Developing Reasonable Assurance:
A Guide to Performing Model-Based
Analysis to Support Municipal Stormwater
Program Planning
Appendices
-------
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
A. EXAMPLE EMERGING MS4 PERMITS AND RAA
REQUIREMENTS
The following sections provide a summary of select MS4 permits throughout the U.S. with varying
requirements for RAAs, including case studies of RAAs performed to address these requirements
(Table A-l).
Table A-1. Appendix A Example MS4 Permits
Section
MS4 Permit
RAA
Approach
Case Study Included
A.1
Los Angeles County Phase I
MS4 Permit
Deterministic
a. Upper Los Angeles River EWMP
b. Santa Monica Bay Jurisdictional Group
2 and 3 EWMP
A.2
Washington Phase I MS4
Permit
Deterministic
Stormwater Retrofit Plan for WRIA 9, King
County, WA
A.3
San Diego Region Phase I
MS4 Permit
Deterministic
Los Penasquitos Watershed Management
Area WQIP and CLRP
A.4
Central Coast California Phase
I and Phase II MS4 Permit
Deterministic
City of Pacific Grove and City of Monterey;
ASBS Low Impact Development Planning and
Future Implementation
A.5
San Francisco Bay Area
Regional MS4 Permit
Deterministic
Tools are described, pilot implementation is
underway.
A.6
Virginia Phase I MS4 Permits
Addressing the Chesapeake
Bay TMDL
Prescriptive
Arlington County Chesapeake Bay TMDL
Action Plan
A.7
Massachusetts General Phase
II MS4 Permits
Prescriptive
Optimal Stormwater Management Plan
Alternatives: A Demonstration Project in
Three Upper Charles River Communities
February 2017
A-l
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
A.1 Los Angeles County Phase I MS4 Permit
Administered by the Los Angeles Regional Water Quality Control Board (LARWQCB), NPDES
Permit No. CAS004001 sets requirements for MS4 discharges of stormwater and non-stormwater by
Los Angeles County Flood Control District (LACFCD), the County of Los Angeles, and 84 cities
within coastal watersheds of Los Angeles County (hereafter referred to as the Los Angeles County
MS4 Permit). This permit covers more than 3,000 square miles of area within Los Angeles County,
including multiple watersheds that each have distinct hydrologic characteristics, beneficial uses, and
water quality impairments (LARWQCB 2012).
The Los Angeles County MS4 Permit incorporates WQBELs to implement wasteload allocations
for 33 TMDLs that address a range of pollutants, including metals, bacteria, toxic pollutants (e.g.,
PCBs, DDT), nutrients, chloride, sediment, and trash. These TMDL wasteload allocations and
associated assumptions and requirements are established as WQBELs in the permit, which are
assigned to those permittees that discharge to the waterbodies addressed by the TMDL. To
demonstrate compliance with WQBELs and receiving water limitations, a permittee has the option
to participate in a Watershed Management Program1 (individually or through collaboration with
other permittees) to identify control measures and BMPs to be implemented over time to meet those
WQBELs and receiving water limitations. A key component of the Watershed Management
Programs is the RAA, for which the permit defines as follows:
"Permittees shall conduct a Reasonable Assurance Analysis for each water body-
pollutant combination addressed by the Watershed Management Program. A
Reasonable Assurance Analysis (RAA) shall be quantitative and performed using a
peer-reviewed model in the public domain. Models to be considered for the RAA,
without exclusion, are the Watershed Management Modeling System (WMMS),
Hydrologic Simulation Program - FORTRAN (HSPF), and the Structural BMP
Prioritization and Analysis Tool (SBPAT). The RAA shall commence with assembly
of all available, relevant subwatershed data collected within the last 10 years,
including land use and pollutant loading data, establishment of quality
assurance/quality control (QA/QC) criteria, QA/QC checks of the data, and
identification of the data set meeting the criteria for use in the analysis. Data on
performance of watershed control measures needed as model input shall be drawn
only from peer-reviewed sources. These data shall be statistically analyzed to
determine the best estimate of performance and the confidence limits on that
estimate for the pollutants to be evaluated. The objective of the RAA shall be to
demonstrate the ability of Watershed Management Programs and EWMPs to ensure
that Permittees' MS4 discharges achieve applicable water quality based effluent
limitations and do not cause or contribute to exceedances of receiving water
limitations."
Prior to adoption of the Los Angeles County Permit, there had been a number of public domain
models developed in the region that supported TMDL source assessments and/or the calculation of
wasteload allocations or were specifically designed to support TMDL implementation and
watershed planning efforts. These models had been developed through a public process, underwent
public and peer reviews, and were publicly available at no cost. The availability and general
acceptance of these modeling systems contributed to the ability of permittees to perform cost-
1 Permittees may also elect to develop an Enhanced Watershed Management Program (EWMP), which are
consistent with Watershed Management Programs but include an evaluation of opportunities for collaboration
among permittees and other partners on multi-benefit regional projects.
A-2
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
effective RAAs to meet permit requirements, as well as aid review of models used to support RAAs
by the permitting authority. Below is a summary of the models referenced in the permit that are
available to support RAAs.
• Hydrologic Simulation Program - FORTRAN (HSPF) (Bicknell et al. 1997) and the
Loading Simulation Program C++ (LSPC), a recoded version of HSPF into C++ (Shen et
al. 2004; Tetra Tech and USEPA 2002; USEPA 2003), were used by the LARWQCB and
EPA to model various watersheds throughout the region to support the development of
TMDLs. The models typically supported the source assessment, linkage analysis, and/or
calculation of load and wasteload allocations for the TMDLs (LARWQCB 2002, 2006, and
2010; LARWQCB and USEPA 2005a, 2005b, and 2011).
• Watershed Management Modeling System (WMMS) was developed by the LACFCD to
support the management of stormwater and improvement of water quality throughout
coastal watersheds of Los Angeles County. WMMS is a public domain platform that
includes a combination of two models: (1) LSPC for dynamic simulation of watershed flows
and pollutant concentrations, and (2) EPA's System for Urban Stormwater Treatment and
Analysis IntegratioN (SUSTAIN) for simulation of BMP processes and optimization and
selection of cost-effective combinations of BMPs to support stormwater management
(LACDPW 2010a, 2010b, and 2011; USEPA 2009).
• Structural BMP Prioritization and Analysis Tool (SBPAT) is a public domain, GIS-based
water quality analysis tool to support the prioritization and selection of BMP project
opportunities in watersheds, including the quantification of the benefits, costs, uncertainties,
and potential risks. SPBAT is a modeling system that utilizes land use event mean
concentrations, EPA-SWMM, data from the EPA/American Society of Civil Engineers
International BMP Database, and a Monte Carlo approach to quantify water quality benefits
and uncertainties (Geosyntec Consultants 2008 and 2012).
To further assist permittee efforts to perform RAAs in the region, the LARWQCB (2014) prepared a
guidance document to provide information and guidelines for RAAs, additional clarification of
permit requirements regarding the RAA, and recommended criteria for permittees to follow in
preparing an appropriate RAA for LARWQCB approval. This guidance specifically addressed
processes for:
1. Identification of water quality priorities to achieved applicable water quality limitations (e.g.,
WQBELs and receiving water limitations);
2. Assessment of current/existing pollutant loading associated with current BMPs;
3. Estimation of required pollutant reductions to meet interim and/or final allowable pollutant
loadings;
4. Selection of strategies, control measures, and BMPs to be implemented to attain needed
reductions;
5. Specification of a schedule of selected watershed management strategies that are sufficient to
result in attainment of corresponding interim and/or final pollutant reductions;
6. Development of a pollutant reduction plan, including:
a. Determination of compliance points to assess MS4 discharges from the area covered
by a Watershed Management Program to the receiving waters;
b. Evaluation of selected management program/BMP performance;
c. Analysis to demonstrate that the selected BMPs cumulatively have reasonable
assurance to meet interim/final requirements;
d. Process for incorporating additional BMPs if water quality milestones are not met as
scheduled.
February 2017
A-3
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
The guidance also provided specific modeling requirements for performing RAAs to support
estimation of current loadings, required load reductions, and analysis of water quality outcomes of
selected BMP options. Within this discussion, the LARWQCB provided a general description of
available models, expectations in terms of model input and output to be reported for the RAA,
critical conditions to be represented by the model for pollutant reduction analyses (90th percentile
flow rates and/or the critical conditions specified in applicable TMDLs), and model calibration
criteria.
The LARWQCB guidance for RAAs is a detailed approach specific to the Los Angeles County
Permit. Development of the guidance was a collaborative effort of permittees, as the LARWQCB
provided generous opportunity for input from permittees on discussions and assumptions included
in the guidance. Much of the discussions within the guidance are meant to provide clarification of
requirements within that permit. Modeling requirements that were documented in the guidance are
meant for experienced modelers, as many of the specifics discussed dealt directly with model input,
output, and statistical approaches for assessing model performance and acceptability of calibration.
However, there are multiple lessons learned from the availability and application of guidance with
the technical specificity provided by the LARWQCB, including the following:
• Information for selection of RAA approaches: Permittees may not have modeling
capability, and all Los Angeles area permittees that chose to follow this compliance
pathway utilized contractors to develop and perform RAAs for the various Watershed
Management Programs. This required the permittees to develop scopes of work, issue
Requests for Proposals, review contractor proposals, and select among qualified contractors
and proposed technical approaches to performing an RAA. The LARWQCB guidance "set
the bar" in terms of the expectation of the RAA, and provided sufficient information to
permittees to guide the selection of RAA technical approaches and contractors.
• Recommended instructions for performing RAAs: With explicit instructions on the permit
expectations and process to be followed in performing RAAs, permittees were provided a
clear roadmap and understanding of the tasks to be completed.
• Clear expectations of LARWQCB to guide approval of RAAs: The guidance provided
permittees a clear understanding of LARWQCB expectations of the RAA and the
documentation needed to accompany the Watershed Management Program. As long as a
permittee followed the guidance, it was provided additional assurance that the LARWQCB
would approve the RAA. Although LARWQCB often provided comments and/or
requested additional information or documentation during its review of draft RAAs, it was
generally found that the guidance served as an important reference throughout the review
process and was a clear indicator of LARWQCB preferences to ensure RAA approval.
Thirty-one Watershed Management Programs and associated RAAs were completed to meet
requirements of the Los Angeles County MS4 Permit and the LARWQCB guidance on RAAs.
These Watershed Management Programs were developed by groups of permittees and individual
permittees, and varied significantly depending on the watershed's water quality priorities,
preferences of the permittees, types of BMPs proposed, and a number of additional factors. The
following provides summaries of RAAs to support Watershed Management Programs for the Upper
Los Angeles River (ULAR WMG 2016) and a group of jurisdictions discharging to Santa Monica
Bay (SMB JG2/JG3 EWMP Group 2016), which varied in terms of models selected to support
RAAs and the pollutants that were addressed.
A-4
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Upper los Angeles River EWMP
An EWMP for the Upper Los Angeles River (ULAR) Watershed Management Area (ULAR EWMP Area)
was developed by the cities of Los Angeles, Alhambra, Burbank, Calabasas, Glendale, Hidden Hills, La
Canada Flintridge, Montebello, Monterey Park, Pasadena, Rosemead, San Fernando, Los Angeles, San
Marino, South Pasadena, and Temple City and the County of Los Angeles/LACFCD. The planning area for
the ULAR EWMP is the largest of all the EWMPs in the Los Angeles region, representing 479 square miles
of watershed. The LSPC and SUSTAIN components of WMMS served as the basis for the modeling system
used to conduct the RAA for the ULAR EWMP (ULAR WMG 2016).
Element 1: The RAA addressed areas identified through discussions with the LARWQCB to be regulated by
the Los Angeles County MS4 Permit. This excluded areas addressed by separate NPDES permits (e.g.,
Caltrans, Industrial General Permit) and federal and state owned lands (e.g., CA Dept. of Fish and Wildlife).
Element 2: The LSPC model of the LA River watershed (available within WMMS) was used to characterize
existing hydrologic and pollutant loading conditions within the watershed. The model provided hourly
simulation of flow and sediment/pollutant concentrations from 1,129 model subwatersheds. This included
modeling of Assessment Areas within and outside of the EWMP area to allow calibration of the model based
on monitoring data collected throughout the watershed (Figure A-l). Model subwatersheds were also
configured to provide separation of municipal jurisdictions to allow characterization of stormwater pollutant
loadings from each jurisdiction. The model was calibrated to meet criteria established by the LARWQCB
guidelines, with results of calibration (Figure A-2 and Table A-2) and characterization of existing pollutant
loads documented within the EWMP. A separate methodology was used to characterize existing non-
stormwater, which typically results from outdoor water use during dry weather conditions. The amount of
non-stormwater generated in each ULAR subwatershed was estimated as the product of (1) the estimated
population based on U.S. census blocks and (2) the estimated per capita outdoor water use.
Element 3: As part of the overall EWMP planning effort, water quality priorities were established based on
WQBELs and receiving water limitations relevant to each of the Assessment Areas (Figure A-l). The
LARWQCB guidelines allow the RAA to be developed with consideration of a "limiting pollutant", or the
pollutant that drives BMP implementation planning (i.e., control measures that address the limiting pollutant
will also address other pollutants). Limiting pollutants were identified in the RAA as zinc and E. coli during
wet weather, and E. coli during dry weather. Based on a critical wet weather condition defined by the RAA
guidelines (90th percentile flow), the LSPC model was used to compare predictions of existing water quality
with numeric targets to determine necessary reductions to attain WQBELs and receiving water limitations
(Figure A-3). For dry weather, the goal of the EWMP was 100% reduction of non-stormwater, which was
quantified during critical drought conditions from late August to September.
Element 4: The EWMP planning efforts identified a number of opportunities for BMPs to be implemented
within each municipal jurisdiction. These included nonstructural programs and structural BMPs, including
low impact development (LID) (e.g., LID ordinance for new and re-development, retrofits of public parcels),
green streets, and regional projects that provide centralized treatment of large drainage areas. The RAA
provided (1) demonstration that pollutant reduction targets will be achieved with BMPs implemented within
each jurisdiction (Figure A-3), and (2) cost-benefit analysis of various BMP scenarios to identify the most
cost-effective implementation plan. Based on discussions with the LARWQCB, nonstructural programs were
assumed to achieve a 5% or 10% pollutant reduction depending on the amount/types of programs to be
implemented within each jurisdiction. For structural BMPs, the RAA provided simulation of stormwater and
non-stormwater flows to the BMPs, and SUSTAIN provided analysis of the volumes reduced to meet
pollutant reduction goals. SUSTAIN also provided cost-optimization of millions of BMP scenarios to
identify the most cost effective combination of BMPs to meet interim and final schedule milestones (Figure
A-4.
Element 5: Results of the RAA were reported in detail using maps, charts, and tables that present for each
jurisdiction the volumes of stormwater to be managed in each subwatershed (Compliance Targets) and the
control measures to achieve those volume reductions (EWMP Implementation Strategy) to meet interim and
final schedule milestones. The resulting Compliance Targets provide a mechanism for tracking and reporting
of progress implementing the EWMP. The EWMP Implementation Strategy provides a framework for
adaptive management and development of more detailed capital improvement planning efforts in the future.
February 2017
A-5
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Sawpit Wash Mo|
I Live Oak Avenue (F194B-R)
m / Santa Anita Wash'
. At longdem Avenue (F193B-R)
'Arcadia Wm^Cb|^|
fe-luw Gr.ind Avi-nue(F317^R)Bj
Eaton Wash Ju wol
AtLoftus Drive (F31B R)
Los Angeles River, (>fl
Above Arroyo Seco (F57C-R)
Waterbody
EWMP Boundary
o Hydrology Gage
¦ Water Quality Gage
Assessment Area
Aliso Wash
Arroyo Seco
Bell Creek
Brown's Canyon Wash
¦I Bull Creek
¦I Burbank Western Channel
Compton Creek
Los Angeles River (Reach 2)
Los Angeles River (Reach 6)
H McCoy-Dry Canyon Creek
Rio Hondo
Tujunga Wash
VerdugoWash
Rio Hondo Above 1
Stuart And Cray Road (F4SB-R)
f Compton Creek \
Near Grcenlcaf Drive (F37B-R),
LOS Angrtr-. River
Mass E meson Station (S10)
. Los Angeles River Below
Wardtow River R(Md (F319-R)
Figure A-1. ULAR EWMP Area, Model Subwatersheds, and Monitoring Stations (ULARWMG 2016).
A-6 February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Modeled (Reach 6013) • Ob*e»
re
T3
c
2
¦
a.
i
a.
Figure A-2. Model Water Quality (total copper) Calibration for Los Angeles River (ULAR WMG 2016).
Table A-2. Summary of Water Quality Calibration Performance for the Upper Los Angeles River Watershed
(ULAR WMG 2016).
Water Quality
Constituent
Sample Count
Model vs. Observed
Load (% Error)
Los Angeles RAA
Guidelines
Performance
Assessment
Total Sediment
80
8.6%
Very Good
Total Copper
54
-20.1%
Good
Total Zinc
54
-27.7%
Fair
Total Lead
49
-32.5%
Fair
E. Coli1
49
-32.1%
Fair
Total Phosphorus
49
-13.1%
Very Good
1 E. coli was assumed to have a 1:1 translator with fecal coliform.
February 2017
A-7
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Identify Control
Represent Measures for Identify
Baseline Metals TMDL Control
Critical and other Water Measures
Conditions Quality Priorities f°r C°H
(U
—
Use LPSC to simulate
watershed rainfall-runoff
for 10-year 90th percentile
critical condition
Use SUSTAIN to identify
control measures to achieve
required reductions based
on zinc control
Use SUSTAIN to determine
additional capacity required
(if any) to retain the critical
bacteria storm
+->
nj
0)
•*-*
0)
£
RAA
1
Required
Reductions
(flow rate, zinc,
bacteria)
LID,
Green Streets,
and Regional BMPs
W
Additional
Regional
BMPs
Unmanaged Areas
(if necessary)
Dry Weather
RAA
1
Use census data and per-
capita outdoor water use
to estimate dry-weather
runoff volume
Route dry-weather runoff
through wet-weatherBMPs
to quantify elimination of
non-stormwater runoff
Determine if additional
non-stormwater control
measures are necessary
-
Dry-Weather
Runoff Volume
Wet-Weather
BMP network
Additional
Non-stormwater
Abatement
(if necessary)
Figure A-3. RAA Process used by the ULAR EWMP to Establish Critical Conditions and Address Water
Quality Priorities (ULAR WMG 2016).
A-8
February 2017
-------
1st: Use cost-optimization to identify BMP solutions to achieve a wide
range of percent pollutant load reductions for each jurisdiction and
each assessment area/watershed.
Los Angeles (Aroyo Seco)
2nd: Determine equitable % pollutant load
reduction needed for each jurisdiction that
will result in RWL attainment (and confirm
using instream validation).
of Regional BMP
to Upstream Jurisdictions)
¦ Regional BMPs (private)
¦ Regional BMPs (Medium)
Regional BMPs (High)
Regional BMPs (Very High)
¦ Green Streets
ill LID (residential)
¦ LID (public)
I LI D (existing/planned)
¦ LID (ordinance)
-Implementation Cost
iad Reduction (limiting pollutant)
3rd: Extract the optimized BMP solution for the required % load
reduction, and it becomes the RAA output and EWMP
implementation plan.
Also, for regional BMPs that have drainage areas in other jurisdictions,
the regional BMP capacity (dotted line) is proportionately allocated to
the upstream jurisdiction's compliance recipe (as such, the regional
BMP and total capacity in the compliance recipe may slightly differ
from the capacities labeled on the cost-optimization curve)
COMPLIANCE
TARGETS:
MEASURABLE AND
ENFORCEABLE
BMP GOAL
For
For
Metals Bacteria
by2028 by 2037
EWMP IMPLEMENTATION PLAN:
APPROACH TO ACHIEVE COMPLIANCE TARGETS.
SUBJECT TO ADAPTIVE MANAGEMENT
(BMP capacityexpressed in units ofacre-feet)
For Metals Attainment by2028
Low-lm pact Development Streets Regional BMPs
For Bacteria
Attainment by
2037
1
3
§ "
•= B
Z &
<
Volume t
(acre-ft)
— o
° s
# o
Ordlnanc
c
CL
~1
a.
cc
to
c
CD
I i2
^ s
M
.2 I.
Private
Total BM
(acre-ft)
Regional
(private)
Total BM
(acre-ft)
640249
22.20
0.62
48%
0.50
-
0.15
0.78
3.36 :
2.99
0.00
0.00
0.00
7.8
0.62
8.4
640349
0.14
0.20
7%
0.14
-
-
0.00
0.00 |
0.00
0.00
0.00
0.00
0.1
020
0.3
640449
0.13
0.05
12%
0.03
-
0.08
0.01
0.00
0.00
0.00
0.00
0.00
0.1
0.05
0.2
640549
1.98
0.23
9%
0.54
0.02
1.37
0.00
13.2J
0.04
0.00
0.00
15.2
023
15.4
640649
5.29
0.08
38%
020
-
-
0.15
4.45
0.00
0.00
0.00
0.00
4.8
0.08
4.9
640749
6.01
0.07
47%
022
0.00
0.24
0.22
2.89
0.00
321
0.00
0 00
6.8
0.07
6.9
640849
1222
0.01
63%
023
-
0.35
0.54
298
0.00
4.82
0.00
0.00
8.9
0.01
8.9
640949
1.40
0.02
37%
0.06
-
0.22
0.01
0.43
0.00
0.10
0.00
0.00
0.8
0.02
0.8
641049
1.94
0.01
48%
0.05
-
008
1.63
0.01
0.42
0.00
000
2.2
0.01
2.2
641149
0.90
0.00
48%
0.02
-
0.02
0.04
0.69 I
0.06
0.00
0.00
0.00
0.8
0.00
0.8
641449
0.02
0.00
64%
0.00
-
0.02
0.00
0.00 |
0.00
0.02
0.00
0.00
0.0
0.00
0.0
641549
1.13
0.01
49%
0.03
-
0.11
0.11
0.00
0.65
0.00
0.00
0.00
0.9
0.01
0.9
641649
0.00
0.00
13%
0.00
-
0.00
0.00
0.00 ;
0.00
0.00
0.00
0.00
0.0
0.00
0.0
642049
0.00
0.00
6%
0.00
-
"
-
0.00 ;
0.00
0.00
0.00
000
0.0
0.00
0.0
Total
53.4
1.3
G
)"
0.0
1.2
3.3
16.4:
I
16.9
8.6
0.0
......
0.0
-------
Santa Monica Bay Jurisdictional Group 2 and 3 EWMP
The Santa Monica Bay (SMB) Jurisdictional Groups 2 and 3 (JG2/JG3) EWMP was developed by the cities
of Los Angeles, Santa Monica, El Segundo, and the County of Los Angeles/LACFCD. The JG2/JG3
planning area for the SMB EWMP covered 39.4 square miles within central region of the SMB watershed,
including the urbanized Dockweiler and Santa Monica subwatersheds, as well as natural open space located
in the Castle Rock, Pulga Canyon, Temescal Canyon, and Santa Monica Canyon subwatersheds (SMB
EWMP Area). The RAA of the SMB EWMP was based on the SBPAT modeling system (SMB JG2/JG3
EWMP Group 2016).
Element 1: Through discussions with the LARWQCB, the EWMP addressed areas regulated by the Los
Angeles County MS4 Permit, excluding federal and state owned lands or areas regulated by other NPDES
permits (e.g., Caltrans, Industrial General Permit) ( ).
Element 2: SBPAT was used characterize existing hydrologic and pollutant loading conditions in the
watershed. SBPAT utilized EPA's Stormwater Management Model (SWMM) to simulate stormwater runoff.
SBPAT linked long-term hydrologic output from SWMM to a stochastic Monte Carlo water quality model to
develop statistical descriptions of stormwater quantity and quality. This approach combines model-predicted
stormwater runoff volumes with land-use based event mean concentrations (EMCs) through a statistical
process to predict existing wet weather pollutant loads. The RAA did not include a quantification of existing
non-stormwater flows or pollutant loads during dry weather. Instead, the EWMP relied on a semi-
quantitative approach to address dry weather that focused on implementation and further monitoring.
Element 3: Parallel to the RAA, EWMP planning efforts established water quality priorities based on
WQBELs and receiving water limitations to address multiple TMDLs and 303(d) impairments for SMB
(DDT, PCBs, debris), adjacent beaches (bacteria), and the Santa Monica Canyon Channel (bacteria and
lead). The RAA for debris was based on demonstration that a number of catch basins will be retrofitted with
screens or inserts to capture debris. Of the water quality priorities identified, only lead and bacteria were
determined to be caused or contributed by MS4 dischargers, requiring an RAA. Based on a critical condition
established by the RAA guidelines, the SBPAT model was used to compare the existing load (90th percentile
flow and concentration) with the target load (90th percentile flow and target concentration) for lead to
determine load reduction goals for wet weather. However, as the target concentration exceeded the critical
concentration, no reduction of lead was determined necessary to address the water quality priority. To
address wet weather bacteria, SBPAT was used to compare existing and target flows and concentrations
during the critical condition to determine wet weather load reductions. To address bacteria during dry
weather, the RAA assumed 100% capture or treatment of non-stormwater discharging from MS4 outfalls that
cause or contribute to the exceedance of water quality objectives for SMB beaches.
Element 4: The EWMP identified multiple opportunities for nonstructural and structural BMPs within each
municipal jurisdiction. The RAA included a process for prioritizing BMPs based on cost, effectiveness, and
feasibility for implementation. Nonstructural BMPs were given highest priority, and addressed in the RAA
based on two methods: (1) a pollutant load reduction of 2.5% to 7.5%was assumed for BMPs that were not
modeled, and (2) SBPAT was used to simulate load reductions associated with public BMP retrofit incentives
(e.g., downspout disconnects) and LID associated with redevelopment. SBPAT was also used to estimate
load reductions associated with a number of green streets, assuming these BMPs were designed to capture
100% of the 85th percentile, 24-hour storm. The RAA included representation of existing, planned, and
proposed regional/centralized BMPs, with SBPAT modeling the BMPs as closely as possible to their actual
or conceptual designs. Figure A-6 provides a summary of the methods used by SBPAT to represent structural
BMPs. No modeling was performed in the RAA to demonstrate dry weather pollutant reductions, rather,
BMPs associated with the diversion, infiltration, or disinfection of non-stormwater was assumed to
completely address pollutant reduction goals.
Element 5: Results of the RAA were reported in tables, charts, and maps that presented the % load
reduction, stormwater volumes to be managed, and capacities of structural BMPs for each BMP type, model
subwatershed, and jurisdiction to meet interim and final schedule milestones. The results of the RAA
provided measurable and quantifiable metrics (e.g., stormwater volumes managed) for future tracking of the
progress of BMP implementation based on adaptive management.
February 2017
A-10
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Figure A-5. SMB EWMP Area and Model Subwatersheds (SMB JG2/JG3 EWMP Group 2016).
February 2017
A-11
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
2) Select Storm S from TMOL
year Y - look up depth of rainfall,
hydrologic response, and BMP
h-ydrologic and hydraulic
performance
1) Continuous simulation
of selected TMDL year Y
3) Estimate pollutant
concentration in Storm S,
from each land use area by
randomly sampling from land
use EMC distributions
6) Estimate BMP
effluent concentration
by randomly sampling
from distributions
5) Apply percent capture and volume loss
to each modeled BMP
-- i
••
4) Calculate total runoff
volume and pollutantload
for each land use; sum to
yield watershed average
concentration for storm S,
J
Catchment
tm 9 •
definition
..
Smallestunit =
• - i
sr«
unique land
»\i
use BMP
vi»
combination
7) Sum bypass and
treated flows to yield _
load, volume and
concentration in Storm
S,
8) Repeat steps 2 through 7 for
all storms sequentially within
TMOL year Y; sum individual
storm results to yield annual
pollutant load for TMOL year Y
9) Repeat steps 2 through 8
for selected TMOl year Y
20.000 times
Figure A-6. Summary of SBPAT Methods to Represent BMPs in the SMB EWMP (SMB JG2/JG3 EWMP Group
2016).
A-12
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
A.2 Washington Phase I MS4 Permit
Issued by the Washington State Department of Ecology (Ecology), the Phase I Municipal
Stormwater Permit for the State of Washington regulates the discharges from MS4s owned or
operated by Clark, King, Pierce and Snohomish Counties; the cities of Seattle and Tacoma and
MS4s owned by public entities located in a Phase I city or county; including the Ports of Seattle and
Tacoma. The MS4 permit requires each of the county permittees to conduct watershed-scale
stormwater planning for one watershed within their jurisdiction. The objective of the watershed-
scale stormwater planning is as follows (WDOE 2012):
".. .identify a stormwater management strategy or strategies that would result in the
hydrologic and water quality conditions that fully support the 'existing uses,' and
designated uses," as those terms are defined in WAC 173-201A-0203, throughout the
stream system."
A critical component of watershed-scale stormwater planning is modeling that demonstrates how
management strategies will result in the necessary hydrologic and water quality conditions to be
attained. This modeling is equivalent to an RAA, and is described in the MS4 permit with the
following excerpts (WDOE 2012):
".. .calibrate a continuous runoff model to the selected watershed to reflect existing
hydrologic, water quality, and biologic (as represented by B-IBI score) conditions."
"...the Permittee will use the model calibrated ...to estimate hydrologic changes
from the historic condition; and predict the future hydrologic, biologic, and water
quality conditions at full build-out under existing or proposed comprehensive land
use management plan(s) for the watershed. Future biologic conditions shall be
estimated by using a correlation of hydrologic metrics with B-IBI4 scores for Puget
Sound Lowland Streams5, or other similar correlation if approved by Ecology.
Future water quality conditions shall be described through estimation of
concentrations of, at a minimum, dissolved copper, dissolved zinc, temperature, and
fecal coliform."
".. .if the estimation .. .predicts water quality standards will not be met, the Permittee
will use the calibrated watershed model to evaluate stormwater management
strategies to meet the standards."
"Share the results of the modeling performed by the Permittee with all other
Permittees in the watershed."
The MS4 permit requires watershed-scale stormwater planning to evaluate the effectiveness of
stormwater management strategies, which must include: changes to development related codes,
rules, standards, and plans; and potential future structural stormwater control projects. However,
management strategies may also include: basin-specific stormwater control requirements for new
and redevelopment; strategies to encourage redevelopment and infill, with an assessment of options
for efficient/effective runoff controls for redevelopment projects (e.g., regional facilities), in lieu of
individual site requirements; or strategies to preserve or improve other factors that influence
3 Washington Administrative Code for water quality standards for surface waters of the State of Washington
4 Benthic Index of Biotic Integrity
5 DeGasperi et al. 2009
February 2017
A-13
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
maintenance of the existing designated uses of the stream (e.g., channel restoration, in-stream
culvert replacement) (WDOE 2012).
To support watershed-scale stormwater planning in Washington, EPA Region 10 (Pacific Northwest
Region) funded four pilot projects in 2010 to provide modeling tools and examples of model
applications that specifically address planning needs in the Puget Sound region. EPA Region 10
provided enhancements to the SUSTAIN model in order to provide direct support in the planning
and cost-optimization of BMPs to address hydrological indicators of ecological health in streams.
Previous versions of SUSTAIN allowed direct comparison to water quality indicators, but
modifications by EPA Region 10 provided direct comparison with duration and peak flows for
various return frequencies, and statistics measuring peak and base flows over a long continuous
period. As a result, SUSTAIN can be used to represent multiple BMPs in the watershed and
evaluate the frequency of exceedance of flow-duration based indicators (Riverson et al. 2014).
100
a, 10
^¦Difference (Area Reduction = 70.6%)
Excluded from Analysis
Post-Developed (Baseline)
Pre-Developed Condition (Target)
BMP Scenario (Optimized)
Target Range (1.025 - 25 cfs)
0)
a
o
CO
0.1
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Percent of time that indicated flow is equaled or exceeded
Figure A-7. EPA Region 10 SUSTAIN Flow Duration Curve Evaluation Factor for BMP Cost-Optimization
(Riverson etal. 2014).
Although each county permittees has identified the watershed for which watershed-scale stormwater
planning will performed, at the time of this investigation those planning efforts were not complete.
However, as part of a project partially funded by EPA Region 10 through the Puget Sound
Watershed Management Assistance Program, King County developed a stormwater retrofit plan
and modeling approach for portions of the Green-Duwamish watershed and Central Puget Sound
watersheds (King County 2013, 2014a, and 2014b). This collaborative effort of King County and
EPA Region 10 serves as an example for how to perform modeling to support watershed-scale
stormwater planning efforts currently underway by county permittees for other watersheds in
Washington. The following case study summarizes the modeling approach used to support the King
County Stormwater Retrofit Plan.
The RAA for the Stormwater Retrofit Plan for Water Resources Inventory Area (WRIA) 9 focused
on modeling to reflect hydrologic, water quality, and biologic health, and improvements that could
result from implementation of stormwater BMPs and LID techniques in existing and future
A-14
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
developed areas. In the Puget Sound Region, critical conditions are generally defined based on
hydrologic metrics used to evaluate Benthic Index of Biotic Integrity, or B-IBI scores as the target. A
B-IBI score aggregates both hydrology and water quality objectives into a single metric that
quantifies benthic health and stream conditions of an existing condition relative to a desired healthy
condition. Using the calibrated watershed model, a "pre-developed" condition is derived by
converting existing land cover to historical conditions (i.e. forested and/or barren land). B-IBI
metrics are then computed for both existing and forested conditions. Examples of the metrics used
for the King County RAA are presented in Table A-3. The metrics derived for the "pre-developed"
scenario become the stormwater improvement goals for the existing condition. The methodology for
addressing the impairment is to identify BMPs that minimize the difference between the existing and
pre-developed condition flow-duration curves.
Table A-3. Hydrologic metrics (B-IBI indicators) applied for the Stormwater Retrofit Plan for Water Resources
Inventory Area 9 (King County 2014a and 2014b)
WRIA 9 Metrics
Description
High Pulse Count (HPC)
The number of times the daily mean flow rate exceeds two
times the long-term mean annual flow rate per water year
High Pulse Range (HPR)
The number of days between the first and the last high pulse
of the water year
PEAK:BASE
Ratio of the peak 2-year return flow to the annual average
winter base flow (i.e. base flow separated from storm flows)
between October and April (Horner 2013)
February 2017
A-15
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Stormwater Retrofit Plan for Water Resources Inventory Area 9
A stormwater retrofit plan was prepared by King County to develop a cost estimate for implementing
stormwater BMPs and LID techniques in existing and future developed areas to improve flow and water
quality and support biological health. The plan addressed 278 square miles of the Green-Duwamish
watershed and portions of the Central Puget Sound watershed that comprise the WRIA 9. Modeling served
as a critical component of the plan, which included a combination of HSPF and SUSTAIN to simulate
hydrology, water quality, and BMP performance, and provide cost-optimization to select cost-effective BMP
and LID strategies throughout the study area to meet water quality and flow-based targets (King County
2013,2014a, and 2014b).
Element 1: As the stormwater retrofit plan was not developed to directly address requirements of the MS4
permit, the plan did not emphasize the separation of MS4 and non-MS4 areas. However, the plan did
exclude portions of the City of Seattle within WRIA 9 as this area is served by a combined sewer and
stormwater system and a combined sewer overflow (CSO) control program was underway within the City.
The plan also excluded portions of WRIA 9 upstream of Howard Hanson Dam as this area is primarily
forested and maintained to protect water supply.
Element 2: An HSPF model of the watershed was configured for hourly simulation of hydrology and
sediment loads for 446 catchments ranging in size from 0.21 to 3,567 acres. The model was configured to
represent existing conditions based on multiple available GIS datasets defining the physical characteristics of
the watershed, including topography, geology, and 2007 land use. The model was calibrated based on
comparison of model output with flow measurements (Figure A-8) and TSS monitoring data collected at
multiple locations in the study area. Model performance was evaluated based on multiple statistical analyses
and comparison to ranges of sediment loading rates reported in literature (King County 2013).
Element 3: Based on spatial projections of future land use, the calibrated HSPF model was reconfigured to
simulate hydrology and sediment loading in the year 2040. Population increase projected by 2040 is expected
to result in the conversion of additional land for urban use, and the redevelopment of previously developed
land for higher density use. SUSTAIN was used to evaluate the management scenarios to meet hydrologic
and water quality indicators. The goals of the SUSTAIN modeling included: (1) optimization of BMPs and
estimation of their effectiveness at reducing flow and sediment loads for hypothetical catchments with unique
combinations of land use and other physical characteristics, (2) scaling of these solutions to the study area for
the 2040 future land use scenario. The BMP solutions were scaled to the study area to achieve maximum
effectiveness, but not beyond conditions for a fully-forested watershed (King County 2014a).
Element 4: SUSTAIN was used to model three management options for a stormwater program, including:
(1) required stormwater management occurring with new and redevelopment, (2) stormwater retrofit of roads
and highways, and (3) stormwater retrofit of non-road unchanged development. Option 1 assumed BMPs are
installed with new and redevelopment (e.g., LID) as outlined in current stormwater guidelines. Option 2
assumed BMPs are installed to capture runoff from local roads and highways (e.g., bioretention and
detention ponds). Option 3 assumed BMPs are installed to control or treat runoff from the remaining
developed area (e.g., cisterns, rain gardens, and detention ponds). Figure A-9 shows an example BMP
routing network used. SUSTAIN provided quantification of the improvement of indicators with incremental
increase of stormwater management for the three options. These scenarios were compared to modeled
existing conditions in 2007 and future conditions in 2040 with no stormwater management, and fully-forested
conditions. Results of the SUSTAIN modeling scenarios were evaluated regarding the amount of BMPs and
their costs and effectiveness at meeting hydrologic and water quality indicators, as compared to scenarios of
fully-forested and existing and future conditions without BMPs. Multiple pollutants and measures of stream
biological health were assessed based on the multiple modeling scenarios, which included comparison of
pollutant loads, percent of time water quality standards were exceeded (Figure A-10), and various metrics for
evaluating B-IBI scores (Figure A-l 1). Model results were also reported in a series of maps that show the
spatial distribution of B-IBI scores for the various model catchments (Figure A-12) (King County 2014a).
Element 5: The plan reported the number, storage volumes, and costs of BMPs for the various model
scenarios for each city/county jurisdiction and catchment in the study area using tables and maps (Figure 4-
7) that provide a usefUl guide for future implementation efforts (King County 2014a and 2014b).
A-l 6
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Seasonal Vea nFo* Rates GaugeUSGS 12112600
Empirical CuWulaliva Distributions for Gauge USGS 12112600
f ,
0 100 200 300
Observed {sfaj
10 10'
Flow Rate (cfs)
Dai y Mean Floiv Rales Gauge USGS 12112SQ0
Annual Volume in Inctias Gai.ge USGS 12112SCO
?oo 4no wm eoo 1000
Observed (c's)
5 10 15 ?0 K 30 35
Observed On}
Figure A-8. Model Hydrology Calibration for Big Soos Creek (King County 2013).
February 2017
A-17
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Agriculture
Light Urban
(LowDtnsfly
RnW«n(MI
* n
Medium Urban
(M«dtunvH»Qh
Demny
I h
~
*
1
Heavy Urban
XoRvncvtbl/lndiAt rijfi
I MM|
m
Pjvwmcnt
O.rrf -J*
I i
Tr»niporUtk>n
B>or«f*ntk>n
Biortltntion
Bior*t«fit ion
Bior#tention
Dior*t*otton
1
1
1
J
1
Detention Storage
Outtot
PervtOUi
II Impervioui too!
h Irnpcrvioi/ypovcmint
Figure A-9. Example Model Configuration and Routing Network for BMPs for the Stormwater Retrofit Plan for
Water Resources Inventory Area 9 (King County 2014a).
2040 No
SWM
2040 Req
SWM
2040 Req
SWM+Roads
2040 Full
SWM
Figure A-1Q. Percent of Time Exceeding Turbidity Standards - Stormwater Retrofit Plan for Water Resources
Inventory Area 9 (King County 2014a).6
6 No SWM = No Stormwater Management; Req SWM = Required Stormwater Management for new and
redevelopment; Req SWM+Roads = Req SWM plus stormwater management for roads and highways; Full
SWM = Full Stormwater Management (Req SWM+Roads plus BMP retrofit for other developed areas). The
ranges of results shown in the figure for each scenario represent the variability of results for the 446 individual
model subwatersheds.
A-18
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
35
Forested 2007 2040 No 2040 Req 2040 Req 2040 Full
Conditions Exist SWM SWM SWM+Roads SWM
Figure A-11. Probability of Improving B-IBI Scores7 above 40% of the Maximum Standards - Stormwater
Retrofit Plan for Water Resources Inventory Area 9 (King County 2014a).
1 High Pulse County (HPC) served as a hydrologic metric to measure B-IBI based on the number of times in a
water year the daily mean flow exceeds twice the long-term daily mean annual flow (Horner 2013). The ranges
of results shown in the figure for each scenario represent the variability of results for the 446 individual model
subwatersheds.
February 2017 A-19
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
pttfe
Washington
2040 Full
Stormwater Management
B-IBI Conditions
| Excellent
I Good
| Fair
] Poor
I Very Poor
I Jurisdictions
Figure A-12. WRIA 9 Study Area Biological Conditions for 2040 with Full Stormwater Management - Based
on the Relationship with HPC and 90% Upper Confidence Limit (King County 2014a)
A-20
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
A.3 San Diego Region Phase I MS4 Permit
Adopted in 2015, the NPDES Permit No. CAS0109266 for discharges from MS4s within the San
Diego Region (San Diego Region MS4 Permit) includes provisions for development of Water
Quality Improvement Plans (WQIPs) for each watershed in the region. The San Diego Region MS4
Permit describes the WQIP as:
.. .an adaptive planning and management process that identifies the highest priority
water quality conditions within a watershed and implements strategies through the
jurisdictional runoff management programs to achieve improvements in the quality
of discharges from the MS4 and receiving waters."
A copermittee has the option to utilize the WQIP to demonstrate compliance with requirements for
discharge prohibitions, receiving water limitations, and WQBELs established in the permit. The
WQBELs were based on the TMDL WLAs. To be eligible for this option, the WQIP must
incorporate multiple planning procedures and meet specific permit requirements, including an RAA
that quantitatively demonstrates that the implementation of water quality improvement strategies
will result in the attainment of the WQBELs within associated implementation schedules
(SDRWQCB 2015).
The San Diego Region MS4 permit includes additional guidance for WQIPs that address WQBELs.
For WQBELs that are not currently met by copermittees, the permit allows a WQIP to include an
analysis, "utilizing a watershed model or other watershed analytical tools," to demonstrate that the
implementation of BMPs will result in the attainment of applicable receiving water or effluent
limitations (SDRWQCB 2015). Prior to the adoption of the San Diego Region Permit in 2015, the
previous MS4 permit for San Diego County MS4s included similar requirements for WQIPs to
address WQBELs. To meet these previous MS4 permit requirements, San Diego County
copermittees have developed WQIPs for multiple watersheds. Most of these WQIPs included RAAs
to address compliance with WQBELs; however, the previous MS4 permit did not include alternative
compliance mechanisms consistent with the new San Diego Region MS4 Permit to enable
permittess to meet discharge prohibitions and receiving water limitations through implementation of
the WQIPs.
Unlike the Los Angeles Region, the San Diego Regional Water Quality Control Board
(SDRWQCB) did not provide detailed descriptions in the San Diego Region MS4 Permit that
defined the expectations of an RAA or specific modeling approaches to be used. Nor did the
Regional Board provide supplemental RAA guidance. This approach allowed flexibility for
permittees to develop technical approaches for RAAs to respond to the permit requirements.
However, lack of guidance from the SDRWQCB also resulted in uncertainty in terms of the
expectations of the SDRWQCB, which was often not realized until WQIPs were completed and
submitted to the SDRWQCB for review. Prior to the development of WQIPs, several copermittees
had developed Comprehensive Load Reduction Plans (CLRPs) to comply with specific
implementation requirements of bacteria TMDLs for multiple beaches and creeks throughout the
region (SDRWQCB 2010) and a sedimentation TMDL for Los Penasquitos Lagoon (SDRWQCB
2012). These CLRPs included modeling approaches that were consistent with RAAs (e.g., LSPC,
SUSTAIN, SBPAT), however, the SDRWQCB did not provide comments on the CLRPs regarding
their sufficiency at demonstrating reasonable assurance for compliance with TMDLs. Regardless, for
those WQIP that benefited from previous CLRPs developed, models used to support the CLRPs
were adapted for RAAs used in the WQIPs.
February 2017
A-21
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
As a result of the above, RAAs performed in the region to support WQIPs varied significantly in the
methods used to interpret WQBELs and report results to demonstrate that BMPs will be sufficient to
provide attainment of applicable receiving water or effluent limitations. Although models used to
support RAAs were typically consistent with those performed in the Los Angeles Region, the use of
the models and reporting of results varied. For example, no criteria were established by the
SDRWQCB regarding the technical expectations of models in simulating existing conditions (e.g.,
calibration statistics). Although the WQIPs document the calibration (and in some cases validation)
of the models used, the metrics for which performance was evaluated vary between WQIPs, as well
as in comparison to WMPs developed in the Los Angeles Region where more specific performance
metrics were established by the LARWQCB.
Eight WQIPs have been completed in the San Diego County to respond to MS4 permit
requirements preceding the 2015 adoption of the San Diego Region MS4 Permit. As the previous
permit emphasized RAAs to justify a WQIP's demonstration of compliance with WQBELs, four of
those WQIPs with applicable WQBELs included robust RAAs. Additional WQIPs (potentially
including RAAs) will be developed by Orange County and Riverside County with their addition to
the San Diego Region MS4 Permit in 2015. Should other copermittees choose to use the WQIP as
an option to demonstrate compliance with discharge prohibitions and/or receiving water limitations
authorized in the 2015 San Diego Region MS4 Permit, other WQIPs may be revised to include
RAAs in the future. An example of the RAA performed to support the Los Penasquitos Watershed
Management Area WQIP and CLRP (LP WMA Responsible Agencies 2015) is discussed on the
following page.
A-22
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
los Penasquitos Watershed Management Area WQIP and CIRP
A WQIP was prepared for the Los Penasquitos Watershed Management Area (WMA) by the cities of San
Diego, Del Mar, Poway, the County of San Diego, and the California Department of Transportation
(Caltrans). The WQIP also serves as a CLRP to demonstrate compliance with specific requirements of the
TMDL for sedimentation in Los Penasquitos Lagoon (SDRWQCB 2012). The WMA addressed by the
WQIP includes 94 square miles of urban land and undeveloped space that drains to the Los Penasquitos
Lagoon. The RAA for the WQIP was based on a combination of models including LSPC and SUSTAIN (LP
WMA Responsible Agencies 2015).
Element 1: Although the San Diego Region MS4 Permit only addresses dischargers from the MS4s, bacteria
and sediment TMDL allocations in the watershed are less clear regarding the distinction of responsibility for
copermittees, other NPDES permits, and nonpoint sources (Figure 4-1). For example, the bacteria TMDL
assigns a single wasteload allocation for the WMA to Phase I MS4s, although non-MS4 areas within the
WMA contribute to bacteria loads. Although Caltrans voluntarily participated in the WQIP, non-MS4 areas
typically do not have responsibility to address the TMDL through reductions of sediment and bacteria. Given
these inconsistencies and lack of clarity on how responsible dischargers are identified in the TMDLs, the
WQIP and associated RAA included the entire WMA in the analysis (including MS4 and non-MS4 areas).
However, pollutant loads for MS4 and non-MS4 areas were differentiated in the RAA to assess load
reduction responsibilities in the WMA. This provided the ability to assess cost implications of copermittees
being held accountable for WMA-wide TMDL pollutant reduction requirements.
Element 2: Prior to the WQIP, an LSPC model of the WMA was available through previous efforts
supporting development of the sediment TMDL and CLRP. The model was refined for use in the RAA, and
included hourly simulation of hydrology and sediment and bacteria loads for multiple model subwatersheds,
including loads from various urban and non-urban land uses, pervious and impervious areas, and instream
sources of sediment resulting from hydromodification. Results of the hydrologic and water quality calibration
were documented in the WQIP, although limited statistical metrics were used to evaluate the performance of
the model as compared to observed water quality data. The model was used to assess relative sediment
(Figure 4-4) and bacteria loads throughout subwatersheds of the WMA to prioritize areas for management.
Element 3: The WQIP established numeric goals to be used to measure progress toward addressing the
highest priority water quality conditions for the WMA: sedimentation and bacteria. The sediment TMDL
established two numeric targets specific to the lagoon and watershed, based on reference conditions related to
the state of the watershed and lagoon in the mid-1970s. For the interpretation of the TMDL watershed target
as a numeric goal for the WQIP, the LSPC model was used to simulate subwatershed sediment loading for
2009 (based on available land use data) and the mid-1970s, and the difference in these loads represented the
sediment load reduction goals for the WQIP. For bacteria, numeric goals were established based on load
reduction calculations performed in the TMDL. These numeric goals, in combination with other quantifiable
goals related to lagoon restoration and the frequency of exceedance of receiving water quality objectives for
indicator bacteria, were summarized for each responsible agency.
Element 4: The RAA was performed based on a combination of LSPC and SUSTAIN to estimate load
reductions associated with BMP implementation strategies within each jurisdiction. The RAA assumed 10%
wet weather reduction resulting from numerous non-modeled nonstructural BMPs, with LSPC simulating
additional reductions associated with catch basin cleaning, downspout disconnection incentive programs,
irrigation runoff reduction, rain barrels, and street sweeping. For each jurisdiction, SUSTAIN was used to
model and provide cost-optimization of wet weather load reductions associated with structural BMPs,
resulting in selection of cost-effective implementation strategies over time to meet interim and final numeric
goals (Figure A-13). Significant dry weather load reductions were assumed to result from nonstructural
BMPs, with additional load reductions expected through the implementation of structural BMPs.
Element 5: Results of the RAA are presented in the WQIP using figures, tables, and schedules that present
the pollutant load reductions associated with the combination of BMPs over time (Figure 4-8). The amount
of structural BMPs required to provide these reductions are not fully documented (e.g., stormwater volumes
managed with individual types of BMPs in each subwatershed), which will put emphasis on annual reports to
present and track ongoing status of BMP implementation and progress towards attainment of numeric goals.
February 2017
A-23
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
60%
° 5
QO O
a 3
c -a
X £
"6 *
f I
II
s*
.t:
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
A.4 Central Coast California Phase I and Phase II MS4 Permit
The State Water Resources Control Board adopted Water Quality Order No. 2013-001-DWQ
NPDES General Permit No. CAS00004, Waste Discharge Requirements for Stormwater
Dischargers from Small Municipal Storm Sewer Systems (Phase II Municipal Stormwater Permit)
on February 5, 2013. The Phase II Municipal Stormwater Permit is applicable to the 38 enrolled
Central Coast Phase II communities. In addition, the Central Coast Regional Water Quality Control
Board adopted Order No. R3-2012-0005 NPDES Permit No. CA0049981 for the Phase I City of
Salinas. As such, all Permittees are required to develop, implement, assess, and report overall
program effectiveness in a way that that quantitatively links stormwater management practices with
water quality improvements in the receiving water. In addition to reporting progress, program
effectiveness assessments inform stormwater quality management decisions to improve managers'
capacity to protect receiving water quality and beneficial uses (Phase II Permit Section E. 14). A
number of the Central Coast Permittees discharge stormwater to a 303d listed receiving waterbody
that either currently has an established TMDL wasteload allocation or may have a formal TMDL
wasteload allocation within the next 5 years, which would trigger TMDL monitoring requirements
(E.13). Tracking and quantification of program effectiveness compliance can address many
additional permit requirements, such as identifying and documenting hydrologic linkages between
the municipal separate storm sewer system (MS4) and the associated receiving waters to which it
discharges (Section E.9); developing inventories of existing programmatic and structural best
management practices (BMPs) implemented and assessing the effectiveness of these BMPs to reduce
pollutant loading to the receiving water (Section E. 11); and establishing a mechanism for practical
information feedback and adaptation actions to optimize resource allocations by prioritizing
maintenance actions and identifying where new BMPs are expected to have the greatest receiving
water benefits (E.14, E.15).
The Central Coast Regional Water Quality Control Board is encouraging an iterative approach to
RAAs by placing practical stormwater planning and accounting tools in the hands of stormwater
managers. A suite of guidance documents and customized stormwater software tools are available to
assist municipalities to more efficiently meet annual MS4 permit requirements. The assessment and
modeling tools are designed for direct use by municipal staff to reduce reliance on high resolution
water quality monitoring data collection, complex software systems, and modelling experts. These
tools were collaboratively developed and continue to be refined with regulatory agencies and
municipal stormwater staff to ensure the needs of both are met. Structural and non-structural BMPs
within the MS4 are assessed regularly by using customized practical monitoring tools that allow
objective performance evaluations for hundreds of structural BMPs fwww.bmpram.com) or
widespread implementation of non-structural BMPs on parcels (www.parcelram.com') or roads (e.g.,
www.tahoeroadram.com'). Anyone with a day of training can become proficient with the
applications and generate precise and meaningful BMP effectiveness data.
Through a shared spatial database, the BMP design and assessment data are inputs to quantify
pollutant load reductions via the stormwater Tool to Estimate Load Reductions (TELR;
www.swtelr.com). TELR is a proprietary, urban hydrology and pollutant loading model purpose-
built to reduce required user expertise, input data needs, and output uncertainty. TELR includes
only the most influential determinants on rainfall-runoff transformation, pollutant generation, and
runoff and pollutant BMP reductions. Simplifying assumptions are used by TELR to reduce the
complexity inherent in water quality and focus stormwater management on the implementation and
maintenance of actions expected to have the greatest water quality benefits. Use of these custom-
built monitoring and modeling tools seamlessly integrates planning and implementation tracking by
turning water quality data into information to evaluate stormwater program effectiveness and
February 2017
A-25
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
identify spatial opportunities for water quality improvements. The online map-based data
management and reporting system generates watershed based results are easy to interpret and
communicate to many. The relative risk of stormwater impacts on receiving water quality is
consistently quantified allowing comparisons of pollutant loading over space and time. The relative
water quality risk determination is informed by regular and continued on-the-ground assessments to
verify that structural and non-structural BMPs are effectively implemented. By incorporating BMP
effectiveness data for actual BMPs on the landscape, the continued commitment to operations and
maintenance to sustain water quality benefits is paramount. Standardization of methods,
terminology, and outputs provide a common language in accessible formats to compare and
aggregate results, greatly improving communication about stormwater quality. The iterative annual
feedback of program progress and current priorities allows stormwater managers and regulators to
adapt management decisions as new information become available. This approach can inform more
detailed research on BMP or program effectiveness and where focused water quality evaluations
may be most valuable.
The following provides a summary of an application of TELR software for the City of Pacific Grove
and Monterey relative to the five basic elements of an RAA.
A-26
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
City of Pacific Grove and City of Monterey: ASBS low Impact
Development Planning and Future Imnlementation
This case study provides a summary of the application the TELR software to a specific municipal planning,
funding, and fUture implementation example; the process, tools and reporting formats are purposely
standardized and scalable to any MS4.
Element 1: The approach is spatially explicit and requires an understanding and documentation of
stormwater routing from the MS4 to receiving waters to which they drain. Online maps serve at the
organizational foundation for data inputs and analysis of results. Each Permittee follows a pragmatic process
to obtain data, assesses actual BMP effectiveness to inform maintenance and track programmatic progress. In
addition, the use of these tools across Central Coast municipalities allows regional comparisons and
collaboration to set and track progress toward achieving TMDLs, stormwater plans, watershed plans, or
other water quality objectives. Figure A-14 presents the catchment delineation and land use distribution for
the Area of Special Biological Significance (ASBS) drainage area.
Element 2: Unmitigated runoff and loading estimates for all MS4 catchments are generated with TELR
using land use characteristics, rainfall, and urban drainage delineations to receiving waters assuming no BMP
implementation. This simplifies the complexity of recreating past landscape development, BMP distribution,
BMP sizing and BMP effectiveness, as well as the uncertainty introduced by making assumptions of such
past conditions. As a precursor to characterizing existing condition, unmitigated results provide a
quantitative baseline from which water quality improvement is measured. Spatial outputs are normalized by
catchment area to show the distribution of relative runoff and pollutant loading problems within the
municipality. Figure A-15 presents TELR automated graphical displays of baseline particulate loading.
Element 3: Each municipality will identify stormwater goals by focusing on achieving and sustaining
effective stormwater and pollutant controls in catchments identified with the greatest relative risk to receiving
water quality. The difference between unmitigated (Figure A-15) and existing conditions (with BMPs)
quantifies the load reduction to receiving waters achieved by BMPs implemented throughout the
municipality. These load reductions provide both spatial and numeric basis to focus the municipality on
implementing water quality improvements that they can affect. Stormwater improvement goals can be
defined by relative runoff and pollutant load reductions from baseline or current conditions using TELR. The
mapped products also provide a watershed context for MS4 discharges to impaired waterbodies to improve
discussions and expectations surrounding TMDL wasteload allocation attainment.
Element 4: TELR planning scenarios allow users to quantify the expected water quality benefits for planned
improvements in priority catchments and to easily communicate these expected benefits to funders,
regulators and the community. In the process of planning to evaluate alternatives and the pursuit of funding
for implementation, the TELR planning scenario can be created to quantify the predicted water quality
benefit of the planned improvements. Figure A-16 display the planned structural BMPs components of the
LID plan (as designed by engineers) within the catchments of the ASBS drainage area. The average annual
particulate (Figure A-16) load reductions predicted are documented for each catchment with a 42.4 T/yr total
load reduction to the ASBS should the project be implemented as designed and BMP maintained. Map
catchment shading in Figure A-16 summarizes the revised catchment prioritization should the LID project be
implemented as designed, illustrating the iterative implementation and planning process.
Element 5: TELR and the supporting assessment tools are made available via an integrated web-based
spatial application (www.2nform.com) and allow quantification of progress based on actual implementation
and effectiveness year after year. Once implementation occurs, BMP RAM is used to inventory and assess
the installed BMPs and these data are used by TELR to quantify the benefits achieved based on both sizing
and observed performance Over time and space the municipality can track and demonstrate that the intended
stormwater improvement goals have been achieved. This iterative approach provides a long-term process by
which priorities are identified, water quality improvement actions are implemented and maintenance is
conducted, progress is quantified and documented, and the new current conditions are then used to inform
the next priorities.
February 2017
A-27
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
^ 1 Milm
0 0.25 0.5
Pacific Gfove ASBS boundary
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
urban stormwater catchments
<3J catchment drainage groupings
particulates: baseline (T/ac/yr)
¦ 0.12-0.13
a o.io-o.ii
¦ 0.08-0.09
J 0.05-0.07
n 0.03 0.04
ASBS particulates totals:
baseline: 71.3 T/yr
Figure A-15. TELR Scenario without BMPs to Generate a Baseline Average Annual Particulate Loads for
Each Catchment.
February 2017
A-29
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
urban stormwater catchments
catchment drainage groupings
f centralized stormwater BMPs
(planned & existing)
decentralized stormwater BMPs
(plan ied] per catchment
particulates (T/ac/yr) reductions (T/yr)
¦ 0.07 0.12 i8
H 0.04 -0.06
¦ 0.03 - 0.04
¦ 0.0? - 0.03 0 2?
¦ ' 0.01 0.02 • <2
• 0
0.5
=M3es
AS6S pofticvtato totals;
basetne: 71.3 T/yr
future: 28.9 T/yr
load reduction: 42.4 T/yr
(59.5% reduced)
:igure A-16. TELR Planning Scenario Results Given the Future Mapped Distribution of Structural BMPs.
A-30
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
A.5 San Francisco Bay Area Regional MS4 Permit and GreenPlan-IT
Issued in 2015, San Francisco Bay Regional Stormwater NPDES Permit (CAS612008) regulates the
discharge of stormwater runoff from MS4s within San Francisco Bay Region that includes Alameda
County, Contra Costa County, Santa Clara County, San Mateo County, and the cities of Fairfield,
Suisun City, and Vallejo. The MS4 permit includes a provision that requires each of the permittees
to develop and implement a Green Infrastructure (GI) Plan within their jurisdiction to include low
impact development drainage design into storm drain infrastructure on public and private lands. The
intention of the plan is described in the permit as:
.serve as an implementation guide and reporting tool during this and subsequent
Permit terms to provide reasonable assurance that urban runoff TMDL wasteload
allocations (e.g., for the San Francisco Bay mercury and PCBs TMDLs) will be met,
and to set goals for reducing, over the long term, the adverse water quality impacts of
urbanization and urban runoff on receiving waters"
The GI Plan must be developed using "a mechanism (e.g., SFEI's GreenPlanIT tool or another tool)
to prioritize and map areas for potential and planned projects, both public and private, on a
drainage-area-specific basis" for implementation by 2020, 2030, and 2040.
The San Francisco regional permit also includes two provisions that require the permittees to
prepare a plan for PCBs and mercury control measure implementation and also conduct an RAA
that demonstrates sufficient control measures will be implemented to attain the PCBs TMDL
wasteload allocations by 2030 and mercury TMDL wasteload allocations by 2028. The permit
provides clear expectations of an RAA but also offers flexibility for permittees to define what
constitutes an acceptable RAA. Los Angeles region guidelines are suggested as a starting point to
conduct RAA for PCBs and mercury for the San Francisco Bay Area in terms of the mechanics of
the analysis, control measure identification, critical condition selection, choice of models, model
calibration criteria, and modeling inputs and outputs. Local watershed characteristics must be
included in the RAA, as well as what has been learned about the distribution, fate, and transport
characteristics of PCBs and mercury.
To support the development of GI Plans and RAAs for PCBs and mercury in the San Francisco Bay
Area, San Francisco Estuary Institute (SFEI), in partnership with San Francisco Estuary Partnership
(SFEP) and a number of City and County agencies and funded by State Water Resources Control
Board (SWRCB) and EPA Region 9, developed a planning tool (GreenPlan-IT) that is specifically
designed to address these planning needs. GreenPlan-IT combines sound science and engineering
principles with GIS analysis and modeling tools to support the cost-effective selection and placement
of GI at the watershed scale and the development of quantitatively-derived watershed master plans
to guide future GI implementation. GreenPlan-IT comprises four standalone tools: (a) a GIS-based
Site Locator Tool that combines the physical properties of different GI types with watershed GIS
information to identify and rank potential GI locations; (b) a Modeling Tool that is built on EPA's
SWMM model to establish baseline conditions and quantify anticipated runoff and pollutant load
reductions from GI sites; (c) an Optimization Tool that uses a cost-benefit analysis to identify the
best combinations of GI types and number of sites within a watershed for achieving flow and/or
load reduction goals; and (d) a tracker tool that is used to track GI implementation and report the
cumulative programmatic outcomes for regulatory compliance and other communication needs.
GreenPlan-IT is currently being applied within a number of city and county areas in the San
Francisco Bay Area to support the development of watershed-scale GI Plans for permit compliance.
February 2017
A-31
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
One example of the tool applications is a case study in the City of San Jose, where the cost-effective
LID scenarios for PCBs control was identified for a 4300-acre priority development area in the
downtown area. The results of the application included a cost-benefit analysis for a range of flow
reduction targets, ranking of sites for specific optimal solutions, and maps showing the distribution
of LID within the focus area under a specific optimal solution portfolio. These results are being used
to identify specific GI projects, support the City's current and future planning efforts, and facilitate
compliance with permit requirements. Similar applications are currently ongoing in the cities of
Oakland, Sunnyvale, and Richmond, as well as in Contra Costa County.
GreenPlan-IT had been developed through a public process with inputs and guidance from
stakeholders and is publicly available for use. The versatility of tool's functions, availability, and
general acceptance of the tool in the region makes it an ideal tool for supporting RAAs. Although, at
present, GreenPlan-IT has not been used for an RAA, the tool meets all the specifications of the Los
Angeles guidance for performing an RAA to demonstrate the compliance with NPDES permits.
Similarly, the Tracker tool is currently being developed and piloted with the City of Richmond with
the express purpose of promoting scalability. The Tracker tool is coupled with the modeling and
optimization tools to ensure internal consistency between the planning phase of GI implementation
and the tracking and crediting phase, an important aspect to ensure high credibility.
A-32
February 2017
-------
A.6 Virginia Phase I MS4 Permits Addressing the Chesapeake Bay
TMDL
Through development of the Chesapeake Bay TMDL for nutrients and sediment, EPA defined
target allocations of nutrient and sediment loads assigned to nineteen major river basins and six
states and the District of Columbia (Bay jurisdictions). Each of the Bay jurisdictions developed a
Phase I Watershed Implementation Plan (WIP) that described how the target allocations would be
achieved. These WIPs included suballocations assigned to wastewater treatment facilities, urban
stormwater, confined animal feeding operations, and nonpoint sources, as well as implementation
strategies. EPA evaluated each WIP to determine if jurisdiction-wide and major river basin target
allocations were met, adequate detail was included to ensure that NPDES permits will be developed
that are consistent with the assumptions of TMDL wasteload allocations, and EPA's expectations of
providing reasonable assurance8 that nonpoint source reductions would be achieved and maintained.
Based on the suballocations and implementation strategies identified in the WIPs, EPA performed
modeling (via HSPF) using the Chesapeake Bay Watershed Model (USEPA 2010a) to evaluate if
jurisdictions met target allocations and if jurisdictional allocations needed to be revised (if pollutant
reductions proposed in WIPs surpassed what was identified by EPA to meet jurisdictional target
allocations) (USEPA 2010b). The resulting Phase I WIPs and associated nutrient and sediment
suballocations assigned MS4 permits form the basis of numeric targets to be incorporated within
those permits for implementing the Chesapeake Bay TMDL.
To address the Chesapeake Bay TMDL, the Phase I WIP for the Commonwealth of Virginia (2010)
identified TMDL allocations by source and river basin. Completed in 2012, the Phase II WIP
provided an opportunity to refine the Phase IWMP, which included an increased commitment to a
phased approach to reducing nutrients and sediment discharging from MS4s (Commonwealth of
Virginia 2012). Multiple Phase I and Phase IIMS4 Permits have since been issued by the Virginia
Department of Environmental Quality (DEQ) that include quantitative pollution reduction
requirements to address the Phase I and Phase II WIPs. These MS4 Permits include a Special
Condition for the Chesapeake Bay TMDL (Special Condition) that requires MS4 operators to
develop a Chesapeake Bay TMDL Action Plan (Action Plan) for submittal to DEQ. To assist MS4s
in the development of Action Plans, DEQ released the Chesapeake Bay TMDL Special Condition
Guidance (Guidance) to support permittees in meeting requirements of the Special Condition, create
consistency in reporting to DEQ, and ensure that compliance and program evaluations are handled
uniformly throughout the Commonwealth (VDEQ 2015). The Guidance states that:
"The Action Plan should provide a review of the current MS4 program, which
demonstrates the permittee's ability to ensure compliance with the Special Condition
and include the means and methods the permittee will use to meet 5.0% of the Level
2 (L2) scoping run reduction for existing development by the end of the first permit
cycle as well as any reductions that may be required for new sources initiating
construction between July 1, 2009 and June 30, 2014 and grandfathered projects that
initiate construction after July 1, 2014. Level 2 implementation equates to an average
8 In the context of the Chesapeake Bay TMDL, "reasonable assurance" was related to sufficient WIP
documentation (by states and the District of Columbia) that nonpoint source controls would be achieved and
maintained and permitting programs would result in point source reductions to meet suballocations identified
in each WIP (USEPA 2010b). This is not equivalent to an RAA in the context of this report, which is related
to MS4 permittee demonstration of reasonable assurance that a watershed or stormwater management plan
will meet MS4 permit requirements and/or numeric goals.
February 2017
A-33
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
reduction of 9.0% of nitrogen loads, 16% of phosphorus loads, and 20% of sediment
loads from impervious regulated acres and 6.0% of nitrogen loads, 7.25% of
phosphorus loads and 8.75% sediment loads from pervious regulated acres beyond
2009 progress loads and beyond urban nutrient management reductions for pervious
regulated acres."
The Guidance outlines multiple steps for completing the Action Plan and methods to be used to
calculate how nutrient and sediment reductions will be met. The methods prescribed are uniform for
all MS4 permits including the Special Condition, and include a simple and straightforward process
for permittees to follow in developing Action Plans. These steps include (VDEQ 2015):
1. Estimate the Size and Extent of the MS4: For development of Action Plans, the permittee
must estimate the size of the MS4 system, eliminating areas regulated by separate MS4
Permits, the general permit or individual permits that address industrial stormwater, forested
lands, agricultural lands, wetlands, and open waters. Permittees must clearly document the
areas within their jurisdiction, and are encouraged to provide maps of MS4 boundaries,
lands served by the MS4, and other lands excluded above. This step is directly related to
Element 1 for an RAA.
2. Estimate the Regulated Urban Impervious and Pervious Acres: Permittees should use best
available judgement, GIS resources, and data to estimate the acreage of regulated pervious
and impervious urban areas served by the MS4 system. Permittees should document the
methodology used to estimate these areas so that DEQ is able to verify an appropriate
method was used. The Guidance also provides links to available aerial imagery that can
support this analysis.
3. Estimate Annual Load of Pollutants Discharged by the MS4 as of2009: Based on results
of Step 2, permittees are to calculate 2009 nutrient and total suspended solids (TSS) loads by
multiplying acreages of regulated pervious and impervious areas by corresponding unit-area
loading rates (lbs/acre/year) listed in the guidance for various watersheds in Virginia. These
"edge-of-field" loading rates were estimated based on previous HSPF model simulations of
the Chesapeake Bay Program Watershed Model Phase 5.3.2 used to support the
development of the nutrient and sediment TMDLs (USEPA 2010). Table A-4 provides an
example calculation sheet and edge-of-field loading rates for urban areas within the Potomac
River Basin, one of four Virginia watersheds addressed by the TMDLs. This step prescribes a
simple approach to address Element 2 for an RAA.
Table A-4. Calculation Sheet for Estimating Existing Source Loads for the Potomac River Basin (VDEQ 2015)
Subsource
Pollutant
Total Existing
Acres Served
by MS4
(06/30/09)
2009 Edge-of-
Field Loading
Rate (Ibs/acre/yr)
Estimated
Pollutant Load
Based on 2009
Progress Run
(Ibs/yr)
Regulated Urban
Impervious
Nitrogen
16.86
Regulated Urban
Pervious
10.07
Regulated Urban
Impervious
Phosphorus
1.62
Regulated Urban
Pervious
0.41
A-34
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Subsource
Pollutant
Total Existing
Acres Served
by MS4
(06/30/09)
2009 Edge-of-
Field Loading
Rate (Ibs/acre/yr)
Estimated
Pollutant Load
Based on 2009
Progress Run
(Ibs/yr)
Regulated Urban
Impervious
Total Suspended
Solids
1,171.32
Regulated Urban
Pervious
175.80
4. Determine Necessary Pollutant Reductions for the First Permit Cycle: Based on the
acreage of regulated pervious and impervious areas estimated in Step 2, permittees are to
calculate nutrient and sediment load reductions required within the first permit cycle. Similar
to Step 3, the Guidance includes unit-area loading rates specific to required load reductions
for various watersheds in Virginia. By multiplying the acreage of regulated pervious and
impervious areas (estimated in Step 2) by these unit-area loading rates, a permittee can easily
determine reductions required by the MS4 for the end of the permit cycle. Table A-5
provides an example sheet and pollutant reduction loading rates for the Potomac River
Basin. Step 4 is the equivalent to Element 3 of an RAA.
Table A-5. Calculation Sheet for Determining Total Pollutant Reductions Required During the Permit Cycle
for the Potomac River Basin (VDEQ 2015)
Subsource
Pollutant
Total Existing
Acres Served
by MS4
(06/30/09)
First Permit
Cycle Required
Reduction
Loading Rate
(Ibs/acre/yr)
Total Reduction
Required First
Permit Cycle
(Ibs/yr)
Regulated Urban
Impervious
Nitrogen
0.08
Regulated Urban
Pervious
0.03
Regulated Urban
Impervious
Phosphorus
0.01
Regulated Urban
Pervious
0.001
Regulated Urban
Impervious
Total Suspended
Solids
11.71
Regulated Urban
Pervious
0.77
5. Calculate Pollutant Reduction Credits: To meet pollutant reductions for the current permit
cycle, permittees should implement those BMPs that are included in the Virginia Stormwater
BMP Clearinghouse ('http://www.vwrrc.vt.edu/swc/) or have been approved by the
Chesapeake Bay Program (Bay Program), as listed in appendices of the Guidance. Based on
methods provided with the Guidance, the permittee is to demonstrate that BMPs
implemented by the permittee will meet pollutant reductions determined in step 4. This is
accomplished based on a two-step process including:
a. Calculate the pollutant load draining to a BMP: This is accomplished based on a
characterization of various land use acreages within a BMP drainage area, in
February 2017
A-35
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
combination with unit-area loading rates assigned for urban (example provided in
Table A-4), forested, and agricultural lands,
b. Calculate BMP Credits: The Guidance outlines methods and assumptions for
calculating pollutant load reductions associated with a number of practices,
including:
• Structural BMPs:
• Land use change: Conversion factors included as
• Urban stream restoration
• Urban nutrient management
• Nutrient trading
• Redevelopment
For example, credits calculated for structural BMPs may be based on (1) BMP
efficiencies reported in the Virginia Stormwater BMP Clearinghouse, or (2) retrofit
performance curves provided the Bay Program. Assumptions for each approach are
included within appendices of the Guidance. Table A-6 provides an example of BMP
pollutant removal efficiencies established by the Bay Program and reported in the
Guidance for use in calculating credits for structural BMPs.
Table A-6. Chesapeake Bay Program BMPs, Established Efficiencies (VDEQ 2015)
Chesapeake Bay Program BMPs
Total
Nitrogen
(%)
Total
Phoshorus
(%)
Wet Ponds and Wetlands
20
45
60
Dry Detention Ponds and Hydrodynamic Structures
5
10
10
Dry Extended Detention Ponds
20
20
60
Infiltration Practices w/o Sand, Veg.
80
85
95
Infiltration Practices w/ Sand, Veg.
85
85
95
Filtering Practices
40
60
80
Bioretention C/D soils, underdrain
25
45
55
Bioretention A/B soils, underdrain
70
75
80
Bioretention A/B soils, no underdrain
80
85
90
Vegetated Open Channels C/D soils, no underdrain
10
10
50
Vegetated Open Channels A/B soils, no underdrain
45
45
70
Bioswale
70
75
80
Permeable Pavement w/o Sand, Veg. C/D soils, underdrain
10
20
55
Permeable Pavement w/o Sand, Veg. A/B soils, underdrain
45
50
70
Permeable Pavement w/o Sand, Veg. A/B soils, no underdrain
75
80
85
Permeable Pavement w/Sand, Veg. C/D soils, underdrain
20
20
55
Permeable Pavement w/Sand, Veg. A/B soils, underdrain
50
50
70
Permeable Pavement w/Sand, Veg. A/B soils, no underdrain
80
80
85
The Guidance provides a detailed and prescriptive methodology for permittees to follow to develop
Action Plans. The result is an "accounting" approach to performing RAAs, utilizing (1) model
A-36
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
results and goals previously established through development of the Chesapeake TMDL and Phase I
and II WIPs, and (2) formally approved assumptions for BMP effectiveness and crediting. The
resulting Action Plans fully address Elements 1 through 4 of an RAA. However, the Action Plans
provide limited information to inform cost-effective implementation and tracking of performance at
meeting pollutant reduction goals (Element 5). For instance, the Guidance assumes that if the BMPs
are implemented as reported in the Action Plans and meet Virginia approved stormwater BMP
standards and specifications (documented by the Virginia Stormwater BMP Clearinghouse), the
BMPs will meet pollutant reduction goals. However, experience has shown that BMP pollutant
removal efficiencies that are often based in literature are not necessarily accurate when compared to
actual implementation. Given local rainfall, soil infiltration, pollutant sources in stormwater, or
considerations in BMP design, the performance can vary greatly and potentially result in an over- or
under-prediction of pollutant reductions. Regardless, given that the Action Plans address the current
permit cycle, and future Action Plans will address interim load reduction targets within subsequent
permit cycles, a permittee will have the opportunity to continuously revisit the assumptions within
past and current Action Plans. This opportunity for iterative and adaptive management can facilitate
an ongoing stormwater management approach to monitor BMPs as they are implemented, benefit
from lessons learned, and modify programs in the future to maximize cost-effectiveness for BMP
implementation to meet load reduction goals over time.
Arlington County was the first municipality to receive a Phase IMS4 Permit (NPDES Permit No.
VA0088579) that included requirements to develop an Action Plan (VDCR 2013). Located within
the Potomac River Basin, the County developed an Action Plan based on Guidance assumptions for
pollutant loading and reduction summarized in Table A-4 and Table A-5, respectively. The
following provides a summary of the Arlington County Action Plan relative to the five basic
elements of an RAA.
February 2017
A-37
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Arlington County Chesapeake Bay TMDI Action Plan
To address the Special Condition of the Arlington County MS4 Permit, Arlington County prepared an
Action Plan that documented how nutrient and sediment reductions would be achieved within the 5-year
permit cycle (2013-2018) to address the Chesapeake Bay TMDL (Arlington County 2015). The Action Plan
was prepared in accordance with methods outlined in the DEQ Guidance (VDEQ 2015).
Element 1: As allowed in the Guidance, the County estimated the physical extent of the MS4 service area.
This included detailed GIS analysis to subtract areas within the County that are addressed by other permitted
facilities (e.g., Virginia Department of Transportation [VDOT]), other state and federal owned land, and
forested and direct drainage to streams (bypassing the MS4). For other permitted facilities, the County opted
to follow a conservative approach to only subtract properties (identified in tax records) or rights-of-way
boundaries (Figure A-17) (the County stated that this approach may be updated in future Action Plans). State
and federal owned land (not permitted) were also identified based on tax records. The County also performed
a GIS analysis of the MS4 system and topography to identify forested areas that drain directly to streams,
non-forested areas that drain directly to streams and do not drain into any portion of the County MS4 (Figure
A-18), and areas that have privately owned storm sewer systems that drain directly to a stream and does not
drain into the County MS4 (Figure A-19). After subtracting these areas, the County MS4 service area was
estimated to be 11,269.4 acres. The County performed further analysis to identify 5,201.3 acres of impervious
area within this MS4 service area.
Element 2: Based on the estimates above for regulated pervious and impervious urban areas within the
County MS4 service area (Element 1), in combination with 2009 edge-of-field loading rates (Table A-4), 2009
loads were estimated to be discharged from the MS4 for nitrogen, phosphorus, and TSS.
Element 3: Consistent with the Chesapeake Bay TMDL and the Phase I and II WIPs, the MS4 permit
requires a 5% reduction of the pollutants of concern within the current permit cycle. As a result, a minimum
5% pollutant reduction served as the planning goal for the Action Plan. To calculate the pollutant loads to be
reduced to meet this goal, the estimates of regulated pervious and impervious urban areas within the County
MS4 service area (Element 1) were multiplied by first permit reduction loading rates (Table A-5).
Element 4: The Action Plan reported load reductions associated with watershed retrofit projects, stream
restoration projects, redevelopment-based reductions, and 2006-2009 "historical BMPs." These "in-place
reductions" represent BMPs implemented and resulting pollutant reductions through 2014. Watershed
retrofit projects included green streets, a municipal facility retrofitted catch basin insert systems, and one
large-scale wetland project. Stream restoration projects included two completed projects, as well as two
projects scheduled for completion within the current permit cycle. A number of historic BMPs were identified
that included bioretention, green roofs, infiltration trenches, pavers, swales, and manufactured filters.
Pollutant reductions associated with watershed retrofit projects, stream restoration projects, and historic
BMPs were estimated based on methods outlined in the Guidance for pollutant removal efficiencies. An
estimate was also performed for the amount of pollutant load reduced through regulated development
activity that occurred in the MS4 service area from 2009 to 2014 and was subject to DEQ's Virginia
Stormwater Management Program (VSMP) Regulations and Arlington's Stormwater Management
Ordinance. A pollutant reduction credit was applied for all regulated development and redevelopment
activity within the MS4 service area that disturbed at least 2,500 square feet of land. These calculations
considered load changes associated with changes in land use/imperviousness, and the incorporation of
BMPs. BMP load reductions were assumed based on removal efficiencies reported by the Guidance and the
Virginia Stormwater Management Handbook (VDCR 2011). The Action Plan summarized the total load
reductions for all practices in place (through 2014) and scheduled for implementation through the current
permit cycle. The 5% load reduction was shown to be met for phosphorus and TSS with BMPs in place, with
all pollutants exceeding the 5% reduction by the end of the permit cycle.
Element 5: The Action Plan includes multiple tables and appendices that list the BMPs in place and to be
implemented and methods used for calculating load reductions. For large scale projects and stream
restoration projects, maps were also provided including drainage areas. Site photographs were also provided
for stream restoration projects. Limited information was provided regarding the locations and designs of
BMPs included for watershed retrofit projects, although such information was not requested by DEQ.
A-38
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Figure A-17. Example of Non-MS4 Permitted Area - Virginia Department of Transportation (Arlington County
2015).
Figure A-18. Example of Non-MS4 Permitted Area - Forested (dark green) and Non-Forested (light green)
Areas that Drain Directly to Streams (Arlington County 2015).
February 2017
A-39
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Figure A-19. Example of Non-MS4 Permitted Area - Privately Owned Storm Sewer System with Direct
Drainage to a Stream (yellow) (Arlington County 2015)
A-40
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
A.7 Massachusetts General Phase II MS4 Permits
In 2016, EPA Region 1 (New England Region) released General Permits for Stormwater Discharges
from Small Municipal Separate Storm Sewer Systems in Massachusetts (General Permits), which
become effective July 1, 2017. The General Permits cover three small MS4s in the Commonwealth
of Massachusetts, including: traditional cities and towns (NPDES Permit No. MAR041000); State,
federal, county, and other publicly owned properties (NPDES Permit No. MAR042000); and State
transportation agencies (except the Massachusetts Department of Transportation - Highway
Division) (NPDES Permit No. MAR043000). The General Permits include new requirements for
discharges to impaired waters with approved TMDLs. For TMDLs that address phosphorus for the
Charles River (MassDEP and USEPA 2007 and 2011) and sixty-six lakes and ponds (MassDEP
1999, 2001, 2002a, 2002b, 2002c, 2002d, 2002e, 2002f, 2002g, 2002h, 2003, 2006, and 2010), the
General Permits include specific requirements for phosphorus reductions from MS4 discharges.
Each permittee that discharges to one of these TMDL waterbodies is required to develop a
Phosphorus Control Plan (PCP) for discharges to Charles River or a Lake Phosphorus Control Plan
(LPCP) for discharges to lakes or ponds. The requirements for PCPs and LPCPs are similar, with
defined schedules for their development (including interim schedules for completion of components
and phases of the PCPs and LPCPs) and implementation of BMPs specified in the plans (with
phased phosphorus reductions over time). Table A-7 summarizes the components, phases, and
schedules of PCPs to address the Charles River TMDL (USEPA 2016a).
Table A-7. PCP Phases, Components, and Schedule to Address Discharges to the Charles River (USEPA
2016a)
Component
Number
Phase 1 of the PCP Component and Milestones
Completion Date
1-1
Legal analysis
2 years after permit
effective date
1-2
Funding source assessment
3 years after permit
effective date
1-3
Define scope of PCP (PCP Area) baseline phosphorus load and
phosphorus reduction requirement and allowable phosphorus
load
3 years after permit
effective date
1-4
Description of Phase 1 planned nonstructural controls
5 years after permit
effective date
1-5
Description of Phase 1 planned structural controls
1-6
Description of operation and maintenance program for structural
controls
1-7
Phase 1 implementation schedule
1-8
Estimated cost for implementing Phase 1 of the PCP
1-9
Complete Written Phase 1 PCP
1-10
Full implementation of nonstructural controls
6 years after permit
effective date
1-11
Performance evaluation
6, and 7 years after
permit effective
date
1-12
1. Performance Evaluation.
2. Full implementation of all structural controls used to
demonstrate that the total phosphorus export rate (Pexp) from the
PCP Area in mass/yr is equal to or less than the applicable
8 years after permit
effective date
February 2017
A-41
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Component
Number
Phase 1 of the PCP Component and Milestones
Completion Date
Allowable Phosphorus Load (Paiiow) plus the applicable
Phosphorus Reduction Requirement (PRR) multiplied by 0.80
Pexp—Pallow'^'(tPRR X 0.80)
1-13
Performance evaluation
9 years after permit
effective date
1-14
1. Performance Evaluation.
2. Full implementation of all structural controls used to
demonstrate that the total phosphorus export rate (Pexp) from the
PCP Area in mass/yr is equal to or less than the applicable
Allowable Phosphorus Load(Paiiow) plus the applicable
Phosphorus Reduction Requirement (PRR) multiplied by 0.75
Pexp—Pallow'*'(tPRR X 0.75)
10 years after
permit effective
date
Component
Number
Phase 2 of the PCP Component and Milestones
Completion Date
2-1
Update legal analysis
As necessary
2-2
Description of Phase 2 planned nonstructural controls
10 years after
permit effective
date
2-3
Description of Phase 2 planned structural controls
2-4
Updated description of operation and maintenance program for
structural controls
2-5
Phase 2 implementation schedule
2-6
Estimated cost for implementing Phase 2 of the PCP
2-7
Complete written Phase 2 Plan
2-8
Performance evaluation
11, and 12 years
after permit
effective date
2-9
1. Performance Evaluation.
2. Full implementation of all structural controls used to
demonstrate that the total phosphorus export rate (Pexp) from the
PCP Area in mass/yr is equal to or less than the applicable
Allowable Phosphorus Load (Paiiow) plus the applicable
Phosphorus Reduction Requirement (PRR) multiplied by 0.65
Pexp—Pallow'^'(tPRR X 0.65)
13 years after
permit effective
date
2-10
Performance evaluation
14 years after
permit effective
date
2-11
1. Performance Evaluation.
2. Full implementation of all structural controls used to
demonstrate that the total phosphorus export rate (Pexp) from the
PCP Area in mass/yr is equal to or less than the applicable
Allowable Phosphorus Load (Paiiow) plus the applicable
Phosphorus Reduction Requirement (PRR) multiplied by 0.50
Pexp—Pallow'^'(tPRR X 0.50)
15 years after
permit effective
date
Component
Number
Phase 3 of the PCP Component and Milestones
Completion Date
3-1
Update legal analysis
As necessary
3-2
Description of Phase 3 planned nonstructural controls
3-3
Description of Phase 3 planned structural controls
A-42
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Component
Number
Phase 1 of the PCP Component and Milestones
Completion Date
3-4
Updated description of operation and maintenance program for
structural controls
15 years after
permit effective
date
3-5
Phase 3 implementation schedule
3-6
Estimated cost for implementing Phase 3 of the PCP
3-7
Complete written Phase 3 Plan
3-8
Performance evaluation
16, and 17 years
after permit
effective date
3-9
1. Performance Evaluation.
2. Full implementation of all structural controls used to
demonstrate that the total phosphorus export rate (Pexp) from the
PCP Area in mass/yr is equal to or less than the applicable
Allowable Phosphorus Load (Paiiow) plus the applicable
Phosphorus Reduction Requirement (PRR) multiplied by 0.30
Pexp—Pallow'^'(tPRR X 0.30)
18 years after
permit effective
date
3-10
Performance evaluation
19 years after
permit effective
date
3-11
1. Performance Evaluation.
2. Full implementation of all structural controls used to
demonstrate that the total phosphorus export rate (Pexp) from the
PCP Area in mass/yr is equal to or less than the applicable
Allowable Phosphorus Load (Paiiow)
P exp—P allow
20 years after
permit effective
date
The General Permits provide detailed descriptions of each of the components of PCPs (Table A-7)
and LPCPs, as well as processes and instructions to follow for their development. A number of the
PCP/LPCP components align directly with the five RAA elements. For example, Component 1-3
for Phase I of the PCP (Table A-7) directly addresses Element 1 for an RAA (Designation of Area
Addressed by Analysis). To perform Component 1-3 of the PCP, the General Permit states that
(USEPA 2016a):
"The permittee shall indicate the area in which it plans to implement the PCP. The
permittee must choose one of the following: (1) to implement its PCP in the entire
area within its jurisdiction (for municipalities this would be the municipal boundary)
within the Charles River Watershed; or (2) to implement its PCP only in the
urbanized area portion of the permittee's jurisdiction within the Charles River
Watershed. The implementation area selected by the permittee is known as the "PCP
Area" for that permittee."
Component 1-3 also addresses Element 2 (Characterization of Existing Conditions) and Element 3
(Determination of Stormwater Improvement Goals) for an RAA. The General Permits include
tables designating phosphorus load reduction requirements for each permittee with MS4 discharges
to Charles River or lakes and ponds addressed by the TMDLs. These load reduction requirements
vary depending on the option selected by each permittee to designate the PMP Area. For example,
for a subset of permittees (Bellingham, Franklin, and Milford) in the Charles River watershed that
select option 2 for designation of the PCP area (confined to the urbanized area portion of the
permittee's jurisdiction), the General Permits specify the baseline phosphorus load, phosphorus
February 2017
A-43
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
reduction requirement (load reduced and percent reduction), and allowable phosphorus load (Table
A-8 ) to be uses for their PCPs. These loading rates are based on methods used in development of the
Charles River TMDL for phosphorus (MassDEP and USEPA 201 IB), with further refinements for
incorporation in the General Permits to account for: (1) revised hydrologic soil groups, and (2)
phosphorus reductions assumed to result from proper use of fertilizers and turf grasses in accordance
with Massachusetts Regulation (USEPA 2016b). Based on these loading rates, a permittee can
characterize phosphorus loadings associated with existing conditions (Element 2), and determine
phosphorus reductions associated with stormwater (Element 3) to set planning goals for the PCP. It
should be noted that the phosphorus loads reported in the General Permits are based on land use
data from 2005, which was the year chosen as the baseline land use for the permit. Permittees can
develop updated assumptions and maps for land uses in their jurisdiction and submit the
information in their annual report for year four of the permit. This information can then be used to
update phosphorus loads for permittees in future permits (USEPA 2016a).
Table A-8. Phosphorus Baseline Load, Reduction Requirement, Allowable Load, and Percent Reduction for
Selected Permittees in the Charles River Watershed (USEPA 2016a)
Community
Baseline
Phosphorus
Load (kg/yr)
Stormwater
Phosphorus
Load Reduction
Requirement
(kg/yr)
Allowable
Phosphorus
Load (kg/yr)
Stormwater
Percent
Reduction in
Phosphorus
Load (%)
Bellingham
947
331
616
35%
Franklin
2,344
818
1,526
35%
Milford
1,486
653
833
44%
The Performance Evaluations associated with Phase 1 Component 1-11 through 1-14, Phase 2
Component 2-8 through 2-11, and Phase 3 Component 3-8 through 3-11 (Table A-7) provide the
equivalent to Element 4 for an RAA (Demonstration that Management Actions will Result in
Attainment of Goals). The General Permits outline processes for performing evaluations of the
effectiveness of the PCP and phosphorus load reductions achieved through the implementation of
structural and non-structural BMPs and tracking increases of phosphorus loads resulting from
development. For non-structural BMPs, the General Permits provides easy-to-use equations for
calculating load reduction "credits" associated with enhanced street sweeping programs, catch basin
cleaning, and organic waste and leaf litter collection programs. As an example, the equation for
calculating phosphorus load reduction credits for street sweeping program is as follows (USEPA
2016a):
Credit sweeping IA swept X PLE IC-land use X PRF
sweeping x AF
where:
Credit sweeping = Amount of phosphorus load removed by enhanced sweeping program
(lb/year),
IA swept = Area of impervious surface that is swept under the enhanced sweeping
program (acres),
PLE IC-land use = Phosphorus Load Export Rate for impervious cover and specified land use
(lb/acre/yr) (provided in table in the General Permits),
A-44
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
AF = Annual Frequency of sweeping. For example, if sweeping does not occur in
Dec/Jan/Feb, the AF would be 9 mo./12 mo. = 0.75. For year-round
sweeping, AF=1.0.
The General Permits also includes a detailed methodology for permittees to use to determine design
storm volume capacities and to calculate phosphorus load reductions for a number of structural
stormwater BMPs (e.g., infiltration basin, bio-filtration practice, porous pavement, wet or dry ponds)
and semi-structural/non-structural BMPs (e.g., impervious area disconnection, conversion of
impervious area to permeable pervious area). The assumptions associated with this methodology
have evolved over years of research performed by EPA Region 1 on the effectiveness of these BMPs
at reducing phosphorus loads. For instance, EPA Region 1 performed modeling to estimate the long-
term cumulative efficiencies of several types of BMPs based on their size and other design
considerations. This modeling utilized typical designs for several types of structural BMPs developed
by the Massachusetts Department of Environmental Protection (MassDEP) (2008). An example
typical design of a water quality swale is shown in Figure A-20. This modeling, based on EPA's
Storm Water Management Model (SWMM) (Huber and Dickenson 1988) and a BMP Decision
Support System (BMPDSS) (Prince George's County 2005a, 2005b, and 2007), was used to provide:
1. Selection of appropriate long-term rainfall record (data and location) that is representative of
major urbanized areas within the New England region.
2. Simulation of hydrograph and pollutant time series using a land-based hydrologic and water
quality model.
3. Simulation of BMP hydraulic and treatment processes in BMP models.
4. Creation of BMP performance curves on the basis of BMP model simulation results.
The modeling included extensive calibration based on local monitoring data to ensure that results
were representative of hydrology, pollutant loading, and BMP performance observed within the
New England region (USEPA 2010c). The resulting BMP performance curves and companion tables
for long-term phosphorus and runoff volume reduction were included within the General Permits to
support the methodology for PCP performance evaluations. An example BMP performance curve
reported in the General Permits for an infiltration trench (with a soil infiltration rate of 0.17 in/hr) is
shown in Figure A-21. These curves provide permittees access to BMP performance estimates that
are based on modeling and tailored to conditions and characteristics of the New England region.
Based on estimates of land use and imperviousness within a BMP drainage area, a permittee can use
phosphorus export rates (lbs/acre/year) reported in the General Permits to estimate the phosphorus
loading to a planned BMP. The permittee can use the BMP performance curves to determine the
pollutant removal associated with a planned BMP, or size the BMP to meet a phosphorus reduction
target (USEPA 2016a).
February 2017
A-45
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
v ftOA0M*r .
PLAN VIEW
PROFILE
Figure A-20. Typical Design for a Water Quality Swale (MassDEP 2008)
To further assist permittees in the development of PCPs and the planning of BMPs, EPA Region 1
has developed multiple tools to calculate the phosphorus load reductions associated with various
str uctural and nonstructural BMPs. For example, prior to release of the General Permits, EPA
Region 1 developed the BMP Performance Extrapolation Tool (BMP-PET). The BMP-PET allowed
a user to enter information on drainage area and BMP characteristics, and the tool accessed the
previously developed BMP performance curves to estimate pollutant removal efficiency or runoff
volume reduction. Alternatively, a user could enter a target BMP removal efficiency and BMP-PET
would estimate the corresponding required BMP size and runoff volume reduction to meet the target
(USEPA 2011b). After release of the General Permits, EPA Region 1 developed the BMP
Accounting and Tracking Tool (BATT), which builds off of BMP-PET to provide tracking of
structural BMP performance in terms of phosphorus reduction (based on BMP performance curves),
A-46
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
with added capability to estimate phosphorus load reductions associated with nonstructural BMPs
and land use conversion (consistent with methods outlined in the General Permits) and generate
reports that summarize all implementation activities for a permittee (Figure A-22). The BATT
reduces the burden on permittees in interpreting the methods outlined in the General Permits (e.g.,
BMP performance curves, equations associated with nonstructural BMPs), providing efficiency in
BMP tracking, performance evaluations, and reporting associated with PCPs (USEPA 2016c).
BMP Performance Curve: Infiltration Trench
(Soil infiltration rate 0.17 in/hr)
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Physical Storage Capacity: Depth of Runoff from Impervious Area
(inches)
•Total Phosphorus
¦Volume
Figure A-21. BMP Performance Curve: Infiltration Trench (USEPA 2016a).
February 2017
A-47
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
BMP AcxounViiK} and Tiactanq Tool (0ATT>
kJd
feT%
BMP Accounting, Tracking, and Reporting Tool
for EPA Region 1 (BATT)
Figure A-22, User Interface for BATT (USEPA 2016c).
To further support permittee efforts to prepare PCPs, EPA Region 1 also developed case studies to
serve as examples for using modeling and cost-optimization techniques for the planning of BMPs,
assessing potential costs for meeting phosphorus reductions through stormwater management, and
the feasibility of establishing funding sources to support the needed stormwater management. Prior
to release of the General Permits and in anticipation of the new requirements for PCPs, EPA Region
1 developed an initial case study applied to three communities in the Upper Charles River watershed
subject to the nutrient TMDL: Bellingham, Franklin, and Milford (Table A-8) (MassDEP and
USEPA 2011B, USEPA and MassDEP 2009). As no PCPs have been completed at the time of this
investigation, no case study could be included within this report to demonstrate methods used by a
permittee to meet requirements of the General Permits. However, additional discussion of the EPA
Region 1 case study and its relevance to the five elements of an RAA are provided in the following
example. It is anticipated that PCPs will have similarities to this case study, however, optimization
is not a required component of PCPs. Regardless, EPA Region 1 has recognized that it will likely be
in a permittee's interest to utilize optimization techniques to assist in the planning of BMPs to be
incorporated within PCPs, and provide assurance that those BMPs are most cost-effective at meeting
phased phosphorus load reduction targets over time. To further support permittee efforts in
developing PCPs, EPA Region 1 has developed a spreadsheet tool that utilizes BMP performance
curves included within the General Permits, as well as optimization techniques available within
EPA's SUSTAIN, to provide an easy-to-use methodology to assist in developing technically sound,
cost-effective PCPs. Released in 2016, this "Opti-Tool" is available as a public-domain tool for all
permittees and other agencies in the New England region to support PCPs and stormwater planning
efforts (USEPA 2016d and 2016e).
EPA Region 1 supported an additional case study to assess: (1) the potential cost of meeting TMDL
phosphorus reduction targets through public and private sector stormwater management, (2) the
A-48 February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
feasibility of establishing a funding sources to serve the three example communities previously
investigated for cost-effective BMP planning (Bellingham, Franklin, and Milford) (USEPA 201 IB).
Findings from this case study informed development of the General Permits, and provides a useful
reference for permittees to support selection of cost-effective BMPs and the Funding Source
Assessment component of the PCP (Phase I, Component 1-2) (Table A-7).
February 2017
A-49
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Optimal Stormwater Management Plan Alternatives: A Demonstration
Project in Three Unner Charles River Communities
Prior to the release of the General Permits and the associated requirements for PCPs, EPA Region 1
performed a study to provide a demonstration and guide to communities in the implementation of the
phosphorus TMDL for the Upper Charles River (MassDEP and USEPA 2011). The objectives of the study
included development of optimized, planning-level-scale stormwater management alternatives for the
communities of Bellingham, Franklin, and Milford, MA. Models used to support the analysis included
EPA's SWMM for simulation of stormwater and phosphorus loads, and BMPDSS for simulation of BMP
phosphorus reductions and cost-optimization for BMP selection (USEPA 2009b, MassDEP 2009).
Element 1: The study included a comprehensive GIS analysis to identify urban and nonurban land uses,
pervious/impervious areas, and hydrologic soil groups within each community jurisdiction. The resulting
combination of land characteristics led to the development of unique hydrologic response units (HRUs) to
support pollutant loading analysis and selection of BMPs to treat runoff. The HRUs allowed the modeling
and pollutant reduction analysis to focus on urban areas within each community.
Element 2: Based on land use and pervious/impervious areas identified in Element 1, phosphorus unit-area
loading rates (kg/ha/yr) were applied to calculate annual average phosphorus loadings from each
community to the Upper Charles River. SWMM was used to provide ten-year, hourly simulation of
phosphorus loads for each HRU, with calibration of SWMM buildup and washoff parameters performed to
closely match annual loading estimates based on unit-area loading rates. As a result, SWMM could be
applied to represent existing phosphorus loading conditions within each proposed BMP drainage area, based
on the combinations of HRUs within those areas.
Element 3: Phosphorus reduction targets for the three communities were based on TMDL reduction targets
(MassDEP and USEPA 2011), prior to adjustment of those targets during preparation of the General Permits
(Table A-8) (USEPA 2016a and 2016b). As a result, reduction targets used in the study for Bellingham,
Franklin, and Milford were assumed to be 52%, 52%, and 57%, respectively. It should be noted that future
PCPs developed for these communities will based reductions on those summarized in Table A-8, as reported
in the General Permits.
Element 4: The study assigned different combinations of BMPs based on 28 management categories that
aggregated areas sharing similar site conditions and suitability for implementing the same type of BMPs. All
BMP site condition requirements and design assumptions were consistent with the Massachusetts
Stormwater Handbook (MassDEP 2008). Figure A-20 shows an example typical design for a water quality
swale modeled in the study. Three model scenarios were investigated that considered runoff from impervious
or pervious HRUs (modeled by SWMM), routed to a system of BMPs for the different management
categories (modeled by BMPDSS), with untreated runoff and overflow from BMPs routed to outlets for the
communities (Figure A-23). These scenarios included: (1) runoff from only impervious HRUs within parcels
is routed to BMPs; (2) consistent with scenario 1, except impervious runoff from the public right-of-way
parcels is treated at downstream centralized facilities (neighborhood BMPs identified with input from the
communities); and (3) consistent with scenario 2, except impervious and pervious runoff from non-public
right-of-way parcels is treated by onsite BMPs. An example model configuration of scenario 1 is depicted in
Figure A-23. The goal of the optimization was to identify the most cost-effective configuration of BMPs for
each scenario to meet TMDL phosphorus load reductions identified in Element 3. BMPDSS provided
simulation of thousands of alternative BMP configurations for each of the three scenarios and predicted
associated phosphorus load reductions and BMP implementation costs. These BMPDSS results were plotted
to demonstrate increasing phosphorus reductions verses costs, from which a cost-effectiveness curve can be
investigated to identify the most cost-effective solution. Figure A-24 provides an example of optimization
results for scenario 1 in Bellingham.
Element 5: The study presented results of the modeling and optimization in various forms, ranging from
summaries of impervious area treated with BMPs and associated total costs, costs per acre, and costs per
pound of phosphorus removed, to more detailed tabulation of the different types of BMPs and associated
level of treatment (area of BMP and runoff depth treated) within key urban and non-urban land use areas.
A-50
February 2017
-------
Guide to Performing Model-Based Analysis to Support Municipal Stormwater Program Planning
Parcels
n
~
HRUs iiihI management
categories IBMPs)
0 ~
BMP-l
EM
HRUl-Impv
"=> n-
HRU2-Iinpv
MC,
777
W//W///W/
¦fss/sfs///u/s
¦S//S/JS////SSS
//S/S//////S//S/S//////S/S
MC.
>
BMP-l
~rj~r'
HRUi-Loipv
MC,
HRUm-Impv
'SSSS
'////
//S/////S/////
'//////////////
'//////////////
\J
MC.
Specific HRl's
Specific management
categories (BMPs)
HRUl-Imuv
BMP-l
HRUl-Impv
BMP-2
HRU 1-Imp v
BMP-X
....
.... —
HRUru-Iiupv
BMP-l
HRUm-Impv
BMP-2
HRUm-Impv
BMP-X
Virtual
outlet
All pervious HRUs
Less likely for onsite
BMP-Iiupv
Tabulation of HRU to management category
Scenario I setap
Figure A-23. Example Model Configuration for a Combination of HRUs and Management Categories in Upper
Charles River Communities (USEPA and MassDEP 2009).
eo
50
«* 40
c
o
r 30
------- |