUSTAINA Framework for Placement of
Best Management Practices in Urban Watersheds to Protect Water Quality. EPA/600/R-09/095.
USEPA, Office of Research and Development, Edison, NJ.
USEPA (U.S. Environmental Protection Agency). 2010a. Chesapeake Bay Phase 5.3 Community
Watershed Model. EPA 903S10002 - CBP/TRS-303-10. USEPA, Chesapeake Bay Program
Office, Annapolis MD.
USEPA (U.S. Environmental Protection Agency). 2010b. Chesapeake Bay Total Maximum Daily Load
for Nitrogen, Phosphorus and Sediment. USEPA Region 3, Philadelphia, PA, and USEPA
Region 2, New York, NY.
USEPA (U.S. Environmental Protection Agency). 2010c. Stormwater Best Management Practices (BMP)
Performance Analysis. Prepared for USEPA Region 1, Boston, MA, by Tetra Tech, Inc.,
Fairfax, VA.
USEPA (U.S. Environmental Protection Agency). 2011a. PCB TMDL Handbook. USEPA, Office of
Wetlands, Oceans and Watersheds, Washington, DC. December 2011.
USEPA (U.S. Environmental Protection Agency). 2011b. Operating Instructions for EPA Region l's
BMP Performance Extrapolation Tool for New England. USEPA Region 1, Boston, MA.
USEPA (U.S. Environmental Protection Agency). 2014. Revisions to the November 22, 2002
Memorandum "Establishing Total Maximum Daily Load (TMDL) Wasteload Allocations (WLAs) for
Storm Water Sources and NPDES Permit Requirements Based on Those WLAs". Memorandum
from EPA Office of Wastewater Management and Office of Wetlands, Oceans and
Watersheds to EPA Water Division Directors, Regions 1-10.
USEPA (U.S. Environmental Protection Agency). 2015. Helpful Practices for Addressing Point Sources
and Implementing TMDLs in NPDES Permits. USEPA Region 9, San Francisco, CA. June
2015.
USEPA (U.S. Environmental Protection Agency). 2016a. General Permits for Stormwater Discharges
from Small Municipal Separate Storm Sewer Systems in Massachusetts. USEPA Region 1, Boston,
MA.
USEPA (U.S. Environmental Protection Agency). 2016b. Overview of Methodology to Calculate Baseline
Stormwater Phosphorus Load Reduction Requirements for Charles River Watershed - Final MA MS4
Permit [Memorandum], USEPA Region 1, Boston, MA.
USEPA (U.S. Environmental Protection Agency). 2016c. BMP Accounting and Tracking Tool (BATT)
- User's Guide. Prepared for USEPA Region 1, Boston, MA, by Tetra Tech, Inc., Fairfax,
VA.
USEPA (U.S. Environmental Protection Agency). 2016d. Fact Sheet: Stormwater Management with
Opti-Tool. USEPA Region 1, Boston, MA.
February 2017
43
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Pro gram Planning
USEPA (U.S. Environmental Protection Agency). 2016e. Opti-Toolfor Stormwater and Nutrient
Management- User's Guide. Prepared for USEPA Region 1, Boston, MA, by Tetra Tech, Inc.,
Fairfax, VA.
USEPA (U.S. Environmental Protection Agency) and MassDEP ((Massachusetts Department of
Environmental Protection). 2009. Optimal Stormwater Management Plan Alternatives: A
Demonstration Project in Three Upper Charles River Communities. Prepared for USEPA Region 1,
Boston, MA, and MassDEP, Worcester, MA, by Tetra Tech, Inc., Fairfax, VA.
VDCR (Virginia Department of Conservation and Recreation). 2011. Virginia Stormwater
Management Handbook, 2nd Edition. VDCR, Richmond, VA.
VDCR (Virginia Department of Conservation and Recreation). 2013. NPDES Permit No. VA0088579
Authorization to Discharge under the Virginia Stormwater Management Program and the Virginia
Stormwater Management Act. VDCR, Richmond, VA.
VDEQ (Virginia Department of Environmental Quality). 2015. Guidance Memo No. 15-2005-
Chesapeake Bay TMDL Special Condition Guidance. VDEQ, Richmond, VA.
WDOE (State ofWashington Department of Ecology). 2012. Phase I Municipal Stormwater Permit -
National Pollutant DSischarge Elmination System and State Waste Discharge General Permit for
Discharges from Large and Medium Municipal Separate Storm Sewer Systems. WDOE, Olympia,
WA.
44
February 2017
------- |