EPA
United States
Environmental Protection
Agency
EPA/600/R-17/469F I May 2018 I www.epa.gov/research
Improving the Resilience of Best Management
Practices in a Changing Environment:
Urban Stormwater Modeling Studies
Office of Research and Development
Washington, D.C.
-------
EPA/600/R- 17/469F
May, 2018
www.epa.gov/research
Improving the Resilience of Best
Management Practices in a Changing
Environment: Urban Stormwater Modeling
Studies
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
1
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
ABSTRACT
The United States Environmental Protection Agency (U.S. EPA) identified a need for improved
understanding of the potential impacts of changes in long term weather conditions on the occurrence and
management of stormwater runoff (U.S. EPA, 2008; 2010; 2012). Accordingly, we conducted continuous
simulation modeling of the hydrology, hydraulics, and water quality discharged from a series of
conceptual development sites using a variety of conventional (or "gray") and green infrastructure (GI)
practices consistent with local stormwater and site design regulations. We assessed the performance of
green and gray stormwater controls under current conditions and a range of potential changes in
precipitation and temperature, and examined how designs could be adapted accordingly. The stormwater
management scenarios covered five types of developed land use in five geographic locations representing
different hydroclimatic regimes throughout the United States. The results and conclusions of the study are
applicable to both new development, redevelopment, and stormwater retrofits.
Preferred citation:
U.S. EPA (Environmental Protection Agency). 2018. Improving the resilience of BMPs in a changing environment:
urban stormwater modeling studies. Office of Research and Development, Washington, DC; EPA/600/R-17/469F.
Available online at http://www.epa.gov/research.
li
-------
TABLE OF CONTENTS
LIST OF TABLES vi
LIST OF FIGURES viii
LIST 01 ABBREVIATIONS xii
AUTHORS, CONTRIBUTORS, AND REVIEWERS xiii
EXECUTIVE SUMMARY xiv
1. INTRODUCTION AND OVERVIEW 1
2. STUDY APPROACH 6
2.1. REGIONS AND LAND USES 6
2.2. STORMWATER MANAGEMENT APPROACHES 8
23. ADAPTATION SIMULATION 10
2.3.1. Analytical Design 10
2.3.2. Best Management Practice (BMP) Performance Measures 11
2.3.2.1. Annual Average Runoff Volume 12
2.3.2.2. Flow Duration Curve 12
2.3.2.3. Annual Average Pollutant Load 14
2.4. HIGH PRECIPITATION SCENARIO 14
2.4.1. High Intensity Change Scenarios 14
2.4.2. Low, Medium, and High Intensity Change Scenarios Used for
Sensitivity Analysis 17
2.4.3. Percentage Difference Scenarios Used for Sensitivity Analysis 17
2.5. SITE ASSUMPTIONS 18
2.5.1. Stormwater Conveyance Representation 18
2.5.1.1. Peak Flow Estimation 18
2.5.1.2. Culvert Sizing 20
2.5.2. Infrastructure Cost Estimates 21
3. MID-ATI.ANTIC SITE: MIXED USE 25
3.1. REGULATORY REQUIREMENTS AFFECTING STORMWATER
MANAGEMENT 25
3.2. STORMWATER MANAGEMENT SCENARIOS 25
3.2.1. Conventional (Gray) Infrastructure 27
3.2.2. Green Infrastructure (GI) with Gray Infrastructure 29
3.2.3. Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) 31
3.3. ADAPTATION SIMULATION 31
3.4. CURRENT AND FUTURE CHANGES IN PRECIPITATION 32
3.5. RESULTS 34
4. MIDWEST SITE: RESIDENTIAL 45
4.1. REGULATORY REQUIREMENTS AFFECTING STORMWATER
MANAGEMENT 45
4.2. STORMWATER MANAGEMENT SCENARIOS 45
4.2.1. Conventional (Gray) Infrastructure 47
4.2.2. Green Infrastructure (GI) with Gray Infrastructure 49
4.2.3. Green Infrastructure (GI) Only 50
in
-------
4,2,4. Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) 52
4.3. ADAPTATION SIMULATION 53
4.4. CURRENT AND FUTURE CHANGES IN PRECIPITATION 53
4.5. RESULTS 57
5. ARID SOUTHWEST SITE: COMMERCIAL 69
5.1. REGULATORY REQUIREMENTS AFFECTING STORMWATER
MANAGEMENT 69
5.2. STORMWATER MANAGEMENT SCENARIOS 69
5.2.1. Conventional (Gray) Infrastructure 70
5.2.2. Green Infrastructure (GI) Only 71
5.3. ADAPTATION SIMULATION 73
5.4. CURRENT AND FUTURE CHANGES IN PRECIPITATION 74
5.5. RESULTS 76
6. SOUTHEAST SITE: ULTRA-URBAN 83
6.1. REGULATORY REQUIREMENTS AFFECTING STORMWATER
MANAGEMENT 83
6.2. STORMWATER MANAGEMENT SCENARIOS 83
6.2.1. Conventional (Gray) Infrastructure 84
6.2.2. Green Infrastructure (GI) with Gray Infrastructure 86
6.3. ADAPTATION SIMULATION 88
6.4. CURRENT AND FUTURE CHANGES IN PRECIPITATION 88
6.5. RESULTS 90
7. PACIFIC NORTHWEST SITE: TRANSPORTATION CORRIDOR/GREEN
STREET 97
7.1. REGULATORY REQUIREMENTS AFFECTING STORMWATER
MANAGEMENT 97
7.2. STORMWATER MANAGEMENT SCENARIOS 97
7.2.1. Green Infrastructure (GI) Only 99
7.3. ADAPTATION SIMULATION 100
7.4. CURRENT AND FUTURE CHANGES IN PRECIPITATION 100
7.5. RESULTS 103
8. DISCUSSION AND CONCLUSIONS 110
8.1. STUDY QUESTION #1 112
8.1.1. Performance Comparison for Stormwater Management Scenarios
Under Current and Future Precipitation Conditions 112
8.1.1.1. Changes in Pretreatment Site Performance 112
8.1.1.2. Changes in Post-treatment Site Performance 116
8.1.2. Sensitivity Analyses to Precipitation Events 126
8.1.2.1. Sensitivity Analysis* Modeled Scenarios 127
8.1.2.2. Sensitivity Analysis* Percentage Change Scenarios 129
8.2. STUDY QUESTION #2 134
8.2.1. Adapting Best Management Practices (BMPs) for Heavy
Precipitation 134
8.2.2. Limiting Factors for Adaptation Optimizations 136
8.3. STUDY QUESTION #3 139
IV
-------
8.3.1. Stormwater Infrastructure Cost 139
8.3.2. Resizing Current Practices versus Adding Green Infrastructure (GI)
to Site 142
8.3.3. Increase in Best Management Practice (BMP) Footprint and
Implications 143
8.4. CONCLUSIONS 148
9. REFERENCES 150
APPENDIX A. FLOW DURATION CURVES 1
APPENDIX B. DETAILED RESULTS 1
B. 1. Simulation Results by Site 1
B. 1.1. Harford County, MD 1
B.1.2. Scott County, MN 21
B.1.3. Maricopa County, AZ 80
B.1.4. Atlanta, GA 91
B.1.5. Portland, OR 103
B.2. Sensitivity Analysis 109
B.2.1. Harford County, MD 109
B.2.2. Scott County, MN 155
v
-------
LIST OF TABLES
Table 1-1. Matrix of regions, locations, land uses, management approaches, and future climate
scenarios 5
Table 2-1. Summary of geographic locations and land use types 7
Table 2-2. Stormwater management approach summary 9
Table 2-3. Climate scenarios for each geographic location 16
Table 2-4. Maryland Stormwater Manual pervious runoff coefficients 19
Table 2-5. Unit cost estimates for modeled practices 23
Table 3-1. Features of adaptation simulation for Harford County, MD 31
Table 3-2. 24-hour precipitation depth percentiles for current conditions and high intensity future
climate scenario at Harford County, MD 34
Table 3-3. Stormwater management and climate scenarios for Harford County, MD 35
Table 3-4. Current and future performance of Harford County, MD site by stormwater management
approach 36
Table 3-5. Comparison of current and future adapted best management practice (BMP) footprints for
Harford County, MD stormwater management scenarios 41
Table 3-6. Comparison of current and future adapted 20-year present value costs for the Harford
County, MD stormwater management scenarios 44
Table 4-1. Features of adaptation simulation for Scott County, MN 53
Table 4-2. 24-hour precipitation depth percentiles for current conditions and low, medium, and high
intensity future climate scenario at Scott County, MN 56
Table 4-3. Stormwater management and climate scenarios for Scott County, MN 58
Table 4-4. Current and future performance of Scott County, MN site by stormwater management
approach for general circulation model (GCM) high intensity scenario 59
Table 4-5. Comparison of current and future adapted best management practice (BMP) footprints for
Scott County, MN stormwater management scenarios 63
Table 4-6. Comparison of current and future adapted 20-year present value costs for the Scott
County, MN stormwater management scenarios 66
Table 5-1. Features of adaptation simulation for Maricopa County, AZ 74
Table 5-2. 24-hour precipitation depth percentiles for current conditions and high intensity future
climate scenario at Maricopa County, AZ 76
Table 5-3. Stormwater management and climate scenarios for of Maricopa County, AZ 77
Table 5-4. Current and future performance of Maricopa County, AZ site by stormwater management
approach 78
Table 5-5. Comparison of current and future adapted best management practice (BMP) footprints for
Maricopa County, AZ stormwater management scenarios 82
Table 5-6. Comparison of current and future adapted 20-year present value costs for the Maricopa
County, AZ stormwater management scenarios 82
Table 6-1. Features of adaptation simulation for Atlanta, GA 88
Table 6-2. 24-hour precipitation depth percentiles for current conditions and high intensity future
climate scenario at Atlanta, GA 90
Table 6-3. Stormwater management and climate scenarios for Atlanta, GA 91
Table 6-4. Current and future performance of Atlanta, GA Site by stormwater management approach
' 92
Table 6-5. Comparison of current and future adapted best management practice (BMP) footprints for
Atlanta, GA stormwater management scenarios 95
Table 6-6. Comparison of current and future adapted 20-year present value costs for the Atlanta, GA
stormwater management scenarios 96
Table 7-1. Features of adaptation simulation for Portland, OR 100
VI
-------
Table 7-2. 24-hour precipitation depth percentiles for current conditions and high intensity future
climate scenario at Portland, OR 102
Table 7-3. Stormwater management and climate scenario for Portland, OR 103
Table 7-4. Current and future performance of Portland, OR site by stormwater management
approach 105
Table 7-5. Comparison of current and future adapted best management practice (BMP) footprints for
Atlanta, GA stormwater management scenarios 108
Table 7-6. Comparison of current and future adapted 20-year present value costs for the Maricopa
County, AZ stormwater management scenarios 108
Table 8 -1. Stormwater management approach summary Ill
Table 8-2. Cost metrics for future climate and stormwater management scenarios 135
Table 8-3. Adaptation optimization limiting factors 138
Vll
-------
LIST OF FIGURES
Figure 1 -1. Modeling framework for study 2
Figure 1-2. Example site layout with green infrastructure (GI) stormwater management approach 3
Figure 1-3. Climate scenarios, with adaptation using larger best management practices (BMPs) 4
Figure 1-4. Climate scenarios, with adaptation using additional best management practices (BMPs). 4
Figure 2-1. Locations of sites selected for analysis 7
Figure 2-2. Flow duration curve (FDC) performance factor 13
Figure 3-1. Mixed-use site layout (Harford County, MD) 26
Figure 3-2. Mixed-use Conventional (Gray) Infrastructure stormwater management scenario (Harford
County, MD) 28
Figure 3-3. Mixed-use Green Infrastructure (GI) with Gray Infrastructure stormwater management
scenario (Harford County, MD) 30
Figure 3-4. Ranked annual precipitation for current and high intensity future climate at Harford
County, MD 33
Figure 3-5. Monthly average precipitation for current conditions and high intensity future climate
scenario at Harford County, MD 33
Figure 3-6. Hourly precipitation recurrence interval for current conditions and high intensity future
climate scenario at Harford County, MD 34
Figure 3-7. Annual site runoff under current climate and future general circulation model (GCM)
scenario by stormwater management approach for Harford County, MD 37
Figure 3-8. Maximum hourly peak flow under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Harford County, MD 37
Figure 3-9. Annual sediment loading rate under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Harford County, MD 38
Figure 3-10. Annual TN loading rate under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Harford County, MD 38
Figure 3-11. Annual TP loading rate under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Harford County, MD 39
Figure 4-1. Residential site layout (Scott County, MN) 47
Figure 4-2. Residential Conventional (Gray) Infrastructure stormwater management scenario (Scott
County, MN) 48
Figure 4-3. Residential Green Infrastructure (GI) with Gray Infrastructure stormwater management
scenario (Scott County, MN) 50
Figure 4-4. Residential Green Infrastructure (GI) Only stormwater management scenario (Scott
County, MN) 52
Figure 4-5. Ranked annual precipitation for current conditions and low intensity future climate
scenario at Scott County, MN 54
Figure 4-6. Ranked annual precipitation for current conditions and medium intensity future climate
scenario at Scott County, MN 55
Figure 4-7. Ranked annual precipitation for current conditions and high intensity future climate
scenario at Scott County, MN 55
Figure 4-8. Monthly average precipitation for current conditions and low/medium/high intensity
future climate scenarios at Scott County, MN 56
Figure 4-9. Hourly precipitation recurrence interval for current conditions and low/medium/high
intensity future climate scenarios at Scott County, MN 57
Figure 4-10. Annual site runoff under current climate and future general circulation model (GCM)
high intensity scenario by stormwater management approach for Scott County, MN 60
vni
-------
Figure 4-11. Maximum hourly peak flow under current climate and future general circulation model
(GCM) high intensity scenario by stormwater management approach for Scott County,
MN 60
Figure 4-12. Annual sediment loading rate under current climate and future general circulation model
(GCM) high intensity scenario by stormwater management approach for Scott County,
MN 61
Figure 4-13. Annual total nitrogen (TN) loading rate under current climate and future general
circulation model (GCM) high intensity scenario by stormwater management approach
for Scott County, MN 61
Figure 4-14. Annual total phosphorous (TP) loading rate under current climate and future general
circulation model (GCM) high intensity scenario by stormwater management approach
for Scott County, MN 62
Figure 5-1. Commercial site layout (Maricopa County, AZ) 70
Figure 5-2. Commercial Conventional (Gray) Infrastructure stormwater management scenario
(Maricopa County, AZ) 71
Figure 5-3. Commercial Green Infrastructure (GI) Only stormwater management scenario (Maricopa
County, AZ) 73
Figure 5-4. Ranked annual precipitation for current conditions and high intensity future climate
scenario at Maricopa County, AZ 75
Figure 5-5. Monthly average precipitation for current conditions and high intensity future climate
scenario at Maricopa County, AZ 75
Figure 5-6. Hourly precipitation recurrence interval for current conditions and high intensity future
climate scenario at Maricopa County, AZ 76
Figure 5-7. Annual site runoff under current climate and future general circulation model (GCM)
scenario by stormwater management approach for Maricopa County, AZ 78
Figure 5-8. Maximum hourly peak flow under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Maricopa County, AZ 79
Figure 5-9. Annual sediment loading rate under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Maricopa County, AZ 79
Figure 5-10. Annual TN loading rate under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Maricopa County, AZ 80
Figure 5-11. Annual TP loading rate under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Maricopa County, AZ 80
Figure 6-1. Ultra-urban site layout (Atlanta, GA) 84
Figure 6-2. Ultra-urban Conventional (Gray) Infrastructure stormwater management scenario
(Atlanta, GA) 86
Figure 6-3. Ultra-urban Green Infrastructure (GI) with Gray Infrastructure stormwater management
scenario (Atlanta, GA) 87
Figure 6-4. Ranked annual precipitation for current conditions and high intensity future climate
scenario at Atlanta, GA 89
Figure 6-5. Monthly average precipitation for current conditions and high intensity future climate
scenario at Atlanta, GA 89
Figure 6-6. Hourly precipitation recurrence interval for current conditions and high intensity future
climate scenario at Atlanta, GA 90
Figure 6-7. Annual site runoff under current climate and future general circulation model (GCM)
scenario by stormwater management approach for Atlanta, GA 92
Figure 6-8. Maximum hourly peak flow under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Atlanta, GA 93
Figure 6-9. Annual sediment loading rate under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Atlanta, GA 93
IX
-------
Figure 6-10. Annual total nitrogen (TN) loading rate under current climate and future general
circulation model (GCM) scenario by storm water management approach for Atlanta, GA.
94
Figure 6-11. Annual total phosphorous (TP) loading rate under current climate and future general
circulation model (GCM) scenario by storm water management approach for Atlanta, GA.
94
Figure 7-1. Green street site layout (City of Portland, 2008) 99
Figure 7-2. Ranked annual precipitation for current conditions and high intensity future climate
scenario at Portland, OR 101
Figure 7-3. Monthly average precipitation for current conditions and high intensity future climate
scenario at Portland, OR 102
Figure 7-4. Hourly precipitation recurrence interval for current conditions and high intensity future
climate scenario at Portland, OR 103
Figure 7-5. Annual site runoff under current climate and future general circulation model (GCM)
scenario by storm water management approach for Portland, OR 105
Figure 7-6. Maximum hourly peak flow under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Portland, OR 106
Figure 7-7. Annual sediment loading rate under current climate and future general circulation model
(GCM) scenario by stormwater management approach for Portland, OR 106
Figure 7-8. Annual total nitrogen (TN) loading rate under current climate and future general
circulation model (GCM) scenario by stormwater management approach for Portland,
OR 107
Figure 7-9. Annual total phosphorous (TP) loading rate under current climate and future general
circulation model (GCM) scenario by stormwater management approach for Portland,
OR 107
Figure 8-1. Percentage change in site runoff (no best management practices [BMPs]) between current
and future climate conditions 113
Figure 8-2. Percentage change in site maximum hourly peak outflow (no best management practices
[BMPs]) between current and future climate conditions 114
Figure 8-3. Percentage change in site sediment load (no best management practices [BMPs]) between
current and future climate conditions 114
Figure 8-4. Percentage change in site TN load (no best management practices [BMPs]) between
current and future climate conditions 115
Figure 8-5. Percentage change in site total phosphorous (TP) load (no best management practices
[BMPs]) between current and future climate conditions 115
Figure 8-6. Best management practice (BMP) percentage reduction of annual runoff under current
and future climate conditions 117
Figure 8-7. Best management practice (BMP) percentage reduction of sediment load under current
and future climate conditions 118
Figure 8-8. Best management practice (BMP) percentage reduction of total nitrogen (TN) load under
current and future climate conditions 119
Figure 8-9. Best management practice (BMP) percentage reduction of total phosphorous (TP) load
under current and future climate conditions 120
Figure 8-10. Normalized site-scale best management practice (BMP) removal of annual runoff under
current and future climate conditions 121
Figure 8-11. Normalized site-scale best management practice (BMP) removal of sediment load under
current and future climate conditions 122
Figure 8-12. Normalized site-scale best management practice (BMP) removal of total nitrogen (TN)
load under current and future climate conditions 122
Figure 8-13. Normalized site-scale best management practice (BMP) removal of total phosphorous
(TP) load under current and future climate conditions 123
x
-------
Figure 8-14. Normalized site-runoff export under current and future climate conditions 124
Figure 8-15. Normalized site-sediment mass export under current and future climate conditions 124
Figure 8-16. Normalized site-total nitrogen (TN) mass export under current and future climate
conditions 125
Figure 8-17. Normalized site-total phosphorous (TP) mass export under current and future climate
conditions 126
Figure 8-18. Change in runoff volume export for Scott County between current and future downscaled
general circulation model (GCM) climate scenarios 127
Figure 8-19. Change in sediment load export for Scott County between current and future downscaled
general circulation model (GCM) climate scenarios 128
Figure 8-20. Change in total nitrogen (TN) load export for Scott County between current and future
downscaled general circulation model (GCM) climate scenarios 128
Figure 8-21. Change in total phosphorous (TP) load export for Scott County between current and
future downscaled general circulation model (GCM) climate scenarios 129
Figure 8-22. Change in runoff volume export for Harford County between current and future
percentage change climate scenarios 130
Figure 8-23. Change in sediment load export for Harford County between current and future
percentage change climate scenarios 130
Figure 8-24. Change in total nitrogen (TN) load export for Harford County between current and future
percentage change climate scenarios 131
Figure 8-25. Change in total phosphorous (TP) load export for Harford County between current and
future percentage change climate scenarios 131
Figure 8-26. Change in runoff volume export for Scott County between current and future percentage
change climate scenarios 132
Figure 8-27. Change in sediment load export for Scott County between current and future percentage
change climate scenarios 132
Figure 8-28. Change in total nitrogen (TN) load export for Scott County between current and future
percentage change climate scenarios 133
Figure 8-29. Change in total phosphorous (TP) load export for Scott County between current and
future percentage change climate scenarios 133
Figure 8-30. Current cost and best management practice (BMP) adaptation cost for Portland, Maricopa
County, Atlanta, and Harford County stormwater management scenarios 141
Figure 8-31. Current cost and best management practice (BMP) Adaptation cost for Scott County
stormwater management scenarios 141
Figure 8-32. Current cost and best management practice (BMP) adaptation cost for Harford County
and Scott County conventional stormwater management scenarios, using different
adaptation approaches 143
Figure 8-33. Percentage change in best management practice (BMP) footprint between current and
adapted future climate for Portland, Maricopa County, and Atlanta stormwater
management scenarios 144
Figure 8-34. Percentage change in best management practice (BMP) footprint between current and
adapted future climate for Harford County stormwater management scenarios 145
Figure 8-35. Percentage change in best management practice (BMP) footprint between current and
adapted future climate for Scott County Conventional and Green Infrastructure
(GI) + Gray stormwater management scenarios 146
Figure 8-36. Percentage change in best management practice (BMP) footprint between current and
adapted future climate for Scott County Green Infrastructure (GI) Only and
Conventional + Distributed GI stormwater management scenarios 147
XI
-------
LIST OF ABBREVIATIONS
ACE
Air, Climate and Energy
Rev
recharge volume
ACF
Apalachicola'ChattahoochecFlint
ROW
right'of-way
BCSD
bias'correction spatial
SF
square feet
disaggregation
SRES
Special Report on Emissions
BMP
best management practice
Scenarios
C
coefficients
SUSTAIN
System for Urban Stormwater
CCSM
Community Climate System Mode
Treatment and Analysis Integration
CF
cubic feet
SWMM
stormwater management model
CMIP
Coupled Model Intercomparison
Tc
timing variable
Project
TN
total nitrogen
CNT
Center for Neighborhood
TP
total phosphorous
Technology
TSS
total suspended solids
Cp
pervious coefficient runoff value
WMO
watershed management
CPv
channel protection volume
organizations
ET
evapotransporation
WQv
water quality volume
FDC
flow duration curve
GCM
general circulation model
GFDL
Geophysical Fluid Dynamics
Laboratory
GI
green infrastructure
GIS
geographic information system
HADCM3
Hadley Centre Coupled Model,
Version 3
HSG
hydrologic soil group
HSPF
Hydrologic Simulation
Program* Fortran
HUC
hydrologic unit code
Hw/D
headwater/pipe diameter ratio
LF
linear feet
LID
low impact development
MPCA
Minnesota Pollution Control
Agency
NARCCAP
North American Regional Climate
Change Assessment Program
NCAR
National Center for Atmospheric
Research
NCDC
National Climatic Data Center
NCEA
National Center for Environmental
Assessment
NO A A
National Oceanic and Atmospheric
Administration
O&M
operation and maintenance
Qp
overbank flood protection
PET
potential evapotranspiration
PFDS
Precipitation Frequency Data
Server
PV
present value
RCM
regional climate models
xii
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
EPA's Office of Research and Development was responsible for producing this report. The report was
prepared by Tetra Tech, Inc. under EPA Contract No. EP-C-12-060. Susan Julius served as the Task
Order Project Officer, providing overall direction and technical assistance, and was a contributing author.
We would like to thank the internal and external reviewers for their valuable comments and insights.
AUTHORS:
Scott C. Job, Tetra Tech, Inc.
Maureen Harris, Tetra Tech, Inc.
Susan H. Julius, U.S. EPA
Jonathan B. Butcher, Tetra Tech, Inc.
J. Todd Kennedy, Tetra Tech, Inc.
Additional contributions were provided by Heather Fisher, Town of Hillsborough, N.C.; Bobby Tucker,
Tetra Tech, Inc.; and Jonathan Smith, Tetra Tech, Inc.
INTERNAL REVIEWERS:
Ashley Allen, U.S. EPA, Office of Water
Robert Brown, ORISE Postdoctoral Research Fellow, U.S. EPA
Robert Goo, U.S. EPA, Office ofWater
John Kemmerer, U.S. EPA Region 9
Karen Metchis, U.S. EPA, Office ofWater
EXTERNAL REVIEWERS:
Shirley E. Clark, Penn State University
David J. Sample, Hampton Roads Agricultural Research and Extension Center
Eric Strecker, Geosyntec Consultants, Inc.
xni
-------
EXECUTIVE SUMMARY
The EPA Office of Water has identified a need for improved understanding of the potential impacts of
future changes in extreme precipitation events on the occurrence and management of stormwater runoff
(U.S. EPA, 2008; 2010; 2012). Accordingly, the EPA National Center for Environmental Assessment
(NCEA) conducted a technical analysis of the performance of conventional ("gray") and natural or semi-
natural ("green") stormwater controls under precipitation and temperature scenarios to examine how those
controls could be re-engineered to control stormwater. This report presents the results of these modeling
studies.
Using continuous simulation modeling of the hydrology, hydraulics, and water quality discharged from a
series of conceptual development sites with a variety of gray and green infrastructure (GI) practices, we
addressed the following questions:
1. How might extreme precipitation events affect the performance of conventional stormwater
infrastructure and GI,
2. How can conventional designs and GI designs be adapted so that a site experiencing extreme
precipitation conditions in the future provides the same performance as the site under current
conditions, and
3. What do the results suggest regarding the adaptation potential of gray and green infrastructure for
increases in extreme precipitation events?
We used the modeling framework shown in Figure ES-1 to address the three questions above. The
Hydrologic Simulation Program* Fortran (HSPF) model (Bicknell et al., 2004) was used to simulate
hourly unit-area time series of runoff and pollutant loads (total suspended solids [TSS], total nitrogen
[TN], and total phosphorous [TP]) from pervious and impervious land. Future scenarios of extreme
precipitation were developed from the current time series using downscaled general circulation models
(GCMs) as well as percent changes in precipitation. The conceptual sites and associated stormwater
management infrastructure were simulated using the System for Urban Stormwater Treatment and
Analysis Integration (SUSTAIN) model, a decision support system and modeling tool. Hourly output is
provided by SUSTAIN for each individual best management practice (BMP) and at the site outlet.
xiv
-------
BMP ET
Climat
-------
Table ES-1. Matrix of regions, locations, land uses, future precipitation scenarios,
and stormwater management approaches
Region
Location
Type
Scenarios
Stormwater management approach
Mid-
Atlantic
Midwest
Arid
Southwest
Southeast
Pacific
Northwest
Harford
County,
MD
Mixed use
GCM high intensity
Minus 10%
Plus 10%
Plus 20%
Conventional (Gray) Infrastructure
GI with Gray Infrastructure
Conventional (Gray) Infrastructure
with distributed GI
Scott
County,
MN
Residential
GCM low intensity
GCM medium intensity
GCM high intensity
Minus 10%
Plus 10%t
Plus 20%
Conventional (Gray) Infrastructure
GI with Gray Infrastructure
GI Only
Conventional (Gray) Infrastructure
with Distributed GI
Maricopa
County,
AZ
Commercial
GCM high intensity
Conventional (Gray) Infrastructure
GI Only
Atlanta,
GA
Ultra-urban
GCM high intensity
Conventional (Gray) Infrastructure
GI with Gray Infrastructure
Portland,
OR
Transportation
corridor
GCM high intensity
GI Only
At two of the geographic locations, additional scenarios were developed that kept the current
conventional BMP configuration intact but added distributed GI practices to provide treatment equivalent
to the current conditions.
For each site location, a "base-case" future precipitation scenario was selected from a pool of ten
downscaled scenarios EPA developed previously (U.S. EPA, 2013; Johnson et al., 2015) to model the
impact of changes in precipitation on hydrology and water quality in 20 watersheds throughout the United
States (referred to in this report as the "20 Watersheds" project). The base-case was chosen as the
scenario with the largest increase in the intensity of large storm events to represent the upper range of
potential impacts. In addition, two types of sensitivity analyses were conducted as part of the study. First,
2 additional downscaled GCM scenarios were selected from the pool of 10 for the Midwestern
site* scenarios representing the lowest change and a medium change in large storm event intensity.
Finally, a sensitivity analysis was conducted at the Midwestern site and the Mid-Atlantic site evaluating
the effects of set percent changes in all precipitation events* minus 10, plus 10, and plus 20% changes in
precipitation volume. Table ES-1 provides a summary of components of the analyses.
The study sites and stormwater management approaches are outlined below in Table ES-2. The sites,
approaches, and applicable regulations are discussed in detail in each of the five site sections in the report
(see Sections 3. to 7. ). Each detailed site representation developed included both the local requirements
as well as the local or state guidance for BMP footprint, volume, and configuration. TR-55 and other tools
were used to develop scoping-level designs for the practices, including details of outlet structures with
xvi
-------
orifices and weirs set to meet stormwater requirements. Soil properties of media in filtering BMPs such as
bioretention and sand filters were represented using properties from design guidance as well as values
from BMP research.
To address impacts to water resources due to changes in precipitation events, each stormwater
management approach was modified by increasing the size (area, or footprint) of the structural
stormwater BMPs to provide treatment equivalent to current conditions. The performance metrics were
defined as follows:
• Annual outflow volume to address stormwater volume treatment requirements,
• Flow duration curve (FDC) to address channel erosion risk and flooding risk, and
• Pollutant mass export (nitrogen, phosphorus, and sediment) to address water quality performance.
Annual runoff volume, pollutant loads, and the flow duration curve were tabulated for each site under
current conditions in an initial SUSTAIN model run. Under projected future precipitation conditions,
annual runoff volume, pollutant loads, and the magnitude of flows in the upper portion of the flow
duration curve generally increased. To find a new configuration that met all of the current condition
metrics, an optimization was performed in SUSTAIN with numerous model runs that incrementally
changed practice sizes. For the flow duration curve, the difference between the current conditions and
future precipitation curves was used to assess increases in peak flows across a range of storms.
Optimization sought to minimize the area between the curves, thus, mimicking the current conditions
hydraulic response. The SUSTAIN optimization included scoping-level estimates of unit-area practice
costs, so the optimal solution was the configuration that was both lowest cost and met all of the target
values of the metrics. As an example, for the Scott County GI with Gray stormwater management
approach, the footprints of the bioretention cells and the dry detention basin were increased under future
precipitation conditions. This increase provided additional hydrologic control and pollutant removal so
that the revised configuration performed as well or better than the site under current precipitation
conditions.
For the Harford County, MD and Scott County, MN stormwater management approaches, the designs
were modified using two different management strategies* increasing the size of the structural practices
(as done for the other locations) and addressing the performance gap by incorporating additional
distributed GI practices into the site. In the latter approach, the current Conventional Infrastructure
configuration was unchanged and distributed infiltration trenches (Harford County, MD) and bioretention
(Scott County, MN) were added to provide treatment equivalent to the current conditions.
XVll
-------
Table ES-2. Stormwater management approach summary
Location
Characteristics
Stormwater
requirements
Stormwater
management
approach
Practices
Harford
County,
MD
Mixed use
20 acres
65%
Completely infiltrate
recharge volume
Treat water quality volume
Conventional
(Gray)
Infrastructure
Surface sand filters, extended dry
detention basin
impervious
for TSS/TP
Channel protection volume
(24-h detention of 1-yr
24-h storm)
Match predeveloped peak
for 10-yr 24-h storm
GI with Gray
Infrastructure
Infiltration trenches, infiltration
basins, permeable pavement, and
dry detention basin
Conventional
(Gray)
Infrastructure with
Distributed GI
Surface sand filters, extended dry
detention basin, distributed
infiltration trenches
Scott
County,
MN
Residential
30 acres
48%
Treat water quality volume
for TSS
Match predeveloped peak
Conventional
(Gray)
Infrastructure
Wet pond
impervious
for 2-yr 24-h storm and
100-yr 24-h storm
GI with Gray
Infrastructure
Distributed bioretention and dry
detention basin
GI Only
Distributed bioretention, permeable
pavement, and impervious surface
disconnection
Conventional
(Gray)
Infrastructure with
Distributed GI
Wet pond, distributed bioretention
Maricopa
County, AZ
Commercial
10 acres
80%
100% retention of the
100-yr 2-h storm event
Conventional
(Gray)
Infrastructure
Detention/infiltration basin
impervious
GI Only
Permeable pavement, cistern
bioretention, and stormwater
harvesting basin
Atlanta,
GA
Ultra-urban
2 acres
90%
Treat water quality volume
for TSS
Channel protection volume
Conventional
(Gray)
Infrastructure
Underground sand filter,
underground dry detention basin
impervious
(24-h detention of 1-yr
24-h storm)
Match predeveloped peak
for 2-yr, 5-yr, 10-yr, 25-yr,
and 100-yr 24-h storm
GI with Gray
Infrastructure
Green roof, permeable pavement,
bioretention, and underground dry
detention basin
Portland,
OR
Transportation
corridor
0.35 acres
89%
impervious
70% TSS reduction
Infiltration of 10-yr 24-h
storm event as practicable
Match predeveloped peak
for 2-yr, 5-yr, 10-yr 24-h
storm
GI Only
Bioretention swales, permeable
pavement
TSS = total suspended solids.
xviii
-------
This study was not exhaustive and considers only a limited number of developed land use types, practice
configurations, and projected future precipitation scenarios. In addition, while the five geographic regions
selected for modeling represent a range of ecoregions and climate types, they do not cover all of the
precipitation conditions found in the United States. Results are also dependent on the representations of
the complex physical processes governing stormwater hydrology and pollutant loading that are
incorporated into the HSPF and SUSTAIN model codes. Like all simulation models, these tools are
simplified approximations of the real world. For most of the sites, the precipitation change scenarios were
selected to represent the upper range of potential impacts. The analysis does provide insights into the
potential impacts of changes in precipitation events on stormwater infrastructure performance, and allows
comparison of how the responses may differ between conventional and GI practices.
RESULTS
Model simulations in five study locations suggest some ranges for the increase in pretreatment total urban
runoff and pollutant loads that may result from changes in precipitation events by midcentury. For overall
post-treatment site-scale performance, simulations using both conventional and GI BMP scenarios
generally remove more runoff volume and pollutant mass under future increases in precipitation and
runoff compared to current conditions. However, overall site export rates of runoff volume and pollutant
mass still increase (i.e., BMP does not remove 100% of the additional runoff/pollutant load resulting from
increased precipitation) despite better volume/mass removal. Changes in large storm event runoff (as
indicated by comparison of FDCs) show that BMPs designed for current conditions will not mitigate
increases in stormwater runoff and associated downstream channel erosion and flooding impacts under
projected future conditions. Thus, there may be a need for adapting site stormwater infrastructure to
future precipitation conditions to protect downstream water resources. Sites may also need to be
configured to be adaptable in the first place to allow for placement of additional stormwater treatment if
needed in the future.
When considering the adaptation of BMPs under future precipitation conditions to achieve the same or
better performance as seen under current conditions, the most difficult performance measure to mitigate
was usually control of large flooding event outflows. Given that control of flooding events is a ubiquitous
requirement throughout the United States, this indicates that current practices will need greater temporary
volume storage and/or reconfiguration of outlet structures to mitigate flooding and channel erosion risk in
locations where the magnitude of extreme events is expected to increase. GI practices that rely on
treatment without volume storage will be at a disadvantage for adaptation to increased precipitation, but
approaches that rely only on adapting conventional practices may not have the flexibility to address
multiple performance objectives.
When comparing the current cost of stormwater management for new development between conventional
and Gl-based approaches, the conventional approaches tended to be more cost-effective than their GI
counterparts. However, when precipitation scenarios with smaller increases in large storm event
intensities are considered, the additional cost of adapting sites using GI approaches tended to be less than
adapting conventional-only approaches. Overall, approaches to stormwater management that combined
both conventional and GI elements tended to have the best combined cost resiliency.
xix
-------
1. INTRODUCTION AND OVERVIEW
The technical analysis described in this report was performed to quantify potential impacts of climate
change on stormwater infrastructure performance. The foundation of the technical analysis is continuous
simulation modeling of the hydrology, hydraulics, and water quality discharged from a series of
conceptual study sites using a variety of conventional (or "gray") and green infrastructure (GI) practices
consistent with local stormwater and site design regulations. The report assesses the performance of green
and gray stormwater controls under current and future climate and offers insights into how designs could
be adapted in the future in response to a changing climate. The results and conclusions of the study are
applicable to both new development, redevelopment, and stormwater retrofits.
The principal questions addressed by the study include:
1. How might extreme precipitation events affect the performance of conventional stormwater
infrastructure and GI,
2. How can conventional designs and GI designs be adapted so that a site experiencing extreme
precipitation conditions in the future provides the same performance as the site under current
conditions, and
3. What do the results suggest regarding the adaptation potential of gray and green infrastructure for
increases in extreme precipitation events?
This report describes the modeling approach and the results of the continuous simulation modeling.
Figure 1-1 summarizes the modeling framework. First, the Hydrologic Simulation Program* Fortran
(HSPF; Bicknell et al., 2004) was used to simulate hourly unit-area time series of runoff and pollutant
loads (total suspended solids [TSS], total nitrogen [TN], and total phosphorous [TP]) from pervious and
impervious land surfaces for 30 years of input meteorology obtained from National Weather Service
monitoring stations. Future climate scenarios were developed from the current time series using
downscaled general circulation model (GCM) output, augmented by percentage change climate sensitivity
analyses. The conceptual urban sites and associated stormwater management infrastructure were
simulated by inputting the HSPF unit-area time series into the System for Urban Stormwater Treatment
and Analysis Integration (SUSTAIN, U.S. EPA, 2009) decision support system and modeling tool.
Hourly output was provided by the model for each individual best management practice (BMP) and at the
site outlet. Note that only surface runoff and associated loads are considered in the analysis because the
purpose of this analysis is to assess climate change impacts on stormwater. It is possible for site BMPs to
increase groundwater outflow and associated pollutants, but the impact on the results would have been
minimal compared to the magnitude of surface runoff and pollutants. For simplicity, high water table
impacts are not considered in this analysis.
1
-------
BMP ET
Climat
-------
Cistern
Permeable
Pavement
Harvesting
Bioretention
Permeable
Pavement
Figure 1-2. Example site layout with green infrastructure (GI) stormwater
management approach.
Each stormwater management approach was modeled under current and a selected set of projected future
climate conditions for the mid-21st century, and the water quantity and quality performance of the site
practices were calculated from modeling results. In an additional model run, the site's practices were
adapted under future climate conditions to achieve the same or better performance as under the current
climate scenario using SLSTAIN s optimization function. Modifications targeted resizing the water
quality treatment and peak flow control BMPs, which are the primary drivers controlling site performance
(see Figure 1-3). When optimization is performed, SUSTAIN is configured to execute numerous runs
(usually in the hundreds to over a thousand) making incremental changes in the BMP configuration to
achieve specified targets (in this case, the targets were equal to the current hydrology and water quality
performance of the site). The optimization scenario with the least cost was selected as the best solution
because there were multiple '"solutions" that achieved the goals of the simulation.
At two of the geographic locations, additional scenarios were developed that kept the current
conventional BMP configuration intact but added distributed GI practices to provide treatment equivalent
to the current conditions climate scenario (see Figure 1-4).
3
-------
Current Climate
Baseline Site
Baseline Loads
Baseline Volume
Baseline Peak Flows
Future Climate
Baseline Site
Higher Loads
HigherVolume
Increased Peak Flows
Future Climate
Adapted Site
Reduced Loads
Reduced Volume
Reduced Peak Flows
Figure 1-3. Climate scenarios, with adaptation using larger best management
practices (BMPs).
Current Climate
Baseline Site
Baseline Loads
Baseline Volume
Baseline Peak Flows
Future Climate
Baseline Site
(JM
Higher Loads
HigherVolume
Increased Peak Flows
Future Climate
Adapted Site
( BMP )
^ ( BMP )
R!\/IP \ ( nn/in ^ S - /
Reduced Loads
Reduced Volume
Reduced Peak Flows
Figure 1-4. Climate scenarios, with adaptation using additional best management
practices (BMPs).
4
-------
A "base-case" future climate scenario was selected for each site location from a pool of ten potential
downscaled GCM climate scenarios developed for the "20 Watersheds" project. The base-case was
chosen as the scenario with the largest increase in the intensity of large storm events in order to estimate
the potential impact of climate change at the upper end of the range of potential climate futures. In
addition, two types of climate sensitivity analysis were conducted as part of the study. First, two
additional downscaled GCM scenarios were selected from the "20 Watersheds" project for the
Midwestern site* scenarios representing the lowest change and a medium change in large storm event
intensity. Finally, a sensitivity analysis was conducted at the Midwestern site and the Mid-Atlantic site
evaluating the effects of set percentage changes in all precipitation events* minus 10, plus 10, and plus
20% changes in hourly precipitation volume. Table 1-1 provides a summary of the components of the
analysis.
Table 1-1. Matrix of regions, locations, land uses, management approaches, and
future climate scenarios
Management approach
Future
Region
State
Land use
Gray
Mixed
GI only
climate
scenarios
Mid-Atlantic
Maryland
Mixed use
X
X
Midwest
Minnesota
Residential
X
X
X
MM
Arid southwest
Arizona
Commercial
X
1
Southeast
Georgia
Ultra-urban
X
X
Pacific northwest
Oregon
Green street
X
1
This report is organized into eight sections beginning with the introduction, followed by details of the
modeling approach, a separate section for each of the five sites, and finally results and discussion to
address the principle study questions.
5
-------
2. STUDY APPROACH
2.1. REGIONS AND LAND USES
The five geographic regions selected for simulation are listed in Table 2-1 and shown in Figure 2-1. These
regions were chosen because they leverage existing EPA climate change research: each of these was
modeled as part of the EPA "20 Watersheds" project, which examined the impact of climate change on
water quality (represented using sediment, TN, and TP) in 20 large (4-digit hydrologic unit code [HUC]
scale) watersheds located throughout the United States, and included several future climate scenarios
representing a range of projected meteorological conditions. Output from this modeling is available down
to an approximately 12-digit HUC scale and for individual unit-area upland land use/land cover types
within each model subbasin. As a result, watershed response model simulations for future climate
scenarios have already been produced for the forcing meteorological data for these regions, which
represent a diversity of geographic settings and climate conditions. The "20 Watersheds" project included
five watersheds that were modeled with HSPF using an hourly time step, which is needed for the
continuous simulation SUSTAIN modeling; No other regions had HSPF models available to use in this
study (the full set of 20 watersheds was modeled using the Soil and Water Assessment Tool [Neitsch et
al., 2011], which runs on a daily time step and, therefore, did not meet our requirements). The
corresponding HSPF model river basins are shown in Figure 2-1.
For each geographic location, a specific municipality or county was selected within or near the river
basin. Following selection of a municipality/county, a specific land use type was chosen for each location.
The decision regarding which municipality/county and land use type to select within each region was
informed by several factors, including presence of an urbanized area, types of development taking place,
and applicable stormwater requirements. For instance, Maricopa County, AZ has stormwater
requirements and performs design review for the greater Phoenix area. The City of Minneapolis, MN is
more or less built-out and limited new residential development is taking place within city limits, so Scott
County on the southern outskirts of the Minneapolis metropolitan area was chosen because active
residential development is taking place in some of the county's communities. Municipality/county and
land use types are shown in Table 2-1.
6
-------
Table 2-1. Summary of geographic locations and land use types
Geographic region
River basin
Municipality/county
Land use type
Mid-Atlantic
Susquehanna River
Harford County, MD
Mixed use
Midwest
Minnesota River
Scott County. MN (near
Minneapolis)
Ultra-urban
Arid southwest
Salt River
Maricopa County, AZ
(surrounds Phoenix)
Commercial
Southeast
ACF rivers
Atlanta. GA
Residential
Pacific northwest
Willamette River
Portland, OR
Transportation
corridor/green street
ACF = Apalachicola-Chattahoochee-Flint.
Portland,
Harford County,
Maryland
Maricopa County,
Arizona
Atlanta,
Georgia
Figure 2-1. Locations of sites selected for analysis.
7
-------
2.2. STORMWATER MANAGEMENT APPROACHES
The study sites and stormwater management approaches are outlined below in Table 2-2. The sites,
approaches, and applicable regulations are summarized in detail in each of the five site sections in the
report. To address impacts of increased stormwater volume, large storm event peak flow, and pollutant
loading due to climate change, each stormwater management approach was modified by increasing the
size (area, or footprint) of the structural stormwater BMPs to provide treatment equivalent to the current
conditions climate scenario. For the Harford County, MD and Scott County, MN stormwater management
approaches, the designs were modified using two different management strategies: increasing the size of
the structural practices (as done for the other locations) and addressing the performance gap by
incorporating additional distributed GI practices into the site. In the latter approach, the current
conventional BMP configuration was unchanged, and distributed infiltration trenches (Harford County,
MD) and bioretention (Scott County, MN) were added to provide treatment equivalent to the current
conditions climate scenario.
8
-------
Table 2-2. Stormwater management approach summary
Region
Location
Type
Characteristics
Stormwater
management
approach
Practices
Mid-Atlantic
Harford
County, MD
Mixed use
20 acres
65% impervious
Conventional (Gray)
Infrastructure
Surface sand filters,
extended dry detention
basin
GI with Gray
Infrastructure
Infiltration trenches,
infiltration basins,
permeable pavement, and
dry detention basin
Conventional (Gray)
Infrastructure with
Distributed GI
Surface sand filters,
extended dry detention
basin, distributed
infiltration trenches
Midwest
Scott County,
MN
Residential
30 acres
48% impervious
Conventional (Gray)
Infrastructure
Wet pond
GI with Gray
Infrastructure
Distributed bioretention
and dry detention basin
GI Only
Distributed bioretention,
permeable pavement, and
impervious surface
disconnection
Conventional (Gray)
Infrastructure with
Distributed GI
Wet pond, distributed
bioretention
Arid
southwest
Maricopa
County, AZ
Commercial
10 acres
80% impervious
Conventional (Gray)
Infrastructure
Detention/infiltration
basin
GI Only
Permeable pavement,
cistern, bioretention, and
stormwater harvesting
basin
Southeast
Atlanta, GA
Ultra-urban
2 acres
90% impervious
Conventional (Gray)
Infrastructure
Underground sand filter,
underground dry detention
basin
GI with Gray
Infrastructure
Green roof, permeable
pavement, bioretention,
and underground dry
detention basin
Pacific
northwest
Portland, OR
Transportation
corridor
0.35 acres
89% impervious
GI Only
Bioretention swales,
permeable pavement
9
-------
2.3. ADAPTATION SIMULATION
2.3.1. Analytical Design
The objectives of this study are to understand:
1. How might extreme precipitation events affect the performance of conventional stormwater
infrastructure and GI,
2. How can conventional designs and GI designs be adapted so that a site experiencing extreme
precipitation conditions in the future provides the same performance as the site under current
conditions, and
What do the results suggest regarding the adaptation potential of gray and green infrastructure for
increases in extreme precipitation events?
To answer the first question, each stormwater management scenario was modeled under current climate
conditions, and the performance of the site practices was calculated from the modeling results. Next, each
management scenario was modeled under future climate conditions and the change in performance
metrics tabulated. Performance metrics used for the analyses are shown in the next subsection.
For the second and third questions, the ultimate objective of the simulation is to answer the question:
"How would the current stormwater managementpractice(s) need to be adapted in order to maintain
current performance under future climate conditions?''' For all sites, performance is evaluated at the site
"outlet," defined as the point to which all areas, BMPs, and conveyances ultimately drain. Therefore, the
objective is to evaluate a site's performance as a whole at meeting performance targets, rather than the
performance of individual BMPs. For sites with multiple BMPs, the goal of the adaptation simulation is
then to determine an optimal combination of BMP areas that result in the site as a whole meeting
performance objectives, or "targets." In the final model run, the site's practices were modified under
future climate conditions to achieve the same or better performance as the current climate scenario.
Modifications targeted resizing the water quality treatment and peak flow control BMPs, which are the
primary drivers controlling site performance. For the Midwest and Mid-Atlantic sites, additional runs
were performed where distributed GI practices were added to the sites rather than increasing the size of
the existing BMPs.
The primary tool for analyzing BMP performance is SUSTAIN, a decision support system and modeling
tool developed by EPA to facilitate selection and placement of BMPs and low-impact development
practices in urban watersheds (U.S. EPA, 2009). SUSTAIN was selected because it is able to do the
following:
• Support continuous simulations.
• Use unit-area runoff and pollutant time series from the continuous simulation watershed models
HSPF or Loading Simulation Program in C++ (LSPC) (Tetra Tech, 2009). When series are
available from a calibrated watershed model, there is no need to parameterize soil properties to
support a long-term simulation.
10
-------
• Simulate the hydrologic and hydraulic conditions of both conventional and GI practices.
• Represent pollutant reduction mechanisms for both conventional and GI practices.
• Run optimizations (i.e., can be used to develop optimal design addressing future projected
climate).
The EPA stormwater management model (SWMM 5) (Rossman, 2010) was also considered for
simulating BMP performance in this project. It is similar to SUSTAIN (indeed, much of SUSTAIN is
based on SWMM 5), and it can be driven by external runoff and pollutograph time series. SWMM has a
fairly well-developed module for representing GI hydrology. SUSTAIN was chosen over SWMM
because one cannot specify variable pollutant removal mechanisms for multiple outflow paths in SWMM.
It is also not designed for running optimizations. HSPF is used to provide unit-area input time series to
SUSTAIN, but was not selected for the BMP simulation because it does not include an optimization
component to perform the adaptation scenarios.
To investigate the adaptation of stormwater practices under future climate conditions, the SUSTAIN
model was executed in its "optimization" mode. When SUSTAIN performs an optimization, the site
model is executed hundreds (or even thousands) of times with incremental variations in the sizes and/or
configurations of the stormwater BMPs. Model results are automatically compared to performance targets
set by the user, and the increase in cost associated with the change in practice dimensions is calculated
using user cost inputs. SUSTAIN uses algorithms to guide the selection of subsequent incremental
variations in practice configuration, using both performance relative to the targets and cost differential.
After numerous model runs are complete, the user can select the best solution* one that achieves all of
the performance targets at the lowest cost. During optimization, practices can be reconfigured using a
number of options available in SUSTAIN. However, increasing practice size (as opposed to adding
volume by changing the stage of an outlet) was used for all the optimizations because many of the BMP
cost metrics used in the analysis are based on area alone. Note that practice costs in the optimization
reflected Present Value Life Cycle Cost (as discussed in Section 2.5.2. ); the opportunity cost of land area
was not included.
2.3.2. Best Management Practice (BMP) Performance Measures
Site performance is summarized in terms of typical stormwater BMP performance metrics from the
SUSTAIN model output and includes the following:
• Annual outflow volume to address stormwater volume treatment requirements,
• Flow duration curve (FDC) to address hydromodification associated with channel erosion risk
and flooding risk, and
• Pollutant mass export (nitrogen, phosphorus, and sediment) to address water quality performance.
The measures work together to assess impacts to site hydrology and water quality. These performance
measures are discussed in detail below.
11
-------
2.3.2.1. Annual Average Runoff Volume
Stormwater BMPs can be designed to serve numerous functions, with one of the most critical functions
being their ability to reduce the volume of stormwater leaving a site. The primary mechanisms of volume
reduction by BMPs include infiltration, evapotranspiration (ET), and storage for reuse. Their relative
importance varies across the numerous BMP types, land uses, site locations, and climate scenarios
investigated.
All of the sites included in this investigation are subject to varying degrees of local regulations for
stormwater volume retention. Annual runoff volume is also a better indicator for changes in long-term
hydrology than a measure related to large storm events. To address the impacts of increased stormwater
volume due to climate change, the annual average flow volume (ft3/year) measure is included in the
adaptation simulation procedure to ensure that a site's performance for stormwater volume reduction
under current climate is maintained under future climate conditions. The objective is for the annual
average flow volume leaving each site under current climate conditions to remain the same (or decrease)
under future climate conditions.
2.3.2.2. Flow Duration Curve
The FDC is a cumulative frequency curve that shows the percentage of time discharges are equaled or
exceeded during a given period (see APPENDIX A. for a detailed discussion of the FDC and its
relevance for this project). For the purposes of this investigation, the flows resulting from the largest
storm events during the 30-year simulation period were investigated. These are the storms associated with
large infrequent flooding events (e.g., a 10-year frequency event), as well as more frequent events
associated with the highest cumulative risk of downstream bank erosion (often called bankfull events,
which typically occur every 1 to 2 years). Using the FDC allows a comparison of current versus future
climate conditions across a range of flows that have the potential to physically alter the channel. For these
reasons, matching the FDC was selected as a performance measure rather than the single largest peak
discharge flow during the 30-year simulation. While the use of the largest peak discharge flow as a
performance measure would help ensure that BMPs provide adequate control of the highest magnitude
flows under future climate conditions, matching the FDC as a performance measure is designed to
maintain BMP performance for multiple flooding events. Many of the locations represented in this study
have peak flow and/or channel protection requirements for stormwater management, so the FDC analysis
is used to address these.
The FDC performance factor is computed by SUSTAIN as the area between two flow duration curves
representing two different hydrologic conditions, within specified lower and upper bounds. For this study,
the two hydrologic conditions evaluated are (1) the FDC under current climate conditions and (2) the
FDC under future climate conditions. The modeler defines upper and lower flow limits (thresholds) that
bound the FDC comparison. Figure 2-2 illustrates an example of how the FDC performance factor is
calculated as the area between the two curves between the lower and upper flow boundaries. In this
12
-------
example, the lower flow boundary is 13.5 cfs, where the 2-year hourly flow crosses the current conditions
FDC.1 The upper flow boundary is set to the highest flow from either FDC in the simulation.
Current with BMPs Future, adapted BMPs 2-yr Hourly Flow
90
80
70
Tj 60
s
O 50
H-
+•»
3
0 40
>>
1 30
I
20
10
0
0.000% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure 2-2. Flow duration curve (FDC) performance factor.
For this investigation, the high flow limit was defined as the highest flow encountered during the
simulation among the climate scenarios, and the low flow limit was varied according to site location. For
Portland, the hourly flow with a 1 -year recurrence from the 30-year simulation was used for the low flow
limit due to limited outflow below this frequency. For Atlanta, the hourly flow with a 0.5-year recurrence
was used to capture a range of flows including bankfull events. For Harford County and Scott County, the
hourly flow with a 2-year recurrence was used as the lower limit to allow the FDC optimization to better
fit large events associated with local peak flow requirements, For Maricopa County, no outflow from the
BMP is predicted under current climate conditions, so FDC optimization was not needed. The SUSTAIN
optimization process tracks the area between the current and future climate curves bounded by the lower
and upper flow limits, and attempts to minimize the area over the course of hundreds of model runs. The
future climate FDC varies according to changed precipitation record and the size and configuration of all
the site practices; surface conditions, practice volume, and runoff timing are contributing factors
controlling the shape of the FDC. When the simulation is complete, the user selects the best FDC
'Flow recurrence is calculated as the reciprocal of the product of flow percentile (from hourly output over the course
of the 30-year simulation) and the number of hours in a year. In the example shown the 2-year flow recurrence
occurs at flow percentile 5.7 n 10 I The ranked flow at this percentile is 13.5 cfs.
13
-------
performance (lowest value) with the corresponding lowest cost as the adapted solution. It is important to
note that the lower flow limits used in the analysis are not the same as design storm event peak flows. The
values are derived from the hourly outflows from the 30-year simulation, and represent recurrence
frequencies (i.e., the 2-year hourly recurrence flow is the 15th highest hourly flow during the 30-year
simulation).
2.3.2.3. Annual Average Pollutant Load
Another principal design function of stormwater BMPs is pollutant load reduction. Across the United
States, local stormwater and water quality regulations mandate specific pollutant reduction goals
(typically driven by total maximum daily loads and/or local water quality management plans). BMPs may
be used to help address these requirements. The primary pollutant removal mechanisms of stormwater
BMPs vary widely by practice type, and include filtration, infiltration, and settling. For some BMP types,
biological uptake and soil adsorption may also be significant pollutant removal pathways.
Under future climate conditions, changes in the depth, intensity, and duration of rainfall are expected to
have a significant impact on the delivery of pollutant loads, affecting both the timing and magnitude. In
most of the climate scenarios used in this project, pollutant loads increase under future climate conditions
(although this is not always the case). To address the impacts of increased pollutant loading due to climate
change, the annual average pollutant load (pound/year) measure is included in the adaptation simulation
procedure to ensure that BMP performance for pollutant load reduction under the current climate is
maintained under future climate conditions. The objective is for the loading of nitrogen, phosphorus, and
sediment (referred to as total suspended solids, or TSS) under current climate conditions to remain the
same, or decrease, under future climate conditions. The HSPF models developed for the ""20 Watersheds"
project included these three pollutants, so the water quality simulation in SUSTAIN was limited to these
constituents.
2.4. HIGH PRECIPITATION SCENARIO
Three groups of future climate scenarios were chosen for this study:
1. Downscaled GCMs with high intensity change for largest storm events, used for all sites,
stormwater management approaches, and adaptation approaches.
2. Downscaled GCMs with low, medium, and high intensity changes for largest storm events for
first climate sensitivity analysis. Applied to Midwest site only.
3. Precipitation volume percentage change scenarios for second climate sensitivity analysis. Applied
to Mid-Atlantic and Midwest sites.
2.4.1. High Intensity Change Scenarios
This analysis drew from a pool of 10 future climate change scenarios from the EPA "20 Watersheds"
study. Six future climate scenarios are from the North American Regional Climate Change Assessment
Program (NARCCAP). NARCCAP scenarios were developed by driving a number of different regional
climate models (RCMs) at a resolution of 50 x 50 km with results from four GCMs from Phase 3 of the
14
-------
Coupled Model Intercomparison Project (CMIP3), (Mearns et al., 2009, 2013).1 All scenarios assume the
Special Report on Emissions Scenarios (SRES) A2 greenhouse gas emissions trajectory (Nakicenovic et
al., 2000). Differences among SRES emissions scenarios, however, are not substantial for the mid-century
time period considered here. The NARCCAP scenarios were selected because they provide higher spatial
and temporal resolution climate change information for the entire contiguous United States and, unlike
the archived data from the parent GCMs, provide the full suite of meteorological variables needed to
implement HSPF simulations that use an energy balance approach to estimate evapotranspiration. In
addition, four scenarios were developed based on statistically downscaled output from the same set of
GCMs used by NARCCAP from the Downscaled CMIP3 and CMIP5 Climate and Hydrology Projections
archive at http://gdo-dcp.ucllnl.org/downscaled cmip projections commonly referred to as
bias-correction spatial disaggregation (BCSD) data but more formally developed using BCSD and
bias-correction constructed analogs temporal disaggregation (Maurer et al., 2007).2 This data set provides
temperature and precipitation on a 1/8-degree (approximately 14 x 10 km at 45°N) horizontal grid and at
a daily time step.
All climate change scenarios were implemented using a change-factor approach (Anandhi et al., 2011) to
modify 30 years of observed local hourly weather data to ensure realistic patterns in time series. Projected
monthly change statistics (change factors) at each weather station were calculated for total precipitation
(%), precipitation above/below the 70th percentile (%), air temperature (°C), relative humidity (°C),
surface downwelling shortwave radiation (%), and wind speed (%). Projected changes in the proportion of
precipitation volume occurring in larger events (i.e., event intensity) were represented by applying
different change factors to events above and below the 70th percentile event (based on daily depth). For
further details, see EPA (2013) and Johnson et al. (2015).
The climate scenario representing the largest increase in precipitation intensity among the 10 scenarios
was selected for modeling climate change impacts to stormwater infrastructure (see Table 2-3). The
scenario with the largest intensity increase was selected to allow the study to characterize an upper bound
for climate change impacts. Intensity was assessed at the National Climatic Data Center (NCDC)
meteorological monitoring station closest to the municipality/county. An important aspect of future
climate is the potential for increases in high intensity precipitation. This is expected for physical reasons
because increased air temperature increases the capacity of the air to hold moisture and potential energy.
For each of the 10 climate scenarios, we calculated both the change in monthly precipitation depth and the
fraction of precipitation contained within events greater than the 70th percentile daily event by comparing
runs of the same GCM for future and historic 30-year time periods. Future climate time series are created
from observed historic time series by applying a multiplicative change factor approach to adjust total
volume and redistributing the fraction of this volume to events above and below the 70th percentile event.
1 Seth McGinnis of the National Center for Atmospheric Research (NCAR) processed the North American Regional
Climate Change Assessment Program (NARCCAP) output into change statistics for use in the watershed modeling.
NCAR is supported by the National Science Foundation.
2We acknowledge the modeling groups and the Program for Climate Model Diagnosis and Intercomparison and the
WCRP's Working Group on Coupled Modeling for their roles in making available the WCRP Coupled Model
Intercomparison Project Phase 3 multimodel data set. Support of this data set is provided by the Office of Science,
U.S. Department of Energy.
15
-------
The scenario with the month with the highest predicted volume of events above the 70th percentile was
chosen as approximating the highest storm event volume among the available scenarios. Because the
same monthly multiplicative factors were used throughout a given scenario, it follows that the scenario
will contain the series of highest storm event volumes, and thus, serve as a proxy for the scenario with the
greatest change in large storm event intensity. Note that this selection is constrained by the available
climate scenarios and other scenarios not contained in the "20 Watersheds" data set are likely to show an
even greater range of storm volumes and intensities at these sites.
Table 2-3. Climate scenarios for each geographic location
Geographic
region
River basin
NCDC station3
Climate scenario
Mid-Atlantic
Susquehanna River
PA 366289 (New Park)
BCSD HADCM3
Midwest
Minnesota River
MN 215435
(Minneapolis/St. Paul
Airport)
Low: NARCCAP GFDL High Res GFDL
Medium: NARCCAP RCM3 GFDL
High: BCSD CCSM
Arid southwest
Salt River
AZ 026840 (Punkin Center)
BCSD GFDL
Southeast
ACF Rivers
GA 096407 (Atlanta
Hartsfield Intl. Airport)
NARCCAP RCM3 GFDL
Pacific
northwest
Willamette River
OR 356749 (Portland KGW
TV)
BCSD GFDL
CCSM = Community Climate System Model. GFDL = Geophysical Fluid Dynamics Laboratory,
HADCM3 = Hadley Centre Coupled Model, Version 3.
aState and cooperative summary of the day identification number.
In terms of the analysis, future climate inputs have been prepared that are representative of local
meteorology by virtue of using local NCDC stations. In addition, the redistribution of precipitation
changes between smaller (<70th percentile) versus larger (>70th percentile) events helps account for not
only changes in volume but also changes in intensity. (Potential changes in frequency or duration of
events independent of volume were not investigated.) The approach is appropriate for the goal of
providing examples of potential impacts of projected climate change on the performance of storm water
BMPs in urban watersheds distributed across the United States; however, the specific results for each
geographic location are in part dependent on the characteristics of the individual meteorological station.
In this project, SUSTAIN used external time series to represent surface runoff and pollutant loads from
land surfaces. These external runoff and pollutant time series were derived from output from the HSPF
watershed model. The HSPF models were modified to generate unit-area time series on an hourly basis
using meteorological inputs corresponding to the station shown in Table 2-3. Separate unit-area time
series were produced for developed pervious and impervious land, representing a continuous simulation
of approximately 30 years of input meteorology derived from weather monitoring stations. The
16
-------
representation of sediment-associated loading of TP from pervious land in the Susquehanna River HSPF
model was corrected from the representation in the previous model version.
The meteorological inputs to HSPF include precipitation, air temperature, and potential
evapotranspiration (PET). All of the meteorological inputs were modified for the future climate scenarios,
including calculation of PET using the Penman-Monteith energy balance method. Therefore, the impact
of changing temperature and PET were included in HSPF's generation of runoff and pollutant
concentrations, which are the inputs to SUSTAIN as stated above.
It is important to note that only surface runoff modeled by HSPF is used as an input to SUSTAIN.
Infiltration and ET from site land surfaces are modeled by HSPF, but are not tracked in this project.
Nevertheless, SUSTAIN simulates infiltration into the underlying soil via BMPs, and also calculates ET
loss directly from BMPs. Daily minimum and maximum air temperature time series from HSPF were
used as inputs to SUSTAIN, which then calculated evaporation and transpiration losses from BMPs using
the Hamon method (Hamon,1961).
2.4.2. Low, Medium, and High Intensity Change Scenarios Used for
Sensitivity Analysis
The Midwest location uses the range of intensity changes to explore the extent of projected impacts
(smallest, average, and largest). The previous section describes how the climate scenario with the highest
change in large storm event intensity was selected from the pool of 10 future climate scenarios. The same
approach was taken here, but in this case, scenarios were selected from the pool of 10 to represent the
lowest, medium, and highest changes in intensity. The highest intensity change scenario was the same as
previously selected, as discussed in the preceding section.
2.4.3. Percentage Difference Scenarios Used for Sensitivity Analysis
The climate scenarios discussed up to this point were derived from a variety of GCM outputs and spatial
downscaling approaches. The future projected precipitation patterns show a great deal of variability with
respect to degree of intensity change by rainfall depth, monthly volume increases and decreases, and
interannual volume changes. While GCM/downscaling approaches provide detailed representations of
possible future conditions, a simpler structured approach assessing system sensitivity to percentage
change in different climatic drivers is also informative. When a series of percentage changes is explored
(e.g., +5, +10, +15%), the resulting responses are more comparable because sources of variation in the
future precipitation are minimized. Other meteorological parameters may also be modified in a similar
manner. Air temperature is often modified using a fixed delta (e.g., +3°F) applied to each observation.
For Harford County, MD and Scott County, MN, graduated precipitation and temperature changes were
applied to the historic records for each site. The graduated climate scenarios were applied to all of the
stormwater management scenarios for both locations. The current precipitation record was modified to
represent potential future climate conditions by applying a graduated set of percentage changes to the
entire precipitation record (across the entire range of hourly precipitation values from a trace to the largest
rainfall value, in other words); therefore, the number of events and event durations remained unchanged.
The percentage change factors were -10 (a decrease in intensity), +10, and +20%(both increases in
17
-------
intensity). Air temperature was adjusted as well using the Clausius-Clapeyron relationship (Clausius,
1850; Clapeyron, 1834), such that a 10% change in precipitation intensity was paired with a 2.6°F change
in air temperature. PET was then recalculated from the modified air temperature and precipitation time
series using the same method employed for the ""20 Watersheds" project. The unit-area runoff HSPF
models were executed with the new meteorological inputs, and the resulting surface runoff time series
used for the climate sensitivity analysis.
2.5. SITE ASSUMPTIONS
There were two elements of design assumptions that were developed generically for all of the
locations/practice scenarios: those related to piped stormwater conveyance and those related to
stormwater infrastructure cost. Site-specific assumptions are detailed in the individual site Sections 3.
through 7. .
2.5.1. Stormwater Conveyance Representation
A simplified pipe-sizing methodology was used to estimate appropriate conveyance capacity for each
site's design storm runoff. For the smaller sites (Atlanta, GA and Maricopa County, AZ), peak flow rates
and pipe sizes were calculated for each delineated subwatershed area. For the larger sites (Scott County,
MN and Harford County, MD), a generic sizing table was developed that matched drainage area threshold
size with adequate pipe diameter. The Portland, OR site was less than one acre and did not include any
piped conveyance. For all sites, the conveyance infrastructure was designed according to the 10-year,
24-hour storm event (typical for municipal storm sewers), unless noted otherwise.
The two primary methods for estimating peak flow and culvert sizes are described below. The different
approaches used between the smaller sites and larger sites are also explained in further detail.
2.5.1.1. Peak Flow Estimation
For smaller watersheds, the Rational method is appropriate for estimating peak discharges for specified
design storms. Although it is considered a crude but efficient method, its level of precision is justified by
the need to select among standard pipe sizes available. The Rational equation is defined as follows:
Q = CIA (2-1)
In which
0 = peak discharge (cfs)
C = composite runoff coefficient for the watershed (dimensionless)
1 = average rainfall intensity (inch/hour) for storm frequency, duration (equal to time of
concentration), and geographic area
A = watershed area (acre)
18
-------
Although runoff coefficients (C) are provided for a large range of land use/cover types and hydrologic
soil group (HSG) conditions, composite C values can also be calculated from a site's total impervious
percentage using the following equation:
C = 0.9(% impervious) + Cp(l — % impervious) (2-2)
Where Cp represents the pervious coefficient runoff value, which is defined according to the
watershed HSG (see Table 2-4).
Table 2-4. Maryland Stormwater Manual pervious runoff coefficients
HSG
Cp
A
0.20
B
0.25
C
0.30
D
0.35
Rainfall intensities were calculated using statistical rainfall depth-duration tables from the National
Oceanic and Atmospheric Administration's (NOAA's) Precipitation Frequency Data Server (PFDS). For
the 10-year frequency storm, rainfall depths were selected for a storm duration equal to the time of
concentration for each drainage area. Rainfall intensity was then calculated by dividing the 10-year
precipitation depth (inch) by the time of concentration (hour).
The timing variable, also referred to as the time of concentration variable (I'c) was assumed to equal
5 minutes for all of the drainage areas in the Atlanta and Maricopa County scenarios. Although actual Tc
values are likely less the 5 minutes, the Rational method uses a minimum Tc value of 5 minutes to
estimate peak discharge. For the larger site scenarios (i.e., MD and MN), time of concentration values
were calculated using the Kirpich equation, formulated as:
K
L3"
.1281
H
(2-3)
In which
Tc = time of concentration (minute)
L = hydraulic length of watershed (feet)
19
-------
H = height (feet) of highest elevation point in watershed above the outlet (elevation
difference)
K = multiplier (dimensionless) associated with the nature of flow path (1.0 for flow in
mixed urban settings)
2.5.1.2. Culvert Sizing
In lieu of Federal Highway Administration culvert capacity charts, an analytical model was used to
determine culvert sizes for the various stormwater management scenarios. For circular pipes operating
under inlet-control conditions, the orifice equation can be expressed as follows:
Q = 0.0437CdD2Jz-^ (2-4)
In which
O = discharge (cfs)
D = pipe diameter (inch)
Z = depth of water above the centerline of pipe entrance (feet)
Cd = coefficient of discharge (dimensionless)
Headwater/pipe diameter ratio (Hw/D) is the headwater depth (height above pipe centerline) divided by
the pipe diameter. For all scenarios, the Hw/D ratio was set to 2, assuming that new developments will be
able to minimize culvert depths for most upland drainage areas. Cd was assumed to equal the default
value, 0.6. Hydraulic Condition was assumed to be under inlet control. For new developments, the
downstream discharge location (e.g., open ephemeral channel or floodplain) will likely yield greater
conveyance capacity than the 10-year peak flow rate.
For each drainage area in the Atlanta and Maricopa County scenarios, the 10-year peak discharge was
calculated using the Rational method and aforementioned input assumptions. Culvert sizes were then
selected using an iterative approach. Using the orifice equation to determine pipe discharge, the pipe
diameter was varied by the standard available pipe sizes to find the minimum pipe size that can convey
the 10-year peak flow.
For the Scott County and Harford County scenarios, the site areas were too large and complex to calculate
culvert sizes for each subwatershed area. Instead, a threshold table was developed to automatically assign
a culvert size based on total drainage area. To develop the threshold table, the Rational equation was
ultimately rearranged to calculate a drainage area size for each culvert size and associated peak discharge
capacity. Because the Tc and average rainfall intensity (based on the Tc value) changes with watershed
size, a separate table was first created that matched time of concentration, rainfall intensity, and drainage
area size. The resolution of this "Rational input table" were based on 1-minute Tc intervals between the
minimum assumed Tc (i.e., 5 minutes) and the maximum calculated Tc for the largest drainage area in the
20
-------
scenario site. Precipitation values were interpolated from NOAA PFDS tables between reported values
(e.g., 5, 10, 15-minute duration intervals). Finally, drainage area thresholds were assumed for each pipe
diameter by iteratively back-calculating the drainage area from the Rational equation, using each pipe's
peak discharge (orifice equation) and the associated rainfall intensity (extracted from the "Rational input
table").
2.5.2. Infrastructure Cost Estimates
Cost estimates were developed for adapting gray and GI practices using a cost/tradeoff analysis. Life
cycle costs were estimated for a 20-year project operation period, assuming that design-build (capital)
costs will occur in the year before the first year of operation and BMP operation and maintenance (O&M)
will occur annually for 20 years. While individual BMPs and BMP types vary in life span, the 20-year
period allows for ease of comparison across the scenarios and reflects a typical planning period for
stormwater management.
BMP life-cycle costs were estimated using literature sources and best professional judgment based on
Tetra Tech project experience. The primary sources were King and Hagan (2011) and the Green Values
Calculator (CNT, 2014). The King and Hagan (2011) model, produced by the University of Maryland
Center for Environmental Science, incorporates BMP cost information into Maryland's Assessment and
Scenario Tool. Their study is a summary of previous regional studies, which were verified or modified
based on interviews with stormwater experts. The Green Values Calculator was developed by the Center
for Neighborhood Technology (CNT), a nonprofit organization with a national scope, in collaboration
with the EPA Office of Wetlands, Oceans, and Watersheds* assessment and Watershed Protection
Division. Similar to King and Hagan (2011), the purpose of the tool is to evaluate performance, costs, and
benefits of GI practices when compared to conventional treatment. Costs were compiled on a national
scale from literature reviews along with information from municipalities, public utilities, and research
institutions. The King and Hagan (2011) model provides cost data on capital as well as operation and
maintenance expenditures on a cost per impervious drainage area basis. The Green Values Calculator
provides cost data based on a whole BMP measure (e.g., cost per square foot of permeable pavement).
Both sources provide sufficient data for approximate life-cycle cost estimates, including capital and O&M
costs.
These sources covered most of the BMPs implemented but were supplemented with other cost data. RS
Means (2016), a construction industry cost database, was used to adjust BMP costs and estimate the cost
of additional infrastructure (e.g., culverts). The cost estimates were developed to reflect national averages.
Insufficient data were available to estimate local or regional differences in costs.
Some adjustments to the cost data methods were necessary in a few cases. For instance, changes to BMP
storage volumes, not drainage areas, are likely to occur between the current and future scenarios. For
BMP costs based on impervious surface drainage area, a unit cost by BMP volume (e.g., cost per cubic
foot) was calculated by dividing the current scenario BMP cost by the current scenario BMP volume.
That unit cost by volume was then applied to the future scenario volume to estimate the BMP cost for the
future scenarios. BMP volume was assumed to be treatment volume, not total excavation volume.
21
-------
Table 2-5 presents the unit costs assumed for the infrastructure costs estimates, indicating the data sources
by type infrastructure. Note that costs are a function of BMP size, not BMP treatment area. Capital costs
reflect both preconstruction (planning, design, and engineering) and construction costs. To estimate the
present value (PV) of annual costs over a BMP's lifetime, the annual O&M costs were discounted at a
rate of 3%. The PV life-cycle costs reflect the sum of the capital and 20-year PV O&M costs.
Stormwater infrastructure costs vary widely and are often site-specific. These unit cost estimates reflect
the best available information and professional judgment on average costs for a comparative analysis at a
national-scale. These costs are not appropriate for use in site-level budget estimates.
22
-------
Table 2-5. Unit cost estimates for modeled practices
Practice
Unit3
Capital
cost ($)
Annual
O&M ($)
20-Year
present value
O&M ($)
Present
value
life-cycle
cost ($)
Reference(s)
Compost amended soils
SF
4
0.40
7
11
RS Means (2016)
Impervious surface
disconnection
SF roof
area
0.10
0.0002
3
3
CNT (2014)
Stonnwater harvesting
basin
SF
21
2.00
30
51
RS Means (2016)
Bioretention
SF
25
2.00
30
55
RS Means (2016);
King and Hagan
(2011)
Bioretention with
underdrain
SF
60
2.00
30
90
King and Hagan
(2011); MPCA (2016)
Bioretention
swales* infiltration
trench hybrid
SF
126
2.00
30
156
RS Means (2016);
King and Hagan
(2011)
Green roof
SF
19
1.00
8
27
CNT (2014)
Permeable pavement
SF
5
0.20
3
8
CNT (2014); King
andHagan(2011)
Permeable pavement with
underdrain
SF
36
0.20
3
39
CNT (2014); King
and Hagan (2011);
MPCA (2016)
Dry detention basin
CF
8
0.20
3
11
King and Hagan
(2011)
Extended dry detention
basin
CF
8
0.20
3
11
King and Hagan
(2011)
Underground dry
detention basin
CF
19
0.20
1
20
CNT (2014)
Wet pond
CF
8
0.20
3
11
King and Hagan
(2011)
Infiltration basin
CF
22
0.20
4
26
King and Hagan
(2011)
Infiltration trench
CF
22
0.20
4
26
King and Hagan
(2011)
Cistern
CF
13
1.00
14
27
CNT (2014), Impact
Infrastructure and
Stantec (2014), and
LIDC (2005)
Sand filter
CF
41
0.90
13
54
King and Hagan
(2011)
Underground sand filter
CF
47
1.00
15
62
King and Hagan
(2011)
23
-------
Table 2 5. Unit cost estimates for modeled practices (Continued)
Practice
Unit*
Capital
cost ($)
Annual
O&M ($)
20-Year
present value
O&M ($)
Present value
life cycle cost
($)
Reference(s)
Concrete drainpipe, 12"
diameter
LF
52
0.50
7
59
RS Means (2016)
Concrete drainpipe, 15"
diameter
LF
58
0.50
7
65
RS Means (2016)
Concrete drainpipe, 18"
diameter
LF
70
0.50
7
77
RS Means (2016)
Concrete drain pipe, 21"
diameter
LF
80
0.50
7
88
RS Means (2016)
Concrete drain pipe, 24"
diameter
LF
96
0.50
7
103
RS Means (2016)
Concrete drain pipe, 27"
diameter
LF
127
0.50
7
134
RS Means (2016)
Concrete drainpipe, 30"
diameter
LF
139
0.50
7
146
RS Means (2016)
Concrete drainpipe, 33"
diameter
LF
158
0.50
7
166
RS Means (2016)
Concrete drainpipe, 36"
diameter
LF
178
0.50
7
185
RS Means (2016)
Concrete drain pipe, 42"
diameter
LF
232
0.50
7
239
RS Means (2016)
Concrete drain pipe, 48"
diameter
LF
269
0.50
7
276
RS Means (2016)
Concrete drain pipe, 54"
diameter
LF
329
0.50
7
336
RS Means (2016)
SF = square feet, CF = cubic feet, LF = linear feet.
aUnits reflect BMP size, not BMP treatment area.
24
-------
3. MID-ATLANTIC SITE: MIXED USE
3.1. REGULATORY REQUIREMENTS AFFECTING STORMWATER
MANAGEMENT
Harford County defers its stormwater requirements to the State of Maryland, which are published in the
Maryland Stormwater Manual (Center for Watershed Protection and Maryland Department of
Environment, 2000). The Maryland Stormwater Manual uses a tiered approach for managing stormwater
at different scales and for different purposes. Requirements are discussed below.
• Rev: The recharge volume addresses impacts to groundwater resulting from development. The
volume must be completely infiltrated on the site. Rev is equal to the product of an area-weighted
value based on HSG, a coefficient based on percentage impervious area, and site area. A
volume-based criterion using a volumetric runoff coefficient is used for structural BMPs and an
area-based criterion is used for nonstructural practices. A variety of structural and nonstructural
practices can be used to meet the Rev requirement.
• WQv: In the region of Harford County, the water quality volume is equal to 1 inch times the site
area times a coefficient based on percentage impervious area. The detention time for treatment is
24 hours. Treatment must achieve 80% TSS removal and 40% TP removal. Rev can be subtracted
from the WQv.
• Cpv: The channel protection criterion requires 24-hour extended detention of the
postdevelopment 1-year 24-hour storm.
• Qp: The overbank flood protection criterion requires peak matching to predeveloped conditions
for the 10-year 24-hour storm event. The Manual notes it is optional and depends on the review
authority. Harford County requires it.
• Credits are given for various forms of impervious surface disconnection. There are specific
minimum flow path length requirements for the credits.
3.2. STORMWATER MANAGEMENT SCENARIOS
The 20-acre mixed-use site (see Figure 3-1) is assumed to have the following characteristics in each of the
scenarios:
• The site is 65% impervious, distributed as follows:
o 41% road, parking, and sidewalk area,
o 24% building area.
• The remaining pervious area (35%) is comprised of lawn/landscaping.
• The HSG percent distribution is based on a regional geographic information system (GIS)
analysis (the portion of Harford County within the Susquehanna River Basin). The HSG
composition is used for design storm event routing calculations to size practices for peak flow
25
-------
control. Predevelopment land cover is assumed to be woods in good condition. It is likely that a
small site would have at most two HSG types of course, but the HSG distribution was used to be
representative of average conditions in the region.
o HSG A: 1%
o HSG B: 71%
o HSG C: 20%
o HSG D: 8%
Legend
Building
Other
Pervious
Roadway
Figure 3-1. Mixed-use site layout (Harford County, MD).
26
-------
Two scenarios have been developed representing different approaches to stormwater management: a
conventional scenario using gray practices and a GI scenario using a combination of green and gray
practices. As noted in Section 5. , the conventional scenario was used as the basis for a fourth scenario in
which distributed GI practices were added to achieve current performance under future climate
conditions. The site's percentage impervious area is sufficiently high that it is not feasible to use only GI
practices to meet the regulatory requirements. The scenarios are described in the following subsections.
3.2.1. Conventional (Gray) Infrastructure
The key design elements in the Conventional (Gray) Infrastructure scenario are as follows:
• Surface sand filters address the Rev and WQv requirements. A portion of the volume is infiltrated
into the soil, meeting the Rev. The remainder of the volume is treated by the sand filter and
discharged via underdrain, meeting the WQv requirement.
• An extended dry detention basin treats the entire site to address the CPv and Qp requirements.
o 24-hour drawdown of CPv is provided via a low flow orifice.
o Peak matching for 10-year 24-hour storms (Qp requirement) is addressed using a weir.
As seen in Figure 3-2 for the Conventional scenario, runoff reaches surface sand filters distributed
throughout the site via overland flow. Underdrain flow and larger storm event overflow from the sand
filters is then conveyed to the extended dry detention basin, which then discharges flow offsite. To
simplify the representation in SUSTAIN, the site sand filters were lumped into six representative sand
filters, one for each drainage area.
27
-------
Legend
| Building
Other
Pervious
Roadway
| | Drainage Areas
BMP Type
I Extended Detention Basin
HHHf Surface Sand Filter
- - - ' BMP Conveyance
Figure 3-2. Mixed-use Conventional (Gray) Infrastructure stormwater management
scenario (Harford County, MD).
For the SUSTAIN simulation, the surface sand filters were sized according to design standards published
in the Man land Stonnwater Manual, taking into account the contributing drainage areas and percentage
impervious area. The design specifications for "perimeter sand filter' (a type of surface sand filter) were
used to represent the configuration. An underlying soil infiltration rate of 0.52 inches/hour was assumed
for the site, reflecting the minimum infiltration rate needed to use infiltrating practices (an analysis of
soils data for the northern portion of Harford County in the Susquehanna River Basin indicates that even
higher infiltration rates are typical in this region). The sand filter media was assumed to achieve pollutant
removal rates of 86% for TSS, 30% for TN, and 60% for TP using published performance values from the
Center for Watershed Protection (2007) and Hirschman et al. (2008).1 Removal was modeled in
SUSTAIN for only the volume that filtered through the sand media and was subsequently discharged via
'The 2014 BMP Performance Summaries published by the International Stonnwater BMP Database (Geosyntec
Consultants and Wright Water Engineers, 2014) became available around the time we developed assumptions for
BMP pollutant removal performance. However, for the most part our values are within the 95% confidence intervals
of the percentage reductions relative to influent-effluent concentrations shown by their performance summaries.
28
-------
the underdrain. Any pollutants carried by runoff infiltrating into the underlying soil were assumed to be
removed completely from the system. While dissolved pollutants may be transported via groundwater and
eventually be exported from the site, this study is focused on stormwater and surface runoff only.
The dry extended detention basin was sized and configured for SUSTAIN using a two-step process:
• The CPv was calculated using design criteria in the Maryland Stormwater Manual.
• A routing spreadsheet was created simulating design storm runoff using a 1-minute time step.
Inflow hydrographs were produced using TR-55 methods (USDA, 1972, 1986) for undeveloped
(forest) and developed conditions. Detention basin outflow was represented using basin
dimension, stage, and outlet characteristics (orifice/weir size and stage). The detention basin size
and outlet characteristics were optimized to allow for release of the CPv over a 24-hour period via
an orifice; then, a weir was used to match the developed site peak outflow from the 10-year
24-hour storm to the predeveloped condition.
The basin was assumed to be earthen and a background infiltration of 0.52 inches/hour was included in
SUSTAIN, allowing for infiltration and removal of runoff and pollutants. ET was modeled in SUSTAIN
for both the sand filters and the detention basin.
3.2.2. Green Infrastructure (Gl) with Gray Infrastructure
The key design elements in the Green with Gray Infrastructure scenario are as follows:
• Permeable pavement is used for the parking areas and sidewalks throughout the site.
• Infiltration Basins (aboveground) and infiltration trenches (below ground) address the Rev, WQv,
and CPv. The roads drain to infiltration basins, while the rooftops drain to the infiltration
trenches. The entire capture volume is infiltrated into the underlying soil.
• A dry detention basin addresses the Qp only.
Figure 3-3 provides the BMP locations and site conveyance for the GI with Gray scenario. Road runoff is
conveyed to the infiltration basins either via curb flow or by culvert. Rooftops drain to adjacent
infiltration trenches. Overflow from large storm events is then conveyed from the infiltration practices via
a separate drainage network to the dry detention basin. Flow is then discharged offsite.
29
-------
Legend
| Building
Other
| Pervious
Roadway
~ Drainage Areas
" - » Road Drainage Conveyance
BMP_Type
|_" Dry Detention Basin
j Infiltration Basin
Infiltration Trench
Permeable Pavement
- - - ' BMP Conveyance
Figure 3-3. Mixed-use Green Infrastructure (GI) with Gray Infrastructure
stormwater management scenario (Harford County, MD).
The process for developing the SUSTAIN simulation representation was similar to the approach used for
the conventional site scenario. The permeable pavement was configured to be consistent with Maryland
Stormwater Manual. No external runoff was directed to the permeable pavement, and the subbase
provided sufficient storage to address the CPv. The infiltration basins and infiltration trenches were sized
to capture the CPv using the Manual's procedures for applying an alternative stormwater management
strategy called "Environmental Site Design," which incorporates low impact development (LID)
principles. Volume in excess of the CPv were routed to a conventional earthen dry detention basin
designed to match postdeveloped peak flow to predeveloped conditions for the 10-year 24-hour storm, as
well as pass larger storm events with a spillway. The detention basin dimensions and outlet configuration
were estimated using a stage-storage-discharge spreadsheet as was done for the conventional site. All of
the practices were assumed to have infiltration rates of 0.52 inches per hour. The only pollutant removal
mechanism is via infiltration from the BMPs. ET was simulated to occur from the detention basin and
from the infiltration basins, but not from the infiltration trenches (which store runoff for infiltration below
30
-------
the ground surface). A small amount of ET from permeable pavement was assumed, equal to 10% of ET
that would normally take place, based on best professional judgment.
3.2.3. Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (Gl)
The objective of this scenario is to address the performance gap between current and future climate by
incorporating additional distributed GI practices into the site. The current conventional BMP (extended
detention basin and surface sand filters) configurations are unchanged and distributed GI practices are
added during the adaptation SUSTAIN run to provide treatment equivalent to the current conditions
climate scenario.
For the Conventional with Distributed GI scenario, infiltration trenches are distributed throughout the
20-acre study site (represented as one aggregate infiltration trench per drainage area). The same basic
design used for the GI with Gray Infrastructure was incorporated, but practices were not sized for a
specific design criterion. Rather, SUSTAIN was allowed to dynamically size the individual trenches until
the site's performance met or exceeded all of the performance measures defined for current conditions as
discussed in Section 2.3.2. .
3.3. ADAPTATION SIMULATION
The objective of the adaptation simulation is to determine the increases in BMP footprint (surface area)
that would be required to maintain current levels of performance under future climate conditions for each
stormwater management scenario. Table 3-1 summarizes the key components of the modeling procedure
for each scenario. In the GI with Gray scenario, all practice types were resized except for permeable
pavement because no additional area was considered to be available for permeable pavement. As
discussed in Section 3.2.3. , the Conventional with Distributed GI scenario consisted of adding distributed
green practices (infiltration trenches) to the site; the conventional practices were not resized.
Table 3-1. Features of adaptation simulation for Harford County, MD
Location
Stormwater management
scenario
Future adaptation
Affected practices
Harford
County, MD
Conventional (Gray) fnfrastructure
Resize practices
Surface sand filters, extended dry
detention basin
Gf with Gray fnfrastructure
Resize practices
tnfiltration trenches, infiltration
basins, dry detention basin
Conventional (Gray) fnfrastructure
with Distributed Gf
Add distributed Gf
to site
Distributed infiltration trenches
31
-------
3.4. CURRENT AND FUTURE CHANGES IN PRECIPITATION
Water-year annual precipitation ranked by current climate totals is shown in Figure 3-4. Annual
precipitation volume is projected to increase for each year of the simulation period; however, the increase
varies from a few inches per year to over 10 inches per year. The overall annual averages are 44.3 inches
for current conditions, and 50.0 inches for future conditions, reflecting a 12.8% increase in volume. A
comparison of current and projected future monthly average precipitation is provided in Figure 3-5.
Precipitation depth increases for some months and decreases in other months. The largest change is seen
in September, where projected average depth increases from about 4 to 7 inches. In addition, daily sums
of precipitation depth were calculated and were used to determine percentiles of 24-hour depth of interest
to stormwater managers (see Table 3-2). While daily sums do not provide a true measure of storm event
depth (storms have variable lengths and may span more than 1 day), they do provide useful information
about expected depths over a 24-hour period. As seen in the table, the change in depth between current
and future ranges from 0.09inches for the 85th percentile to 0.43 inches for the 99th percentile.
While the first two figures provide an indication of changes in overall precipitation volume, they do not
speak to changes in storm event volume and intensity, which was used as the basis for the selection of the
future climate scenario among 10 candidate scenarios. Figure 3-6 shows the highest hourly precipitation
volumes in the current and future precipitation time series, plotted by recurrence interval in years in the
30-year simulation period. It is important to note that the depths shown (1) are not storm event depths but
rather hourly precipitation values and (2) may not reflect the true distribution of hourly depths due to the
use of only 30 years of meteorological history. However, the figure provides a useful way to visualize
volume/intensity changes for the largest events resulting from projected climate change. There is an
approximately 1.5-fold increase in hourly precipitation depth across the recurrence range. The figure also
provides an indication in the change in frequency for a given depth. The depth corresponding to 10-year
recurrence under current conditions is projected to take place at a 2-year recurrence under future climatic
conditions.
32
-------
Annual Precipitation (30-yr record)
80
¦ Current ¦ Future
70'
60
50
.£ 40
30
20
10
0
rlNfO'tl/liIlNCOlJlOrlNrO'tulyDNMCnOrlrNlrO'tuluDNMaiO
-------
Table 3-2. 24-hour precipitation depth percentiles for current conditions and high
intensity future climate scenario at Harford County, MD
Percentile
Current conditions
24-h depth (in)
Future climate 24-h
depth (in)
Change (+/-in)
85th
0.81
0.90
+0.09
90th
1.03
1.15
+0.12
95th
1.39
1.58
+0.19
99th
2.33
2.76
+0.43
Hourly Precipitation
4
3.5
3
2.5
.£ 2
1.5
1
0.5
0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
Simulation Recurrence Interval (years)
Figure 3-6. Hourly precipitation recurrence interval for current conditions and high
intensity future climate scenario at Harford County, MD.
3.5. RESULTS
SUSTAIN was run under the following conditions for each stormwater management scenario:
• Current climate, site without stormwater management/BMPs
• Future climate, site without stormwater management/BMPs
• Current climate, site with stormwater management/BMPs
• Future climate, site with stormwater management/BMPs
• Future climate, site with BMPs adapted to meet current hydrology and water quality performance
Current
Future
34
-------
As shown in Table 3-3, 11 sets of SUSTAIN runs were performed for a combination of three stormwater
management approaches and four climate scenarios. The Conventional with Distributed GI approach is a
variation of the Conventional approach, where the adaptation to meet current performance used addition
of GI components to the site, rather than resizing practices already in place as was done for the
Conventional approach. Because the minus 10% change climate scenario resulted in a reduction to all
performance measures, there was no need to perform an adaptation run because current performance was
already met. As a result, no SUSTAIN run was needed for the Conventional with Distributed GI approach
because it is identical to the Conventional approach prior to adaptation.
Table 3-3. Stormwater management and climate scenarios for Harford County, MD
Stormwater management
approach
GCM high
intensity
Minus 10%
Plus 10%
Plus 20%
Conventional
X
X
X
X
GI + Gray
X
X
X
X
Conventional with Distributed GI
X
X
X
A full presentation of the results of all the runs is provided in APPENDIX B. . For brevity, the results in
this section focus on a few topics of interest to stormwater managers: (1) a comparison of the site
performance with BMPs between current and future climate conditions, (2) the increases in BMP
footprints needed to offset impacts of climate change when BMPs are adapted using SUSTAIN
optimization, and (3) a comparison of current stormwater infrastructure costs to future costs when BMPs
are adapted to offset impacts of climate change.
For the comparison of the site performance with BMPs between current and future climate conditions, the
downscaled future GCM (high intensity change) scenario was selected for the comparison. A discussion
of other topics of interest are provided in the general conclusions Section 8. , including changes in
pretreatment site performance, changes in post-treatment site performance, climate scenario sensitivity
analysis, and adapting BMPs under future climate to meet current performance.
Rather than comparing the performance of the stormwater management approaches independent of
climate change (i.e., how much better does one perform than the other under current conditions), this
study focuses on how the stormwater management approaches compare relative to climate change. Table
3-4 provides current and future performance for the stormwater management approaches, normalized to
area. Note that there is no numeric measure of change in the FDC between current and future climate, so
the highest hourly peak flow during the simulation is presented as a proxy for large storm event response.
Figure 3-7 through Figure 3-11 present each metric graphically from Table 3-4. For annual average site
runoff, the increase in runoff for the Conventional approach at 3.92 inches is more than double the runoff
increase for GI + Gray at 1.88 inches. This indicates the GI + Gray approach was better at disposing of
additional runoff due to changes in future precipitation volume than the Conventional approach,
35
-------
suggesting that the GI + Gray approach is more resilient to climate change for this measure. The same is
not true for the maximum hourly peak flow, where both approaches gained 0.67 cfs/acre, as well as for
the sediment loading rate, where both sites increased by a bit less than 0.1 ton/acre/year. The GI + Gray
approach does appear to be somewhat more resilient for the nutrient loading rates, with increases
somewhat less than for the Conventional approach.
Table 3-4. Current and future performance of Harford County, MD site by
stormwater management approach
Stormwater management
approach
Current
Future
Change
Runoff (inch/yr)
Conventional
7.04
10.96
+3.92
GI + Gray
1.52
3.40
+1.88
Maximum hourly peak flow (cfs/ac)
Conventional
1.12
1.80
+0.67
GI + Gray
0.85
1.52
+0.67
Sediment (ton/ac/yr)
Conventional
0.12
0.20
+0.09
GI + Gray
0.04
0.11
+0.08
TN (lb/ac/yr)
Conventional
2.74
4.34
+1.60
GI + Gray
0.64
1.58
+0.94
TP (lb/ac/yr)
Conventional
0.32
0.51
+0.18
GI + Gray
0.07
0.18
+0.11
36
-------
12.00
10.00
8.00
6.00
4.00
2.00
0.00
Annual Site Runoff
10.96
Conventional Gl + Gray
¦ Current ¦ Future
Future
Current
Figure 3-7. Annual site runoff under current climate and future general circulation
model (GCM) scenario by stormwater management approach for Harford County,
MD.
Maximum Peak Flow
1.80
2.00
1.50
1.00
0.50
0.00
Conventional Gl + Gray
¦ Current ¦ Future
Future
Current
Figure 3-8. Maximum hourly peak flow under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Harford County, MD.
37
-------
0.25
0.20
0.15
0.10
0.05
0.00
Annual Sediment Loading Rate
0.20
Conventional Gl + Gray
¦ Current ¦ Future
Future
Current
Figure 3-9. Annual sediment loading rate under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Harford County, MD.
Annual TN Loading Rate
0.00
Conventional Gl + Gray
¦ Current ¦ Future
Figure 3-10. Annual TN loading rate under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Harford County, MD.
38
-------
Annual TP Loading Rate
051^_
0.60
0.50
1—
0.40
0.32 |
018^_
M 1
0.30
.a
0.20
Future
O07^_
0.10
Current
0.00
Conventional
GI + Gray
¦ Current
¦ Future
Figure 3-11. Annual TP loading rate under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Harford County, MD.
Design Results - General Circulation Model (GCM) High Intensity
For the Conventional (Gray) Infrastructure scenario, the optimal solution resulted in an increase in size
for both the extended dry detention basin and the surface sand filters. The detention basin needs to be
more than triple its original size, while the sand filters show an increase of nearly 50%. The combined
increase in BMP footprint is equal to about 7% of the site area, about 1.4 acres.
The GI with Gray Infrastructure scenario uses a smaller detention basin for large flooding event
mitigation and transfers much of the stormwater control to infiltration practices. For this adaptation,
SUSTAIN favored the infiltration practices for resizing, with infiltration basins quadrupling in size, and
the infiltration basins more than tripling in size. On the other hand, the detention basin has a 130%
increase, more than double the size under current climate conditions. The combined increase in BMP
footprint amounts to nearly 10% of the site area, over 2 acres. Note that permeable pavement was not
included as an adaptation practice because it was already implemented to the maximum practical extent.
The Conventional Infrastructure with Distributed GI scenario differs from the previous ones in that GI
practices are added to the site rather than resizing the practices already present. To offset the impacts of
climate change, nearly 100,000 square feet (over 2 acres) of infiltration trenches must be added to the site
to mitigate future climate change impacts under the GCM high intensity future climate scenario.
Design Results - Intensity Change Plus 10%
In the Conventional (Gray) Infrastructure scenario, the SUSTAIN optimization selected the extended dry
detention basin only for adaptation. This outcome is somewhat surprising and follows a different pattern
than the adaptations to the GCM high intensity future climate scenario, where both sand filters and the
39
-------
detention basin were selected for resizing during the optimization as being the most cost-effective
solution. The required basin footprint for future climate adaptation reflects a near doubling of size.
In the GI with Gray Infrastructure scenario, the SUSTAIN optimization targeted resizing only the
infiltration basins and infiltration trenches, the opposite outcome as seen for the Conventional scenario.
This difference is likely due in part to sediment and TP loads being the limiting factor, resulting in the
selection of practices with the best infiltration capacity and load reduction. The required infiltration basin
footprint reflects nearly a 50% increase, and nearly a 40% increase is required in the infiltration trench
footprint.
For the Conventional (Gray) Infrastructure with Distributed GI scenario, the addition of 15,351 square
feet of distributed infiltration trenches would be required to maintain current BMP performance. This
footprint represents approximately 1.8% of the total site area.
Design Results - Intensity Change Plus 20%
In the Conventional (Gray) Infrastructure scenario, the extended dry detention basin footprint must
increase by a factor of 2.8 for future climate adaptation. In the GI with Gray Infrastructure scenario, the
required infiltration basin footprint reflects a more than doubling in size, and nearly a 60% increase is
required in the infiltration trench footprint. For the Conventional (Gray) Infrastructure with Distributed GI
scenario, the addition of 32,514 square feet of distributed infiltration trenches would be required to
maintain current performance. This footprint represents approximately 3.7% of the total site area.
Table 3-5 summarizes the increases in BMP footprints for the Harford County stormwater management
scenarios that would be required to maintain current performance under future climate conditions. The
current and adapted footprints are presented in terms of both actual square feet of practice as well as
percentage of overall site area. The latter is provided as a means of comparing the current and future
adapted sizes relative to the site area (20 acres) for this particular development type (mixed use). Results
are discussed separately for each of the future climate scenarios modeled for the Harford County site.
40
-------
Table 3-5. Comparison of current and future adapted best management practice
(BMP) footprints for Harford County, MD stormwater management scenarios
Future
climate
scenario
Stormwater
management
scenario
Practice
Current
Future adapted
%
Increase in
footprint
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint as
% of site
area
GCM high
intensity
Percentage
difference
plus 10%
Conventional
(Gray)
Infrastructure
Extended dry
detention basin
25,000
2.9
81,250
9.3
225
47
130
306
224
0
0
0
Surface sand
filters
10,119
1.2
14,840
1.7
GI with Gray
Infrastructure
Dry detention
basin
10,000
1.1
23,000
2.6
Infiltration basin
12,858
1.5
52,155
6.0
Infiltration
trench
14,800
1.7
47,954
5.5
Permeable
pavement
201,242
23.1
201,242
23.1
Conventional
(Gray)
Infrastructure
with Distributed
GI
Extended dry
detention basin
25,000
2.9
25,000
2.9
Surface sand
filters
10,119
1.2
10,119
1.2
Distributed
infiltration
trenches
0
0
95,869
11.0
Conventional
(Gray)
Infrastructure
Extended dry
detention basin
25,000
2.9
25,000
2.9
0
98
Surface sand
filters
10,119
1.2
20,023
2.3
GI with Gray
Infrastructure
Dry detention
basin
10,000
1.1
10,000
1.1
0
47
38
0
Infiltration basin
12,858
1.5
18,943
2.2
Infiltration
trench
14,800
1.7
20,435
2.3
Permeable
pavement
201,242
23.1
201,242
23.1
Conventional
(Gray)
Infrastructure
with Distributed
GI
Extended dry
detention basin
25,000
2.9
25,000
2.9
0
0
Surface sand
filters
10,119
1.2
10,119
1.2
Distributed
infiltration
trench
0
0.0
15,351
1.8
41
-------
Table 3-5. Comparison of current and future adapted best management practice (BMP)
footprints for Harford County, MD stormwater management scenarios (Continued)
Current
Future adapted
Future
Stormwater
Footprint as
Footprint as
%
climate
management
Footprint
% of site
Footprint
% of site
Increase in
scenario
scenario
Practice
SF
area
SF
area
footprint
Percentage
Conventional
Extended dry
25,000
2.9
25,000
2.9
0
difference
(Gray)
detention basin
plus 20%
Infrastructure
Surface sand
filters
10,119
1.2
28,043
3.2
177
GI with Gray
Dry detention
10,000
1.1
10,000
1.1
0
Infrastructure
basin
Infiltration basin
12,858
1.5
27,846
3.2
117
Infiltration
14,800
1.7
23,350
2.7
58
trench
Permeable
201,242
23.1
201,242
23.1
0
pavement
Conventional
Extended dry
25,000
2.9
25,000
2.9
0
(Gray)
detention basin
Infrastructure
with Distributed
GI
Surface sand
filters
10,119
1.2
10,119
1.2
0
Distributed
0
0.0
32,514
3.7
—
infiltration
trench
Cost Results - General Circulation Model (GCM) High Intensity
For the Conventional (Gray) scenario, the cost of adaptation (based on 20-year present value) is estimated
to increase by $6.47 million, or 122%, compared with the current cost. This is equivalent to a cost of
adaptation of $0.32 million per acre of site area.
The cost of adaptation for the GI with Gray scenario is estimated to increase by $6.99 million, or 136%.
On a cost per site-acre basis, the estimated cost of adaptation is $0.35 million per acre of site area.
Implementing distributed green practices (infiltration trenches) to address the performance gap between
current and future climate comes at an estimated cost increase of $10.56 million, an increase of 199%.
The increase in cost per acre of site is estimated to be $0.53 million for the Conventional with Distributed
GI scenario.
42
-------
Cost Results - Intensity Change Plus 10%
The cost of adaptation for the Conventional (Gray) scenario is estimated to be an increase of
$2.69 million, which reflects a 51% increase in cost. This is equivalent to a cost of adaptation of
$0.13 million per acre of site area.
For the GI with Gray scenario, the estimated cost of adaptation is a $1.09 million increase compared to
the current cost, or an increase in cost of 21%. The increase in cost per acre of site is estimated to be
$0.05 million.
Implementing distributed green practices (infiltration trenches) to address the performance gap between
current and future climate comes at an estimated cost increase of $1.62 million, an increase of 30%. On a
cost per site acre basis, the estimated cost of adaptation is $0.08 million per acre of site area.
For all three stormwater management scenarios, the cost of adaptation for the percentage change plus
10% scenario is significantly less than the cost for the GCM high intensity change scenario. The reason
adaptation costs are so high for the GCM high intensity change scenario is due to the large increase in
storm event volume for the largest storm events, as seen in Figure 3-6. The largest hourly rainfall depth
increases from about 2 to 3.5 inches, an increase of 75%.
Cost Results - Intensity Change Plus 20%
For the Conventional (Gray) scenario, the estimated cost of adaptation is a $4.89 million increase
compared to the current cost, or an increase in cost of 92%. The increase in cost per acre of site is
estimated to be $0.24 million.
The cost of adaptation for the GI with Gray scenario is estimated to be an increase of $2.13 million,
which reflects a 41 % increase in cost. This is equivalent to a cost of adaptation of $0.11 million per acre
of site area.
Implementing distributed green practices (infiltration trenches) to address the performance gap between
current and future climate comes at an estimated cost increase of $3.55 million, an increase of 67%. On a
cost per site acre basis, the estimated cost of adaptation is $0.18 million per acre of site area.
For all three stormwater management scenarios, the cost of adaptation for the percentage change plus
20% scenario is less than the cost for the GCM high intensity change scenario, but more than for the
percentage change plus 10% scenario.
Table 3-6 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for all of the Harford County stormwater management scenarios. Refer to
Section 2.5.2. of the report for a discussion on how the infrastructure cost estimates were developed. Also
provided are the increase in cost, both in dollars and percentage, and the increase in cost per acre of site.
These three metrics represent three alternative methods for evaluating the cost of adaptation, which is
effectively the increase in cost between the current and future adapted climate scenarios.
43
-------
Table 3-6. Comparison of current and future adapted 20-year present value costs
for the Harford County, MD stormwater management scenarios
Future
climate
scenario
Stormwater
management
scenario
Current cost
(20-Year
present value,
$millions)
Future adapted
cost (20-Year
present value,
$millions)
Increase in
cost (20-Year
present value,
$millions)
%
Increase in
cost
Increase per
acre of site
($millions)
GCM high
intensity
Conventional
(Gray)
Infrastructure
5.31
11.78
6.47
122
0.32
GI with Gray
Infrastructure
5.15
12.15
6.99
136
0.35
Conventional
(Gray)
Infrastructure with
Distributed GI
5.31
15.87
10.56
199
0.53
Percentage
difference
plus 10%
Conventional
(Gray)
Infrastructure
5.31
8.00
2.69
51
0.13
GI with Gray
Infrastructure
5.15
6.24
1.09
21
0.05
Conventional
(Gray)
Infrastructure with
Distributed GI
5.31
6.93
1.62
30
0.08
Percentage
difference
plus 20%
Conventional
(Gray)
Infrastructure
5.31
10.17
4.86
92
0.24
GI with Gray
Infrastructure
5.15
7.29
2.13
41
0.11
Conventional
(Gray)
Infrastructure with
Distributed GI
5.31
8.86
3.55
67
0.18
44
-------
4. MIDWEST SITE: RESIDENTIAL
4.1. REGULATORY REQUIREMENTS AFFECTING STORMWATER
MANAGEMENT
At the highest level, development is regulated under the Minnesota Pollution Control Agency's (MPCA)
statewide stormwater permit. Local requirements are also common and highly variable, with many
locations falling under the jurisdiction of local watershed management organizations (WMOs). The Scott
County WMO has requirements in addition to the state permit requirements.
Relevant MPCA stormwater requirements include:
• Retention of 1 inch of runoff from impervious surfaces is required (the WQv). The volume shall
be infiltrated, evaporated, or reused on site.
• If retention is not possible (e.g., clay soils with low infiltration rates), then treatment BMPs must
be used to remove 80% of TSS from the WQv. The WQv must be discharged within 48 hours.
• Wet ponds qualify for meeting the TSS removal requirement without need for another treatment
BMP. If wet ponds are used, the design requirements are as follows:
o Permanent pool of 1,800 ft3 per acre of drainage
o Maximum discharge d5.66 cfs per acre of pond surface area
o Depth 3 to 10 feet
• Stormwater credits are given for various GI practices. The WQv can be reduced, and in some
cases, an adjusted runoff curve number can be used for large storm event peak flow calculations.
There are a number of restrictions on using the credit based on contributing impervious area,
receiving pervious area, HSG, etc.
Relevant Scott County WMO stormwater requirements include:
• If detention is not possible, treatment BMPs must be used to remove 80% of TSS.
• Predevelopment peak matching is required for the 2-year, 10-year, and 100-year 24-hour events.
Predevelopment conditions are defined as "woods in good condition" for purposes of performing
stormwater routing calculations.
4.2. STORMWATER MANAGEMENT SCENARIOS
The 30-acre Residential site (see Figure 4-1) is assumed to have the following characteristics in each of
the scenarios:
• A combination of single-family homes and townhomes occupy the site. The density is about 6.5
units per acre.
45
-------
The site is 48% impervious, distributed as follows:
o 16% road area
o 22% house area
o 7% driveway area
o 3% sidewalk area
The remaining pervious area (52%) is comprised of lawn/landscaping.
The entire site is assumed to be composed of HSG D soils; this was done to allow one of the
modeled geographic locations to have poor infiltration rates. In addition, D soils are common in
this region due to poorly drained soils of glacial origin. Due to limited infiltration capacity, the
infiltration requirement is assumed to be waived. This assumption was made to allow for a
conventional scenario that did not incorporate GI practices. The HSG composition is also used for
design storm event routing calculations to size practices for peak flow control. Predevelopment
land cover is assumed to be woods in good condition.
The site must meet the more restrictive MPCA requirements for WQv and the discharge time
period.
The effects of frozen conditions on BMP performance are not modeled specifically in SUSTAIN.
However, the input runoff time series from HSPF do account for snowfall, development of
snowpack, and snow melt timing.
46
-------
Legend
g Building
Driveway/Sidewalk
Pervious
Roadway
Figure 4-1. Residential site layout (Scott County, MN).
Three stormwater management scenarios have been developed representing different approaches to
stormwater management: a conventional scenario using gray practices, a scenario using a combination of
green and gray practices, and a scenario using only GI practices. As noted in Section 2.2. , the
conventional scenario was used as the basis for a fourth scenario in which distributed GI practices were
added to achieve current or better performance under future climate conditions. The scenarios are
described in the following subsections.
4.2.1. Conventional (Gray) Infrastructure
The key design elements represented in the Conventional (Gray) Infrastructure scenario are as follows:
• The entire site drains to one point and is treated by a wet pond.
• The WQv is discharged from a low-flow orifice over a period of approximately 48 hours.
• The predevelopment peak-matching requirements are met for 2-year, 10-year, and 100-year
24-hour storms using a weir in the wet pond with extra volume storage above the WQv storage.
47
-------
As shown in Figure 4-2 for the Conventional scenario, site drainage is directed to a storm drain sewer
system that conveys flow to the wet pond. Flow is then discharged from the wet pond off the site.
For the SUSTAIN simulation, a routing spreadsheet was created simulating design storm runoff using the
same approach as for the Harford County detention basins. The wet pond dimension and permanent pool
volume were set-based MPCA design requirements. A low-flow orifice was configured to discharge the
WQv within 48 hours. For the Scott County WMO peak matching requirements, a weir was added and
optimization used to size the weir so that design storm event peak flows matched predeveloped
conditions. The entire design configuration was transferred to SUSTAIN. Infiltration from the pond was
set to zero, but ET was configured to occur from the pond water surface. For wet ponds, pollutant
removal can be modeled using decay rates, which reduce the ambient concentration in the permanent
pool. Decay rates were identified through successive model runs that mimicked published percentage
removal values of 80% for TSS, 30% for TN, 50% for TP (Center for Watershed Protection, 2007;
Hirschman et al., 2008).
Legend
BMPType
Driveway/Sidewalk
Pervious
Roadway
—» Site Drainage Conveyance
] Drainage Areas
Figure 4-2. Residential Conventional (Gray) Infrastructure stormwater
management scenario (Scott County, MN).
48
-------
4.2.2. Green Infrastructure (Gl) with Gray Infrastructure
The key design elements represented in the GI with Gray Infrastructure scenario are as follows:
• Distributed bioretention treats the WQv.
• Bioretention is designed as follows:
o 18-inch ponding depth
o 4 8-hour drawdown time
o 3 feet of media
o An underdrain is used because the soils have a low infiltration capacity.
• A dry detention basin is used for peak matching for 2-year, 10-year, and 100-year 24-hour storms.
Figure 4-3 shows the BMP locations and stormwater conveyance for the GI with Gray Scenario. Site
runoff is directed to five bioretention cells either by the storm drain sewer system or via grass channels
(swales) where drainage patterns allow. Flow discharged from the bioretention cells (either via the
underdrain or bypass flow during larger events) is then routed to a single centralized dry detention basin
serving the entire site for large event peak flow reduction. Any overflow is discharged offsite from the dry
detention basin.
For the SUSTAIN configuration, the bioretention cells were configured to store and treat the WQv
associated with each contributing drainage area according to MPCA guidelines. A nominal infiltration
rate of 0.06 inches per hour was used to represent infiltration from a 3-inch rock layer below the
underdrain. The bioretention media was assumed to achieve pollutant removal rates of 78% for TSS, 57%
for TN, and 63% for TP, using published performance values from Center for Watershed Protection
(2007) and Tetra Tech (2014). Removal was modeled in SUSTAIN for only the volume that filtered
through the bioretention media and was subsequently discharged via the underdrain. The detention basin
configuration was determined using the same approach as for the conventional site, except there was no
permanent pool nor WQv orifice. The 0.06 inches/hour infiltration rate was used also for the detention
basin. ET was modeled in SUSTAIN for both the bioretention and the dry detention basin.
49
-------
Legend
3] Building
~ Driveway/Sidewalk
Pervious
Roadway
Site Drainage Conveyance
Drainage Areas
BMPType
Bioretention
_J Dry Detention
=-=¦ BMP Conveyance
Figure 4-3. Residential Green Infrastructure (GI) with Gray Infrastructure
stormwater management scenario (Scott County, MN).
4.2.3. Green Infrastructure (GI) Only
The key design elements represented in the GI Only scenario are as follows:
• Penneable pavement is used for the sidewalks. No adjacent areas drain to the sidewalks.
Driveways were assumed to use conventional paving surfaces.
• Rooftop downspout disconnection is used in select areas where there is sufficient pervious
surface to meet the design criteria.
• Bioretention is used with a modified design to address the peak flow matching requirements:
o Additional storage in the bioretention performs peak matching for 2-year. 10-year, and
100-year 24-hour storms using a weir.
o 12-inch ponding depth (modification to allow for peak matching using additional storage
above ponding depth).
50
-------
o 4 8 -hour drawdown time.
o 3 feet of media.
o An underdrain is used because the soils have a low infiltration capacity.
The GI Only scenario differs from the previous two scenarios, each of which provided centralized control
of large storm event peak flows. In the GI Only scenario, peak flow and flood control management are
addressed in a distributed fashion. Each of the bioretention cells includes additional storage and a control
structure to capture and release large storm event volumes closer to the source. As shown in Figure 4-4,
many rooftops are configured to discharge to adjacent pervious areas. Because the site is composed of D
soils, the pervious areas receiving the roof runoff must have their soils amended by compost to improve
infiltration capacity (as required by MPCA). Minor grading may also be needed to ensure that runoff is
well dispersed and overland flow maintained. The sidewalks are comprised of pervious concrete and have
sufficient storage in an underlying stone layer to store the 100-year storm event volume. As shown in the
figure, street culverts or grass channels are still needed to convey flow to the bioretention cells. Flow
from the bioretention cells is then conveyed offsite via a separate drainage system.
For the SUSTAIN simulation, the following assumptions were used to develop the model configuration:
• Each bioretention cell was configured to capture and treat the WQv from its drainage area. The
same configuration as was used for the GI with Gray Infrastructure site, including the use of
underdrains and percentage removal of treated pollutants.
• Additional storage was added to each bioretention cell above the WQv to address peak flow
reduction requirements. The stage-storage-discharge routing spreadsheet, as discussed previously,
was reconfigured for each individual drainage area. Flows in excess of the WQv were discharged
gradually from weirs to perform the peak matching to undeveloped conditions.
• Permeable pavement areas were not assumed to have underdrains. While underdrains are
typically used when infiltration rates are very low, the use of permeable pavement was restricted
to sidewalks, which would likely have lateral infiltration as well as vertical infiltration. A small
amount of ET was assumed, equal to 10% of ET that would normally take place.
• The areas receiving disconnected roof runoff were sized at an approximate 1:1 ratio (i.e., 1,000 ft2
of impervious surface drained to 1,000 ft2 of pervious surface). The receiving pervious areas were
assumed to have compost-amended soils. The infiltration rate was increased from
0.06 inches/hour to 0.15 inches/hour (a 2.5 x increase) based on a literature review of infiltration
rate changes in compost amended soils (Harrison et al., 1997; Carmen, 2015; Brown and Cotton,
2011; Eusufzai and Fujii, 2012).
51
-------
Legend
Building
Driveway/Sidewalk
~ Pervious
Roadway
- - - ¦ Site Drainage Conveyance
I I Drainage Areas
BMPType
Bioretention
Downspout Disconnection
- - - 1 BMP Conveyance
Figure 4-4. Residential Green Infrastructure (GI) Only stormwater management
scenario (Scott County, MN).
4.2.4. Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI)
The objective of this scenario is to address the performance gap between current and future climate by
incorporating additional distributed GI practices into the site. The current conventional BMP (wet pond)
configuration is unchanged and distributed GI practices are added to provide treatment equivalent to the
current conditions climate scenario. The Scott County. MN location is unique in that three distinct future
climate scenarios representing low, medium, and high intensity changes (as discussed in Section 2.4.2. )
are evaluated.
For the Conventional with Distributed GI scenario, bioretention areas are distributed throughout the
30-acre study site. The bioretention design is the same configuration used in the GI with Gray scenario:
52
-------
• 18-inch ponding depth.
• 48 -hour drawdown time.
• 3 feet of media.
• An underdrain is used because the soils have a low infiltration capacity.
4.3. ADAPTATION SIMULATION
The objective of the adaptation simulation is to determine the increases in BMP footprint (surface area)
that would be required to maintain or exceed current performance under future climate conditions for
each stormwater management scenario. Table 4-1 summarizes the key components of the modeling
procedure for each scenario. Although the GI Only scenario includes permeable pavement and impervious
surface disconnection, only the distributed bioretention practices were resized in the adaptation simulation
because (1) permeable pavement is already implemented in 100% of sidewalk areas, and its expansion to
include residential driveways and streets was ruled impractical due primarily to maintenance concerns,
and (2) impervious surface disconnection is already implemented to the maximum extent practicable in
this scenario for residential rooftops and disconnection of additional impervious surface is not considered
feasible.
Table 4-1. Features of adaptation simulation for Scott County, MN
Location
Stormwater management scenario
Future adaptation
Affected practices
Scott County,
MN
Conventional (Gray) Infrastructure
Resize practices
Wet pond
GI with Gray Infrastructure
Resize practices
Distributed bioretention and dry
detention basin
GI Only
Resize practices
Distributed bioretention
Conventional (Gray) Infrastructure
with Distributed GI
Add distributed GI
to site
Distributed bioretention
4.4. CURRENT AND FUTURE CHANGES IN PRECIPITATION
As discussed in Section 4.1. , three future climate scenarios were selected for simulation reflecting a range
in changes in intensity the lowest intensity change, a medium intensity change, and the highest intensity
change (note that the highest intensity change was used for all the other locations). Figure 4-5, Figure 4-6,
and Figure 4-7 provide the ranked projected annual precipitation totals for the low, medium, and high
intensity future scenario compared to current climate conditions. Under the low intensity scenario,
projected annual precipitation volume decreases during nearly all the years. The medium and high
intensity scenarios show a somewhat variable increase across all years. Average annual precipitation is
30.1 inches for current climate and is projected to be 28.7 inches for future low, 33.3 inches for future
medium, and 33.6 inches for future high, corresponding to changes of-4.4, 10.7, and 11.6% respectively.
53
-------
For monthly average precipitation depth (see Figure 4-8), the changes across future scenarios are highly
variable by month, notably in July. In addition, daily sums of precipitation depth were calculated and used
to determine percentiles of 24-hour depth of interest to stormwater managers (see Table 4-2). While daily
sums do not provide a true measure of storm event depth (storms have variable lengths and may span
more than 1 day), they do provide useful information about expected depths over a 24-hour period. For
the future low intensity scenario, there is a decrease across the board for all percentiles. The future
medium intensity and future high intensity scenarios have comparable depth increases across the
percentiles, with the future high showing a larger increase for the 99th percentile.
Figure 4-9 provides a comparison of the highest hourly precipitation volumes. The low intensity scenario
projects a decrease in intensity compared with current climate for all hours except the single largest
precipitation depth in the 30-year time series. The medium intensity scenario is only slightly higher than
current conditions, while the high intensity scenario has a projected increase of about 1.25 to 1.4 times
that of the current conditions.
Annual Precipitation (30-yr record)
50
¦ Current ¦ Future Low
45
40
Figure 4-5. Ranked annual precipitation for current conditions and low intensity
future climate scenario at Scott County, MN.
54
-------
Annual Precipitation (30-yr record)
50
45
40
35
30
.E 25
20
15
10
5
0
I Current ¦ Future Medium
Hrvim^-LDiDr^-ooCTiOtH
(Mm'sfiniDr^-ooCTiOt-iiMm'sfinvDr^-oooio
HrlHHHHrlrlNNMlNNNNfNNNfO
Rank
¦5
Figure 4-6. Ranked annual precipitation for current conditions and medium
intensity future climate scenario at Scott County, MN.
50
45
40
35
30
.E 25
20
15
10
5
0
Annual Precipitation (30-yr record)
I Current ¦ Future High
-------
Monthly Average Precipitation
..Aflilllllii.
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
¦ Current ¦ Future Low ¦ Future Medium ¦ Future High
Figure 4-8. Monthly average precipitation for current conditions and
low/medium/high intensity future climate scenarios at Scott County, MN.
Table 4-2. 24-hour precipitation depth percentiles for current conditions and low,
medium, and high intensity future climate scenario at Scott County, MN
Percentile
Current
conditions
24-h depth in
Future low
24-h depth
in
Future low
change
+/-in
Future
medium
24-h depth
in
Future
medium
change
+/-in
Future high
24-h depth
in
Future
high
change
+/-in
85th
0.50
0.48
-0.02
0.56
+0.06
0.55
+0.05
90th
0.67
0.65
-0.02
0.75
+0.08
0.76
+0.09
95th
1.01
0.96
-0.05
1.17
+0.16
1.12
+0.11
99th
1.94
1.77
-0.17
2.27
+0.33
2.33
+0.40
56
-------
Hourly Precipitation
5
4.5
4
3.5
3
.£ 2.5
2
1.5
1
0.5
0
Simulation Recurrence Interval (years)
Figure 4-9. Hourly precipitation recurrence interval for current conditions and
low/medium/high intensity future climate scenarios at Scott County, MN.
4.5. RESULTS
SUSTAIN was run for the following conditions for each stonnwater management scenario:
• Current climate, site without stormwater management/BM Ps
• Future climate, site without stormwater management/BMPs
• Current climate, site with stormwater management/B VI Ps
• Future climate, site with stormwater management/BMPs
• Future climate, site with BMPs adapted to meet current hydrology and water quality performance
As shown in Table 4-3, 22 sets of SUSTAIN runs were performed for a combination of four stormwater
management approaches and six climate scenarios. The Conventional with Distributed GI approach is a
variation of the Conventional approach, where the adaptation to meet or exceed current performance used
the addition of GI components to the site, rather than resizing practices already in place as was done for
the Conventional approach. Because the downscaled GCM low intensity and minus 10% change climate
scenarios both resulted in a reduction in all performance measures, there was no need to perform
adaptation runs; the current performance was already met. As a result, no SUSTAIN run was needed for
the Conventional with Distributed GI approach for these two climate scenarios because it is identical to
the Conventional approach prior to adaptation.
Current
Future Low
Future Medium
Future High
57
-------
Table 4-3. Stormwater management and climate scenarios for Scott County, MN
Stormwater
management approach
GCM low
intensity
GCM
medium
intensity
GCM high
intensity
Minus 10%
Plus 10%
Plus 20%
Conventional
X
X
X
X
X
X
GI + Gray
X
X
X
X
X
X
GI Only
X
X
X
X
X
X
Conventional with
Distributed GI
X
X
X
X
A full presentation of the results of all the runs is provided in APPENDIX B. . For brevity, the results in
this section focus on a few topics of interest to stormwater managers: (1) a comparison of the site
performance with BMPs between current and future climate conditions, (2) the increases in BMP
footprints needed to offset impacts of climate change when BMPs are adapted using SUSTAIN
optimization, and (3) a comparison of current stormwater infrastructure costs to future costs when BMPs
are adapted to offset impacts of climate change.
For the comparison of the site performance with BMPs between current and future climate conditions, the
downscaled future GCM (high intensity change) scenario was selected for the comparison. A discussion
of other topics of interest are provided in the general conclusions Section 8. , including changes in
pretreatment site performance, changes in post-treatment site performance, climate scenario sensitivity
analysis, and adapting BMPs under future climate to meet current performance.
Rather than comparing the performance of the stormwater management approaches independent of
climate change (i.e., how much better does one perform than the other under current conditions), this
study focuses on how the stormwater management approaches compare relative to climate change. Table
4-4 provides current and future performance for the stormwater management approaches, normalized to
area. Note that there is no numeric measure of change in the FDC between current and future climate, so
the highest hourly peak flow during the simulation is presented as a proxy for large storm event response.
Figure 4-10 through Figure 4-14 present each metric graphically from Table 4-4.
58
-------
Table 4-4. Current and future performance of Scott County, MN site by stormwater
management approach for general circulation model (GCM) high intensity scenario
Stormwater management approach
Current
Future
Change
Runoff (inch/yr)
Conventional
13.04
14.61
+1.57
GI + Gray
10.36
12.03
+1.67
GI Only
7.71
9.23
+1.52
Maximum hourly peak flow (cfs/ac)
Conventional
2.40
3.45
+1.05
GI + Gray
2.38
3.43
+1.05
GI Only
2.35
3.41
+1.06
Sediment (ton/ac/yr)
Conventional
0.123
0.193
+0.070
GI + Gray
0.248
0.360
+0.112
GI Only
0.199
0.300
+0.102
TN (lb/ac/yr)
Conventional
6.95
7.44
+0.49
GI + Gray
3.64
4.07
+0.44
GI Only
2.39
2.77
+0.38
TP (lb/ac/yr)
Conventional
0.69
0.79
+0.10
GI + Gray
0.49
0.59
+0.10
GI Only
0.36
0.46
+0.10
59
-------
Annual Site Runoff
Future High
Current
Conventional
¦ Current ¦ Future High
Figure 4-10. Annual site runoff under current climate and future general circulation
model (GCM) high intensity scenario by stormwater management approach for
Scott County, MN.
Maximum Hourly Peak Flow
Future High
Current
Conventional
¦ Current ¦ Future High
Figure 4-11. Maximum hourly peak flow under current climate and future general
circulation model (GCM) high intensity scenario by stormwater management
approach for Scott County, MN.
60
-------
Annual Sediment Loading Rate
Future High
Current
Conventional
¦ Current ¦ Future High
Figure 4-12. Annual sediment loading rate under current climate and future general
circulation model (GCM) high intensity scenario by stormwater management
approach for Scott County, MN.
Annual TN Loading Rate
Future High
Current
Conventional
¦ Current ¦ Future High
Figure 4-13. Annual total nitrogen (TN) loading rate under current climate and
future general circulation model (GCM) high intensity scenario by stormwater
management approach for Scott County, MN.
61
-------
Annual TP Loading Rate
0.80
,.0.60
lo 0.40
.n
0.20
0.00
¦ Current ¦ Future High
Figure 4-14. Annual total phosphorous (TP) loading rate under current climate and
future general circulation model (GCM) high intensity scenario by stormwater
management approach for Scott County, MN.
As discussed in Section 3.5. , for Harford County the GI + Gray stormwater management approach
appeared to be more resilient to climate change in terms of raw increase in annual stormwater runoff and
nutrient load export. An examination of the results of the three stormwater management approaches for
Scott County does not reveal a similar trend. The increases in annual runoff, highest hourly peak flow,
sediment loads, and nutrient loads are all similar. In other words, the stormwater management approach
had little effect on changes in site runoff and pollutant loading. The reason for the difference between
Harford County and Scott County is not known, but may be related to the low permeability of the soils for
the Scott County site.
Table 4-5 summarizes the increases in BMP footprints for the Scott County stormwater management
scenarios that would be required in order to maintain or exceed current performance under future climate
conditions. The current and adapted footprints are presented both in terms of actual square feet of practice
as well as percentage of overall site area. The latter is provided as a means of comparing the current and
future adapted sizes relative to the site area (30 acres) for this particular development type (residential).
Results are discussed separately for each of the future climate scenarios modeled for the Scott County
site.
Future High
Current
Conventional
62
-------
Table 4-5. Comparison of current and future adapted best management practice
(BMP) footprints for Scott County, MN stormwater management scenarios
Current Future adapted
/O
Future Stormwater Footprint Footprint increase
climate management Footprint as % of Footprint as % of in
scenario scenario Practice SF site area SF site area footprint
GCM
medium
intensity
Conventional (Gray)
Infrastructure
Wet pond
32,670
2.5
107,811
8.3
230
GI with Gray
Infrastructure
Bioretention
34,848
2.7
58,848
4.5
69
Dry detention basin
26,136
2.0
32,336
2.5
24
GI Only
Bioretention (modified)
43,275
3.3
71,675
5.5
66
Rooftop downspout
disconnection
94,901
7.3
94,901
7.3
0
Permeable pavement
39,390
3.0
39,390
3.0
0
Conventional (Gray)
Infrastructure with
Distributed GI
Wet pond
32,670
2.5
32,670
2.5
0
Distributed bioretention
0
0.0
18,280
1.4
-
GCM high
intensity
Conventional (Gray)
Infrastructure
Wet pond
32,670
2.5
128,066
9.8
292
GI with Gray
Infrastructure
Bioretention
34,848
2.7
93,286
7.1
168
Dry detention basin
26,136
2.0
123,136
9.4
371
GI Only
Bioretention (modified)
43,275
3.3
111,735
8.6
158
Rooftop downspout
disconnection
94,901
7.3
94,901
7.3
0
Permeable pavement
39,390
3.0
39,390
3.0
0
Conventional (Gray)
Infrastructure with
Distributed GI
Wet pond
32,670
2.5
32,670
2.5
0
Distributed bioretention
0
0.0
56,770
4.3
-
Percentage
difference
plus 10%
Conventional (Gray)
Infrastructure
Wet pond
32,670
2.5
107,484
8.2
229
GI with Gray
Infrastructure
Bioretention
34,848
2.7
70,348
5.4
102
Dry detention basin
26,136
2.0
26,136
2.0
0
GI Only
Bioretention (modified)
43,275
3.3
80,405
6.2
86
Rooftop downspout
disconnection
94,901
7.3
94,901
7.3
0
Permeable pavement
39,390
3.0
39,390
3.0
0
Conventional (Gray)
Infrastructure with
Distributed GI
Wet pond
32,670
2.5
32,670
2.5
0
Distributed bioretention
0
0.0
17,500
1.3
-
63
-------
Table 4 5. Comparison of current and future adapted best management practice (BMP)
footprints for Scott County, MN stormwater management scenarios (Continued)
Future
climate
scenario
Stormwater
management
scenario
Practice
Current
Future adapted
%
Footprint
SF
Footprint
as % of
site area
Footprint
SF
Footprint
as % of
site area
increase
in
footprint
Percentage
difference
plus 20%
Conventional (Gray)
Infrastructure
Wet pond
32,670
2.5
172,171
13.2
427
139
163
GI with Gray
Infrastructure
Bioretention
34,848
2.7
83,348
6.4
Dry detention basin
26,136
2.0
68,636
5.3
GI Only
Bioretention (modified)
43,275
3.3
117,601
9.0
172
0
0
Rooftop downspout
disconnection
94,901
7.3
94,901
7.3
Permeable pavement
39,390
3.0
39,390
3.0
Conventional (Gray)
Infrastructure with
Distributed GI
Wet pond
32,670
2.5
32,670
2.5
0
Distributed bioretention
0
0.0
30,500
2.3
Design Results - General Circulation Model (GCM) Medium Intensity
The Conventional (Gray) Infrastructure scenario uses an adapted wet pond that is 3.3 times larger than the
wet pond under current climate conditions. Due to the presence of D soils at the site, the wet pond
infiltration rate is negligible. As a result, a large size is needed under future climate conditions to increase
ET to allow the adapted wet pond to meet the runoff volume criterion.
For the GI with Gray Infrastructure scenario, the adapted site would need a 69% increase in bioretention
area and a 24% increase in the dry detention basin area, equivalent to 2.3% of the site area.
For the GI Only Infrastructure scenario, only bioretention was modified for future climate adaptation.
Permeable pavement and rooftop downspout disconnection were not modified for two reasons:
(1) permeable pavement is already implemented in 100% of sidewalk areas, and its expansion to include
residential driveways and streets was considered impractical due primarily to maintenance concerns, and
(2) impervious surface disconnection is already implemented to the maximum extent practicable in this
scenario for residential rooftops, and disconnection of additional impervious surface is not considered
feasible. The increase in size for the adapted bioretention was 66%, or 2.2% of the site area.
The Conventional + Distributed GI Infrastructure scenario adaptation would require the addition of
18,280 square feet of bioretention (roughly 1.4% of the total site area).
64
-------
Design Results - General Circulation Model (GCM) High Intensity
For the Conventional (Gray) Infrastructure scenario, the adapted wet pond is nearly four times larger than
the current wet pond. The adapted size is larger for this climate scenario than for the GCM medium
intensity change scenario due to greater precipitation volume. The GI with Gray Infrastructure scenario
resulted in an adapted bioretention footprint that was approximately 2.7 times larger than the current
footprint, and the adapted dry detention basin footprint that was approximately 4.7 times larger than the
current footprint. For the GI Only Infrastructure scenario, the adapted bioretention footprint was 158% of
the size under current conditions, or 5.3% of the site area. The Conventional + Distributed GI
Infrastructure scenario adaptation would require the addition of 56,770 square feet of bioretention
(roughly 4.3% of the total site area).
Design Results - Intensity Change Plus 10%
Adaptation for the Conventional (Gray) Infrastructure scenario would require the wet pond footprint to
increase by nearly 3.3 times, which is analogous to the GGM medium intensity change scenario. For the
GI with Gray Infrastructure scenario, no increase in the dry detention basin footprint was required, but the
bioretention footprint would need to more than double in size. Adaptation for the GI Only scenario would
require an 86% increase in bioretention footprint. When distributed GI is added to the Conventional
(Gray) Infrastructure scenario for adaptation, the required bioretention footprint of 17,500 square feet
would comprise approximately 1.3% of the total site area.
Design Results - Intensity Change Plus 20%
Adaptation for the Conventional (Gray) Infrastructure scenario would require the wet pond footprint to
increase by nearly 4.3 times, greater even that for the GCM high intensity change scenario. For the GI
with Gray Infrastructure scenario, the dry detention basin footprint would need to increase by 163%, and
the bioretention footprint would need to increase by 139%. Adaptation for the GI Only scenario would
require a 172% increase in bioretention footprint. When distributed GI is added to the Conventional
(Gray) Infrastructure scenario for adaptation, the required bioretention footprint of 30,500 square feet
would comprise approximately 2.3% of the total site area.
Table 4-6 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for all of the Scott County storm water management scenarios. Refer to Section 2.5.2.
of the report for a discussion on how the infrastructure cost estimates were developed. Also provided are
the increase in cost, both in dollars and percentage, and the increase in cost per acre of site. These three
metrics represent three alternative methods for evaluating the cost of adaptation, which is effectively the
increase in cost between the current and future adapted climate scenarios.
65
-------
Table 4-6. Comparison of current and future adapted 20-year present value costs
for the Scott County, MN stormwater management scenarios
Future
climate
scenario
Stormwater
management
scenario
Current cost
(20-yr
present
value,
$millions)
Future
adapted cost
(20-yr present
value,
$millions)
Increase in
cost (20-yr
present
value,
$millions)
%
Increase
in cost
Increase per
acre of site
($millions)
GCM
medium
Conventional (Gray)
Infrastructure
3.05
8.99
5.94
195
0.30
intensity
GI with Gray
Infrastructure
4.92
7.37
2.46
50
0.12
GI Only
8.51
11.80
3.29
39
0.16
Conventional (Gray)
Infrastructure with
Distributed GI
3.05
4.69
1.65
54
0.08
GCM high
intensity
Conventional (Gray)
Infrastructure
3.05
10.59
7.54
248
0.38
GI with Gray
Infrastructure
4.92
14.82
9.90
201
0.50
GI Only
8.51
16.44
7.93
93
0.40
Conventional (Gray)
Infrastructure with
Distributed GI
3.05
8.16
5.11
168
0.26
Percentage
difference
Conventional (Gray)
Infrastructure
3.05
8.96
5.92
194
0.30
plus 10%
GI with Gray
Infrastructure
4.92
8.11
3.20
65
0.16
GI Only
8.51
12.76
4.25
50
0.21
Conventional (Gray)
Infrastructure with
Distributed GI
3.05
4.62
1.58
52
0.08
Percentage
difference
Conventional (Gray)
Infrastructure
3.05
14.08
11.03
362
0.55
plus 20%
GI with Gray
Infrastructure
4.92
11.31
6.40
130
0.32
GI Only
8.51
17.12
8.61
101
0.43
Conventional (Gray)
Infrastructure with
Distributed GI
3.05
5.79
2.75
90
0.14
66
-------
Cost Results - General Circulation Model (GCM) Medium Intensity
For the Conventional (Gray) scenario, the cost of adaptation is estimated to increase by $5.94 million, or
195%, compared with the current cost. This is equivalent to a cost of adaptation of $0.30 million per acre
of site area.
For the GI with Gray scenario, the cost of adaptation is estimated to increase by $2.46 million, or 50%
compared to the current cost. The increase in cost per acre of site is estimated to be $0.12 million.
The GI Only scenario adaptation to future climate resulted in a $3.29 million increase in cost, a 39%
increase. When normalized to site area, the increase is estimated to be $0.16 million per acre.
Implementing distributed green practices (bioretention) to address the performance gap between current
and future climate comes at an estimated cost increase of $ 1.65 million, an increase of 54%. On a cost per
site acre basis, the estimated cost of adaptation is $0.08 million per acre of site area.
Cost Results - General Circulation Model (GCM) High Intensity
For the Conventional (Gray) scenario, the cost of adaptation is estimated to increase by $7.54 million, or
248% compared to the current cost. This is equivalent to a cost of adaptation of $0.38 million per acre of
site area.
For the GI with Gray scenario, the cost of adaptation is estimated increase by $9.90 million, or 201%
compared to the current cost. On a cost per site acre basis, the estimated cost of adaptation is
$0.50 million per acre of site area.
For the GI Only scenario, the cost of adaptation was estimated as a $7.93 million increase over the current
cost, or an increase of 93%. This is equivalent to an increase of $0.40 million per acre of site area.
Implementing distributed green practices (bioretention) to address the performance gap between current
and future climate comes at an estimated cost increase of $5.11 million, an increase of 168%. The
increase in cost per acre of site is estimated to be $0.26 million for the Conventional with Distributed GI
scenario.
Cost Results - Intensity Change Plus 10%
For the Conventional (Gray) scenario the cost of adaptation is estimated to increase by $5.92 million, or
194% compared to the current cost. This is equivalent to a cost of adaptation of $0.30 million per acre of
site area.
For the GI with Gray scenario, the cost of adaptation is estimated to increase by $3.20 million increase, or
65% compared to the current cost. The increase in cost per acre of site is estimated to be $0.16 million.
For the GI Only scenario, the cost of adaptation was estimated as a $4.25 million increase over the current
cost, or an increase of 50%. This is equivalent to an increase of $0.21 million per acre of site acre.
67
-------
Implementing distributed green practices (infiltration trenches) to address the performance gap between
current and future climate comes at an estimated cost increase of $1.58 million, an increase of 52%. On a
cost per site acre basis, the estimated cost of adaptation is $0.08 million per site area.
Cost Results - Intensity Change Plus 20%
For the Conventional (Gray) scenario, the cost of adaptation is estimated to increase by $ 11.03 million, or
362% compared to the current cost. The increase in cost per acre of site is estimated to be $0.55 million.
The cost of adaptation for the GI with Gray scenario is estimated to increase by $6.40 million, or 130%
compared to the current cost. This is equivalent to a cost of adaptation of $0.32 million per acre of site
area.
For the GI Only scenario, the cost of adaptation is estimated to increase by $8.61 million, or 101%
compared to the current cost. This is equivalent to an increase of $0.43 million per acre of site area.
Implementing distributed green practices (infiltration trenches) to address the performance gap between
current and future climate costs an additional $2.75 million, or a 90% increase compared to current costs.
On a cost per site acre basis, the estimated cost of adaptation is $0.14 million per acre of site area.
68
-------
5. ARID SOUTHWEST SITE: COMMERCIAL
5.1. REGULATORY REQUIREMENTS AFFECTING STORMWATER
MANAGEMENT
Maricopa County has design requirements for new development:
• Retention of runoff from the 100-year 2-hour storm event (as shown in the Maricopa County
Hydrology Manual) is required. Volume shall be infiltrated, evaporated, or reused on site. The
100-year 2-hour storm event depth varies widely across Maricopa County; a value of 2.8 inches
was selected from the Maricopa County Hydrology Design Manual as being typical for the
portion of Maricopa County close to the selected weather station.
• If retention is not possible, then the following requirements apply:
o The first flush volume must be treated. The volume is defined as 0.5 inch of uniform
runoff from site.
o Predevelopment peak matching is required for the 2-, 10-, 50-, and 100-year storm
events. The ordinance does not state the duration, so a 2-hour event was assumed based
on the storm event duration for the retention standard.
5.2. STORMWATER MANAGEMENT SCENARIOS
The 10-acre Commercial shopping center site (see Figure 5-1) is assumed to have the following
characteristics in each of the scenarios:
• The site is 80% impervious, distributed as follows:
o 30% building
o 50% pavement
• The remaining pervious area (20%) is comprised of native vegetation/landscaping.
• The HSG percentage distribution is based on a GIS analysis of soils in Maricopa County. The
HSG composition is used for sizing practices to meet the retention requirement.
HSG
A: 1%
HSG
B: 79%
HSG
C: 8%
HSG
D: 12 %
69
-------
Legend
| Building
~ Pervious
Vehicle
Figure 5-1. Commercial site layout (Maricopa County, AZ).
Two scenarios have been developed representing different approaches to stormwater management* a
conventional scenario using a retention basin and a GI scenario using alternative green practices. (Note
that "retention basin" is the nomenclature used by Maricopa County; it is more commonly called an
infiltration basin.) The scenarios are described in the following subsections.
5.2.1. Conventional (Gray) Infrastructure
The key design elements in the Conventional (Gray) Infrastructure scenario are as follows:
• Entire site is treated by a retention basin. The basin provides full infiltration of the required
volume.
• The peak-matching requirement is automatically met by meeting the retention requirement.
As shown in Figure 5-2 for the Conventional scenario, site runoff is conveyed to a single infiltration
basin. The basin has a relatively large surface area to minimize the depth stored during the 100-year
70
-------
event, thus allowing the entire volume to infiltrate within 36 hours per design requirements. An
emergency spillway provides discharge for events exceeding the 100-year design event.
Legend
Building
Pervious
| Vehicle
a Inlet
- - - Site Drainage Conveyance
BMPType
¦ I Infiltration Basin
BMP Drainage Areas
Figure 5-2. Commercial Conventional (Gray) Infrastructure storm water
management scenario (Maricopa County, AZ).
For SUSTAIN, the retention basin was sized using design guidance from Maricopa County. Because the
basin was designed to fully infiltrate the design storm (100-year 2-hour event), no additional routing
calculations were needed. A large spillway was included for volumes exceeding the design capacity. An
infiltration rate of 0.7 inches/hour was used, based on the rate needed to fully infiltrate the design volume
within 36 hours. An investigation into GIS soil survey properties in the portion of Maricopa County
associated with the selected meteorological station showed that infiltration rates in excess of
0.7 inches/hour are common. ET was also modeled in SUSTAIN from the retention basin.
5.2.2. Green Infrastructure (Gl) Only
The key design elements in the GI Only scenario are as follows:
71
-------
• The GI practices are configured to meet the retention requirement.
• The Pima County (Tucson, AZ) LID Manual is used to provide design guidance for the GI
practices because Maricopa County does not currently have design guidance for GI.
• In the parking areas, approximately 50% of the pavement is permeable. Runoff from adjacent
conventional pavement and pervious areas flows onto the permeable pavement.
• The remaining paved area drains to bioretention.
• The entire roof drains to a large cistern.
• It is not possible to guarantee that the cistern will be completely empty prior to the 100-year
2-hour storm event, so cistern overflow is routed to a stormwater harvesting basin. A stormwater
harvesting basin is a shallow vegetated basin with storage that provides for infiltration. The
stormwater harvesting basin is assumed to have one-half the storage capacity of the cistern.
• Runoff captured and stored by the cistern is used to irrigate landscaping in the stormwater
harvesting basin. Water is released at a slow constant rate; the entire cistern, if full, would take
about 60 days to empty completely. The application rate is based on applying approximately
1.3 inches/week of irrigation to the stormwater harvesting basin.
Figure 5-3 provides the BMP locations and drainage network for the GI Only scenario. Permeable
pavement (using pervious concrete or asphalt) is used for site parking areas; the permeable pavement
fully addresses the 100-year storm event capture volume using a stone storage layer below the pavement
matrix. The pavement surrounding the building is of conventional design, and its drainage is conveyed to
the two bioretention cells either via culvert or sheet flow. The bioretention cells are configured to fully
store the required 100-year storm event volume. The rooftop drains to a large cistern, which is also
configured to store the entire 100-year event volume. Overflow from the cistern is routed to the
stormwater harvesting basin. If the cistern whose water can be used to irrigate the stormwater harvesting
basin at a low, constant rate is completely full, a pump is assumed to fully drain it within about 6 days.
For the SUSTAIN configuration, the Pima County LID Manual provided the primary source for design
guidelines. Due to the high underlying infiltration rates, underdrains were not used for any of the GI
components. Infiltration rates were set to 0.7 inches/hour for all practices except the cistern. ET was also
modeled for bioretention and the stormwater harvesting basin. A small amount of ET from permeable
pavement was assumed, equal to 10% of ET that would normally take place. No ET was assumed to take
place for the water stored in the cistern.
72
-------
Legend
Building
Pervious
Vehicle
0 Inlet
- - - Road Drainage Conveyance
BMPType
Bioretention
Cistem
Permeable Pavement
- Stormwater Harvesting Basin
| | BMP Drainage Areas
- - - ¦ BMP Conveyance
Figure 5-3. Commercial Green Infrastructure (GI) Only stormwater management
scenario (Maricopa County, AZ).
5.3. ADAPTATION SIMULATION
The objective of the adaptation simulation is to determine the increases in BMP footprint (surface area)
that would be required to maintain or exceed current performance under future climate conditions for
each stormwater management scenario. Table 5-1 summarizes the key components of the modeling
procedure for each scenario.
73
-------
Table 5-1. Features of adaptation simulation for Maricopa County, AZ
Stormwater
Future
Location
management scenario
adaptation
Affected practices
Maricopa
Conventional (Gray)
Resize
Infiltration basin
County, AZ
Infrastructure
practices
GI Only
Resize
Permeable pavement, bioretention, cistern.
practices
stormwater harvesting basin
5.4. CURRENT AND FUTURE CHANGES IN PRECIPITATION
Maricopa County is located in the arid Southwest, where annual precipitation volume is much lower than
at the other locations in the study. The climate station selected for the modeling is a short distance outside
Maricopa County to the northeast. The Salt River watershed selected for the "20 Watersheds" project
borders Maricopa County, and the climate station was the closest available. Annual precipitation across
Maricopa County is highly variable, ranging from less than 5 to over 20 inches per year.1 Annual average
precipitation at the climate station used in this analysis is over 18 inches per year, which is at the high end
of the range for the county.
The annual precipitation comparison shown in Figure 5-4 reveals that projected future conditions are
highly variable, with increases seen in some years and decreases in other years. (Note that 29 years were
used in the Maricopa County SUSTAIN simulations rather than the 30 years used for the other locations.)
Average annual totals are 18.4 inches for current conditions and 19.6 inches for future conditions,
reflecting a change of 6.5%. Monthly changes are somewhat variable, with the largest changes in January
(see Figure 5-5). In addition, daily sums of precipitation depth were calculated and were used to
determine percentiles of 24-hour depth of interest to stormwater managers (see Table 5-2). While daily
sums do not provide a true measure of storm event depth (storms have variable lengths and may span
more than 1 day), they do provide useful information about expected depths over a 24-hour period. The
change in depth between current and future ranges from no change for the 85th percentile to 0.80 inches
for the 99th percentile. Note that 0.80 inches is the single largest change for the 99th percentile across all of
the geographic locations.
The comparison of the highest hourly precipitation volumes shown in Figure 5-6 indicates a steadily
increasing gap between current and future intensity, ranging from 1.2/ at the 1-year recurrence interval to
as much as 1.8* at the highest recurrence interval.
1http://www.fcd.maricopa.gov/Weather/Rainfall/raininfo.aspx.
74
-------
Annual Precipitation (29-yr record)
40
¦ Current ¦ Future
35
30
Figure 5-4. Ranked annual precipitation for current conditions and high intensity
future climate scenario at Maricopa County, AZ.
Monthly Average Precipitation
4.5
IiIl.-iIiJJi
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
¦ Current ¦ Future
Figure 5-5. Monthly average precipitation for current conditions and high intensity
future climate scenario at Maricopa County, AZ.
75
-------
Table 5-2. 24-hour precipitation depth percentiles for current conditions and high
intensity future climate scenario at Maricopa County, AZ
Percentile
Current conditions
24-h depth (in)
Future climate
24-h depth (in)
Change (+/-in)
85th
0.71
0.71
0.00
90th
0.91
0.94
+0.02
95th
1.16
1.57
+0.41
99th
1.99
2.79
+0.80
Hourly Precipitation
6
5
4
.£ 3
2
1
0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
Simulation Recurrence Interval (years)
Figure 5-6. Hourly precipitation recurrence interval for current conditions and high
intensity future climate scenario at Maricopa County, AZ.
5.5. RESULTS
SUSTAIN was run for the following conditions for each stormwater management scenario:
• Current climate, site without stormwater management/BMPs
• Future climate, site without stormwater management/BMPs
• Current climate, site with stormwater management/BMPs
• Future climate, site with stormwater management/BMPs
• Future climate, site with BMPs adapted to meet current hydrology and water quality performance
Current
Future
76
-------
As shown in Table 5-3, two sets of SUSTAIN runs were performed for a combination of two stormwater
management approaches and one climate scenario.
Table 5-3. Stormwater management and climate scenarios for of Maricopa
County, AZ
GCM high
Stormwater management approach
intensity
Conventional
X
GI + Gray
X
A full presentation of the results of all the runs is provided in APPENDIX B. . For brevity, the results in
this section focus on a few topics of interest to stormwater managers: (1) a comparison of the site
performance with BMPs between current and future climate conditions, (2) the increases in BMP
footprints needed to offset impacts of climate change when BMPs are adapted using SUSTAIN
optimization, and (3) a comparison of current stormwater infrastructure costs to future costs when BMPs
are adapted to offset impacts of climate change.
For the comparison of the site performance with BMPs between current and future climate conditions, the
downscaled Future GCM (high intensity change) scenario was selected for the comparison. A discussion
of other topics of interest are provided in the general conclusions Section 8. , including changes in
pretreatment site performance, changes in post-treatment site performance, climate scenario sensitivity
analysis, and adapting BMPs under future climate to meet current performance.
Rather than comparing the performance of the stormwater management approaches independent of
climate change (i.e., how much better does one perform than the other under current conditions), this
study focuses on how the stormwater management approaches compare relative to climate change. Table
5-4 provides current and future performance for the stormwater management approaches, normalized to
area. Note that there is no numeric measure of change in the FDC between current and future climate, so
the highest hourly peak flow during the simulation is presented as a proxy for large storm event response.
Figure 5-7 through Figure 5-11 present each metric graphically from Table 5-4.
77
-------
Table 5-4. Current and future performance of Maricopa County, AZ site by
stormwater management approach
Stormwater management
approach
Current
Runoff (inch/yr)
Conventional
0.000
0.075
+0.075
GI Only
0.000
0.065
+0.065
Maximum hourly peak flow (cfs/ac)
Conventional
0.000
0.673
+0.673
GI Only
0.001
0.640
+0.638
Sediment (ton/ac/yr)
Conventional
0.000
0.018
+0.018
GI Only
0.000
0.049
+0.049
TN (lb/ac/yr)
Conventional
0.000
0.013
+0.013
GI Only
0.000
0.012
+0.011
TP (lb/ac/yr)
Conventional
0.0000
0.0005
+0.0005
GI Only
0.0000
0.0057
+0.0057
Annual Site Runoff
0.000
Conventional Gl Only
¦ Current ¦ Future
Figure 5-7. Annual site runoff under current climate and future general circulation
model (GCM) scenario by stormwater management approach for Maricopa County,
AZ.
78
-------
Maximum Hourly Peak Flow
0.800
0,600
0,400
0.200
0,000
0.673
0.640
0.000
0.001
Future
Current
Conventional Gl Only
¦ Current ¦ Future
Figure 5-8. Maximum hourly peak flow under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Maricopa County, AZ.
Annual Sediment Loading Rate
f
ra
0.050
0.040
0.030
0.020
0.010
0.000
0.049
0.018
0.000
0.000
Future
Current
Conventional Gl Only
¦ Current ¦ Future
Figure 5-9. Annual sediment loading rate under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Maricopa County, AZ.
79
-------
Annual TN Loading Rate
n m ^
ra
-----
.a
0.015
0.010
0.005
0.000
nm?
Future
0.000 ,
0.000 ,
Current
Conventional Gl Only
¦ Current ¦ Future
Figure 5-10. Annual TN loading rate under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Maricopa County, AZ.
Annual TP Loading Rate
ro
Si
0.0060
0.0050
0.0040
0.0030
0.0020
0.0010
0.0000
Future
Current
Conventional Gl Only
¦ Current ¦ Future
Figure 5-11. Annual TP loading rate under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Maricopa County, AZ.
As discussed in the Results sections for Harford County, MD and Scott County, MN, the resiliency of the
stormwater management approaches relative to each other can be assessed by analyzing the increase in
runoff, peak flow, and pollutant loading due to projected climate change. The first thing apparent when
looking at the Maricopa County results is that there is practically no runoff from either stormwater
80
-------
management approach under current conditions. This is not surprising because the design standard calls
for zero discharge up to the 100-year storm event, and there are only 30 years of meteorology in the HSPF
and SUSTAIN simulations. Nevertheless, as a result of climate change there is a small amount of
discharge* less that one tenth of an inch per year on average. The amount of runoff is comparable for the
Conventional and GI Only approaches, as is the maximum hourly peak flow and TN loading rate. There
is, however, a substantially larger increase in the sediment and TP loading rates for the GI Only scenario.
The future climate simulation for Maricopa County contains an especially intense precipitation event
resulting in a short period of high sediment erosion. TP is represented as sediment-associated, so both
were elevated in runoff during the storm event. The reason that the GI Only approach captures less of the
sediment and phosphorus load increases is not known. It is important to note that while the GI + Gray
approach has a much higher increase in the two loading rates, the changes are still very small.
Table 5-5 summarizes the increases in BMP footprints for the Maricopa County storm water management
scenarios that would be required to maintain or exceed current performance under future climate
conditions. The current and adapted footprints are presented both in terms of actual square feet of practice
as well as percentage of overall site area. The latter is provided as a means of comparing the current and
future adapted sizes relative to the site area (10 acres) for this particular development type (commercial).
The Conventional (Gray) Infrastructure scenario showed a 44% increase in infiltration basin size (the sole
practice) to address future climate change impacts under the GCM high intensity change climate scenario.
This represents an increase of 5.0% of the site area, or 0.5 acres. On the other hand, the combined
increase in area for the four practices modeled under the GI Only stormwater management scenario is
16.2%, or 1.62 acres.
Table 5-6 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for all of the Maricopa County stormwater management scenarios. Refer to
Section 2.5.2. of the report for a discussion on how the infrastructure cost estimates were developed. Also
provided are the increase in cost, both in dollars and percentage, and the increase in cost per acre of site.
These three metrics represent three alternative methods for evaluating the cost of adaptation, which is
effectively the increase in cost between the current and future adapted climate scenarios.
For the Conventional (Gray) scenario, the cost of adaptation is estimated to increase by $2.04 million, or
43% compared to the current cost. This is equivalent to a cost of adaptation of $0.20 million per acre of
site area.
The cost of adaptation for the GI Only scenario is estimated to increase by $2.35 million, or 59%
compared to the current cost. On a cost per site acre basis, the estimated cost of adaptation is
$0.23 million per acre of site area. Interestingly, while the area increase for the GI Only scenario is nearly
three times that of the Conventional scenario, the cost increases are nearly equivalent.
81
-------
Table 5-5. Comparison of current and future adapted best management practice
(BMP) footprints for Maricopa County, AZ stormwater management scenarios
Current
Future adapted
Future
climate
scenario
Stormwater
management
scenario
Practice
Footprint
SF
Footprint
as % of site
area
Footprint
SF
Footprint
as % of site
area
% Increase
in
footprint
GCM high
intensity
Conventional
(Gray)
Infrastructure
Infiltration
basin
49,997
11.5
71,776
16.5
44
GI Only
Permeable
pavement
86,382
19.8
124,482
28.6
44
Bioretention
13,405
3.1
24,125
5.5
80
Cistern
2,495
0.6
3,564
0.8
43
Stormwater
harvesting
basin
32,034
7.4
53,034
12.2
66
Table 5-6. Comparison of current and future adapted 20-year present value costs
for the Maricopa County, AZ stormwater management scenarios
Future
climate
scenario
Stormwater
management
scenario
Current cost
(20-yr present
value,
$millions)
Future adapted
cost (20-yr
present value,
$millions)
Increase in cost
(20-yr present
value,
$millions)
% Increase
in cost
Increase per
acre of site
($millions)
GCM high
intensity
Conventional
(Gray)
Infrastructure
4.79
6.83
2.04
43
0.20
GI Only
3.98
6.33
2.35
59
0.23
82
-------
6. SOUTHEAST SITE: ULTRA-URBAN
6.1. REGULATORY REQUIREMENTS AFFECTING STORMWATER
MANAGEMENT
Atlanta has recently enacted a progressive stormwater ordinance that promotes use of GI practices, and it
includes a substantial retention requirement. There is, however, a WQ treatment alternative if a site
cannot meet the retention requirement. The Atlanta ordinance retains the focus on tiered volume and
control requirements used in the Georgia Stormwater Manual.
• Retention of runoff from the first inch of rainfall is required (WQv). The calculation of runoff
incorporates a volumetric runoff coefficient based on site impervious area. The volume must be
infiltrated, evaporated, or reused on site.
• If retention is not possible, then the site practices must provide treatment to remove 80% of TSS
for the WQv.
• A CPv is required as well. The CPv is equal to the runoff from 1-year 24-hour storm event, and
must be discharged over a 24-hour period.
• Overbank flood protection and extreme flood protection* predevelopment peak matching is
required for the 2-, 5-, 10-, 25-, and 100-year 24-hour storm events.
6.2. STORMWATER MANAGEMENT SCENARIOS
The 2-acre ultra-urban site (see Figure 6-1) is assumed to have the following characteristics in each of the
scenarios:
• The site is 90% impervious, distributed as follows:
o 45% building
o 40% driving surfaces (parking, entry road, loading dock)
o 5% sidewalk
• The remaining pervious area (10%) is comprised of lawn/landscaping.
• The HSG percent distribution is based on a GIS analysis of soils within Atlanta. The HSG
composition is used for design storm event routing calculations to size practices for peak flow
control. Predevelopment land cover is assumed to be woods in good condition.
o HSG A: 0%
o HSG B: 77%
o HSG C: 8%
o HSG D: 15%
83
-------
Figure 6-1. Ultra-urban site layout (Atlanta, GA).
Two scenarios have been developed representing different approaches to stormwater management* a
conventional scenario using gray practices and a GI scenario using a combination of green and gray
practices. The site's percentage impervious area is sufficiently high that it is not feasible to use only GI
practices to meet the regulatory requirements. The scenarios are described in the following subsections.
6.2.1. Conventional (Gray) Infrastructure
The ultra-urban nature of the site means that pervious area is extremely limited for BMP placement. As a
result, the BMP components are assumed to be located underground. The key design elements in the
Conventional (Gray) Infrastructure scenario are as follows:
• The entire site is treated by an underground sand filter that meets the WQv requirement. The
underground sand filter is assumed to be constructed as an enclosed vault with no contact with the
underlying soil. The sand filter is located underneath a parking lot, so the design is not amenable
84
-------
to infiltration into the underlying soil. As a result, it is assumed that the infiltration requirement is
waived.
• Underground detention is used to address the CPv and flooding requirements:
o The CPv is discharged using a low flow orifice.
o A weir is used for peak-matching requirements (2-, 5-, 10-, 25-, and 100-year 24-hour
storms).
As shown in Figure 6-2 for the Conventional scenario, site runoff is routed to the underground sand filter
via culverts. Drainage and overflow from the sand filter are routed to the underground detention basin.
Runoff is then discharged off the site.
For the SUSTAIN configuration, design guidance in the Georgia Stormwater Manual was used. The sand
filter was sized to capture and treat the WQv (calculated from site impervious area), with excess runoff
discharged from a spillway to the detention basin. Underdrain outflow was also routed to the detention
basin. The sand filter media was assumed to achieve pollutant removal rates of 86% for TSS, 30% for
TN, and 60% for TP using published performance values from Center for Watershed Protection (2007)
and Hirschman et al. (2008). Removal was modeled in SUSTAIN for only the volume that filtered
through the sand media. For the detention basin, the CPv was estimated using procedures from the
Georgia Stormwater Manual. A routing spreadsheet was used to configure a weir to achieve
predevelopment peak flow matching. No pollutant removal was assigned to the detention basin. Both the
sand filter and the detention basin were assumed to be enclosed in concrete vaults, so no infiltration or ET
was modeled.
85
-------
Legend
| Building
Pervious
Sidewalk
j Vehicle
| | Drainage Areas
9 inlet
- - - Site Drainage Conveyance
BMP Type
[Mill Underground Sand Filter
j Underground Detention
Figure 6-2. Ultra-urban Conventional (Gray) Infrastructure stormwater
management scenario (Atlanta, GA).
6.2.2. Green Infrastructure (Gl) with Gray Infrastructure
The key design elements in the GI with Gray Infrastructure scenario are as follows:
• The site GI practices meet the retention requirement.
• The building has an extensive green roof that covers about 78% of the roof area (35% of the total
site area).
• Bioretention is incorporated into the site pervious area, and is configured to treat the WQv.
• Penneable pavement is used for 62.5% of the driving surface (25% of the total site area).
• Permeable pavement is used for the entire sidewalk area.
• Underground detention is used to address the remaining CPv and flood protection requirements.
86
-------
The GI with Gray scenario BMPs and conveyance are shown in Figure 6-3. Pervious concrete or asphalt
is used in the parking areas, and the storage layer fully captures the 100-year design storm depth. The
green roof captures the water quality volume within its soil media, and excess runoff is discharged with
the runoff from the remainder of the rooftop. Runoff from the roof and from the conventional pavement is
conveyed to the bioretention cells north and south of the building. Overflow from the bioretention cells is
routed to the underground detention basin, and then offsite.
Legend
| Building
~ Pervious
[ Sidewalk
[ Vehicle
- - - Roof Drainage Conveyance
| | Drainage Areas
BMP Type
: Bioretention
/ Green Roof
Permeable Pavement
¦ Underground Detention
- - - r BMP Conveyance
Figure 6-3. Ultra-urban Green Infrastructure (GI) with Gray Infrastructure
stormwater management scenario (Atlanta, GA).
For the SUSTAIN configuration, practice dimensions and properties were based on design criteria and
guidance from the Georgia Stormwater Manual. The green roof soil media depth was assumed to be
6 inches, with soil moisture holding properties based on Palla et al. (2008), Schneider (2011), and
Latshaw et al. (2009). Both the permeable pavement and bioretention were assumed to have underdrains.
For both the bioretention and the permeable pavement, 4 inches of stone base were assumed to lie below
the underdrain, which allowed for a fraction of the site runoff to be infiltrated. Infiltration rates for
bioretention and penneable pavement were assumed to be 0.1 inches/hour, consistent with compacted
87
-------
B soils. ET was assumed for bioretention. A small amount of ET from permeable pavement was assumed,
equal to 10% of ET that would normally take place. The bioretention media was assumed to achieve
percentage pollutant removal rates of 78% for TSS, 57% for TN, and 63% for TP using published
performance values from Center for Watershed Protection (2007) and Tetra Tech (2014). Removal was
modeled in SUSTAIN for only the volume that filtered through the bioretention media and was
subsequently discharged via the underdrain. A routing spreadsheet was used as described for previous site
scenarios to develop the configuration for the underground detention basin to address the remaining CPv
and to meet peak flow matching requirements.
6.3. ADAPTATION SIMULATION
The objective of the adaptation simulation is to determine the increases in BMP footprint (surface area)
that would be required to maintain or exceed current performance under future climate conditions for
each stormwater management scenario. Table 6-1 summarizes the key components of the modeling
procedure for each scenario. In the GI with Gray Infrastructure scenario, permeable pavement is modeled
but was not modified in the adaptation simulation because the ultra-urban site layout does not allow for
expansion of this practice.
Table 6-1. Features of adaptation simulation for Atlanta, GA
Location
Stormwater management
scenario
Future
adaptation
Affected practices
Atlanta,
GA
Conventional (Gray)
Infrastructure
Resize practices
Underground sand filter, underground dry
detention basin
GI with Gray Infrastructure
Resize practices
Bioretention, underground dry detention basin,
green roof
6.4. CURRENT AND FUTURE CHANGES IN PRECIPITATION
For the Atlanta, GA climate scenarios, the changes in average annual precipitation show an increase for
all years with a low degree of variability (see Figure 6-4). Projected average annual depth increases from
55.7 to 59.4 inches, or by 6.6%. As seen in Figure 6-5, projected monthly precipitation increases in some
months and decreases in other months. In addition, daily sums of precipitation depth were calculated and
were used to determine percentiles of 24-hour depth of interest to stormwater managers (see Table 6-2).
While daily sums do not provide a true measure of storm event depth (storms have variable lengths and
may span more than 1 day), they do provide useful information about expected depths over a 24-hour
period. As seen in the table, the change in depth between current and future ranges from 0.10 inches for
the 85th percentile to 0.41 inches for the 99th percentile.
The plot of highest hourly precipitation volumes (see Figure 6-6) shows an increase of about 1.2x
between the 1-year and 10-year recurrence intervals; for the two highest hourly values (15-year and
30-year recurrence), the increase is about 1.5 x.
88
-------
Annual Precipitation (30-yr record)
90
80
70
60
50
40
30
20
10
0
I Current ¦ Future
rHrvim'd-LniDr^-oomOiH
r\im*fru-)<£>r--ooOcirNim*tu-)<£>r-*Qoa^o qj
HHHrlrlrlHrl(N(NN(N(N(\(N(N(N(NfO UO
Rank
¦5
Figure 6-4. Ranked annual precipitation for current conditions and high intensity
future climate scenario at Atlanta, GA.
Monthly Average Precipitation
lllllllllill
.E 4
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
¦ Current ¦ Future
Figure 6-5. Monthly average precipitation for current conditions and high intensity
future climate scenario at Atlanta, GA.
89
-------
Table 6-2. 24-hour precipitation depth percentiles for current conditions and high
intensity future climate scenario at Atlanta, GA
Percentile
Current conditions
24-h depth (in)
Future climate
24-h depth (in)
Change (+/-in)
85th
1.04
1.14
+0.10
90th
1.28
1.40
+0.12
95th
1.71
1.87
+0.17
99th
2.66
3.07
+0.41
Hourly Precipitation
Current
Future
Simulation Recurrence Interval (years)
Figure 6-6. Hourly precipitation recurrence interval for current conditions and high
intensity future climate scenario at Atlanta, GA.
6.5. RESULTS
SUSTAIN was run for the following conditions for each stormwater management scenario:
• Current climate, site without stormwater management/BMPs
• Future climate, site without stormwater management/BMPs
• Current climate, site with stormwater management/BMPs
• Future climate, site with stormwater management/BMPs
• Future climate, site with BMPs adapted to meet current hydrology and water quality performance
90
-------
As shown in Table 6-3, two sets of SUSTAIN runs were performed for a combination of two stormwater
management approaches and one climate scenario.
Table 6-3. Stormwater management and climate scenarios for Atlanta, GA
Stormwater management approach
GCM high
intensity
Conventional
X
GI Only
X
A full presentation of the results of all the runs is provided in APPENDIX B. . For brevity, the results in
this section focus on a few topics of interest to stormwater managers: (1) a comparison of the site
performance with BMPs between current and future climate conditions, (2) the increases in BMP
footprints needed to offset impacts of climate change when BMPs are adapted using SUSTAIN
optimization, and (3) a comparison of current stormwater infrastructure costs to future costs when BMPs
are adapted to offset impacts of climate change.
For the comparison of the site performance with BMPs between current and future climate conditions, the
downscaled future GCM (high intensity change) scenario was selected for the comparison. A discussion
of other topics of interest are provided in the general conclusions Section 8. , including changes in
pretreatment site performance, changes in post-treatment site performance, climate scenario sensitivity
analysis, and adapting BMPs under future climate to meet current performance.
Rather than comparing the performance of the stormwater management approaches independent of
climate change (i.e., how much better does one perform than the other under current conditions), this
study focuses on how the stormwater management approaches compare relative to climate change. Table
6-4 provides current and future performance for the stormwater management approaches, normalized to
area. Note that there is no numeric measure of change in the FDC between the current and future climate,
so the highest hourly peak flow during the simulation is presented as a proxy for large storm event
response. Figure 6-7 through Figure 6-11 present each metric graphically from Table 6-4.
91
-------
Table 6-4. Current and future performance of Atlanta, GA Site by stormwater
management approach
Stormwater management
approach
Current
Future
Change
Runoff (inch/yr)
Conventional
43.97
47.28
+3.31
GI + Gray
15.14
16.98
+1.84
Maximum hourly peak flow (cfs/ac)
Conventional
0.24
0.51
+0.26
GI + Gray
0.92
1.50
+0.59
Sediment (ton/ac/yr)
Conventional
0.45
0.56
+0.11
GI + Gray
0.55
0.70
+0.16
TN (lb/ac/yr)
Conventional
18.61
19.20
+0.59
GI + Gray
7.07
8.00
+0.93
TP (lb/ac/yr)
Conventional
1.05
1.09
+0.03
GI + Gray
0.63
0.72
+0.09
50.00
40.00
> 30.00
£ 20.00
10.00
0.00
Figure 6-7. Annual site runoff under current climate and future general circulation
model (GCM) scenario by stormwater management approach for Atlanta, GA.
Annual Site Runoff
47.28
43.97
Future
Current
Conventional Gl + Gray
¦ Current ¦ Future
92
-------
Maximum Hourly Peak Flow
1.50,
2.00
1.50
1.00
0.50
0.00
Future
Current
Conventional Gl + Gray
¦ Current ¦ Future
Figure 6-8. Maximum hourly peak flow under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Atlanta, GA.
0.80
0.60
0.40
0.20
0.00
Annual Sediment Loading Rate
0.70,
Future
Current
Conventional Gl + Gray
¦ Current ¦ Future
Figure 6-9. Annual sediment loading rate under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Atlanta, GA.
93
-------
Annual TIM Loading Rate
19.20
ra
20.00
15.00
10,00
5.00
0.00
Future
Current
Conventional Gl + Gray
¦ Current ¦ Future
Figure 6-10. Annual total nitrogen (TN) loading rate under current climate and
future general circulation model (GCM) scenario by stormwater management
approach for Atlanta, GA.
1.20
1.00
0,80
0.60
0.40
0.20
0.00
Annual TP Loading Rate
1.09
ro
-Q
Future
Current
Conventional Gl + Gray
¦ Current ¦ Future
Figure 6-11. Annual total phosphorous (TP) loading rate under current climate and
future general circulation model (GCM) scenario by stormwater management
approach for Atlanta, GA.
As discussed in the previous Results sections for the individual sites, the resiliency of the stormwater
management approaches relative to each other can be assessed by an analysis of the increase in runoff,
peak flow, and pollutant loading due to projected climate change. For annual average site runoff, the
increase in runoff for the Conventional approach at 3.31 inches is nearly double the runoff increase for
94
-------
GI + Gray, at 1.84 inches. This indicates the GI + Gray approach was better at disposing of additional
runoff due to changes in future precipitation volume than the Conventional approach, suggesting that the
GI + Gray approach is more resilient to climate change for this measure. The same is not true for the other
measures, where the GI + Gray has a larger increase in maximum hourly peak flow, as well as a larger
increase in all of the pollutant loading rates.
Table 6-5 summarizes the increases in BMP footprints for the Atlanta, GA stormwater management
scenarios that would be required to maintain or exceed current performance under future climate
conditions. The current and adapted footprints are presented both in terms of actual square feet of practice
as well as percentage of overall site area. The latter is provided as a means of comparing the current and
future adapted sizes relative to the site area (2 acres) for this particular development type (ultra-urban).
Table 6-5. Comparison of current and future adapted best management practice
(BMP) footprints for Atlanta, GA stormwater management scenarios
Current
Future adapted
%
Future
climate
scenario
Stormwater
management
scenario
Practice
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint as
% of site
area
Increase
in
footprint
GCM high
intensity
Conventional
(Gray)
Underground
sand filter
2,500
2.9
4,600
5.3
84
Infrastructure
Underground
dry detention
basin
5,000
5.7
6,200
7.1
24
GI with Gray
Infrastructure
Bioretention
with underdrain
2,810
3.2
3,934
4.5
40
Underground
dry detention
basin
2,500
2.9
3,300
3.8
32
Green roof
30,492
35.0
35,184
40.4
15
Permeable
pavement
26,136
30.0
26,136
30.0
0
For the Conventional (Gray) Infrastructure scenario, the optimal solution resulted in an increase in size
for both the underground dry detention basin and the underground sand filter. The combined increase in
BMP footprint is equal to about 3.8% of the site area, or about 3,300 square feet. One outcome of the
optimization was that increases in runoff volume and TN under the GCM high intensity future climate
scenario could not be fully mitigated by increasing BMP footprints. For runoff volume, this is not
surprising because both practices are underground and concrete-lined, so there is no mechanism to reduce
runoff volume. However, the outcome is surprising for TN because increasing the area (and thus the
treatment volume) of the sand filter does improve TN mass removal. The reason is that the increase in TN
mass under the future climate scenario is greater than the ability of the sand filter to remove TN mass
95
-------
even at full treatment capacity. This may be an artifact of how TN removal is represented in the
SUSTAIN model (fixed percentage reduction of mass), but it does suggest that gray practices with limited
pollutant removal mechanisms may be at a disadvantage for mitigating climate change impacts.
The GI with Gray Infrastructure scenario uses a different approach for meeting regulatory storm water
requirements, and the adapted site under future climate conditions is able to meet all of the targets,
including runoff and TN reduction. The combined increase in BMP footprint is equal to 7.6% of the site
area, about 6,600 square feet. Most of the increase is due to a larger green roof footprint. Note that
permeable pavement was considered to be implemented at the maximum practical footprint, so it was not
included in the adaptation optimization.
Table 6-6 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for all of the Atlanta, GA stormwater management scenarios. Refer to Section 2.5.2.
of the report for a discussion on how the infrastructure cost estimates were developed. Also provided are
the increase in cost, both in dollars and percentage, and the increase in cost per acre of site. These three
metrics represent three alternative methods for evaluating the cost of adaptation, which is effectively the
increase in cost between the current and future adapted climate scenarios.
Table 6-6. Comparison of current and future adapted 20-year present value costs
for the Atlanta, GA stormwater management scenarios
Future
climate
scenario
Stormwater
management
scenario
Current cost
(20-yr present
value,
$millions)
Future adapted
cost (20-yr
present value,
$millions)
Increase in cost
(20-yr present
value,
$millions)
% Increase
in cost
Increase per
acre of site
$millions
GCM high
intensity
Conventional
(Gray)
Infrastructure
1.38
2.27
0.89
64
0.09
GI with Gray
Infrastructure
2.31
2.60
0.29
13
0.03
For the Conventional (Gray) scenario, the estimated cost of adaptation is a $0.89 million increase
compared to the current cost, or an increase in cost of 64%. This is equivalent to a cost of adaptation of
$0.09 million per acre of site area.
The cost of adaptation for the GI with Gray scenario is estimated to be an increase of $0.29 million,
which reflects a 13% increase in cost. On a cost per site acre basis, the estimated cost of adaptation is
$0.03 million per acre of site area. While the adaptation cost increase for the GI with Gray scenario is
significantly less than the cost for the Conventional scenario, the cost of GI with Gray under current
climate is significantly more to start with than the cost of the Conventional stormwater management.
96
-------
7. PACIFIC NORTHWEST SITE: TRANSPORTATION
CORRIDOR/GREEN STREET
7.1. REGULATORY REQUIREMENTS AFFECTING STORMWATER
MANAGEMENT
Portland has a stormwater ordinance that emphasizes use of retention and GI practices. The ordinance
requires infiltration of the 10-year 24-hour storm event to the maximum extent practicable. There is a
70% TSS reduction required for discharging systems (where the infiltration requirement cannot be met).
There is also a predevelopment peak matching requirement for the 2-, 5-, and 10-year 24-hour storm
events, depending on where the site discharges.
7.2. STORMWATER MANAGEMENT SCENARIOS
The approach for the Portland scenario differs from the other locations; the stormwater management
scenario reflects a specific style of stormwater management for which Portland has gained recognition in
the stormwater management profession* the green street. Practitioners use land adjacent to roads as an
opportunity to retrofit practices into the urban landscape. GI elements placed in medians and along
rights-of-way (ROWs) are used to address water quality treatment and stormwater volume reduction
goals. Green street projects in Portland tend not to fall under the city's requirements because the city is
implementing them in road rights-of-way that are exempt from postconstruction stormwater requirements.
The city does attempt to meet the requirements, but it is not always possible due to site limitations.
There are numerous green street case studies and master plans the city has published. One of the master
plans is for a district encompassing several city blocks called the Gateway Urban Renewal Area (City of
Portland, 2008). Rather than providing street-by-street designs, the report presents several "typologies"
based on the ROW width. After reviewing the typologies, the 68-foot ROW Stormwater Curb Extension
typology was selected for modeling representation. This typology lies in the middle of the range of ROW
widths, and it comes the closest to fully meeting the infiltration requirement among the typologies.
The 68" ROW Stormwater Curb Extension typology is shown in Figure 7-1. For the SUSTAIN modeling,
one side of the street was modeled because each street side is a mirror image of the other. Based on
design parameters in the master plan, the green street site has the following characteristics:
• Site area:
o 30,800 ft2
o 1.1% street trees
o 2.2% permeable pavement
o 8.1 bioretention/infiltration trench surface area
o 88.6% impervious surface area
97
-------
• Bioretention/infiltration trench configuration:
o 6 inches of storage above the soil media
o Soil media depth 6 inches
o Drain rock depth 5 feet
The master plan discusses soil characteristics, stating that the area is composed of well-drained loams and
silt loams with infiltration rates exceeding 6 inches per hour below 4 feet. Tests by local staff confirmed
infiltration rates of 2 inches per hour or greater below the top compacted soil layer.
98
-------
-StrOTt Tr##
¦Pgrous Pov^f
¦SlorfiwAter Curb
.Extension i. yyk
¦SlOffliwoSer CuftJ ExlCftsien
Porous Pan«e
'Street free ,
¦Inlet Foreboy
Outlet
Basin Area/Flow
Illustrative Street Plan
19'
Parking
and
Travel Lane
19'
Parking
and
Travel Lane
Curb
11.5' Stormwater
Curb Ext. w/
Drain Rock'
Storage
8'
Sidewalk
2.5'
-Bldg.
Zone
Clearance
Varies
11.5' Stormwater
Curb Ext. w/
Drain Rock1
Storage
68'
Right of Way
Figure 7-1. Green street site layout (City of Portland, 2008).
7.2.1. Green Infrastructure (Gl) Only
The key design elements in the GI Only (green street) scenario are as follows:
99
-------
• Each side of the street is a mirror image, and the assumed flow direction shown in Figure 7-1 is
away from the centerline of the road. For simplicity, the SUSTAIN scenario is built using one
side of the street. As a result, the site area is 15,400 ft2 (about 0.35 acre).
• Bioretention occupies 1,247 ft2, located in bump-outs along the street. The stored runoff
infiltrates completely into the soil; no underdrain is used.
• Permeable pavement occupies 339 ft2, and is located between street parking and the sidewalk.
The site is modeled as a series of adjacent connected drainage areas as shown in Figure 7-1. Within each
drainage area, flow is routed proportionately to the street trees, permeable pavement, and bioretention.
Overflow from the street trees and permeable pavement is routed to bioretention in the same drainage
area. When the capacity of bioretention is exceeded, flow is routed to the downstream bioretention.
Overflow from the most downstream bioretention is routed off the site.
For the SUSTAIN configuration, the areas and depths (storage, soil, and drain rock) were specified as
given in the site plan dimension shown above. Infiltration into the underlying soil from all of the practices
was assumed to be 2 inches per hour. Infiltration was assumed to be the primary removal mechanism, so
no additional pollutant removal was modeled. ET was assumed for the bioretention. A small amount of
ET from permeable pavement was assumed, equal to 10% of ET that would normally take place.
7.3. ADAPTATION SIMULATION
The objective of the adaptation simulation is to determine the increases in BMP footprint (surface area)
that would be required to maintain or exceed current performance under future climate conditions for
each stormwater management scenario. Table 7-1 summarizes the key components of the modeling
procedure for the Portland GI Only scenario. Note that permeable pavement was also modeled as part of
the GI Only (green street) scenario for Portland. However, this practice was not modified in the
adaptation simulation because expansion of the permeable pavement footprint is not feasible given the
current site layout. Further, permeable pavement areas only account for approximately 2% of the area of
the green street and do not receive significant runoff. Rather, they serve more of an aesthetic function.
Table 7-1. Features of adaptation simulation for Portland, OR
Location
Stormwater management
scenario
Future
adaptation
Affected
practices
Portland, OR
GI Only
Resize practices
Bioretention
7.4. CURRENT AND FUTURE CHANGES IN PRECIPITATION
The Portland, OR future climate scenario reflects a deviation from the previous locations, where the
scenarios with the highest increase in intensity also showed an overall volume increase. For Portland (see
Figure 7-2), the precipitation volume is actually projected to decrease, from 42.0 to 39.7 inches (-5.5%
drop). As seen in Figure 7-3, monthly precipitation volume decreases in most months from current to
100
-------
future conditions, although increases are seen during two of the winter months. In addition, daily sums of
precipitation depth were calculated and used to determine percentiles of 24-hour depth of interest to
stormwater managers (see Table 7-2). While daily sums do not provide a true measure of storm event
depth (storms have variable lengths and may span more than 1 day), they do provide useful information
about expected depths over a 24-hour period. For the 85th through 95th percentiles, the change in 24-hour
depth actually decreases. However, there is an increase of 0.14 inches for the 99th percentile, indicating a
modest increase in intensity for the very largest events.
In terms of highest hourly precipitation volumes (see Figure 7-4), projected future intensity more or less
tracks current intensity, with the highest increase about 1.16x.
Annual Precipitation (30-yr record)
70
60
50
40
30
20
10
0
I Current ¦ Future
r-lrMrfl'Si-LniDI^OOOlO
HHridririririMNININNNNNNNrn
Rank
>
<
Figure 7-2. Ranked annual precipitation for current conditions and high intensity
future climate scenario at Portland, OR.
101
-------
Monthly Average Precipitation
9
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
¦ Current ¦ Future
Figure 7-3. Monthly average precipitation for current conditions and high intensity
future climate scenario at Portland, OR.
Table 7-2. 24-hour precipitation depth percentiles for current conditions and high
intensity future climate scenario at Portland, OR
Percentile
Current conditions
24-h depth (in)
Future climate
24-h depth (in)
Change (+/-in)
85th
0.53
0.49
-0.04
90th
0.67
0.62
-0.05
95th
0.91
0.90
-0.01
99th
1.51
1.65
+0.14
102
-------
Hourly Precipitation
1.2
1
0.8
.£ 0.6
0.4
0.2
0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
Simulation Recurrence Interval (years)
Figure 7-4. Hourly precipitation recurrence interval for current conditions and high
intensity future climate scenario at Portland, OR.
7.5. RESULTS
SUSTAIN was run for the following conditions for each stormwater management scenario:
• Current climate, site without stormwater management/BMPs
• Future climate, site without stormwater management/BMPs
• Current climate, site with stormwater management/BMPs
• Future climate, site with stormwater management/BMPs
• Future climate, site with BMPs adapted to meet current hydrology and water quality performance
As shown in Table 7-3, one set of SUSTAIN runs was performed for a combination of one stormwater
management approach and one climate scenario.
Table 7-3. Stormwater management and climate scenario for Portland, OR
Stormwater management
approach
GCM high
intensity
GI Only
X
Current
Future
103
-------
A full presentation of the results of all the runs is provided in APPENDIX B. . For brevity, the results in
this section focus on a few topics of interest to stormwater managers: (1) a comparison of the site
performance with BMPs between current and future climate conditions, (2) the increases in BMP
footprints needed to offset impacts of climate change when BMPs are adapted using SUSTAIN
optimization, and (3) a comparison of current stormwater infrastructure costs to future costs when BMPs
are adapted to offset impacts of climate change.
For the comparison of the site performance with BMPs between current and future climate conditions, the
downscaled future GCM (high intensity change) scenario was selected for the comparison. A discussion
of other topics of interest are provided in the general conclusions Section 8. , including changes in
pretreatment site performance, changes in post-treatment site performance, climate scenario sensitivity
analysis, and adapting BMPs under future climate to meet current performance.
Rather than comparing the performance of the stormwater management approaches independent of
climate change (i.e., how much better does one perform than the other under current conditions), this
study focuses on how the stormwater management approaches compare relative to climate change. Table
7-4 provides current and future performance for the stormwater management approaches, normalized to
area. Note that there is no numeric measure of change in the FDC between current and future climate, so
the highest hourly peak flow during the simulation is presented as a proxy for large storm event response.
Figure 7-5 through Figure 7-9 present each metric graphically from Table 7-4.
The Northwest site has one stormwater management approach, so multiple approaches are not available
for comparison. What stands out from the results is the small increase in measures between current and
future climate conditions* much smaller, for the most part, than the increases reported for the other sites.
This is not due to the lower site area used for this site because all of the measures are normalized to area.
The reason the changes are small is likely that climate models generally predict lower changes in
precipitation intensity relative to most other locations in the United States. The future climate scenario
selected for this geographic location had the highest large storm event intensity change among ten
candidate future climate scenarios. As shown in Table 7-2, the increase in the 99th percentile daily rainfall
volume is only 0.14 inches, considerably less than the 99th percentile values shown for the other sites
corresponding to the future climate scenario with the highest large storm event intensity change.
104
-------
Table 7-4. Current and future performance of Portland, OR site by storm water
management approach
Stormwater management
approach
Current
Future
Change
Runoff (inch/yr)
GI Only
0.052
0.088
+0.036
Maximum hourly peak flow (cfs/ac)
GI Only
0.26
0.49
+0.22
Sediment (ton/ac/yr)
GI Only
0.0014
0.0015
+0.0002
TN (lb/ac/yr)
GI Only
0.027
0.039
+0.012
TP (lb/ac/yr)
GI Only
0.0034
0.0042
+0.0008
Annual Site Runoff
^ 0.088 J
0.100
> 0.052 I
Future
£
c 0.050
Current
0.000
GI Only
¦ Current ¦ Future
Figure 7-5. Annual site runoff under current climate and future general circulation
model (GCM) scenario by stormwater management approach for Portland, OR.
105
-------
Maximum Hourly Peak Flow
0.60
0.40
0.20
0.00
Future
Current
Gl Only
¦ Current ¦ Future
Figure 7-6. Maximum hourly peak flow under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Portland, OR.
Annual Sediment Loading Rate
0.0015
0.0014
Current
0.0012
Gl Only
¦ Current ¦ Future
Figure 7-7. Annual sediment loading rate under current climate and future general
circulation model (GCM) scenario by stormwater management approach for
Portland, OR.
106
-------
Annual TN Loading Rate
A 0.039 J
0.040
^0027*
>
ro
Future
_Q
0.020
Current
0.000
Gl Only
¦ Current ¦ Future
Figure 7-8. Annual total nitrogen (TN) loading rate under current climate and
future general circulation model (GCM) scenario by stormwater management
approach for Portland, OR.
Annual TP Loading Rate
0.00421
ro
-Q
0.0060
0.0040
0.0020
0.0000
0.0034
Future
Current
Gl Only
¦ Current ¦ Future
Figure 7-9. Annual total phosphorous (TP) loading rate under current climate and
future general circulation model (GCM) scenario by stormwater management
approach for Portland, OR.
Table 7-5 summarizes the increases in BMP footprints for the Portland, OR stormwater management
scenarios that would be required to maintain or exceed current performance under future climate
conditions. The current and adapted footprints are presented both in terms of actual square feet of practice
as well as percentage of overall site area. The latter is provided as a means of comparing the current and
107
-------
future adapted sizes relative to the site area for this particular development type (transportation corridor).
Permeable pavement was not resized as part of the adaptation. Bioretention provides almost all of the
water quantity and quality treatment for the site due to their large storage volumes and high infiltration
capacity. The increase in bioretention footprint is 36%, or 2.9% of the site area. Interestingly, total runoff
volume under the GCM high intensity change climate scenario actually decreases compared to current
climate conditions. However, the volume discharged from the BMPs increases under future climate
compared to current climate. The reason is that the site is designed to capture the equivalent of a 10-year
24-hour storm event, so runoff only occurs during the largest of storm events. While overall future runoff
decreases, the intensify of the largest events increases, leading to an increase in discharge, nutrient loads,
and large event peak flows.
Table 7-5. Comparison of current and future adapted best management practice
(BMP) footprints for Atlanta, GA stormwater management scenarios
Current
Future adapted
Future
climate
scenario
Stormwater
management
scenario
Practice
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint as
% of Site
area
%
Increase in
footprint
GCM high
intensity
GI Only
Bioretention
swale
1,239
8.0
1,681
10.9
36
Permeable
pavement
345
2.2
345
2.2
0
Table 7-6 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for all of the Portland, OR stormwater management scenarios. Refer to Section 2.5.2.
of the report for a discussion on how the infrastructure cost estimates were developed. Also provided are
the increase in cost, both in dollars and percentage, and the increase in cost per acre of site. These three
metrics represent three alternative methods for evaluating the cost of adaptation, which is effectively the
increase in cost between the current and future adapted climate scenarios.
Table 7-6. Comparison of current and future adapted 20-year present value costs
for the Maricopa County, AZ stormwater management scenarios
Future
climate
scenario
Stormwater
management
scenario
Current cost
(20-yr present
value,
$millions)
Future adapted
cost (20-yr
present value,
$millions)
Increase in cost
(20-Yr present
value,
$millions)
% Increase
in cost
Increase per
acre of site
($millions)
GCM high
intensity
GI Only
0.20
0.27
0.07
35
0.20
108
-------
The cost of adaptation for the GI Only scenario is estimated to be an increase of $0.07 million, which
reflects a 35% increase in cost. On a cost per site acre basis, the estimated cost of adaptation is
$0.20 million per acre of site area.
109
-------
8. DISCUSSION AND CONCLUSIONS
Each of the individual site sections concluded with a discussion of results centered on a comparison of
site performance under current and projected future climate conditions. This section takes a broader view
and looks at results across all of the sites to interpret what can be learned about climate change impacts to
stormwater management. The discussion is organized around the central study questions provided in the
Introduction and repeated below, and concludes with a summary of results and some additional insights:
1. How might extreme precipitation events affect the performance of conventional stormwater
infrastructure and GI,
2. How can conventional designs and GI designs be adapted so that a site experiencing extreme
precipitation conditions in the future provides the same performance as the site under current
conditions, and
3. What do the results suggest regarding the adaptation potential of gray and green infrastructure for
increases in extreme precipitation events?
For reference, Table 8-1 lists each site analyzed in this study, along with the site's characteristics,
stormwater management requirements, and stormwater management approaches and practices.
110
-------
Table 8-1. Stormwater management approach summary
Region
Location
Characteristics
Stormwater
requirements
Stormwater
management
approach
Practices
Mid-Atlantic
Harford
County, MD
Mixed use
20 acres
65% impervious
Completely infiltrate
recharge volume
Treat water quality
volume for TSS/TP
Channel protection
volume (24-h
detention of 1-yr 24-h
storm)
Match predeveloped
peak for 10-yr 24-h
storm
Conventional
(Gray)
Infrastructure
Surface sand filters,
extended dry detention
basin
GI with Gray
Infrastructure
Infiltration trenches,
infiltration basins,
permeable pavement,
and dry detention basin
Conventional
(Gray)
Infrastructure with
Distributed GI
Surface sand filters,
extended dry detention
basin, distributed
infiltration trenches
Midwest
Scott
County, MN
Residential
30 acres
48% impervious
Treat water quality
volume for TSS
Match predeveloped
peak for 2-yr 24-h
storm and 100-yr
24-h storm
Conventional
(Gray)
Infrastructure
Wet pond
GI with Gray
Infrastructure
Distributed bioretention
and dry detention basin
GI Only
Distributed bioretention,
permeable pavement,
and impervious surface
disconnection
Conventional
(Gray)
Infrastructure with
Distributed GI
Wet pond, distributed
bioretention
Arid
southwest
Maricopa
County, AZ
Commercial
10 acres
80% impervious
100% retention of the
100-yr 2-h storm
event
Conventional
(Gray)
Infrastructure
Detention/infiltration
basin
GI Only
Permeable pavement,
cistern, bioretention
and stormwater
harvesting basin
Southeast
Atlanta, GA
Ultra-urban
2 acres
90% impervious
Treat water quality
volume for TSS
Channel protection
volume (24-h
detention of 1-yr 24-h
storm)
Match predeveloped
peak for 2-yr, 5-yr,
10-yr, 25-yr, and
100-yr 24-h storm
Conventional
(Gray)
Infrastructure
Underground sand filter,
underground dry
detention basin
GI with Gray
Infrastructure
Green roof, permeable
pavement, bioretention,
and underground dry
detention basin
111
-------
Table 8 1. Stormwater management approach summary (Continued)
Region
Location
Characteristics
Stormwater
requirements
Stormwater
management
approach
Practices
Pacific
northwest
Portland,
OR
Transportation
corridor
0.35 acres
89% impervious
70% TSS reduction
infiltration of 10-yr
24-h storm event as
practicable
Match predeveloped
peak for 2-yr, 5-yr,
10-yr 24-h stonn
GI Only
Bioretention swales,
permeable pavement
8.1. STUDY QUESTION #1
How might extreme precipitation events affect the performance of conventional stormwater
infrastructure and green infrastructure (GI)?
To answer the first question, each stormwater management scenario was modeled under current climate
conditions, and the performance of the site practices was calculated from modeling results. Next, each
scenario was modeled under future climate conditions and the change in performance tabulated.
Performance is presented first for the downscaled high intensity GCM climate scenarios. Next, results of
the future climate sensitivity analysis are provided.
8.1.1. Performance Comparison for Stormwater Management
Scenarios Under Current and Future Precipitation Conditions
This subsection presents results showing projected changes to site performance due to climate change.
The future climate scenarios presented here are limited to the downscaled GCM scenarios representing
the largest increase in precipitation intensity (i.e., the low and medium intensity scenarios for Scott
County are not included). The reason for focusing on the high intensity climate scenarios is to allow for a
more equivalent comparison between locations and stormwater management-stormwater treatment
scenarios. It is important to note that these results represent potential conditions under future climate,
notably using climate scenarios with the largest storm intensity change among a population of ten climate
scenarios. Future climate impacts could be less, or more extreme. In the end, the results are intended to
show sensitivity to plausible future conditions that will stress these stormwater systems.
Site performance measures used in this analysis include annual runoff, maximum peak outflow during the
30-year simulation, and pollutant loads for sediment, TN, and TP. Results are first presented for each site
without the impact of BMPs* in other words, how climate change could affect the site as a whole. Next,
site performance is explored taking the benefits of BMPs into account.
8.1.1.1. Changes in Pretreatment Site Performance
The analysis focuses on changes in runoff volume, maximum peak flow, and pollutant mass loading from
the site land surfaces prior to any reductions due to BMPs. Percentage change in runoff ranges from -6.7
112
-------
to 12.5% (see Figure 8-1). The change is negative for Portland due to lower precipitation and higher
summer ET under the future downscaled GCM climate scenario. For maximum hourly peak outflow
during the 30-year simulations, the percentage change ranges from 6.3 to 90.8% (see Figure 8-2). While
the change in peak flow at Portland is the lowest among the locations, it is positive rather than negative;
this indicates that while overall precipitation volume decreased under the future climate scenario, the
intensity of the largest storm events did increase.
Percentage changes in loads range from 1.5 to 26.7% in most cases, except for Portland where the
percentage changes are negative (see Figure 8-3, Figure 8-4, and Figure 8-5). A major exception is
Maricopa County, where sediment more than doubles and TP more than triples. There is one storm event
with wet antecedent conditions where precipitation doubles under future climate, and a large increase in
pervious runoff depth results. Surface and rill erosion have a nonlinear increase with runoff depth, so a
large mass of sediment and bound phosphorus are exported. It is possible that the model prediction
represents an extremely rare occurrence. If the storm is omitted from the analysis, the percentage increase
drops to 58.4% for sediment and 32.0% for TP. Note that percentage changes in loads reflect the entire
30-year simulation, whereas the percentage changes in maximum outflow are calculated from the single
highest hour from the 30-year simulation and, therefore, tend to be larger than the load increases.
Percent Change in Site Runoff (no BMPs)
Portland Maricopa Atlanta Harford Scott County
County County
Figure 8-1. Percentage change in site runoff (no best management practices [BMPs])
between current and future climate conditions.
113
-------
Percent Change in Maximum Outflow (no BMPs)
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
90.8%
63%
Portland
Maricopa
County
54.0%
Atlanta
70.7%
Harford
County
41.7%
Scott County
Figure 8-2. Percentage change in site maximum hourly peak outflow (no best
management practices [BMPs]) between current and future climate conditions.
Percent Change in Sediment Load (no BMPs)
115.8%
120%
100%
80%
60%
40%
20%
0%
-20%
Portland Maricopa
County
6.8%
Atlanta
24.3%
m
26.7%
Harford
County
Scott County
Figure 8-3. Percentage change in site sediment load (no best management practices
[BMPs]) between current and future climate conditions.
114
-------
Percent Change in TN Load (no BMPs)
25%
20%
15%
10%
5%
0%
-5%
-10%
23.1%
-6.4%
Port and
5.6%
2.6%
5.2%
Maricopa
County
Atlanta
Harford
County
Scott County
Figure 8-4. Percentage change in site TN load (no best management practices
[BMPs]) between current and future climate conditions.
Percent Change in TP Load (no BMPs)
300%
250%
200%
150%
100%
50%
0%
-50%
I
.34.17
0
1.5% 2.4%
.5.2%
Portland
Maricopa
County
Atlanta
Harford
County
Scott County
Figure 8-5. Percentage change in site total phosphorous (TP) load (no best
management practices [BMPs]) between current and future climate conditions.
The load simulated in HSPF from the land surface is largely a function of runoff depth, although load
buildup may be exhausted on impervious land. Thus, the sediment and TP loads tend to be strongly
sensitive to changes in intensity. It is really the intensity of runoff, not precipitation that matters* the
difference being especially important in northern sites where there is a change in the snow
accumulation/melt regime.
115
-------
8.1.1.2. Changes in Post-treatment Site Performance
A series of charts is presented in this subsection showing performance of BMPs and overall site export,
accounting for BMP treatment. The focus of the analysis is on comparing performance for currently
implemented BMPs under current versus future climate conditions. Both conventional and GI site-based
scenarios are shown. Performance for the adaptation scenarios (where BMPs are resized or distributed GI
is added to a site to reduce runoff volume, loads, and the highest runoff rates to match current
performance) is not discussed because performance with adapted BMPs is the same or better than current
performance and does not provide insight into how climate change affects BMP performance.
Three sets of analyses are shown:
• BMP percentage reductions in volume and mass
• Unit-area volume and mass reductions from BMPs
• Unit-area post-treatment site export
It is important to note that these results reflect an exploration of the range of climate impacts on BMP
performance. While five different geographic regions are presented, the design of the study does not lend
itself to making inferences about regional variation in BMP response to climate change. Each location
represents a different type of land use, climate conditions, soils and infiltration rates, as well as other
factors. BMP selection is driven largely by local and state design requirements. However, some trends are
evident.
The first set of figures shows percentage reductions in volume (see Figure 8-6) and mass (see Figure 8-7,
Figure 8-8, and Figure 8-9) due to the combined effects of all the BMPs at each site. Percentage reduction
of runoff volume and pollutant mass tends to decrease under future climate conditions. Bypass increases
under future climate conditions with high projected changes in large storm event intensity, while the BMP
footprints and configurations are unchanged. As a result, overall percentage effectiveness decreases.
Portland and Maricopa County are an exception to the trend seen at other locations. In many cases there is
near 100% reduction under both current and future conditions due to the design criteria/goals for these
site-based scenarios.
116
-------
BMP Percent Reduction of Annual Runoff
100%
¦ Current ¦ Future
Figure 8-6. Best management practice (BMP) percentage reduction of annual runoff
under current and future climate conditions.
117
-------
BMP Percent Reduction of Sediment
100%
¦ Current ¦ Future
Figure 8-7. Best management practice (BMP) percentage reduction of sediment load
under current and future climate conditions.
118
-------
BMP Percent Reduction of TN
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
¦ Current ¦ Future
Figure 8-8. Best management practice (BMP) percentage reduction of total nitrogen
(TN) load under current and future climate conditions.
119
-------
BMP Percent Reduction of TP
100%
¦ Current ¦ Future
Figure 8-9. Best management practice (BMP) percentage reduction of total
phosphorous (TP) load under current and future climate conditions.
The difference in current to future percentage reduction ranges are shown in the bullets that follow. The
ranges suggest that site-scale future percentage reduction performance typically declines but not by a
large margin.
• 0.2 to -7.4% for annual runoff
• -0.1 to -9.6% for sediment
• 0 to -5.8% for TN
• 0 to-4.5 % for TP
The second set of figures shows unit-area volume (see Figure 8-10) and mass reductions (see Figure 8-11,
Figure 8-12, and Figure 8-13) from BMPs, in terms of feet/year and pounds/acre/year. The results were
normalized to site area to facilitate comparison between sites and regions. While overall percentage
reduction tends to decrease under future climate conditions (as seen in the previous set of figures), the
magnitude of the volume and sediment mass removal tends to increase. This means that while the BMPs
are removing a lower percentage of volume/mass, they do remove a greater quantity of volume/mass.
This is largely due to the increased volume/mass input, but also depends on assumptions about how
BMPs remove mass in the SUSTAIN configurations. The effect is less pronounced for TN and TP, and
not surprisingly volume and mass removal decreases at Portland where future runoff decreases.
120
-------
BMP Removal of Annual Runoff Volume
l Current ¦ Future
Figure 8-10. Normalized site-scale best management practice (BMP) removal of
annual runoff under current and future climate conditions.
BMP Removal of Sediment Mass
4500
4000
3500
3000
2500
2000
1500
1000
500
oS
¦ llfli
o°
* J
kO1
px 6 Jvpv (y . O
<^" V X <^" vX ^
/ / / ^
-------
Figure 8-11. Normalized site-scale best management practice (BMP) removal of
sediment load under current and future climate conditions.
BMP Removal of TN Mass
40
35
30
& 25
20
¦
~ IB
10
5
¦ - - ¦
¦
1
¦
¦
\ m¦ ¦
u
y ^ • ,/ s «¦
/-cF* r cF"
,0
-------
BMP Removal of TP Mass
5
¦ Current ¦ Future
Figure 8-13. Normalized site-scale best management practice (BMP) removal of
total phosphorous (TP) load under current and future climate conditions.
The difference in volume/mass removal from current to future climate conditions is relatively small, but
the difference in unit-area rates varies widely between regions/locations. This discrepancy is due to
regional differences in developed site loading rates and large storm event precipitation depths, as well as
variation in BMP perfonnance for various types of practices.
The third set of figures shows unit-area post-treatment site export of runoff (see Figure 8-14) and
pollutant mass (see Figure 8-15, Figure 8-16, and Figure 8-17). The results were nonnalized to site area to
facilitate comparison between sites and regions. Even though site practices remove more mass under
future conditions, the overall site export rates of volume/mass increases under future conditions. The
percentage increase in export load is very high for some scenarios (e.g., sediment mass export for Harford
County GI plus Gray nearly triples), but the absolute increase (future minus current) is fairly stable.
123
-------
Annual Runoff Depth Export (Site with BMPs)
3.5
3
2,5
k-
< 2
£
1.5
1
0.5
0
IdJl
.Cr cF -scF <£ <£ ,cr (£ . O
cS ^ . x oS . x ^ . x
& ^ J* ve> <£¦ ^ a
/ #
&
<*• ^
I Current ¦ Future
Figure 8-14. Normalized site-runoff export under current and future climate
conditions.
Sediment Annual Mass Export (Site with BMPs)
1600
1400
1200
1000
"if 800
&
600
400
200
0 ~
^ JS*
fy . ..X ^ <$¦* ^ (3-X
X5 0° ,c?9 o0<> c/ xo<-
-------
TN Annual Mass Export (Site with BMPs)
20
I Current ¦ Future
Figure 8-16. Normalized site-total nitrogen (TN) mass export under current and
future climate conditions.
125
-------
TP Annual Mass Export (Site with BMPs)
1.2
¦ Current ¦ Future
Figure 8-17. Normalized site-total phosphorous (TP) mass export under current and
future climate conditions.
8.1.2. Sensitivity Analyses to Precipitation Events
The sensitivity analyses focus on the most critical measure of site performance* overall volume and mass
export from the site. Sensitivity analyses were conducted using two different sets of variable future
climate conditions. The first set used a range of three large storm event intensity changes (low, medium,
and high) selected from 10 of the downscaled GCM climate scenarios used in the "20 Watersheds"
project, as discussed in Section 2.4.2. . SUSTAIN runs were performed for the Scott County stormwater
management scenarios and are shown below in Section 8.1.2.1. . The second set used a range of three
percentage changes in precipitation depths (-10, +10, and +20%) relative to current precipitation, as
discussed in Section 2.4.3. . SUSTAIN runs were performed for the Harford County and Scott County
stormwater management scenarios, and are shown below in Section 0.
For all of the results, the change in volume and mass between current and future climate conditions,
which provides a measure of resilience, is shown for each stormwater management scenario (rather than
showing current next to future as was done in Section 8.1.1.2. ). Results are grouped in the figures by
different approaches to stormwater management (i.e., Conventional vs. Gl-based) to facilitate
comparison. All of the results were normalized to site area to facilitate comparison between sites and
regions.
126
-------
Note that these results focus on how future climate conditions hypothetically affect performance of
currently implemented BMPs, and how resilient those BMPs are to changes in volume and intensity.
SUSTAIN optimizations for adapting practices to meet current measures were also performed for all of
the future climate scenarios in the sensitivity analyses and those results are presented in Section 8.2. .
8.1.2.1. Sensitivity Analysis* Modeled Scenarios
For change in runoff volume export, (see Figure 8-18) there is not much difference between the scenarios.
Change in sediment load export response (see Figure 8-19) is variable; the GI plus Gray stormwater
management scenario has the highest change across the three GCM intensities. For TN (see Figure 8-20),
the GI plus Gray and GI Only scenarios appear to be progressively more resilient, while for TP (see
Figure 8-21), the pattern vanes between the low, medium, and high intensity scenarios. These results
suggest there is no overall discernible pattern in degree of resiliency between the Conventional, GI plus
Gray, and GI Only scenarios when examining changes in site export across a range of intensity changes in
future precipitation using the downscaled GCM climate scenarios.
Change in Runoff Volume Export (Site with BMPs)
Medium
¦ Conventional «GI with Gray ¦GlOnly
Figure 8-18. Change in runoff volume export for Scott County between current and
future downscaled general circulation model (GCM) climate scenarios.
127
-------
Change in Sediment Load Export (Site with BMPs)
¦ Conventional ¦ Gl with Gray ¦ Gl Only
Figure 8-19. Change in sediment load export for Scott County between current and
future down scaled general circulation model (GCM) climate scenarios.
Change in TN Load Export (Site with BMPs)
Low Medium High
¦ Conventional ¦ Gl with Gray ¦ Gl Only
Figure 8-20. Change in total nitrogen (TN) load export for Scott County between
current and future downscaled general circulation model (GCM) climate scenarios.
128
-------
Change in TP Load Export (Site with BMPs)
0,15
0.1
0.05
ra
£ 0
-0,05
-0.1
Figure 8-21. Change in total phosphorous (TP) load export for Scott County
between current and future down scaled general circulation model (GCM) climate
scenarios.
8.1.2.2. Sensitivity Analysis® Percentage Change Scenarios
As noted previously, the sensitivity analysis using percentage change in precipitation was conducted for
stormwater management scenarios for Harford County and Scott County. Harford County results are
shown in Figure 8-22 (runoff volume), Figure 8-23 (sediment), Figure 8-24 (TN), and Figure 8-25 (TP).
For Harford County, the GI plus Gray scenario has a smaller change in export than the Conventional
scenario across the board for all parameters across the range of future climate percentage changes. This
suggests that the GI plus Gray stormwater management scenario is more resilient to changes in future
climate conditions than the Conventional scenario, at least when percentage-change future conditions are
modeled. Scott County results are shown in Figure 8-26 (runoff volume), Figure 8-27 (sediment), Figure
8-28 (TN), and Figure 8-29 (TP). For Scott County, the difference between the site-based approaches is
smaller than for Harford County, but the Conventional scenario tends to have the highest change in
export, the GI plus Gray scenario tends to be in the middle, and the GI Only scenario tends to be lowest.
An exception is sediment where GI plus Gray is the highest. Interestingly, the patterns for both locations
and all the parameters are carried through to the -10% future climate scenario. In other words, the
negative degree of change tends to be less for approaches using elements of GI. While the Conventional
scenarios have a greater decrease in runoff and loads (suggesting better performance), the Gl-based
scenarios show less change (i.e., greater resilience), which may actually be a benefit in cases where
downstream baseflow needs to be maintained.
The results of the analysis for the percentage change climate scenarios at both sites suggest that GI is
more resilient in terms of mitigating increases in runoff and loads. However, the same conclusion was not
reached for the Scott County analysis using a range of intensity changes among downscaled GCM climate
models. This difference suggests that results regarding resilience are sensitive to the assumptions used to
¦ Conventional ¦ GI with Gray ¦ GI Only
129
-------
generate future climate scenarios. It is important to note that these results reflect a limited set of locations
and site/BMP characteristics.
Change In Runoff Volume Export (Site with BMPs)
-0.2
Minus 10 Plus 10 Plus 20
¦ Conventional ¦ Gl with Gray
Figure 8-22. Change in runoff volume export for Harford County between current
and future percentage change climate scenarios.
Change in Sediment Load Export (Site with BMPs}
150
100
u
TO
o
-100
Minus 10 Plus 10
Plus 20
¦ Conventional ¦ Gl with Gray
Figure 8-23. Change in sediment load export for Harford County between current
and future percentage change climate scenarios.
130
-------
Change in TN Load Export (Site with BMPs)
-1
Minus 10 Plus 10 Plus 20
¦ Conventional BGI with Gray
Figure 8-24. Change in total nitrogen (TN) load export for Harford County between
current and future percentage change climate scenarios.
Change in TP Load Export (Site with BMPs)
-o.i
Minus 10 Plus 10' Plus 20
¦ Conventional ¦ Gl with Gray
Figure 8-25. Change in total phosphorous (TP) load export for Harford County
between current and future percentage change climate scenarios.
131
-------
Change in Runoff Volume Export (Site with BMPs)
0,25
0,2
0,15
0,1
> 0.05
0
nrfi
1^
-U.ib
Minus 10 Plus 10
Plus 20
¦ Conventional ¦ Gl with Gray ¦ Gl Only
Figure 8-26. Change in runoff volume export for Scott County between current and
future percentage change climate scenarios.
Change in Sediment Load Export (Site with BMPs)
-150
Minus 10 Plus 10 Plus 20
¦ Conventional ¦ Gl with Gray ¦ Gl Only
Figure 8-27. Change in sediment load export for Scott County between current and
future percentage change climate scenarios.
132
-------
Change in TN Load Export (Site with BMPs)
1.5
1
0.5
ra
£. 0
-0.5
-1
Minus 10 Plus 10 Plus 20
¦ Conventional ¦ Gl with Gray ¦ Gl Only
Figure 8-28. Change in total nitrogen (TN) load export for Scott County between
current and future percentage change climate scenarios.
0.25
0.2
0.15
0.1
0.05
0
-0.05
-0.1
-0.15
Change in TP Load Export (Site with BMPs}
Minus 10 Plus 101 Plus 20
¦ Conventional ¦ Gl with Gray ¦ Gl Only
Figure 8-29. Change in total phosphorous (TP) load export for Scott County
between current and future percentage change climate scenarios
133
-------
B.2. STUDY QUESTION #2
How can conventional designs and GI designs be adapted so that a site experiencing extreme
precipitation conditions in the future provides the same performance as the site under current
conditions?
For all sites, performance is evaluated at the site "outlet," defined as the point to which all areas, BMPs,
and conveyances ultimately drain. Therefore, the objective is to evaluate a site's performance as a whole
at meeting performance targets, rather than the performance of individual BMPs. For sites with multiple
BMPs, the goal of the adaptation simulation is then to determine the optimal combination of BMP areas
that result in the site as a whole meeting performance objectives, or "targets." Within the SUSTAIN
optimization framework, the site practices were modified under future climate conditions to achieve the
same or better performance as the current climate scenario. Modifications targeted resizing the water
quality treatment and peak flow control BMPs, which are the primary drivers controlling site
performance. The SUSTAIN model performed hundreds of, and in some cases over 1,000, separate
simulations with unique resized BMP configurations to find the optimum solution, which was defined as
a configuration meeting or exceeding all of the performance objectives simultaneously at the least cost.
First, results are provided showing the cost of adapting BMPs under future climate conditions to meet or
exceed performance metrics under current climate conditions. Next, limiting factors for the adaptation
runs are discussed.
8.2.1. Adapting Best Management Practices (BMPs) for Heavy
Precipitation
Results of the adaptation model runs are summarized in Table 8-2 for all combinations of sites,
stormwater management approaches, and climate scenarios. Note that climate scenarios resulting in a
reduction in all performance metrics are not presented because they already meet all of the objectives:
Current cost of stormwater infrastructure reflects the 20-year present value of the capital cost and O&M
for new development. Following the adaptation simulations under future climate conditions, the 20-year
present value was recalculated for resized/adapted BMPs. Note that the adapted cost reflects new
development cost (i.e., the cost for new construction of the adapted BMPs) rather than retrofit costs of
changing BMP configurations on an already developed site. The reason for using the same basis for
calculating costs is twofold; first, it allows the results to be more comparable, and second, retrofit cost
data tend to be highly variable and difficult to generalize from literature values. The current cost was
subtracted from the adapted total cost to obtain the adaptation cost increase. Both metrics were
normalized to contributing impervious area. The last column shows the percentage increase in cost for
adaptation relative to current cost.
134
-------
Table 8-2. Cost metrics for future climate and stormwater management scenarios
Cost metric
Adaptation
Stormwater
Current cost
cost increase
Climate
management
($/impervious
($/impervious
% increase in
Location
scenario
scenario
acre)
acre)
cost
Downscaled GCM high intensity climate scenarios
Maricopa
Downscaled
Conventional
599,248
255,095
43
County, AZ
GCM (high
intensity)
GI Only
497,924
293,403
59
Atlanta, GA
Conventional
767,699
494,727
64
GI with Gray
1,281,819
162,050
13
Portland, OR
GI Only
623,934
219,453
35
Harford
Conventional
408,415
497,355
122
County, MD
GI with Gray
396,483
537,965
136
Scott
Conventional
211,546
523,833
248
County, MN
GI with Gray
341,375
687,650
201
GI Only
590,973
550,952
93
Sensitivity analysis climate scenarios
Harford
Plus 10%
Conventional
408,415
206,662
51
County, MD
GI with Gray
396,483
83,849
21
Plus 20%
Conventional
408,415
373,822
92
GI with Gray
396,483
163,940
41
Scott
Downscaled
Conventional
211,546
412,610
195
County, MN
GCM (medium
intensity)
GI with Gray
341,375
170,608
50
GI Only
590,973
228,319
39
Plus 10%
Conventional
211,546
410,815
194
GI with Gray
341,375
221,875
65
GI Only
590,973
294,965
50
Plus 20%
Conventional
211,546
766,021
362
GI with Gray
341,375
444,388
130
GI Only
590,973
597,650
101
135
-------
Table 8 2. Cost metrics for future climate and stormwater management
scenarios (Continued)
Cost metric
Location
Climate
scenario
Stormwater
management
scenario
Current cost
($/impervious
acre)
Adaptation
cost increase
($/impervious
acre)
% increase in
cost
Conventional with distributed GI stormwater management scenarios
Harford
County, MD
Downscaled
GCM (high
intensity)
Conventional with
distributed GI
408,415
812,054
199
Plus 10%
408,415
124,540
30
Plus 20%
408,415
273,255
67
Scott
County, MN
Downscaled
GCM (medium
intensity)
211,546
114,250
54
Downscaled
GCM (high
intensity)
211,546
354,813
168
Plus 10%
211,546
109,375
52
Plus 20%
211,546
190,625
90
8.2.2. Limiting Factors for Adaptation Optimizations
During the SUSTAIN optimizations, which of the five performance measures were the most limiting
(i.e., hardest to achieve)? What does this suggest about adapting practices to future climate conditions? As
discussed in Section 2.2. , each stormwater management scenario was modified (via increase in current
practice sizes or addition of GI practices) so that the overall site stormwater performance was the same or
better under future climate than under current conditions. At most locations, annual runoff volume and
pollutant loads increased under future climate conditions, so SUSTAIN optimization found the most
cost-effective way to modify BMPs to return the site to current annual runoff volume and pollutant load
export values. In addition, the flow regime changed for the largest runoff values (corresponding to
flooding and downstream bankfull flows) between current and future conditions, so the SUSTAIN
optimization sought to minimize the difference across a range of flows between current and future
conditions as exhibited by FDCs. The goal of the optimizations was to meet all five metrics
simultaneously; the result is that the performance improvement "overshot" some of the metrics while
seeking to meet all of the metrics. When reviewing optimization, it became clear that certain metrics were
the most limiting (i.e., costliest to achieve). Table 8-3 provides a listing of the limiting metrics for each
location, stormwater management approach, and climate scenario (note that the minus 10% climate
scenarios and the low intensity downscaled GCM scenario for Scott County resulting in improvements in
all metrics, so those climate scenarios were not included in the BMP adaptation runs). In many cases, only
136
-------
one metric was the limiting factor, while in other cases, multiple metrics were limiting factors. Some
interesting findings are:
1. Meeting the FDC metric was the limiting or colimiting factor in over 80% of the optimization
runs. This indicates that control of flood event runoff volumes is generally the most difficult
objective to meet when adapting site BMPs to future climate conditions. Practices that can
address flood event volume control are, therefore, a critical component of adaptation to climate
change, assuming there is a substantial increase in large storm event intensity.
2. The Scott County Conventional stormwater management scenarios that focused on resizing
current practices were always limited by reduction of annual runoff volume. The reason is that the
site used a single practice* a wet detention pond* to meet all of the regulatory stormwater
requirements. Due to an assumption of poorly infiltrating soils (plus the need to maintain a
permanent pool), there was essentially no modeling of infiltration from the bottom of the pond.
The only mechanism for the pond to decrease annual runoff under future climate conditions was
evaporation from the pond surface. This required a large increase in pond surface area.
3. The Atlanta Conventional stormwater management scenario used two practices that were
assumed to be located underground and encased in concrete. As a result, no infiltration or
evaporation was modeled from the practices. This meant that there was no mechanism to address
increase in runoff volume. As a result, the runoff volume metric was excluded from the
adaptation analysis.
4. Optimization runs for the Atlanta Conventional stormwater management scenario resulted in
another interesting outcome* it was not possible (at least in the simulation as modeled) for the
practices to be resized to reduce TN export to the current metric. As a result, TN was also
excluded from the adaptation analysis. TN removal was modeled as a fixed percentage of runoff
filtering through an underground sand filter. Because the removal rate did not change because of
resizing the practice, the only way to increase removal was to limit large event bypass from the
sand filter* in other words, convert untreated bypass runoff to treated filtered runoff. A
simulation was performed in which the sand filter was tripled in size, leading to zero bypass and
100% treatment of the future runoff. Even so, the treated mass under future conditions exceeded
the sum of treated and bypass current mass.
137
-------
Table 8-3. Adaptation optimization limiting factors
Adaptation optimization limiting factor
Location
Climate
scenario
Stormwater management scenario
Annual
runoff
FDC
Sediment
TN
TP
Maricopa
Downscaled
Conventional1
X
County,
AZ
GCM (high
intensity)
GI Only1
X
Atlanta,
Downscaled
Conventional2
X
X
GA
GCM (high
intensity)
GI with Gray
X
Portland,
OR
Downscaled
GCM (high
intensity)
GI Only
X
X
Harford
Downscaled
Conventional
X
County,
MD
GCM (high
intensity)
GI with Gray
X
Conventional with Distributed GI
X
Plus 10%
Conventional
X
GI with Gray
X
Conventional with Distributed GI
X
Plus 20%
Conventional
X
GI with Gray
X
Conventional with Distributed GI
X
X
X
138
-------
Table 8 3. Adaptation optimization limiting factors (Continued)
riimatp
Adaptation optimization limiting factor
Location
X ' ¦¦¦¦¦til-V-
scenario
Stormwater management scenario
Annual
FDC
Sediment
TN
TP
Scott
Downscaled
Conventional
X
County,
MN
GCM (medium
intensity)
GI with Gray
X
X
GI Only
X
X
Conventional with Distributed GI
X
X
Downscaled
Conventional
X
X
GCM (high
intensity)
GI with Gray
X
X
GI Only
X
X
Conventional with Distributed GI
X
X
Plus 10%
Conventional
X
GI with Gray
X
X
GI Only
X
X
Conventional with Distributed GI
X
X
Plus 20%
Conventional
X
GI with Gray
X
X
GI Only
X
X
Conventional with Distributed GI
X
X
Objective was to achieve zero outflow.
Annual runoff and TN targets not met.
8.3. STUDY QUESTION #3
What do the results suggest regarding the adaptation potential of gray and green infrastructure
for increases in extreme precipitation events?
This question asks what bigger picture conclusions can be made regarding how adapting BMPs to climate
change differs between green and gray stormwater management approaches. Does one tend to cost more
than the other? Is one approach more adept at addressing particular performance metrics? These and other
questions are addressed in the sections that follow.
8.3.1. Stormwater Infrastructure Cost
How do current stormwater infrastructure costs and additional costs for adaption compare between
Conventional versus Gl-based stormwater management scenarios? What does this say about various
approaches to stormwater management? In the graphs that follow, stormwater infrastructure costs for
meeting current regulatory requirements are shown, along with the cost of adapting site BMPs to meet
139
-------
current performance metrics. The two costs (current cost and additional cost of adaptation) are shown in
different colors and stacked to provide total infrastructure cost. All of the stormwater management
scenarios where current practices were resized to adapt to future conditions are shown, including results
for the climate sensitivity analyses. (Results comparing resizing practices vs. adding distributed GI are
explored in the next subsection.) Figure 8-30 provides results for Portland, Maricopa County, Atlanta, and
Harford County, and Figure 8-31 shows all the results for Scott County.
In general, the original (current) cost of stormwater infrastructure using GI practices is more expensive on
a per impervious acre basis than the equivalent scenario using only conventional practices. The cost for
the GI Only scenario for Maricopa County is less, but that may be due in part to limited cost data for
representing an infiltration basin (used in the Conventional scenario) in an arid environment leading to an
overestimation of the current cost.
However, the cost of adaptation is frequently less for approaches using GI compared to the
Conventional-only approaches. For Maricopa County, the GI Only adaption cost is higher than for
Conventional, but the net cost (current plus adaptation) is less for the GI Only scenario than for the
Conventional scenario. For Atlanta, the adaptation cost is much less for the GI plus Gray approach,
although the combined cost is somewhat higher than the Conventional cost. For Harford County, the
adaptation and combined costs of GI plus Gray are lower than Conventional for the percentage change
future climate scenarios, but slightly higher for the downscaled GCM scenario using high storm event
intensity change. For Scott County, the GI plus Gray scenario has both the lower adaptation and
combined cost (compared to Conventional and GI Only) for the percentage change and medium intensity
downscaled GCM scenarios, but the trend is not held for the high intensity downscaled GCM scenario.
Combined costs are highest for the GI Only approach.
These results suggest two trends:
1. Approaches that use a combination of conventional and GI components tend to have greater cost
resiliency compared to approaches relying on only conventional or only GI practices* in other
words, the increase in cost of maintaining current performance under future climate is less than
for conventional-only or GI Only approaches, which indicates combined conventional/GI
approaches are better equipped (i.e., more resilient) for adaptation. This greater resilience likely
reflects the combined advantages of having practices that better address large flooding events
(such as wet ponds and detention basins) with GI practices that provide most holistic treatment of
volume and pollutants.
2. However, GI practices appear to be at a disadvantage in some cases when there is a large
projected increase in the most extreme precipitation events. The adaptation optimization forced
GI components (which tend to be more expensive on a unit basis) to be larger to provide
sufficient volume control for the highest runoff events to meet the FDC metric. Given the
importance of flood control for stormwater management, this outcome is realistic.
140
-------
$1,600,000
$1,400,000
$1,200,000
$1,000,000
$800,000
$600,000
$400,000
$200,000
$-
,ov
/V
^ J?
xS^ tp v"
* -/ /
. -v
& &
fU
£ ^
>Q A°
C V
<*> (S'
^ x
-------
8.3.2. Resizing Current Practices versus Adding Green Infrastructure
(Gl) to Site
Two different approaches to adapting a site under future climate to address increase in performance
measures were modeled for Harford County and Scott County Conventional stormwater management
scenarios: resizing currently implemented practices versus adding distributed GI to the site. How do the
results compare, and what do they say about the two different approaches? Adaptation optimizations were
performed for all the Harford County and Scott County stormwater management and climate scenarios. A
comparison of performance improvement is not relevant; the adaptation optimizations ensure that current
performance levels will be achieved regardless of the approach. What is more interesting is a comparison
of the adaptation costs.
Figure 8-32 provides the results using the same stacked-cost format from the previous subsection. The
trends shown in the results are consistent with those seen before. Adding distributed GI to a site to adapt
to climate change is generally less expensive than resizing conventional practices* again, the approach
that combines conventional and GI practices has the greatest resiliency. However, this is not the case for
Harford County for the high intensity downscaled GCM climate scenario. In this case, so much additional
volume control is needed that the higher cost of GI outstrips a simple resizing of the less expensive
extended detention basin. For Scott County, the distributed GI approach remains less expensive for the
high intensity downscaled GCM climate scenario, but this is driven in part by the large footprint needed
by the wet pond to provide sufficient evaporation to control the runoff volume increase. In addition, the
GI adaptation cost is highest among the future climate scenarios for the high intensity downscaled GCM.
142
-------
$1,400,000
$1,200,000
$1,000,000
$800,000
$600,000
$400,000
$200,000
$-
& &
C x° 4> >?
k
<^' &
& .a--
I Current Cost
I Adaptation Increase
Figure 8-32. Current cost and best management practice (BMP) adaptation cost for
Harford County and Scott County conventional stormwater management scenarios,
using different adaptation approaches.
8.3.3. Increase in Best Management Practice (BMP) Footprint and
Implications
To adapt stormwater BMPs to address climate impacts, practices were either resized or distributed GI was
added to the sites. BMPs take up physical space, and the SUSTAIN optimizations provided future
practice dimensions. How do changes in BMP footprints (relative to the entire site) compare between
scenarios and locations? Are the increases realistic? BMP footprints as a percentage of overall site area
are shown below for current and future adapted stormwater management scenarios. All climate scenarios
are shown, and adaptation results using distributed GI are provided as well. Due to the number of
stormwater management scenarios, the figures are broken into four groups: Portland, Maricopa County,
and Atlanta (see Figure 8-33); Harford County (see Figure 8-34); Scott County Conventional and GI plus
Gray (see Figure 8-35); and Scott County GI Only and Conventional plus Distributed GI (see Figure
8-36).
143
-------
Stormwater BMP Percent of Site Area
¦ Current ¦ Adapted
Figure 8-33. Percentage change in best management practice (BMP) footprint
between current and adapted future climate for Portland, Maricopa County, and
Atlanta stormwater management scenarios.
144
-------
Stormwater BMP Percent of Site Area
40%
35%
30%
25%
20%
15%
10%
5%
0%
J
0^
<6°°
-------
Stormwater BMP Percent of Site Area
20%
18%
¦ Current ¦ Adapted
Figure 8-35. Percentage change in best management practice (BMP) footprint
between current and adapted future climate for Scott County Conventional and
Green Infrastructure (GI) + Gray stormwater management scenarios.
146
-------
Stormwater BMP Percent of Site Area
20%
18%
16%
14%
12%
10%
8%
6%
4%
2%
0%
A0
^ it ^ .O
J Jul
&
•$
s?
A°
rZ>
-------
approaches is always associated with the high intensity downscaled GCM climate scenario. This
corresponds to the previous conclusion that large increases in the most intense rainfall leads to a larger
practice footprint to provide sufficient storm event volume control. This trend is less evident for Scott
County, where both the high intensity downscaled GCM and the plus 20% change climate scenarios show
the largest footprint increase.
Consideration should be given to how realistic it is to implement some of the adaptation scenarios. For
instance, the wet pond in the Scott County Conventional scenario must be tripled or even quadrupled in
size to provide sufficient control of runoff volume. While sufficient pervious surface is technically
available for expansion, it would be difficult if not impossible in practice to achieve this expansion given
the site layout. This suggests that distributed solutions are a better option when the alternative is to tear
out roads and properties. However, were there a large increase in high intensity storm events, as noted
previously, GI alone might not be suitable for peak flow volume control. In the Harford County
adaptation scenario using distributed GI, the infiltration trenches occupy nearly half of the available
remaining pervious area on the site.
Another factor to consider is how site design and stormwater management are typically conducted.
Development often maximizes the site footprint to meet the goals of the project, which are to make a
profit, or at least to use the available space. Stormwater management is generally minimized to just
comply with current regulation, and little if any thought is given to setting aside space for climate change
resiliency. As a result, if a site is to be adapted to future climate, other site uses (e.g., parking, amenities)
may need to be converted to stormwater management.
8.4. CONCLUSIONS
Model simulations in five study locations suggest potential ranges for altered total urban runoff and
pollutant loads under mid-century climate. Using climate scenarios with larger increases in storm event
precipitation intensity, the percentage increases in volumes and loads were generally between 1.5 and
26.7%.
For overall post-treatment site-scale performance, simulations using both conventional and GI BMP
scenarios generally remove more runoff volume and pollutant mass under future climate conditions
(increased precipitation and runoff) compared to current conditions. However, overall site export rates of
runoff volume and pollutant mass still increase (i.e., BMP does not remove 100% of the additional
runoff/pollutant load due to climate change) despite better volume/mass removal. Changes in large storm
event runoff (as indicated by comparison of FDCs) indicate that BMPs designed for current conditions
will likely fail to mitigate downstream increases in stormwater runoff and associated downstream channel
erosion and flooding impacts under projected future conditions. Thus, there is likely a need to adapt site
stormwater infrastructure to future climate conditions to protect downstream water resources.
Considering the adaptation of BMPs under future climate conditions to achieve the same or better
performance as seen under current climate, the model simulations show that the most difficult
performance measure to mitigate was usually control of large flooding event outflows. Because control of
flooding events is a ubiquitous requirement throughout the United States, currently built practices will
148
-------
need greater temporary volume storage and/or reconfiguration of outlet structures to mitigate flooding and
channel erosion risk in locations where the magnitude of extreme events is likely to increase. GI practices
that rely on treatment without volume storage will be at a disadvantage for climate change adaptation, but
approaches that rely only on adaptation of conventional practices may not have the flexibility to address
multiple performance objectives. For instance, the conventional practices for the Atlanta ultra-urban site
could not be adapted to address runoff volume increase or fully mitigate the increase in TN load.
Likewise, the stormwater wet pond for the Scott County residential site provided poor annual volume
reduction and thus was resized excessively in the adaptation scenarios to address annual runoff increases.
Comparing the current cost of stormwater management for new development between conventional and
Gl-based approaches, the conventional approaches tended to be more cost effective than their GI
counterparts. However, when climate scenarios with smaller increases in large storm event intensities are
considered, the additional cost of adapting sites using GI approaches tended to be less than adapting the
conventional-only approaches. Overall, approaches to stormwater management that combined both
conventional and GI elements tended to have the best combined cost resiliency. This was further reflected
in the stormwater management scenarios that added distributed GI to a conventional approach site versus
resizing the conventional practices. Again, the combination of conventional and GI practices had better
cost resiliency; however, the trend did not hold up for many of the climate scenarios with the highest
projected changes in intensities for large storm events. In these cases, GI was at a disadvantage for
providing temporary detention storage needed to mitigate flooding risk.
Projections of future seasonal average increases in air temperature are relatively consistent between
various climate models, but changes in precipitation regime are much more uncertain. There would be a
"regret cost" if practices were dramatically up-sized in anticipation of climate changes that did not
actually occur. GI may have an advantage in flexibility because it typically has a shorter design life before
rehabilitation is required, so it would be possible to commit less in the present and use a more incremental
approach as climate evolves.
An important issue to consider is the flexibility of different types of practices, regardless of whether gray
or green. On an already-developed site, it will likely be difficult to add more area or types of practices,
especially if all of the developable area is occupied. Adding dispersed GI may be considerably easier at a
later date than resizing hard structures. However, it may be possible to use the existing footprint of BMPs
and excavate them to provide more storage and treatment* something that is not explored in this study.
This option is less likely on sites with a low elevation gradient. Another option is to build flexibility into
site design, setting aside space for potential future BMP addition and/or expansion. Regardless of how it
is addressed, flexibility is a key factor to consider, especially because changing climates may result in
changes to the environment downstream of development sites, which could then lead to changes in policy
and management decisions.
149
-------
9. REFERENCES
Anandhi, A, Frei, A; Pierson, DC; Schneidennan, EM; Zion, MS; Lounsbury, D; Matonse, AH. (2011).
Examination of change factor methodologies for climate change impact assessment. Water Resour Res
47(W03501): doi:10.1029/2010WR009104.
Bicknell, BR; Imhoff, JC; Kittle, JL, Jr.; Jobes, TH; Donigian, AS, Jr. (2004). HSPF version 12 user's manual. Aqua
Terra Consultants, Mountain View, California.
Brown S; Cotton, M. (2011). Changes in soil properties and carbon content following compost application: results
of on-fann sampling. Comp Sci Util 19(2):87-96.
Carmen, N. B. 2015. Volume Reduction Provided by Eight Disconnected Downspouts in Durham, North Carolina
with and without Soil Amendments. MS thesis. North Carolina State University.
CNT (Center for Neighborhood Technology). (2014). Green values stonnwater toolbox: national stonnwater
management calculator, detailed cost sheet. Center for Neighborhood Technology. Accessed November 2017
http://greenvalues.cnt.org/national/cost detail.php.
CWP (Center for Watershed Protection). (2007). National pollutant removal performance database, version 3.
Center for Watershed Protection, Ellicott City, MD. http://www.stonnwaterok.net/CWP%20Documents/CWP-
07%20Natl%20Pollutant%20Removal%20Perform%20Database.pdf.
CWP/MDE (Center for Watershed Protection and Maryland Department of Enviromnent). (2000). 2000 Maryland
stonnwater design manual, volumes I & II. MDE, Baltimore, MD.
http://mde.marvland.gov/programs/Water/StonnwaterManagementProgram/Pages/stonnwater design.aspx.
City of Portland. (2008). Gateway green streets master plan. Bureau of Enviromnental Services, Portland, OR.
https://scholarsbank.uoregon.edu/xmlui/bitstream/liandle/1794/9784/Portland Gateway Green Streets 2008.pdf
:seauence=l.
Clapeyron, MC. (1834). Memoire sur la puissance motrice de la chaleur. J l'Ecole Polytech (in French) 23:153-190.
Clausius, R. (1850). Ueber die bewegende Kraft der Wanne und die Gesetze, welche sich daraus fur die Wannelelire
selbst ableiten lassen [On the motive power of heat and the laws which can be deduced therefrom regarding the
theory of heat]. Annal Physik (in Gennan) 155:500-524.
Eusufzai, MK; Fujii, K. (2012). Effect on organic matter amendment on hydraulic and pore characteristics of a clay
loam soil. Open J Soil Sci 2:372-381. http://www.SciRP.org/iournal/oiss.
Geosyntec Consultants and Wright Water Engineers. (2014). International Stonnwater Best Management Practices
(BMP) database pollutant category statistical summary report* solids, bacteria, nutrients, and metals. December
2014. http://www.bmpdatabase.org/index.htm.
Hamon, WR. (1961). Estimating potential evapotranspiration. J Hydraul Div (Proc Am Soc Civil Engin)
87:107-120.
Harrison, RB; Grey, MA; Henry, CL; Xue, D. (1997). Field test of compost amendment to reduce nutrient runoff.
Final Report. University of Washington, College of Forest Resources.
http://depts.wasliington.edu/esnn311/Winter%202016/Documents/05 FIELD%20TEST%200F%20COMPQST
%20AMENDMENT%20TO%20REDUCE%20NUTRIENT%20RUNQFF Harrison%20et%20al 1997.pdf
Hirscliman, D; Collins, K; Schueler, T. (2008). Technical memorandum: the runoff reduction method. Center for
Watershed Protection and Chesapeake Stonnwater Network, Ellicott City, MD.
http://www.vwnc.vt.edu/swc/documents/CWP TecliMemo VRRM 20080418.
150
-------
Impact Infrastructure, LLC and Stantec. (2014). AutoCASE Beta Testing Project. Evaluation of GI/LID Benefits in
the Pima County Enviromnent. http://webcms.pima.gov/UserFiles/Servers/Server_6/File/Goverrnnent/Flood
Control/FloodplainManagement/Low Impact Development/autocase-testing-final-report-2014071 l.pdf
Johnson, T; Butcher, J; Deb, D; Faizullabhoy, M: Hummel, P; Kittle, J; McGinnis, S; Mearns, LO; Nover, D;
Parker, A; Sarkar, S; Srinivasan, R; Tuppad, P; Warren, M; Weaver, C.; Witt, J. (2015). Modeling streamflow
and water quality sensitivity to climate change and urban development in 20 U.S. watersheds. J Am Water
Resourc Assoc 51(5):1321—1341. doi: 10.1111/1752-1688.12308.
http://onlinelibrarv.wilev.com/doi/10. Ill 1/1752-1688.12308/full.
King, D; Hagan, P. (2011). Costs of stormwater management practices in Maryland counties. Draft final report.
Prepared for Maryland Department of the Enviromnent Science Services Administration. Prepared by the
University of Maryland Center for Environmental Science. Technical Report Series Co. TS-626-11. Accessed
November 2014.
http://www.mde.state.md.us/programs/Water/TMDL/TMDLImplementation/Documents/King Hagan Stonnwat
er%20Cost%20Report%20to%20MDF. Fina1%20Draft 120ct2Q 11 .pdf.
Latshaw, K; Fitzgerald, J; Sutton, R. (2009). Analysis of green roof growing media porosity. RURALS: Rev
Undergrad Res Agricul Life Sci 4(1): Article 2.
http://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1034&context=rurals.
LIDC (Low Impact Development Center). (2005). Low impact development for big box retailers. Prepared for U.S.
Enviromnental Protection Agency, Office of Drinking Water by LIDC, Beltsville, MD.
http://www.envirothon.org/pdf/2012/08 LID big box retailers.pdf.
Maurer, EP; Brekke, L; Pruitt T; Duffy, PB. (2007), Fine-resolution climate projections enhance regional climate
change impact studies. Eos Trans. AGU, 88(47), 504.
Mearns, LO; Sain, S; Leung, R; Bukovsky, M; McGinnis, S; Biner, S; Caya, D; Arritt, RW; Gutowski, W; Takle, E;
Snyder, M; Jones, RG; Nunes, AMB; Tucker, S; Herzmann, D; McDaniel, L; Sloan, L. (2013). Climate change
projections of the North American Regional Climate Change Assessment Program (NARCCAP). Clim Change
120:965-975.
Mearns, LO, Gutowski, WJ; Jones, R; Leung, R; McGinnis, S; Nunes, A; Qian, Y. (2009). A regional climate
change assessment program for North America. EOS, Trans Am Geophys Union 90(8):311-312.
MPCA (Minnesota Pollution Control Agency). (2016). Minnesota stormwater manual: Cost-benefit considerations
for bioretention. Accessed January 2016. http://stormwater.pca.state.mn.us/index.php/Cost-
benefit considerations for bioretention.
Nakicenovic, N; Alcamo, J; Davis, G; et al. (2000). Special Report on Emissions Scenarios (SRES), A special report
of Working Group III of the Intergovernmental Panel on Climate Change. Cambridge University Press,
Cambridge, UK. https://www.ipcc.cli/pdf/special-reports/emissions scenarios.pdf.
Neitsch, SL; Arnold, JG; Kiniry, JR; Williams, JR. (2011). Soil and water assessment tool—theoretical
documentation, version 2009. Texas Water Resources Institute Technical Report No. 406. Texas A&M
University System, College Station, TX. http://swat.tamu.edu/media/99192/swat2009-theorv.pdf.
Palla, A; Berretta, C; Lanza, LG; La Barbera, P. (2008). Modelling storm water control operated by green roofs at
the urban catchment scale. Paper presented at the 11th International Conference on Urban Drainage, Edinburgh,
Scotland, UK, 2008.
https://web.sbe.hw.ac.uk/staffprofiles/bdgsa/llth International Conference on Urban Drainage CD/ICUD08/
pdfsZ245.pdf.
151
-------
Rossman, LA. (2010). Stormwater management model user's manual version 5.0. [EPA/600/R-05/040], U.S.
Enviromnental Protection Agency, Water Supply and Water Resources Division, National Risk Management
Research Laboratory. Cincinnati, OH. http://www.owp.csus.edu/LIDTool/Content/PDF/SWMM5Manual.pdf.
RS Means Online. (2016). Site work and landscape cost data online. The Gordian Group. Accessed January 2017.
http://www.rsmeansonline.com/.
Schneider, D. 2011. Quantifying Evapotranspiration from a Green Roof Analytically. CEE thesis, Villanova
University, Department of Civil and Enviromnental Engineering.
Tetra Tech. (2009). Loading simulation program in C++ (LSPC) version 3.1 user's manual. Fairfax, VA.
http://dpw.lacountv.gov/wmd/wimns/docs/LSPC-UserManual.pdf.
Tetra Tech. (2014). Memorandum: Missouri nutrient framework: stormwater BMP nutrients and cost. Prepared for
EPA.
USD A. (Department of Agriculture). (1972). National Engineering Handbook, Section 4, Hydrology, Chapter 16,
Hydrographs. Soil Conservation Service, Washington, D.C.
USDA (Department of Agriculture) (1986). Urban Hydrology for Small Watersheds. Technical Release 55. Natural
Resources Conservation Service, Washington, DC.
U.S. EPA (Enviromnental Protection Agency). (2008). National water program strategy: response to climate change.
[EPA 800-R-08-001]. Office of Water, Washington, DC.
http://www.allianceforwaterefficiencv.org/uploadedFiles/Resource Center/Librarv/water resources/National-
Water-Program-Strategv-Response-to-Climate%20Change.pdf.
U.S. EPA (Enviromnental Protection Agency). (2009). SUSTAIN~a framework for placement of best management
practices in urban watersheds to protect water quality. [EPA/600/R-09/095], National Risk Management
Research Laboratory, Cincinnati, OH.
U.S. EPA (Enviromnental Protection Agency). (2010). National water program strategy: response to climate change.
Key action update for 2010-2011. [EPA 800/R-10/002], Office of Water, Washington, DC.
U.S. EPA (Enviromnental Protection Agency). (2012). National water program 2012 strategy: response to climate
change. Office of Water, Washington, DC. https://www.epa.gov/sites/production/files/2015-
03/documents/epa 2012 climate water strategy full report final.pdf.
U.S. EPA (Enviromnental Protection Agency). (2013). Watershed modeling to assess the sensitivity of streamflow,
nutrient and sediment loads to potential climate change and urban development in 20 U.S. watersheds.
[EPA/600/R-12/058F], Office of Research and Development, National Center for Enviromnental Assessment,
Washington, DC. https://cfpub.epa.gov/ncea/global/recordisplav.cfm?deid=256912.
152
-------
APPENDIX A FLOW DURATION CURVES
Flow duration curves analyze the cumulative frequency of historic flow data over a specified period.
These curves relate flow values to the percentage of time a flow rate is equal to or exceeded by all the
values in the record. The use of "percentage of time" provides a uniform scale ranging between 0 and
100. Thus, the full range of flows is considered. Low flows are usually exceeded, while flood flows are
exceeded infrequently. Sometimes flow duration curve analyses consider a subset of the entire curve
between lower and upper bounds (as this study does).
A basic flow duration curve runs from high to low along the x-axis, as illustrated in Figure A-l. The
x-axis represents the duration amount, or "percentage of time," as in a cumulative frequency distribution.
The v-axis represents the flow value (e.g., cubic feet per second [cfs]) associated with that "percentage of
time" (or duration). The v-axis is generally shown on a log scale.
Salt Creek near Greenview, IL
Flow Duration Curve
USGSGage: 05582000
100000
10000
440 cfs
954 cfs
1000
Fbw
is
o
130 cfs
100
High
Flows
Moist
Conditions
Low
Flows
Dry
Conditions
100
Flow Duration Interval (%)
USGS Flow D at a 1, S04 square miles
Figure A-l. Example flow duration curve for a perennial stream using daily average
flow.
A-l
-------
Flow duration curves are sorted from the highest value to the lowest (see Figure A-2). Using this
convention, flow duration intervals are expressed as a percentage, with zero corresponding to the highest
stream discharge in the record (i.e., flood conditions) and 100 to the lowest (i.e., drought conditions).
Thus, a flow duration interval of 60 associated with a stream discharge of 440 cfs implies that 60% of all
observed daily average stream discharge values equal or exceed 440 cfs. The generalized formula for a
flow duration curve is as follows:
p = 100 x (M- [n + 1])
(A-l)
p = the probability that a given flow will be equaled or exceeded (percentage of time)
M= the ranked position of the observation (dimensionless)
n = the number of events for the period of record (dimensionless)
While flow duration curves shown in the example represents a perennial water body, they can be used for
any type of drainage with outflow. When applied to a development site, the flow is generally comprised
of surface runoff during storm events (assuming the site is small enough that there is no baseflow
emerging in the site drainage ways). An example of a flow duration curve representative of a
development site is shown in Figure A-2. The flow data were taken from one of the Harford County site
simulations discussed in the main report. Note that flow drops to the minimum on the scale below the
20th percentile (zero flow cannot be plotted on a log scale), which indicates that during the majority of the
time, there is no outflow from the site.
Flow duration curve for development site
1 nn
10
1
Sf o.i
5
£ 0.01
0.001
0.0001
0.00001
0'
%
20% 40% 60% 80% 100%
Figure A-2. Flow duration curve for development site.
A-2
-------
As discussed in the main document, the analysis makes extensive use of flow duration curves for the very
highest site outflows corresponding to flooding storm events. As a result, the flow duration curves shown
in the report and in APPENDIX B. focus on the very highest flows of interest at the far left side of the
curve. An example is shown in Figure A-3. Note the extremely small percentages on the x-axis of the
plot. The v-axis uses a standard scale rather than a log scale to facilitate comparison of flow duration
curves from the various simulation scenarios.
Hourly Outflow (cfs)
i-iNJUJ£>(-ncri""*JQQ
oooooooooo
b
Flow Duration Curve for extreme runoff events
10% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure A-3. Example flow duration curve showing only the most extreme runoff
values.
A-3
-------
APPENDIX B DETAILED RESULTS
This section provides detailed results of the simulation modeling. First, simulation results by site are
presented and focus on current versus future performance within each green or gray scenario. This
information is followed by sensitivity analysis results at two sites.
For all results presented below, the "current with BMPs" label reflects current climate conditions with
current best management practice (BMP) configurations. The "future with BMPs" label represents future
climate conditions with current BMP configurations. Finally, the "future, adapted BMPs" label reflects
future climate conditions with practices resized ("adapted," according to the results of the System for
Urban Stormwater Treatment and Analysis Integration [SUSTAIN] optimization) to maintain current
performance.
B.1. SIMULATION RESULTS BY SITE
B.1.1. Harford County, MD
B.1.1.1. Conventional (Gray) Infrastructure
As discussed in Section 4.2. of the report, the selected future climate scenario for Harford County, MD
resulted in an across the board increase in precipitation depth and intensity, with the largest storm events
having the largest increase in hourly precipitation depth. This trend translates into the increase in annual
runoff volume between the current and future climate scenarios seen in Figure B-l. This figure presents
the partitioning of runoff volume among the three runoff fate pathways: infiltration, evapotranspiration
(ET), and outflow for the Harford County Conventional (Gray) Infrastructure stormwater management
scenario under the simulated current and future climate conditions.
B-l
-------
Runoff Fate
u
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
47.28
48.16
58.36
¦ ET
1.42
1.58
4.10
¦ Outflow
60.45
68.01
11.74
18,27
5.54
Figure B-l. Current and future partitioning of runoff fate for the Harford County,
MD Conventional (Gray) Infrastructure scenario.
Under current climate with BMPs (third bar in Figure B-l), approximately 78% of annual runoff is lost to
infiltration and 2% to ET with the remaining 19% discharging from the site as outflow. Under future
climate with BMPs (fourth bar in Figure B-l), about 71% of runoff is lost to infiltration and 2% to ET. A
greater fraction of runoff volume (27%) is lost to outflow. Increases in rainfall volume and large storm
event intensity associated with projected climate change result in a greater runoff volume overall, leading
to a larger fraction of runoff being discharged rather than infiltrated. When the surface sand filter and
extended dry detention basin footprints are increased to adapt to future climate conditions (last bar in
Figure B-l), the increase in BMP surface area results in an increased fraction of runoff portioning to
infiltration (86%) as well as ET (6%). Only 8% of annual runoff is converted to outflow when BMPs are
adapted for future climate.
Annual average sediment loads for current and future climate conditions (see Figure B-2) exhibit similar
behavior to annual runoff volumes. This is not surprising because sediment load is closely tied to rainfall
intensity such that an increase in intensity, as is predicted for Harford County, will promote greater
sediment wash-off. Increased sediment concentrations combined with increased runoff volumes result in
increased sediment loads in future climate. In the current climate conditions, the combined influence of
the conventional site practices results in a 79% reduction in annual sediment load. In the future without
any modification, the reduction declines to 70%. The surface sand filters and extended dry detention basin
have been sized according to performance criteria that are based on current climate conditions, and the
reduction in performance demonstrates that the current sizing is not adequate to maintain performance m
B-2
-------
future climate. When the BMP footprints are increased for future climate adaptation, the annual sediment
load reduction improves to 90%.
Annual Average Sediment Load
30,000 27^2
25,000
I I
I 15,000
8,169
10,000
I I
J ^ Jp J ^
0
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-2. Current and future performance for annual average sediment load,
Harford County, MD Conventional (Gray) Infrastructure scenario.
Because nitrogen fate and transport in the environment is complex, a direct connection between annual
average total nitrogen (TN) load (pounds/year) and runoff volume cannot readily be made. However, the
key observations from Figure B-3 are that (1) annual average TN loads are predicted to increase under
future climate and (2) the Harford County conventional practices are highly effective at managing TN. In
the current climate conditions, the combination of the surface sand filters and extended dry detention
basin achieves an overall 88% load reduction for TN on an annual basis. Under future climate conditions,
TN load reduction decreases to 83%. With the future adapted BMP footprints, the annual TN load
reduction increases to nearly 95%.
B-3
-------
Annual Average TN Load
501.1
600
500
400
300
200
86.9
100 25.8
474.4
54.9
J
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-3. Current and future performance for annual average total nitrogen (TN)
load, Harford County, MD Conventional (Gray) Infrastructure scenario.
Figure B-4 shows that a small increase in annual average total phosphorous (TP) load (pounds/year) is
predicted under future climate compared to the current climate for the untreated Conventional (Gray) site.
The conventional practices are highly effective at reducing annual average TP load. Their combined TP
load reduction is greater than 93% in the current climate. In future climate without any resizing, the TP
load reduction decreases to 89%. With the increased adapted BMP footprints, the annual average TP load
reduction is improved to 97%.
Annual Average TP Load
100 88.37 90 49
90
80
70
60
50
40
30
20 | 6,47
10
2.75
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-4. Current and future performance for annual average total phosphorous
(TP) load, Harford County, MD Conventional (Gray) Infrastructure scenario.
B-4
-------
As discussed in Section 5.2.2. of the report, the approximate 2-year hourly flow based on the 30-year
hourly outflow record for the Harford County Conventional (Gray) Infrastructure scenario was used to
bound the flow duration curve (FDC) analysis. This is represented by the dashed vertical line in Figure
B-5. This figure presents the FDC results for the current and future BMP scenarios. The objective of the
BMP adaptation was to resize the surface sand filters and extended dry detention basin to minimize the
difference between the "current with BMPs" (blue line) and "future with BMPs" (gray line) FDCs from
the approximate 2-year hourly flow (lower bound) to the highest hourly peak flow. The future adapted
FDC (dashed red line) reflects the resulting increased (adapted) conventional BMP footprints.
Comparison of the current and future adapted FDCs suggests that the adapted BMPs are effective at
reducing the highest peak flows (upper end of the curve) and flows in the vicinity of the 2-year hourly
flow, with variable performance in between. Overall, the future adapted BMPs produce a flow duration
response that reasonably reproduces the current BMP performance.
40
35
30
u
I25
1 20
0
>-
1 15
o
X
10
5
0
O.OC
Current with BMPs
Future with BMPs — — Future, adapted BMPs
¦2-yr Hourly Flow
-
)0% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-5. Flow duration curve (FDC) evaluation for current and future climate,
Harford County, MD Conventional (Gray) Infrastructure scenario.
Maximum hourly peak flow (see Figure B-6) was not a performance measure for the adaptation exercise,
but results are provided for discussion. These results also provide additional insight into the FDC
evaluation in Figure B-5. The observed increase in maximum hourly peak flow between the current and
future untreated scenarios aligns well with expectations based on the current and future climate
comparisons provided in Section 4. of the report for Harford County. Projected increases in precipitation
intensity and depth translate into increases in the maximum peak flow leaving the site. In the
B-5
-------
Conventional (Gray) Infrastructure scenario, peak flow reduction is primarily provided by the extended
dry detention basin. Comparison of the "current with BMPs" and "future with BMPs" scenarios
demonstrates that without resizing, the BMPs are unable to mitigate the increase in peak flow between the
current and future climate. When the footprints are increased for future climate adaptation, the maximum
hourly peak flow is reduced below the "current with BMPs" scenario (17.70 cfs versus 22.50 cfs,
respectively). This impact is observed in Figure B-5 where the "current with BMPs" and "future, adapted
BMPs" curves diverge at the highest hourly outflow value (peak flow).
Max Peak Flow
58.18
60
50
40
•C 30
20
10
35.92
17.70
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-6. Current and future performance for maximum hourly peak flow,
Harford County, MD Conventional (Gray) Infrastructure scenario.
Figure B-7 provides an alternate means of comparing the performance of the current and future adapted
BMPs with respect to outflow volume. Examining these results on a monthly basis provides additional
insight into BMP behavior, and helps verify that they align with expectations given the climate scenario
comparisons provided in Section 4.2. of the report. The current and future monthly average precipitation
comparison for Harford County demonstrated that the increase in future precipitation compared to current
is greatest in July, September, November, and December, with September exhibiting overwhelmingly the
greatest increase of all months. These predicted increases in monthly precipitation result in the
corresponding increases in monthly outflow volume seen in Figure B-7. Comparison of the "current with
BMPs" and "future with BMPs" graphs suggests that prior to adaptation, the conventional BMPs are not
highly effective at mitigating increased runoff volumes under future climate conditions; in almost all
months, the future (not adapted) outflow volumes are higher than the current outflow volumes, indicating
a decrease in volume reduction effectiveness. With the adapted BMP sizes, monthly outflows are well
below the future baseline, and lower than the "current with BMPs" monthly outflows in every month but
September. The increased footprints of the adapted surface sand filters and extended detention basin
B-6
-------
provide increased surface area for infiltration and ET as well as greater storage volume to mitigate the
increase in future outflow volume.
Monthly Outflow Volume
§4
¦V 3
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
¦ Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-7. Current and future performance for monthly outflow volume, Harford
County, MD Conventional (Gray) Infrastructure scenario.
Table B-l summarizes the increases in BMP footprints for the Harford County Conventional (Gray)
Infrastructure scenario that would be required to maintain current performance under fixture climate
conditions. The current and adapted footprints are presented both in terms of actual square feet of practice
as well as a percentage of overall site area. The latter is provided as a means of comparing the current and
future adapted sizes relative to the site area (20 acres) for this particular development type (mixed use).
Table B-l. Comparison of current and future adapted best management practice
(BMP) footprints, Harford County, MD Conventional (Gray) Infrastructure
scenario
Storm water
Practice
Current
Future adapted
% increase in
footprint
management
scenario
Footprint
SF
Footprint as
% of site area
Footprint
SF
Footprint as
% of site area
Conventional
(Gray)
Infrastructure
Extended dry
detention basin
25,000
2.9%
81,250
9.3%
225%
Surface sand filters
10,119
1.2%
14,840
1.7%
47%
B-7
-------
B.1.1.2. Green Infrastructure (Gl) with Gray Infrastructure
In the Green Infrastructure (GI) with Gray Infrastructure stormwater management scenario, the
combination of green (infiltration basins, infiltration trenches, and permeable pavement) and gray (dry
detention basin) practices is highly effective at managing site runoff. Over 92% of annual runoff is
infiltrated, 3% is converted to ET, and the remaining fraction (4%) is converted to outflow under current
climate conditions. Under future climate, the effectiveness of the Green and Gray practices decreases,
with 88% of annual runoff being infiltrated, 3% converted to ET, and approximately 9% leaving the site
as outflow. With the increased BMP footprints in the future adapted scenario, the infiltration fraction is
increased to greater than 91% and ET to over 6%. Less than 3% of annual runoff volume is converted to
outflow.
Runoff Fate
u
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
53.52
57.49
59.54
¦ ET
1.93
2.19
4.19
¦ Outflow
57.99
65.36
2.54
5.67
1.62
Figure B-8. Current and future partitioning of runoff fate for the Harford County,
MD Green Infrastructure (GI) with Gray Infrastructure scenario.
The practices in the GI with Gray Infrastructure scenario are highly effective at managing sediment. On
an annual basis, the sediment load reduction for the site in the current climate is nearly 93%. Under future
climate prior to adaptation, annual sediment load reduction decreases to 83%. With the adapted BMP
footprints, future sediment load reduction improves to nearly 95%.
B-8
-------
Annual Average Sediment Load
30,000
25,000
20,000
15,000
10,000
5,000
27,303
4,590
1,554
1,453
Current Future Current Future with Future,
Untreated Untreated with BMPs 6MPs adapted
IB MPs
Figure B-9. Current and future performance for annual average sediment load,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure scenario.
The future adapted practices in the GI with Gray Infrastructure scenario are highly effective at treating
TN and mitigating the increased annual average TN load under future climate conditions. The current
climate TN load reduction for the site is approximately 97%, which decreases to 93% under future
climate. The adapted BMP footprints achieve an annual average TN load reduction of 98%.
Annual Average TN Load
483.1
>
.c
500
450
400
350
300
250
200
150
100
50
0
454.8
12.7
31.6
Current
Untreated
Future
Untreated
Current with Future with
BMPs BMPs
9.8
Future,
adapted
BMPs
Figure B-10. Current and future performance for annual average total nitrogen
(TN) load, Harford County, MD Green Infrastructure (GI) with Gray
Infrastructure scenario.
B-9
-------
The combination of Green and Gray practices is very effective at reducing annual average TP load.
Current performance achieves a reduction of 99% for the site, which decreases slightly to 98% before
adaptation in future climate conditions. Increasing the practice footprints for future adaptation improves
the site's annual average TP load reduction to nearly 99%.
Annual Average TP Load
0
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-ll. Current and future performance for annual average total phosphorous
(TP) load, Harford County, MD Green Infrastructure (GI) with Gray
Infrastructure scenario.
Figure B-12 presents the flow duration curves for the current, future, and future adapted BMP scenarios,
and demonstrates that the increased BMP footprints adapted for future climate achieve a very similar flow
duration response to the current climate BMP configurations in the evaluated range of flows.
B-10
-------
35
30
¦ST 25
H-
U
1 20
5
3
o 1C
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Flourly Flow
15
k.
3
O
X 10
c,
"" ~ ~** -»«
-s
~~
•
0
0.0(
D0% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-12. Flow duration curve (FDC) evaluation for current and future climate,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure scenario.
In the Harford County GI with Gray Infrastructure scenario, peak flow reduction functions are primarily
provided by the dry detention basin. Under the current climate, the conventional practices reduce the
maximum hourly peak flow for the site by nearly 50% (from 34 to 17 cfs). Under future climate
conditions, without any practice resizing, the maximum hourly peak flow reduction is 46%, with a
maximum hourly peak flow of 30.5 cfs. With adaptation, the practices are able to reduce the future hourly
peak flow to 14.3 cfs, which is lower than the current hourly peak flow of 17 cfs.
B-ll
-------
Max Peak Flow
56.38
0
Current Future Current with Future with Future,
Untreated Untreated 8MPs 8MPs adapted
BMPs
Figure B-13. Current and future performance for maximum hourly peak flow,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure scenario.
Comparison of the annual monthly outflow volumes for the GI with Gray Infrastructure stormwater
management scenario shows that the future adapted BMPs are highly effective at reducing monthly
outfall volumes from the site; volumes are lower in the "future, adapted BMPs" scenario than the "current
with BMPs" scenario for all months except September, which is the month in which the greatest increase
in future runoff volumes is predicted to occur.
Monthly Outflow Volume
3
2.5
-)
z
SI
+¦»
c
o
£ 1-5 J
u f
ro
1 a
1
0.5 /\ 1 k
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-14. Current and future performance for monthly outflow volume, Harford
County, MD Green Infrastructure (GI) with Gray Infrastructure scenario.
B-12
-------
Table B-2 summarizes the increases in BMP footprints for the Harford County GI with Gray
Infrastructure scenario that would be required to maintain current performance under future climate
conditions.
Table B-2. Comparison of current and future adapted best management practice
(BMP) footprints, Harford County, MD Green Infrastructure (GI) with Gray
Infrastructure scenario
Stormwater
Practice
Current
Future adapted
% increase in
footprint
management
scenario
Footprint
SF
Footprint as
% of site area
Footprint
SF
Footprint as %
of site area
GI with Gray
Infrastructure
Dry detention
basin
10,000
1.1%
23,000
2.6%
130%
306%
224%
0%
Infiltration
basin
12,858
1.5%
52,155
6.0%
Infiltration
trench
14,800
1.7%
47,954
5.5%
Permeable
pavement
201,242
23.1%
201,242
23.1%
B.1.1.3. Conventional (Gray) Infrastructure with Distributed Green Infrastructure
(GI)
The partitioning of runoff rate for the Conventional (Gray) Infrastructure with Distributed GI scenario is
identical to the Conventional (Gray) Infrastructure scenario for all climate scenarios except for "future,
adapted BMPs" (fifth bar in Figure B-15). This is the case for all results presented in this subsection.
Figure B-15 demonstrates that when distributed infiltration trenches are added to the site, without any
resizing of the conventional practices, over 95% of annual runoff volume is infiltrated. About 2% of
runoff volume is converted to ET, and the remaining fraction (3%) is outflow.
B-13
-------
Runoff Fate
u
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
47.28
48.16
64.99
¦ ET
1.42
1.58
1.09
¦ Outflow
60.45
68.01
11.74
18.27
1.92
Figure B-15. Current and future partitioning of runoff fate for the Harford County,
MD Conventional (Gray) Infrastructure with Distributed Green Infrastructure (GI)
scenario.
The addition of distributed infiltration trenches to the site results in an annual average sediment load
reduction that is greater than the "current with BMPs" scenario. With adaptation, the practices reduce the
annual average sediment load by greater than 96%. Comparatively, the load reduction for the "current
with BMPs" scenario is approximately 79%, and approximately 70% for the "future with BMPs" (not
adapted) scenario.
B-14
-------
Annual Average Sediment Load
30,000 2
25,000 2^am I
I I
I 15,000
8,169
10,000
I I 4,610
p p p | 983
0
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-16. Current and future performance for annual average sediment load,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) scenario.
As noted in the Conventional (Gray) Infrastructure scenario discussion, future climate conditions result in
a decrease in performance ("future with BMPs") for both annual average TN and TP load reduction
compared to current performance. The adaptation simulation results suggest that implementing distributed
infiltration trenches achieves a very high TN and TP load reduction on an annual basis. The "future,
adapted BMPs" TN load reduction is greater than 98%, and the TP load reduction is nearly 99%.
B-15
-------
Annual Average TN Load
600
500
400
300
200
100
501.1
474.4
86.9
54.9
J
8.9
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-17. Current and future performance for annual average total nitrogen
(TN) load, Harford County, MD Conventional (Gray) Infrastructure with
Distributed Green Infrastructure (GI) scenario.
Annual Average TP Load
100
90
80
70
60
50
40
30
20
10
88.37
90.49
6.47
10.13
0.94
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-18. Current and future performance for annual average total phosphorous
(TP) load, Harford County, MD Conventional (Gray) Infrastructure with
Distributed Green Infrastructure (GI) scenario.
The addition of distributed infiltration trenches ("future, adapted BMPs") results in a reasonably similar
flow duration response to the "current with BMPs" scenario within the evaluated range of flows. The
observed divergence between the two curves in the uppermost range of flows is most likely a result of
seasonal difference in storm event patterns between the current and future climate scenarios.
B-16
-------
40
35
30
£
u
i25
1 20
0
5-
1 15
o
X
10
5
0
o.ot
Current with BMPs
Future with BMPs — — Future, adapted BMPs
-2-yr Hourly Flow
\ * N
\ >»
-N.
N.
X
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-19. Flow duration curve (FDC) evaluation for current and future climate,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) scenario.
Figure B-20 demonstrates that the addition of distributed green practices (infiltration trenches) enables the
site to almost completely mitigate the increase in peak flow predicted under future climate. Maximum
hourly peak flow is reduced from approximately 36.0 cfs ("future with BMPs") to approximately 23.5 cfs
("future, adapted BMPs") when infiltration trenches are added to the site. The "future, adapted BMPs"
hourly peak flow is approximately 1 cfs higher than the "current with BMPs" hourly peak flow.
B-17
-------
Max Peak Flow
i
60
58.18
50
40
35^92
30
¦ J J
u
Current Future
Untreated Untreated
Current with Future with Future,
BMPs BMPs adapted
BMPs
Figure B-20. Current and future performance for maximum hourly peak flow,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) scenario.
Monthly outflow volumes are presented for the Conventional (Gray) Infrastructure with Distributed GI
stormwater management scenario in Figure B-21. These results demonstrate that there is almost no
outflow from the site when distributed green practices are added. With the addition of the infiltration
trenches, monthly outflow volume is lower for all months in the future adapted scenario compared to the
"current with BMPs" scenario, even in September, when the greatest increase in future outflow volume is
likely to occur.
B-18
-------
Monthly Outflow Volume
7
Jan
Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-21. Current and future performance for monthly outflow volume, Harford
County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) scenario.
Table B-3 summarizes the increases in BMP footprints for the Harford County Conventional (Gray)
Infrastructure with Distributed GI scenario that would be required to maintain current performance under
future climate conditions. The footprint of infiltration trenches required for adaptation would comprise
approximately 11% of the total site area.
Table B-3. Comparison of current and future adapted best management practice
(BMP) footprints, Harford County, MD Conventional (Gray) Infrastructure with
Distributed Green Infrastructure (GI) scenario
Practice
Current
Future adapted
% increase in
footprint
management
scenario
Footprint
SF.
Footprint as
% of site area
Footprint
SF
Footprint as %
of site area
Conventional
(Gray)
Infrastructure
with
Distributed
GI
Extended dry
detention basin
25,000
2.9%
1.2%
25,000
2.9%
0%
Surface sand filters
10,119
10,119
1.2%
0%
Distributed
infiltration trenches
0
0%
95,869
11.0%
-
B-19
-------
B.1.1.4. Cost Estimation
Table B-4 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for all three Harford County stormwater management scenarios. Refer to Section 6.5 of
the report for a discussion of how the infrastructure cost estimates were developed. Also provided are the
increase in cost, both in dollars and percentage, and the increase in cost per acre of site. These three
metrics represent three alternative methods for evaluating the cost of adaptation, which is effectively the
increase in cost between the current and future adapted climate scenarios.
Table B-4. Comparison of the current and future estimated 20-year present value
costs for the Harford County, MD stormwater management scenarios
Location
Stormwater
management
scenario
Current cost
20-yr present
value, Smillions
Future adapted
cost
20-yr present
value, Smillions
Increase in cost
20-yr present
value, Smillions
% increase
in cost
Increase per
acre of site
Smillions
Harford
County,
MD
Conventional
(Gray)
fnfrastructure
5.3 f
f f .78
6.47
122
0.32
Gf with Gray
fnfrastructure
5. f 5
f 2. f 5
6.99
136
0.35
Conventional
(Gray)
fnfrastructure
with
Distributed Gf
5.3 f
f 5.8
f0.56
199
0.53
For the Conventional (Gray) scenario, the cost of adaptation (based on 20-year present value) is estimated
to increase by $6.47 million, or 122% compared to the current cost. This is equivalent to a cost of
adaptation of $0.32 million per acre of site area.
The cost of adaptation for the GI with Gray scenario is estimated to be an increase of $6.99 million,
which reflects a 136% increase in cost. On a cost per site acre basis, the estimated cost of adaptation is
$0.35 million per acre of site area.
Implementing distributed green practices (infiltration trenches) to address the performance gap between
the current and future climate comes at an estimated cost increase of $10.56 million, an increase of 199%.
The increase in cost per acre of site is estimated to be $0.53 million for the Conventional with Distributed
GI scenario.
B-20
-------
B.1.2. Scott County, MN
B.1.2.1. Conventional (Gray) Infrastructure
Future Low Intensity
The precipitation analysis for the Scott County, MN future low intensity climate scenario predicted an
overall decrease in annual precipitation depth across the 30-year simulation. The monthly average
precipitation depth across the simulation period is predicted to increase in some months compared to the
current climate; however, on an annual basis, this is offset by the decreases in precipitation depth in other
months. Precipitation intensity is predicted to decrease throughout the entire 30-year assessment period,
except in the single highest hour of precipitation, resulting in a maximum hourly peak flow in the future
low intensity climate scenario that is higher than in the current climate (see Figure B-26). Comparison of
the monthly outflow volumes for the current climate and future low intensity climate scenarios
(see Figure B-27) shows higher monthly outflow volumes in the future for some months and lower for
others. The discrepancy is within ±1 acre-feet/month except in June and July, in which the future outflow
volumes decrease by approximately 1.1 acre-feet/month and 2.7 acre-feet/month, respectively. On an
annual basis, outflow is lower under the future low intensity climate scenario compared to the current
climate.
Due to the predicted decrease in precipitation depth and intensity, the overall runoff volume and outflow
volume also decrease between the current and future low intensity climate conditions (see Figure B-22).
The decrease in site outflow in the future low intensity climate scenario results in decreased sediment,
TN, and TP loads compared to current climate conditions (seeFigure B-23, Figure B-24, and Figure B-25,
respectively). Because there was no decrease in the performance of the conventional (gray) practice (wet
pond) between the current and future climate, this scenario was not investigated for adaptation.
B-21
-------
Runoff Fate
£
i
u
ro
35
30
25
20
15
10
5
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
¦ Infiltration
0.00
0.00
¦ ET
1.46
1.76
¦ Outflow
34.10
31.65
32.59
29.85
Figure B-22. Current and future partitioning of runoff fate for the Scott County,
MN Conventional (Gray) Infrastructure (low intensity) scenario.
Annual Average Sediment Load
38.376
>
_Q
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
7,370
4,460
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-23. Current and future performance for annual average sediment load,
Scott County, MN Conventional (Gray) Infrastructure (low intensity) scenario.
B-22
-------
Annual Average TN Load
0
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-24. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Conventional (Gray) Infrastructure (low intensity)
scenario.
Annual Average TP Load
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-25. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Conventional (Gray) Infrastructure (low intensity)
scenario.
B-23
-------
Max Peak Flow
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-26. Current and future performance for maximum hourly peak flow, Scott
County, MN Conventional (Gray) Infrastructure (low intensity) scenario.
Monthly Outflow Volume
£3
£
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs
Figure B-27. Current and future performance for monthly outflow volume, Scott
County, MN Conventional (Gray) Infrastructure (low intensity) scenario.
B-24
-------
Future Medium Intensity
In the Scott County, MN future medium intensity climate scenario, annual precipitation depth is predicted
to increase in all 30 years of the simulation period compared to the current climate. Monthly average
precipitation is predicted to increase or remain approximately the same for all months, and the intensity of
precipitation is predicted to increase slightly compared to the current climate. Figure B-28 indicates an
increase in outflow (32.59 cfs to 36.48 cfs) if the current conventional practice (wet pond) is not resized.
40
35
30
25
s-
d? 20
u
15
10
5
Runoff Fate
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
0.00
0.00
0.00
¦ ET
1.46
1.66
5.49
¦ Outflow
34.10
38.19
32.59
36.48
32.55
Figure B-28. Current and future partitioning of runoff fate for the Scott County,
MN Conventional (Gray) Infrastructure (medium intensity) scenario.
The adaptation simulation targeted increasing the wet pond footprint until future outflow was the same or
less than the current outflow. With future adaptation, the increased wet pond footprint enables outflow to
be reduced below the "current with BMPs" outflow due to a larger proportioning of runoff to ET.
Infiltration is not a runoff fate pathway for the wet pond in the Scott County Conventional (Gray)
Infrastructure scenario due to poorly infiltrating soils, as discussed in Chapter 4 of the report.
The current conventional practice (wet pond) achieves an annual average sediment load reduction of
nearly 81%. Without resizing, the performance under the future medium intensity climate scenario
declines slightly, with a 79% sediment load reduction. The future adapted wet pond footprint increases
the load reduction to 95% and maintains the future (medium intensity) annual sediment load
(2,296 pound/year) below the current climate load (7,370 pound/year). Because the practice resizing for
the future medium intensity climate was driven by the required reduction in outflow volume (see Figure
B-25
-------
B-28), the future adapted BMP annual average loads for sediment (see Figure B-29), TP (see Figure B-
30). and TN (see Figure B-31) are all well below the '"current with BMPs" loads.
Annual Average Sediment Load
50,000
45,000
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
47,146
9,883
7,370
m fl
2,296
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-29. Current and future performance for annual average sediment load,
Scott County, MN Conventional (Gray) Infrastructure (medium intensity) scenario.
Annual Average TN Load
350 298.6 317 2
300
250
- 200
- 150
100
50
136.4
I
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-30. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Conventional (Gray) Infrastructure (medium
intensity) scenario.
B-26
-------
Annual Average TP Load
47.49
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-31. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Conventional (Gray) Infrastructure (medium
intensity) scenario.
The TN load reductions achieved by the current, future (not adapted), and future adapted conventional
practice (wet pond) in the medium intensity climate scenario are 30, 29, and 57%, respectively.
The TP load reductions achieved by the current, future (not adapted), and future adapted conventional
practice (wet pond) in the medium intensity climate scenario are 50, 47, and 74%, respectively.
As discussed above, the increase in footprint of the wet pond for adaptation in the future medium intensity
climate scenario was primarily driven by the required reduction in outflow volume that would be
necessary to maintain the current climate outflow. As a result, the increased wet pond size produces a
"future, adapted BMPs" flow duration curve that is reduced well below the "current with BMPs" curve
for almost the entire range of flows evaluated. Although good for peak flow reduction, decreasing the
outflow to this extent could have implications for stream baseflow or other ecological considerations.
B-27
-------
80
70
60
J2
U
I50
1 40
0
>
1 30
O
X
20
10
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
¦2-yr Hourly Flow
V \
\
V
\
\
\ 3
>
t -
— —
-
D0% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-32. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Conventional (Gray) Infrastructure (medium intensity) scenario.
Figure B-33 indicates that the future adapted wet pond results in a reduction in maximum hourly peak
flow due to the increased sizing of the practice to maintain current performance for outflow.
Max Peak Flow
7^ m ;l89 74|5
11 1 J]
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-33. Current and future performance for maximum hourly peak flow, Scott
County, MN Conventional (Gray) Infrastructure (medium intensity) scenario.
80
70
60
50
•8 40
30
20
10
0
B-28
-------
The simulated future medium intensity climate condition is predicted to alter both the timing and
magnitude of outflow throughout the year. To some extent, the adapted conventional practice (wet pond)
is able to more closely reproduce monthly outflow volumes under the current climate. However, monthly
outflow volumes in 6 months (January. April, May, September, November, and December) are still higher
than in the current climate. On an annual basis, the future adapted practice produces a lower outflow
volume overall compared to the current practice.
Monthly Outflow Volume
6
0
Jan Feb Mar Apr Mav Jun Jul Aug Sep Oct Nov Dec
Current with BMPs "Future with 8MPs Future, adapted BMPs
Figure B-34. Current and future performance for monthly outflow volume, Scott
County, MN Conventional (Gray) Infrastructure (medium intensity) scenario.
Table B-5 summarizes the increases in BMP footprints for the Scott County Conventional (Gray)
Infrastructure (medium intensity) scenario that would be required to maintain current performance under
future climate conditions. The adapted wet pond is 3.3 times larger than the current wet pond.
B-29
-------
Table B-5. Comparison of current and future adapted best management practice
(BMP) footprints, Scott County, MN Conventional (Gray) Infrastructure (medium
intensity) scenario
Stormwater
management
scenario
Climate
scenario
Practice
Current
Future adapted
% increase
in
footprint
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint as
% of site
area
Conventional (Gray)
Infrastructure
Future medium
intensity
Wet pond
32,670
2.5%
107,811
8.3%
230%
Future High Intensity
In the future high intensity climate scenario, annual precipitation depth is predicted to increase in all years
across the 30-year simulation period. Monthly average precipitation is predicted to increase or remain
approximately the same as in the current climate, with the greatest increase concentrated in the summer
months (June, July, and August), in which most of the annual precipitation occurs due to summer storms
with relatively high intensity and depth. These trends result in the increase in runoff, and consequently
outflow, between the current and future high intensity scenarios. Increasing the wet pond footprint
increases the proportioning of runoff to ET, allowing the conventional site to maintain the current climate
outflow performance under future high intensity climate conditions.
Similar to the Conventional (Gray) Infrastructure medium intensity scenario, the adaptation simulation
targeted increasing the wet pond footprint until future outflow was the same or less than the current
outflow (see Table B-6). The increase in wet pond footprint required to maintain the current outflow in
future high intensity climate resulted in a BMP that is larger than would be required to maintain current
performance for the other measures (annual average load for sediment, TP, and TN), as seen in the figures
below. The adapted wet pond footprint reduces simulated pollutant loads (see Figure B-36, Figure B-37,
and Figure B-38) well below current performance.
B-30
-------
Runoff Fate
40
35
30
25
20
15
10
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
0.00
0.00
0.00
¦ ET
1.46
1.73
6.79
¦ Outflow
34,10
38.30
32.59
36.52
31.33
Figure B-35. Current and future partitioning of runoff fate for the Scott County,
MN Conventional (Gray) Infrastructure (high intensity) scenario.
Annual Average Sediment Load
48,603
50,
45,
40,
35,
30,
25,
20,
15,
10,
5,
.000
000
.000
000
,000
000
,000
000
,000
000
7,370
R
11,591
2,642
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-36. Current and future performance for annual average sediment load,
Scott County, MN Conventional (Gray) Infrastructure (high intensity) scenario.
B-31
-------
Annual Average TN Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-37. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Conventional (Gray) Infrastructure (high intensity)
scenario.
Annual Average TP Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-38. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Conventional (Gray) Infrastructure (high intensity)
scenario.
The flow duration curve analysis (see Figure B-39) suggests that the adapted wet pond is able to
reasonably match the outflow response of the current wet pond, with some deviation at the upper end due
to the practice's inability to fully mitigate the increase in maximum hourly peak flow (see Figure B-40)
under future high intensity climate conditions.
B-32
-------
Current with BMPs Future with BMPs — — Future, adapted BMPs 2-yr Hourly Flow
120
100
80
60
40
20
0
0.000% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-39. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Conventional (Gray) Infrastructure (high intensity) scenario.
Max Peak Flow
87.18
ilill
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-40. Current and future performance for maximum hourly peak flow, Scott
County, MN Conventional (Gray) Infrastructure (high intensity) scenario.
120
100
80
ti 60
40
20
0
B-33
-------
To some extent, the adapted conventional practice (wet pond) is able to more closely reproduce monthly
outflow volumes under the current climate. However, monthly outflow volumes are still higher in several
months under the future adapted scenario compared to the current performance. On an annual basis, the
increases are outweighed by the months in which there are decreases in outflow volume.
Monthly Outflow Volume
7
4
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs Future, adapted SMPs
Figure B-41. Current and future performance for monthly outflow volume, Scott
County, MN Conventional (Gray) Infrastructure (high intensity) scenario.
Table B-6 summarizes the increases in BMP footprints for the Scott County Conventional (Gray)
Infrastructure (high intensity) scenario that would be required to maintain current performance under
future climate conditions. The adapted wet pond is nearly four times larger than the current wet pond.
Table B-6. Comparison of current and future adapted best management practice
(BMP) footprints, Scott County, MN Conventional (Gray) Infrastructure (high
intensity) scenario
Current
Future adapted
Sto rm water
Footprint as
Footprint as
% increase
management
Climate
Footprint
% of site
% of site
in
scenario
scenario
Practice
SF
area
SF
area
footprint
Conventional (Gray)
Future high
Wet pond
32.670
2.5%
128,066
9.8%
292%
Infrastmcture
intensity
B-34
-------
B.1.2.2. Green Infrastructure (Gl) with Gray Infrastructure
Future Low Intensity
Due to the predicted decrease in precipitation depth and intensity in the future low intensity climate
scenario, the overall runoff volume and outflow volume decrease compared to the current climate
conditions (see Figure B-42). The decrease in site outflow in the future low intensity climate scenario
results in decreased sediment load, TN load, and TP load compared to the current climate conditions (see
Figure B-43, Figure B-44, and Figure B-45, respectively). Because there was no decrease in the
performance of the green (bioretention) and gray (dry detention basin) practices between the current and
future low intensity climate, this scenario was not investigated for adaptation.
Runoff Fate
40
35
30
25
20
15
10
5
0
I Infiltration
IET
I Outflow
Current
Untreated
35.68
Future
Untreated
33.16
Current with
BMPs
7.45
2.32
25.90
Future with
BMPs
7.13
2.60
23.44
Figure B-42. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) with Gray Infrastructure (low intensity) scenario.
B-35
-------
Annual Average Sediment Load
39J78
40,000 ^ 33
35,000
30,000
25,000
20,000 I 14^95
io^w
10,000
V V BP v
0
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-43. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (low
intensity) scenario.
Annual Average TN Load
0
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-44. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) with Gray Infrastructure
(low intensity) scenario.
B-36
-------
Annual Average TP Load
43.08
45
40
35
30
> 25
£ 20
15
10
5
0
40.97
14.84
12.98
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Figure B-45. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Green Infrastructure (GI) with Gray Infrastructure
(low intensity) scenario.
80
70
60
50
•S 40
30
20
10
0
Max Peak Flow
54.10
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-46. Current and future performance for maximum hourly peak flow, Scott
County, MN Green Infrastructure (GI) with Gray Infrastructure (low intensity)
scenario.
B-37
-------
-t2
£
Monthly Outflow Volume
Jan Feb Mar Apr May Jun
Current with BMPs
Jul Aug Sep Oct Nov Dec
Future with BMPs
Figure B-47. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) with Gray Infrastructure (low intensity)
scenario.
Future Medium Intensity
The predicted increase in annual precipitation depth and intensity in the Scott County future medium
intensity climate scenario results in an increase in outflow between the current climate (25.90 cfs) and
future climate (29.85 cfs) when the Green and Gray practices are not resized for adaptation.
With future adaptation, the increased green (bioretention) and gray (dry detention basin) practice
footprints reduce the volume of outflow below the "current with BMPs" scenario due to a larger
proportioning of runoff to ET and infiltration. As discussed in Chapter 4 of the report, the soils in the
Scott County study site have low infiltration capacity, so outflow remains the dominant runoff pathway,
even when practice sizes are increased.
B-38
-------
Runoff Fate
i
u
ro
40
35
30
25
20
15
10
5
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
7.45
7.54
10.53
¦ ET
2.32
2.57
3.87
¦ Outflow
35.68
39.95
25.90
29.85
25.56
Figure B-48. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) with Gray Infrastructure (medium intensity)
scenario.
The future adapted Green and Gray practices achieve an annual average sediment load that is lower than
the current climate load. The corresponding sediment load reductions for the current, future, and future
adapted BMP scenarios are 63, 60, and 70%, respectively.
B-39
-------
Annual Average Sediment Load
48,700
50,000
45,000
40,000
35,000
30,000
25,000 19i65^
20,000 14/~~
15,000
10,000
5,000
0
14,895 14,458
sis
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-49. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (medium
intensity) scenario.
The future adapted bioretention and dry detention basin combine to reduce annual average TN load by
70%. This is greater than the current and future (not adapted) reductions of 65 and 63%, respectively.
Annual Average TN Load
332.6
350
300
250
- 200
^ 150 109.1 ^3 gg4
100
50
312.9
fllft
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-50. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) with Gray Infrastructure
(medium intensity) scenario.
B-40
-------
The future adapted Green and Gray practices achieve an annual average TP load that is lower than the
current climate load. The corresponding TP load reductions for the current, future, and future adapted
BMP scenarios are 66, 62, and 70%, respectively.
Annual Average TP Load
48.67
50
45
40
35
30
25
20
15
10
18.55
14.84 14.70
¦ la
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-51. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Green Infrastructure (GI) with Gray Infrastructure
(medium intensity) scenario.
The flow duration curve analysis (see Figure B-52) for the Scott County GI with Gray Infrastructure
(medium intensity) scenario suggests that the adapted practices are able to reasonably match the outflow
response of the current practices within the evaluated range of flows.
B-41
-------
80
70
60
£
u
~ so
140
0
5-
1 30
O
X
20
10
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr H0urly Flow
%
\ \
v\
\ \
\ \
^—,
1
1
1
D0% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-52. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (medium
intensity) scenario.
Although maintaining maximum hourly peak flow at current performance is not an objective of the
adaptation simulation, Figure B-53 indicates that the adapted Green and Gray practices are able to reduce
hourly peak flow somewhat compared to the future (not adapted) practice sizing.
B-42
-------
Max Peak Flow
66
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-53. Current and future performance for maximum hourly peak flow, Scott
County, MN Green Infrastructure (GI) with Gray Infrastructure (medium
intensity) scenario.
Figure B-54 suggests that the adapted bioretention and dry detention basin practices are able to more
closely match the monthly outflow volumes under the current climate, although outflow is greater in
several months, particularly in May, November, and December.
B-43
-------
Monthly Outflow Volume
5
4
3.5
2.5
1.5
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-54. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) with Gray Infrastructure (medium
intensity) scenario.
Table B-7 summarizes the increases in BMP footprints for the Scott County GI with Gray Infrastructure
(medium intensity) scenario that would be required to maintain current performance under future climate
conditions.
Table B-7. Comparison of current and future adapted best management practice
(BMP) footprints, Scott County, MN Green Infrastructure (GI) with Gray
Infrastructure (medium intensity) scenario
Stormwater
management
scenario
Climate
scenario
Practice
Current
Future adapted
% increase
in
footprint
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint
as % of
site area
GI with Gray
Infrastnicture
Future medium
intensity
Bioretention
34,848
2.7%
58,848
4.5%
69%
Diy detention
basin
26,136
2.0%
32,336
2.5%
24%
Future High Intensity
The increase in precipitation depth and intensity in the Scott County future high intensity climate scenario
results in the observed increase in runoff, and consequently outflow, compared to the current climate. The
B-44
-------
comparison of the "current with BMPs" and "future with BMPs" runoff fates indicates that the current
practice sizing in the GI with Gray Infrastructure scenario is not sufficient to mitigate the increase in
runoff volume, with a greater fraction of runoff partitioning to outflow.
Increasing the green (bioretention) and gray (dry detention basin) practice footprints increases the
proportioning of runoff to infiltration and ET, allowing the site to maintain the current climate outflow
performance under future high intensity climate conditions. However, as discussed above, due to the poor
infiltration capacity of the soils in the Scott County study site, outflow remains the dominant runoff fate
pathway.
Runoff Fate
45
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
7.45
7.31
13.11
¦ ET
2.32
2.67
8.08
¦ Outflow
35.68
40.06
25.90
30.08
18.89
Figure B-55. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) with Gray Infrastructure (high intensity) scenario.
The following figures demonstrate that the adapted green (bioretention) and gray (dry detention basin)
practices in the Scott County GI with Gray Infrastructure scenario are able to mitigate the increases in
annual average sediment (see Figure B-56), TN (see Figure B-57), and TP (see Figure B-58) load under
future high intensity climate conditions.
B-45
-------
Annual Average Sediment Load
50,187
60,000
50,000
40,000
30,000 21,612
20,000 1 14^895
10,000
0
14,895 13,186
mum
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-56. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (high
intensity) scenario.
Annual Average TN Load
350
300
250
200
- 150
100
50
329.1
312.9
1
I
109.1 12^1
\ ¦ 1
82.1
1 ¦
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-57. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) with Gray Infrastructure
(high intensity) scenario.
B-46
-------
Annual Average TP Load
50
45
40
35
30
25
20
15
10
46.41
17.82
11.85
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-58. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Green Infrastructure (GI) with Gray Infrastructure
(high intensity) scenario.
Figure B-59 presents the flow duration curves for the current, future, and future adapted BMP scenarios,
and demonstrates that the increased BMP footprints adapted for future high intensity climate achieve a
very similar flow duration response to the current climate BMP configurations in the evaluated range of
flows.
B-47
-------
120
100
'S 80
1
1 60
O
_>
o 40
X
20
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Hourly Flow
\X X
vV
— —
-
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-59. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (high
intensity) scenario.
As was observed in the FDC comparison (see Figure B-59), the increased bioretention and dry detention
basin footprints are able to mitigate the increase in maximum hourly peak flow between the current and
future high intensity climate conditions.
B-48
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-60. Current and future performance for maximum hourly peak flow, Scott
County, MN Green Infrastructure (GI) with Gray Infrastructure (high intensity)
scenario.
Comparison of the monthly outflow volume for the current and future BMP scenarios indicates that the
increased practice sizes are very effective at reducing monthly outflow volume compared to the future
(not adapted) practices. Monthly outflow volumes with the future adapted practices are very similar to, or
lower than, the current climate outflows for all months except July.
B-49
-------
Monthly Outflow Volume
7
6
4
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-61. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) with Gray Infrastructure (high intensity)
scenario.
Table B-8 summarizes the increases in BMP footprints for the Scott County GI with Gray Infrastructure
(high intensity) scenario that would be required to maintain current performance under future climate
conditions. The adapted bioretention footprint is approximately 2.7 times larger than the current footprint,
and the adapted dry detention basin footprint is approximately 4.7 times larger than the current footprint.
Table B-8. Comparison of current and future adapted best management practice
(BMP) footprints, Scott County, MN Green Infrastructure (GI) with Gray
Infrastructure (high intensity) scenario
Stormwater
management
scenario
Climate
scenario
Practice
Current
Future adapted
Increase
in
footprint
Footprint
SF
Footprint as
% of site area
Footprint
SF
Footprint as
% of site
area
GI with Gray
Infrastructure
Future high
intensity
Bioretention
34,848
2.7
93,286
7.1
168
Dry detention
basin
26,136
2.0
123,136
9.4
371
B-50
-------
B.1.2.3. Green Infrastructure (Gl) Only
Future Low Intensity
Due to the predicted decrease in precipitation depth and intensity in the future low intensity climate
scenario, the overall runoff volume and outflow volume decrease compared to the current climate
conditions (see Figure B-62). The decrease in site outflow in the future low intensity climate scenario
results in decreased sediment load, TN load, and TP load compared to the current climate conditions (see
Figure B-63, Figure B-64, and Figure B-65, respectively). Because there was no decrease in the
performance of the GI only practices between current and future low intensity climate, this scenario was
not investigated for adaptation.
Runoff Fate
35
30
25
20
15
10
5
0
I Infiltration
IET
I Outflow
Current
Untreated
34.78
Future
Untreated
32.31
Current with
BMPs
13.54
1.95
19.28
Future with
BMPs
12.92
2.25
17.14
Figure B-62. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) Only (low intensity) scenario.
B-51
-------
Annua! Average Sediment Load
38.990
40,000
35,000
30,000
25,000
I—
> 20,000
_Q '
15,000
10,000
5,000
0
11.914
Current Future Current with
Untreated Untreated BMPs
7,855
Future with
BMPs
Figure B-63. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) Only (low intensity) scenario.
Annual Average TN Load
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-64. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) Only (low intensity)
scenario.
B-52
-------
Annual Average TP Load
Current with Future with
BMPs BMPs
Current
Untreated
Future
Untreated
Figure B-65. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Green Infrastructure (GI) Only (low intensity)
scenario.
Max Peak Flow
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-66. Current and future performance for maximum hourly peak flow, Scott
County, MN Green Infrastructure (GI) Only (low intensity) scenario.
B-53
-------
Monthly Outflow Volume
4
3.5
3
2.5
2
£
ra 1.5
1
0.5
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs
Figure B-67. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) Only (low intensity) scenario.
Future Medium Intensity
The predicted increase in annual precipitation depth and intensity in the Scott County future medium
intensity climate scenario results in an increase in outflow between the current (19.28 cfs) and future
(22.81 cfs) climate when the green practices (bioretention, rooftop downspout disconnection, and
permeable pavement) are not resized for adaptation.
With future adaptation, the increased practice footprints reduce the volume of outflow below the "current
with BMPs" scenario due to a larger proportioning of runoff to ET and infiltration. As discussed in
Chapter 4 of the report, the soils in the Scott County study site have low infiltration capacity, so outflow
remains the dominant runoff pathway, even for this GI Only scenario and even when practice sizes are
increased.
B-54
-------
Runoff Fate
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
13,54 13.97 16.74
1.95 2.18 3.54
34.78 38.96 19.28 22.81 18.68
Figure B-68. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) Only (medium intensity) scenario.
The following figures demonstrate that the adapted green practices in the Scott County GI Only scenario
are able to mitigate the increases in annual average sediment (see Figure B-69). TN (see Figure B-70),
and TP (see Figure B-71) load under future high intensity climate conditions.
40
35
30
1_
25
>
£
20
6
ro
15
10
5
0
¦ Infiltration
¦ ET
¦ Outflow
B-55
-------
Annual Average Sediment Load
47,832
50,000
45,000
40,000
35,000
30,000
¦ 16'148
20,000 ^ 914
15,000
10,000
5,000
0
11,914 11,358
lia
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-69. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) Only (medium intensity) scenario.
Annual Average TN Load
350
300
250
- 200
^ 150
100
50
324.4
305.1
71.6
82.3
62.5
I 1
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
Current Future
Untreated Untreated
Figure B-70. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) Only (medium intensity)
scenario.
B-56
-------
Annual Average TP Load
48.12
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-71. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Green Infrastructure (GI) Only (medium intensity)
scenario.
The flow duration curve analysis (see Figure B-72) for the Scott County GI Only (medium intensity)
scenario suggests that the adapted practices are able to reasonably match the outflow response of the
current practices within the evaluated range of flows.
13-57
-------
80
70
60
£
u
~ so
140
0
5-
1 30
O
X
20
10
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr H0urly Flow
\\
\ ^
\
* ** -
D0% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-72. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Green Infrastructure (GI) Only (medium intensity) scenario.
Figure B-73 indicates that, although maximum hourly peak flow was not targeted as part of the adaptation
simulation, the green practices, when adapted to meet the other performance measures, are only able to
reduce the hourly peak flow by about 0.5 cfs. This result may suggest a lower ability of GI to mitigate
hourly peak flows compared to Conventional (Gray) Infrastructure (which is typically designed
specifically to address peak matching requirements).
B-58
-------
Max Peak Flow
66
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-73. Current and future performance for maximum hourly peak flow, Scott
County, MN Green Infrastructure (GI) Only (medium intensity) scenario.
Figure B-74 indicates the adapted green practices for the Scott County future medium intensity climate
scenario are able to achieve a reasonable match to the current climate monthly outflow volumes. Monthly
outflow volumes with the future adapted practices are very similar to, or lower than the current climate
outflows for all months except May, November, and December.
B-59
-------
Monthly Outflow Volume
4
3.5
3
2.5
2
£
ra 1.5
1
0.5
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-74. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) Only (medium intensity) scenario.
Table B-9 summarizes the increases in BMP footprints for the Scott County GI Only (medium intensity)
scenario that would be required to maintain current performance under future climate conditions. Only
bioretention was modified for future climate adaptation. Permeable pavement and rooftop downspout
disconnection were not modified for two reasons: (1) permeable pavement is already implemented in
100% of sidewalk areas, and its expansion to include residential driveways and streets was ruled
impractical due primarily to maintenance concerns; and (2) impervious surface disconnection is already
implemented to the maximum extent practicable in this scenario for residential rooftops, and
disconnection of additional impervious surface is not considered feasible.
B-60
-------
Table B-9. Comparison of current and future adapted best management practice
(BMP) footprints, Scott County, MN Green Infrastructure (GI) Only (medium
intensity) scenario
Stormwater
management
scenario
Climate
scenario
Practice
Current
Future adapted
%
increase in
footprint
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint as
% of site
area
GI Only
Future
medium
intensity
Bioretention
(modified)
43,275
3.3
71,675
5.5
66
0
0
Rooftop downspout
disconnection
94,901
7.3
94,901
7.3
Permeable pavement
39,390
3.0
39,390
3.0
Future High Intensity
The increase in precipitation depth and intensity in the Scott County future high intensity climate scenario
results in the observed increase in runoff, and consequently outflow, compared to the current climate. The
comparison of the "current with BMPs" and "future with BMPs" runoff fates indicates that the current
practice sizing in the GI Only scenario is not sufficient to mitigate the increase in runoff volume, with a
greater fraction of runoff partitioning to outflow.
Increasing the green practice footprints increases the proportioning of runoff to infiltration and ET,
allowing the site to maintain the current climate outflow performance under future high intensity climate
conditions. As discussed above, due to the poor infiltration capacity of the soils in the Scott County study
site, outflow is an important runoff fate pathway. However, the increase green practice footprints in the
future adapted scenario increase the proportioning of runoff to infiltration enough to make it the dominant
runoff fate pathway in future high intensity climate.
B-61
-------
Runoff Fate
40
35
30
25
<:
£ 20
u
15
10
u
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
13.54
13.70
18.11
¦ ET
1.95
2.27
5.63
¦ Outflow
34.78
39.06
19.28
23.09
15.32
Figure B-75. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) Only (high intensity) scenario.
Figure B-76, Figure B-77, and Figure B-78 demonstrate that with the increased practice sizes, the GI
Only site is able to mitigate the increased sediment, TN, and TP loads due to fixture climate impacts.
Annual Average Sediment Load
49,300
50,000
45,000
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
18,023
11,914 fl 11,169
B IS
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-76. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) Only (high intensity) scenario.
B-62
-------
Annual Average TN Load
350
300
250
- 200
- 150
100
50
321.0
305.1
83.1
1
1
52.3
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-77. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) Only (high intensity)
scenario.
Annual Average TP Load
50 45"82
45
40
35
30
25
20 I 13.85
15 | 9.02
10
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-78. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Green Infrastructure (GI) Only (high intensity)
scenario.
The FDC evaluation for the Scott County GI Only (high intensity) scenario indicates that the green
practices alone are able to achieve a reasonably close flow response to the current climate FDC within the
evaluated range of flows. However, there is a discrepancy between the highest hourly peak flows,
indicating the adapted practices are unable to mitigate the increase in the highest flows due to climate
change (high intensity).
B-63
-------
120
100
'S 80
1
1 60
O
_>
o 40
X
20
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Hourly Flow
\ \
\
\
s.
-
_
— —
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-79. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Green Infrastructure (GI) Only (high intensity) scenario.
Figure B-80 provides additional insight into the behavior seen in the uppermost range of flows in the flow
duration curve analysis. The GI Only practices, even with adaptation, do not significantly reduce the
maximum hourly peak flow under the future climate (high intensity) compared to the original practice
sizes ("future with BMPs").
B-64
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-80. Current and future performance for maximum hourly peak flow, Scott
County, MN Green Infrastructure (GI) Only (high intensity) scenario.
Figure B-81 indicates that with resizing, die future adapted GI Only practices are successful at mitigating
increased monthly outflow volumes under future climate. The future adapted monthly outflows are lower
than, or very close to, the current monthly outflows for all months except July.
Monthly Outflow Volume
o
J 3
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-81. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) Only (high intensity) scenario.
B-65
-------
Table B-10 summarizes the increases in BMP footprints for the Scott County GI Only (high intensity)
scenario that would be required to maintain current performance under future climate conditions. As
discussed above, rooftop downspout disconnection and permeable pavement were not selected for
adaptation in the Scott County GI Only scenarios.
Table B-10. Comparison of current and future adapted best management practice
(BMP) footprints, Scott County, MN Green Infrastructure (GI) Only (high
intensity) scenario
Stormwater
management
scenario
Climate
scenario
Practice
Current
Future adapted
%
increase in
footprint
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint as
% of site
area
GI Only
Future
high
intensity
Bioretention
(modified)
43,275
3.3
111,735
8.6
158
0
0
Rooftop downspout
disconnection
94,901
7.3
94,901
7.3
Permeable pavement
39,390
3.0
39,390
3.0
B.1.2.4. Conventional (Gray) Infrastructure with Distributed Green Infrastructure
(GI)
Future Low Intensity
The Conventional (Gray) Infrastructure scenario was not investigated for adaptation in the future low
intensity climate scenario because there was no decrease in the performance of the conventional (Gray)
practice (wet pond) between the current and future climate. Therefore, the future low intensity climate
scenario was also not investigated for adaptation through the addition of distributed GI practices as there
was no performance gap to address.
Future Medium Intensity
The change in flow and pollutant-related performance for the Scott County Conventional (Gray) scenario
due to future climate impacts was discussed in Section B. 1.2.1. The purpose of the Conventional (Gray)
Infrastructure with Distributed GI scenario is to implement distributed GI practices without resizing the
conventional (Gray) practice as a means of future climate adaptation.
Figure B-82 indicates that the addition of distributed bioretention to the conventional site is able to reduce
outflow below the current climate outflow by increasing the partitioning of runoff to infiltration and ET.
Due to the poorly infiltrating soils on the Scott County site, outflow remains the dominant runoff fate
pathway, even with the addition of infiltrating bioretention.
B-66
-------
Runoff Fate
¦
ro
40
35
30
25
20
15
10
5
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
0.00
0.00
3.10
¦ ET
1.46
1.66
2.55
¦ Outflow
34.10
38.19
32.59
36.48
32.49
Figure B-82. Current and future partitioning of runoff fate for the Scott County,
MN Conventional (Gray) Infrastructure with Distributed Green Infrastructure (GI)
(medium intensity) scenario.
Figure B-83 indicates that the distributed bioretention practices combined with the wet pond are able to
achieve high load reductions for sediment, and allow the site to meet current loading for annual average
sediment load without requiring resizing of the existing wet pond.
B-67
-------
Annua! Average Sediment Load
50,000
45,000
40,000
35,000
30,000
^ 25,000
- 20,000
15,000
10,000
5,000
0
47,146
Current
Future
9,883
7,370
6,979
¦1 ¦
WWW
Current
Future with
Future,
with BMPs
BMPs
adapted
BMPs
Figure B-83. Current and future performance for annual average sediment load,
Scott County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (medium intensity) scenario.
The following figures suggest that the distributed bioretention practices combined with the wet pond are
able to achieve modest load reductions for I N and TP, and allow the site to meet current loading for
annual average sediment load without requiring resizing of the existing wet pond.
Annual Average TN Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-84. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Conventional (Gray) Infrastructure with Distributed
Green Infrastructure (GI) (medium intensity) scenario.
B-68
-------
Annual Average TP Load
47.49
41.79
50 4L79
i 11 ft 11
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-85. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Conventional (Gray) Infrastructure with Distributed
Green Infrastructure (GI) (medium intensity) scenario.
The FDC evaluation for the Scott County Conventional (Gray) Infrastructure with Distributed GI
(medium intensity) climate scenario indicates that with the addition of distributed bioretention practices,
the site is able to achieve a reasonably similar response to the current FDC in the evaluated range of
flows.
B-69
-------
80
70
60
£
u
~ so
140
0
5-
1 30
O
X
20
10
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Flourly Flow
v\
¦ ¦ ¦ ¦ M
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-86. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (medium intensity) scenario.
As discussed above, maximum hourly peak flow was not targeted as a performance criterion for the
adaptation simulation. However, it appears that the addition of distributed bioretention for medium
intensity climate adaptation does not significantly reduce maximum hourly peak flow below the "future
with BMPs" hourly peak flow.
B-70
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-87. Current and future performance for maximum hourly peak flow, Scott
County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (medium intensity) scenario.
Figure B-88 demonstrates that the addition of distributed bioretention practices combined with the
existing wet pond results in monthly outflow volumes that are more similar to the current performance.
Although outflow is higher in some months, on an annual basis, the future adapted scenario produces a
lower outflow volume than the "current with BMPs" scenario.
B-71
-------
Monthly Outflow Volume
Figure B-88. Current and future performance for monthly outflow volume, Scott
County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (medium intensity) scenario.
o
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs Future, adapted BMPs
Table B-l 1 summarizes the increases in BMP footprints for the Scott County Conventional (Gray)
Infrastructure with Distributed GI (medium intensity) scenario that would be required to maintain current
performance under future climate conditions. Adaptation would require the addition of 18,280 square feet
of bioretention (roughly 1.4% of the total site area).
Table B-ll. Comparison of current and future adapted best management practice
(BMP) footprints, Scott County, MN Conventional (Gray) Infrastructure with
Distributed Green Infrastructure (GI) (medium intensity) scenario
Stormwater
management
scenario
Climate
scenario
Practice
Current
Future adapted
% increase
in
footprint
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint as
% of site
area
Conventional (Gray)
Infrastructure with
Distributed GI
Future
medium
intensity
Wet pond
32,670
2.5
32,670
2.5
0
Distributed
bioretention
0
0.0
18,280
1.4
-
B-72
-------
Future High Intensity
The change in flow and pollutant-related performance for the Scott County Conventional (Gray) scenario
due to future climate impacts was discussed in Section B. 1.2.1. . The purpose of the Conventional (Gray)
Infrastructure with Distributed GI scenario is to implement distributed GI practices without resizing the
conventional (Gray) practice as a means of future climate adaptation.
Figure B-89 indicates that the addition of distributed bioretention to the conventional site is able to reduce
outflow below the current climate outflow by increasing the partitioning of runoff to infiltration and ET.
However, due to the poorly infiltrating soils on the Scott County site, outflow remains the dominant
runoff fate pathway, even with the addition of infiltrating bioretention.
Runoff Fate
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
0.00
0.00
9.49
¦ ET
1.46
1.73
4.63
¦ Outflow
34.10
38.30
32.59
36.52
24.16
Figure B-89. Current and future partitioning of runoff fate for the Scott County,
MN Conventional (Gray) Infrastructure with Distributed Green Infrastructure (GI)
(high intensity) scenario.
The wet pond alone achieves a reasonably high load reduction for annual average sediment load, even
with future (high intensity) climate impacts. The current load reduction is 81% and the future (before
adaptation) load reduction is 76%. Without resizing the wet pond, the addition of distributed bioretention
practices improves the future load reduction to 90%, enabling the site to maintain current performance for
sediment load reduction under projected future climate conditions.
B-73
-------
Annual Average Sediment Load
48,603
50,000
45,000
40,000
35,000
30,000
25,000
20,000
15,000
10,000 I 5^32_
5,000
0
11,591
7,370
m m
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-90. Current and future performance for annual average sediment load,
Scott County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (high intensity) scenario.
The addition of distributed bioretention in the future high intensity climate scenario was primarily driven
by the required decrease in outflow needed to maintain current outflow reduction performance. Because
outflow reduction was the driving mechanism, the resulting bioretention footprint is larger than would be
required to address pollutant loading alone. As a result, the future adapted TN and TP loads are much
lower than loads under the current climate conditions.
Annual Average TN Load
350 298.6 313-9
300
250
200
- 150
100
50
68.0
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-91. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Conventional (Gray) Infrastructure with Distributed
Green Infrastructure (GI) (high intensity) scenario.
B-74
-------
Annual Average TP Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-92. Current and future performance for annual average total phosphorous
(TP) load, Scott County, MN Conventional (Gray) Infrastructure with Distributed
Green Infrastructure (GI) (high intensity) scenario.
The FDC analysis for the future high intensity scenario suggests that although the addition of distributed
bioretention practices is able to maintain the current climate outflow volume, these practices as designed
are not effective at reducing the highest hourly peak flow rates. As a result, there is divergence between
the "current with BMPs" and "future, adapted BMPs" flow duration curves, particularly for the highest
outflows.
B-75
-------
120
100
'S 80
1
1 60
O
_>
o 40
X
20
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Hourly Flow
\
\
\
\
"S.
^
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-93. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (high intensity) scenario.
Figure B-94 indicates that the addition of distributed bioretention to the conventional site does not result
in a significant decrease in the maximum hourly peak flow under future (high intensity) climate.
B-76
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-94. Current and future performance for maximum hourly peak flow, Scott
County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (high intensity) scenario.
As noted above, although the distributed bioretention practices, when combined with the existing wet
pond, are not successful at mitigating the highest outflows under future high intensity climate, these
practices are effective at managing monthly outflow volume. The "future, adapted BMPs" monthly
outflow volumes are consistently lower or approximately the same as the "current with BMPs" monthly
outflow volumes for all months.
B-77
-------
Monthly Outflow Volume
7
4
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-95. Current and future performance for monthly outflow volume, Scott
County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (high intensity) scenario.
Table B-12 summarizes the increases in BMP footprints for the Scott County Conventional (Gray)
Infrastructure with Distributed GI (high intensity) scenario that would be required to maintain current
performance under future climate conditions. Adaptation would require the addition of 56,770 square feet
of bioretention (roughly 4.3% of the total site area).
Table B-12. Comparison of current and future adapted best management practice
(BMP) footprints, Scott County, MN Conventional (Gray) Infrastructure with
Distributed Green Infrastructure (GI) (high intensity) scenario
Stormwater
management
scenario
Climate
scenario
Practice
Current
Future adapted
% increase
in
footprint
Footprint
SF
|
rin
f si
ea
t as
te
Footprint
SF
Footprint as
% of site
area
Conventional (Gray)
Infrastructure with
Distributed GI
Future
high
intensity
Wet pond
32,670
2.5
32,670
2.5
0
Distributed
bioretention
0
0.0
56,770
4.3
-
B-78
-------
B.1.2.5. Cost Estimation
Future Medium Intensity
Table B-13 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for all four Scott County stormwater management scenarios under future medium
intensity climate, as well as the percentage increase in cost (current to future adapted) and increase
(millions of dollars) per acre of site.
Table B-13. Comparison of the current and future estimated 20-year present value
costs for the Scott County, MN stormwater management scenarios, future medium
intensity climate
Location
Climate
scenario
Stormwater
management
scenario
Current cost
20-yr present
value,
$millions
Future adapted
cost (20-yr
present value,
$millions)
Increase in
cost (20-yr
present value,
$millions)
% increase
in cost
Increase
per acre of
site
$millions
Scott
County,
MN
Future
medium
intensity
Conventional
(Gray)
Infrastructure
3.05
8.99
5.94
195
0.30
GI with Gray
Infrastructure
4.92
7.37
2.46
50
0.12
GI Only
8.51
11.80
3.29
39
0.16
Conventional
(Gray)
Infrastructure
with
Distributed GI
3.05
4.69
1.65
54
0.08
Future High Intensity
Table B-14 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for all four Scott County stormwater management scenarios under future high intensity
climate, as well as the percentage increase in cost (current to future adapted) and increase (millions of
dollars) per acre of site.
B-79
-------
Table B-14. Comparison of the current and future estimated 20-year present value
costs for the Scott County, MN stormwater management scenarios, future high
intensity climate
Location
Climate
scenario
Stormwater
management
scenario
Current
cost (20-yr
present
value,
$millions)
Future
adapted cost
20-yr present
value,
$millions
Increase in
cost (20-yr
present
value,
$millions)
% increase
in cost
Increase
per acre of
site
$millions
Scott
County,
MN
Future
high
intensity
Conventional
(Gray)
Infrastructure
3.05
10.59
7.54
248
0.38
GI with Gray
Infrastructure
4.92
14.82
9.90
201
0.50
GI Only
8.51
16.44
7.93
93
0.40
Conventional
(Gray)
Infrastructure
with Distributed
GI
3.05
8.16
5.11
168
0.26
B.1.3. Maricopa County, AZ
B.1.3.1. Conventional (Gray) Infrastructure
As discussed in Section 4.2. of the report, the simulated future climate scenario for Maricopa County, AZ
resulted in an overall increase in precipitation depth and intensity. However, when current and future
precipitation depths are compared on a monthly basis across the simulation period, monthly average
precipitation actually decreases in 7 months (February-June, September, and December) and only
increases substantially in 3 months (January, October, and November). The increases are small in July
and August. However, on an annual basis, the increases in precipitation depth and intensity outweigh the
decreases, and the net result is a modest overall increase in both metrics. The increase in precipitation
depth and intensity in the future climate scenario results in an increase in annual runoff volume (acre-feet
per year) (see Figure B-96).
B-80
-------
Runoff Fate
u
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
9.79
10,58
9.70
¦ ET
2.54
2.53
3.48
¦ Outflow
12.66
13.62
0.00
0.06
0.00
Figure B-96. Current and future partitioning of runoff fate for the Maricopa
County, AZ Conventional (Gray) Infrastructure scenario.
The infiltration basin in the Maricopa County Conventional (Gray) Infrastructure scenario achieves zero
outflow under the current climate conditions. Approximately 79% of runoff is lost to infiltration, and the
remaining fraction (21%) to ET. Under future climate conditions, although the annual runoff volume
increases, only a very small fraction (less than 0.5%) of runoff is converted to outflow. Increasing the
surface area of the infiltration basin for climate adaptation results in zero site outflow, with annual runoff
partitioning 73% to infiltration and 26% to ET. Increasing the footprint of the infiltration basin increases
the surface area available for infiltration and ET. In this arid region, ET comprises a relatively large
fraction of the overall water balance under both current and future climate conditions.
Because there is zero site outflow on an annual basis under the "current with BMPs'' scenario, the
corresponding annual average sediment load is zero. Under future climate conditions, the annual average
sediment load increases to 363 pound/year, a result of a more than doubling of the annual sediment load
under future climate conditions ("future untreated'') compared to the current climate ("current untreated'').
Increasing the surface area of the infiltration basin reduces the sediment load from the site to zero.
B-81
-------
Annual Average Sediment Load
35,000
30,000
25,000
- 20,000
- 15,000
10,000
5,000
0
31,022
14,377
d
Current
Untreated
363
Future Current Future with
Untreated with BMPs BMPs
Future,
adapted
BMPs
Figure B-97. Current and future performance for annual average sediment load,
Maricopa County, AZ Conventional (Gray) Infrastructure scenario.
Under future climate conditions, annual TN load is predicted to increase by approximately
3.5 pound/year. The infiltration basin is very effective at reducing TN load from the site. Increasing the
footprint enables the basin to achieve zero outflow under future climate conditions ("future, adapted
BMPs").
20
18
16
14
12
10
Annual Average TN Load
18,81
0.00
0.13
0.00
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-98. Current and future performance for annual average total nitrogen
(TN) load, Maricopa County, AZ Conventional (Gray) Infrastructure scenario.
B-82
-------
Annual average TP load from the site also increases under the future climate by a large factor. Without
resizing, the infiltration basin is still highly effective at reducing TP loading, with an annual average TP
load of 0.005 pound/year. Increasing the footprint for future adaptation results in zero annual TP load.
Annual Average TP Load
1.027
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-99. Current and future performance for annual average total phosphorous
(TP) load, Maricopa County, AZ Conventional (Gray) Infrastructure scenario.
A flow duration curve analysis was not performed for the Maricopa County Conventional (Gray)
Infrastructure scenario. Because there is zero outflow under the "current with BMPs" scenario, a current
climate flow duration curve could not be plotted. The objective of the adaptation simulation was,
therefore, to increase the footprint of the infiltration basin until zero outflow was achieved.
Figure B-100 presents the maximum hourly peak flow results and demonstrates that hourly peak flow
nearly doubles between the current and future untreated climate conditions. As discussed above, the
infiltration basin completely eliminates outflow through infiltration and evapotranspiration under the
current climate conditions. Without resizing, the basin is unable to main its current performance under
future climate conditions, and hourly peak flow increases from zero to 6.7 cfs. Increasing the basin size
reduces hourly peak flow to 0 cfs.
B-83
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-IOO. Current and future performance for maximum hourly peak flow,
Maricopa County, AZ Conventional (Gray) Infrastructure scenario.
Figure B-101 provides an alternate means of analyzing the infiltration basin's performance at managing
runoff volume on a monthly basis. Under future climate, due to increased runoff volumes, the infiltration
basin experiences outflow in the months of January, October, and November. The future adapted BMP
size eliminates outflow in all months.
B-84
-------
Monthly Outflow Volume
0.035
0.03
0.025
-C
c
o
0.02
±T
0.015
ra
0.01
0.005
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
¦ Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-101. Current and future performance for monthly outflow volume,
Maricopa County, AZ Conventional (Gray) Infrastructure scenario.
Table B-15 summarizes the increase in infiltration basin footprint for the Maricopa County Conventional
(Gray) Infrastructure scenario that would be required to maintain current performance under future
climate conditions.
Table B-15. Comparison of current and future adapted best management practice
(BMP) footprints, Maricopa County, AZ Conventional (Gray) Infrastructure
scenario
Stormwater
management
scenario
Practice
Current
Future adapted
% increase
in
footprint
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint as
% of site
area
Conventional (Gray)
Infrastructure
Infiltration basin
49,997
11.5
71,776
16.5
44
B.1.3.2. Green Infrastructure (Gl) Only
In current climate conditions, the combination of green practices (bioretention, cistern, permeable
pavement, and stormwater infiltration basin) in the Maricopa County GI Only scenario achieves near-zero
outflow. Outflow comprises a small fraction of the overall runoff volume fate. The fraction of outflow
increases slightly under future climate without practice resizing. With the future adapted BMP footprints,
B-85
-------
the surface area available for infiltration and ET increases, enabling the site to achieve a slightly lower
outflow volume than in the "current with BMPs" scenario. In the "future, adapted BMPs" scenario, the
fraction of ET is greater than 28%, compared to 20% in the "current with BMPs" scenario.
Runoff Fate
0
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
9.59
10.44
9.20
¦ ET
2.43
2.39
3.69
¦ Outflow
12.35
13.34
0.000042
0.054
0.000039
Figure B-102. Current and future partitioning of runoff fate for the Maricopa
County, AZ Green Infrastructure (GI) Only scenario.
The practices in the GI Only scenario combine to achieve a very high reduction for annual average
sediment load. Under future climate conditions, the green practices reduce sediment load by greater than
99%. Their effectiveness is reduced to approximately 97% under future climate without resizing.
Increasing the practice footprints reduces the annual average sediment load to 0.3 pound/year, which is
below the current load.
B-86
-------
Annual Average Sediment Load
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
35,471
14,399
9
Current
Untreated
0,50
985
0.28
Future Current Future with Future,
Untreated with BMPs 8MPs adapted
BMPs
Figure B-103. Current and future performance for annual average sediment load,
Maricopa County, AZ Green Infrastructure (GI) Only scenario.
A slight increase in annual average TN load is observed when the current BMPs are not resized under
future climate. Increasing the practice footprints for future climate adaptation achieves a minimal annual
TN load that is equivalent to the "current with BMPs" scenario.
Annual Average TN Load
20
18
16
14
12
10
18.50
0.00009
0.12
0.00009
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-104. Current and future performance for annual average total nitrogen
(TN) load, Maricopa County, AZ Green Infrastructure (GI) Only scenario.
B-87
-------
Annual Average TP Load
1.391
1.4
1.2
1
t. 0.8
- 0.6
0.4
0.302
0.2
n nnn 0.057
0
Current Future
Current with Future with
Future,
Untreated Untreated
BMPs BMPs
adapted
BMPs
Figure B-105. Current and future performance for annual average total
phosphorous (TP) load, Maricopa County, AZ Green Infrastructure (GI) Only
scenario.
An FDC analysis was not performed for the Maricopa County GI Only scenario due to both the "current
with BMPs" and "future with BMPs" scenarios producing minimal outflow. As an alternate means of
comparing flow-based performance, the maximum hourly peak flow for all climate scenarios is shown
below. The future adapted BMP sizes are effective at mitigating the increase in hourly peak flow under
future climate and reducing it below the "current with BMPs" value.
B-88
-------
Max Peak Flow
43.06
0
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-106. Current and future performance for maximum hourly peak flow,
Maricopa County, AZ Green Infrastructure (GI) Only scenario.
Under future climate, due to increased runoff volumes, the green practices in the GI Only scenario
experience outflow in the months of January, October, and November. The future adapted BMP sizes
achieve effectively zero monthly outflow volume in all months.
B-89
-------
Monthly Outflow Volume
0.035
0.03
0.025
| 0-02
E
—,
0.015
u
ro
0.01
0.005
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-107. Current and future performance for monthly outflow volume,
Maricopa County, AZ Green Infrastructure (GI) Only scenario.
Table B-16 summarizes the increases in BMP footprints for the Maricopa County GI Only scenario that
would be required to maintain current performance under future climate conditions.
Table B-16. Comparison of current and future adapted best management practice
(BMP) footprints, Maricopa County, AZ Green Infrastructure (GI) Only scenario
Current
Future adapted
Stormwater
management
scenario
Practice
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint as
% of site
area
% increase
in
footprint
GI Only
Permeable pavement
86,382
19.8
124,482
28.6
44
Bioretention
13,405
3.1
24,125
5.5
80
Cistern
2,495
0.6
3,564
0.8
43
Stormwater harvesting basin
32,034
7.4
53,034
12.2
66
B.1.3.3. Cost Estimation
Table B-17 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for both Maricopa County storm water management scenarios. Also provided are the
increase in cost, both in dollars and percentage, and the increase in cost per acre of site.
B-90
-------
Table B-17. Comparison of the current and future estimated 20-year present value
costs for the Maricopa County, AZ stormwater management scenarios
Location
Stormwater
management
scenario
Current cost
20-yr present
value, Smillions
Future adapted
cost (20-yr present
value, Smillions)
Increase in cost
20-yr present
value, Smillions
% increase
in cost
Increase per
acre of site
Smillions
Maricopa
County,
AZ
Conventional
(Gray)
Infrastructure
4.79
6.83
2.04
43
0.20
GI Only
3.98
6.33
2.35
59
0.23
B.1.4. Atlanta, GA
B.1.4.1. Conventional (Gray) Infrastructure
The precipitation analysis for Atlanta, GA (see Section 4.2. of the report) demonstrated an across the
board increase in annual precipitation depth across all 30 years of the simulation period when compared
with the current climate. Precipitation intensity is also predicted to increase, with the greatest increase
occurring among the largest events. The result is an overall increase in runoff volume, which can be seen
in Figure B-108.
Runoff Fate
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
0.00
0.00
0.00
¦ ET
0.00
0.00
0.00
¦ Outflow
7.33
7.88
7.33
7.88
7.88
Figure B-108. Current and future partitioning of runoff fate for the Atlanta, GA
Conventional (Gray) Infrastructure scenario.
B-91
-------
The practices in the Atlanta Conventional (Gray) Infrastructure scenario are unique in that they are both
underground. As a result, there is no opportunity for ET, and the practices are not designed to be
infiltrating. The underground sand filter provides water quality treatment through filtration, and the
underground detention basin addresses peak flow and flooding requirements. Runoff is discharged after
temporary storage and treatment. By nature of these practices, the runoff fate pathways do not change
significantly between the current and future climate scenarios (i.e., increasing their footprint will not
increase the proportioning of runoff fate into pathways other than outflow).
Although the Atlanta study site is ultra-urban with 90% impervious, there is still an opportunity for
greater sediment loading due to wash-off from increased precipitation depth and intensity. Under current
climate conditions, the conventional practices achieve a combined sediment load reduction of 82%. This
is reduced slightly to 79% under future climate. With resizing, the combination of the sand filter and
detention basin achieve a sediment load reduction of 85%.
Annual Average Sediment Load
12,000
10,000
8,000
6,000
4,000
2,000
10,146
10,839
1,817
Current Future Current
Untreated Untreated with BMPs
2,242
l,b9b
Future with
Future,
BMPs
adapted
BMPs
Figure B-109. Current and future performance for annual average sediment load,
Atlanta, GA Conventional (Gray) Infrastructure scenario.
B-92
-------
Annual Average TN Load
Current Future Current with Future with Future,
Untreated Untreated 8MPs 8MPs adapted
BMPs
Figure B-110. Current and future performance for annual average total nitrogen
(TN) load, Atlanta, GA Conventional (Gray) Infrastructure scenario.
With their current sizing, the conventional practices are unable to meet the current annual average TP
load of 2.11 pound/year. With future adaptation, the increased BMP footprints are able to reduce the TP
load to 2.10 pound/year, which is slightly below the current load of 2.11 pound/year.
Annual Average TP Load
6 5.15 5.23
5
n
4
13
2.11 2.17 2.10
n
1
I.
1
1 ¦ 0 ^
Current
Untreated
Future
1 Untreated
Current with Future with Future,
BMPs BMPs adapted
BMPs
Figure B-lll. Current and future performance for annual average total
phosphorous (TP) load, Atlanta, GA Conventional (Gray) Infrastructure scenario.
B-93
-------
Figure B-l 12 presents the flow duration curves for the current, future, and future adapted BMP scenarios
and demonstrates that the increased BMP footprints adapted for future climate achieve a very similar flow
duration response to the current climate BMP configurations in the evaluated range of flows.
1.2
i
Current with BMPs
1/2-yr Hourly Flow
j.
u 0.8
1
1 0.6
O
>¦
§ 0.4
X
0.2
0
0.0(
A
¦ ¦ ¦ — i ¦ ¦ ¦
- » *
)0% 0.005% 0.010% 0.015% 0.020% 0.025%
Percent of time flow is equaled or exceeded
Figure B-l 12. Flow duration curve (FDC) evaluation for current and future climate,
Atlanta, GA Conventional (Gray) Infrastructure scenario.
Peak flow management in the Conventional (Gray) Infrastructure scenario for Atlanta is primarily
provided by the underground detention basin, which is designed to be effective at peak flow reduction.
The increase in hourly peak flow between the "current with BMPs" and "future with BMPs" scenarios
indicates that with the current practice sizing, the current hourly peak flow of 0.49 cfs cannot be
maintained. Increasing the BMP footprints for future adaptation reduces the hourly peak flow to 0.48 cfs.
B-94
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-113. Current and future performance for maximum hourly peak flow,
Atlanta, GA Conventional (Gray) Infrastructure scenario.
As discussed above, the underground practices (sand filter, detention basin) comprising the Conventional
(Gray) Infrastructure scenario are not infiltrating; they provide runoff treatment and temporary storage to
address flooding control requirements prior to discharging. Therefore, the monthly outflow comparison
plot illustrates that there is virtually no difference in monthly outflow volume between the "future, with
BMPs" and "future, adapted BMPs" site conditions. Adaptation via resizing does not achieve additional
volume control for these conventional practices.
B-95
-------
Monthly Outflow Volume
1.2
0.2
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-114. Current and future performance for monthly outflow volume,
Atlanta, GA Conventional (Gray) Infrastructure scenario.
Table B-18 summarizes the increases in BMP footprints for the Atlanta Conventional (Gray)
Infrastructure with practices adapted for future conditions in order to maintain current performance (with
the exception of TN and runoff volume) under future climate conditions.
Table B-18. Comparison of current and future adapted best management practice
(BMP) footprints, Atlanta, GA Conventional (Gray) Infrastructure scenario
Stormwater
management
scenario
Practice
Current
Future adapted
% increase
in
footprint
Footprint
SF
Footprint as
% of site
area
Footprint
SF
Footprint as
% of site
area
Conventional
(Gray)
Infrastructure
Underground sand filter
2,500
5,000
2.9
4,600
5.3
84
Underground dry detention
basin
5.7
6,200
7.1
24
B.1.4.2. Green Infrastructure (Gl) with Gray Infrastructure
The partitioning of runoff volume fate in the Atlanta GI with Gray Infrastructure scenario is markedly
different from the Conventional (Gray) Infrastructure scenario due to the addition of green practices
(bioretention, green roof, permeable pavement) that facilitate ET and infiltration of runoff and combine
with the gray practice (underground detention) to provide peak flow and flooding management. In the
B-96
-------
"current with BMPs"' condition, 44% of runoff is infiltrated, 21% of is converted to ET, and the remaining
fraction (34%) is discharged as outflow. In the "future with BMPs" site condition, the partitioning is
similar: 43, 21, and 36%, respectively. The increased BMP footprints in the future adapted scenario
slightly increase the partitioning into infiltration and ET due to increased surface area; 45% of runoff is
infiltrated. 24% is evapotranspired, and 31% is discharged as outflow.
Runoff Fate
m
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
3.25
3.38
3.56
¦ ET
1.56
1.68
1.91
¦ Outflow
7.34
7.89
2.52
2.83
2.42
Figure B-115. Current and future partitioning of runoff fate for the Atlanta, GA
Green Infrastructure (GI) with Gray Infrastructure scenario.
Figure B-l 16 presents a comparison of the annual average sediment load from the GI with Gray
Infrastructure site scenario under the five simulated climate conditions and demonstrates that, with
adaptation, the combination of BMPs is able to reduce the future sediment load below the current climate
sediment load.
B-97
-------
Annual Average Sediment Load
12 000 10,850
iz'uuu 10.157
10,000
8,000
6,000
4,000 2,181 2,819
2,000
2,181 _ 2,023
3 8 I
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-116. Current and future performance for annual average sediment load,
Atlanta, GA Green Infrastructure (GI) with Gray Infrastructure scenario.
Figure B-l 17 and Figure B-l 18 present the simulated performance for TN and TP load reduction. These
results indicate that the combination of practices in the GI with Gray Infrastructure scenario can
successfully maintain their current performance when their footprints are increased for future adaptation
The TN load reductions in the "current with BMPs," "future with BMPs," and "future, adapted BMPs"
scenarios are 73, 70, and 76%, respectively. The TP load reductions are 76, 73, and 79%, respectively.
B-98
-------
Annual Average TN Load
Current Future Current with Future with Future,
Untreated Untreated 8MPs 8MPs adapted
BMPs
Figure B-117. Current and future performance for annual average total nitrogen
(TN) load, Atlanta, GA Green Infrastructure (GI) with Gray Infrastructure
scenario.
Annual Average TP Load
:
3
1
¦ ¦ m
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-118. Current and future performance for annual average total
phosphorous (TP) load, Atlanta, GA Green Infrastructure (GI) with Gray
Infrastructure scenario.
Figure B-l 19 indicates that when resized for future climate adaptation, the Green and Gray practices
achieve a flow duration response that is very similar to the current climate response.
B-99
-------
Current with BMPs Future with BMPs — — Future, adapted BMPs 1/2-yr Flourly Flow
3.5
0
0.000%
0.005% 0.010% 0.015% 0.020%
Percent of time flow is equaled or exceeded
0.025%
Figure B-119. Flow duration curve (FDC) evaluation for current and future climate,
Atlanta, GA Green Infrastructure (GI) with Gray Infrastructure scenario.
Under current climate conditions, the combination of Green and Gray practices reduces the maximum
hourly peak flow from 4.6 to 1.8 cfs. In the future, without resizing, the practices are still effective at
decreasing peak flow. However, increasing their footprints in the "future, adapted BMPs" scenario results
in a future hourly peak flow (1.9 cfs) that is only slightly higher than the "current with BMPs" hourly
peak flow.
B-100
-------
Max Peak Flow
8 ^2
: _ I
4 3^0
1.83 1_94
¦ III I
0
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-120. Current and future performance for maximum hourly peak flow,
Atlanta, GA Green Infrastructure (GI) with Gray Infrastructure scenario.
The monthly outflow volume comparison plot indicates variable performance of the adapted BMPs in
maintaining the current climate outflow volumes on a monthly basis. Although the "future, adapted
BMPs" outflows are lower than the "future with BMPs" outflows in every month, the adapted BMP
outflows are higher than the "current with BMPs" outflows in some months and lower in others.
However, as the runoff fate plot demonstrated, the adapted practices achieve a lower annual outflow
volume overall compared to the current practices.
B-101
-------
Monthly Outflow Volume
0,5
0.45
0.4
0.35
| 0.3
o
,£ 0.25
c£
}j 0.2
H3
0.15
0.1
0.05
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-121. Current and future performance for monthly outflow volume,
Atlanta, GA Green Infrastructure (GI) with Gray Infrastructure scenario.
Table B-19 summarizes the increases in BMP footprints for the Atlanta Conventional (Gray)
Infrastructure with scenario that would be required to maintain current performance under future climate
conditions.
Table B-19. Comparison of current and future adapted best management practice
(BMP) footprints, Atlanta, GA Green Infrastructure (GI) with Gray Infrastructure
scenario
Stormwater
management
Current
Practice
GI with Gray
Infrastructure
Bioretention with underdrain
2.810
3.934
Underground dry detention
basin
2.500
3.300
Green roof
30.492
35.184
Permeable pavement
26.136
26.136
B-102
-------
B. 1.4.3. Cost Estimation
Table B-20 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for both Atlanta stormwater management scenarios. Also provided are the increase in
cost, both in dollars and percentage, and the increase in cost per acre of site.
Table B-20. Comparison of the current and future estimated 20-year present value
costs for the Atlanta, GA stormwater management scenarios
Location
Stormwater
management
scenario
Current cost
20-yr present
value, Smillions
Future adapted
cost (20-yr present
value, Smillions)
Increase in cost
20-yr present
value, Smillions
% increase
in cost
Increase per
acre of site
Smillions
Atlanta,
GA
Conventional
(Gray)
Infrastructure
1.38
2.27
0.89
64
0.09
GI with Gray
Infrastructure
2.31
2.60
0.29
13
0.03
B.1.5. Portland, OR
B.1.5.1. Green Infrastructure (Gl) Only
As discussed in the precipitation analysis (see Section 4.2. of the report), Portland is unique in that the
future climate simulation predicts an overall decrease in annual rainfall depth. On a monthly basis,
average precipitation is predicted to decrease in all months except January, May, and December. In terms
of intensity, the smallest, most frequent storms are predicted to increase in intensity. The intensity during
the largest events remains approximately the same in the future climate as in the current climate. Between
the smallest, most frequent and largest, least frequent events, the future rainfall intensity is sometimes
higher than the current climate and sometimes lower.
Figure B-122 illustrates the overall decrease in total runoff volume between the current and future climate
conditions as a result of the overall decrease in annual precipitation. The small size of the Portland GI
Only site (0.35 acre) results in a very small runoff volume overall (1.1 acre-feet/year in "current
untreated") compared to the other investigation sites. The green practices (bioretention swales, permeable
pavement) that comprise the Portland GI Only site are highly infiltrating. Almost all of the runoff is
infiltrated, with a small fraction (less than 4% in "current with BMPs") converted to ET and an even
smaller fraction converted to outflow. Although it cannot be seen in the chart because outflow accounts
for such a small fraction of the runoff balance, the outflow volume actually increases between the
"current with BMPs" scenario to the "future with BMPs" from 0.0015 acre-feet/year to
0.0026 acre-feet/year. This increase in outflow volume is due to the increase in precipitation intensity for
large storm events, resulting in a greater fraction of the runoff being discharged rather than infiltrated.
With the adapted BMP footprints, the fraction of ET is increased due to the larger practice surface areas,
resulting in a lower fraction of outflow.
B-103
-------
Runoff Fate
0.0
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
1.07
0.99
0.98
¦ ET
0.041
0.040
0.052
¦ Outflow
1.11
1.04
0.0015
0.0026
0.0015
Figure B-122. Current and future partitioning of runoff fate for the Portland, OR
Green Infrastructure (GI) Only scenario.
Due to the slight increase in outflow in future climate, there is a small increase in annual average
sediment load between the "current with BMPs'' and "future with BMPs" scenarios. Increasing the
practice footprints reduces the future adapted load (0.62 pound/year) below the current load
(0.96 pound/year).
B-104
-------
Annual Average Sediment Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-123. Current and future performance for annual average sediment load,
Portland, OR Green Infrastructure (GI) Only scenario.
The green practices in the Portland GI Only scenario combine to achieve very high load reductions for
TN and TP on an annual basis. In all three scenarios with BMPs, the TN and TP loads are reduced by
99% or greater. The future adapted practice sizes result in lower TN and TP loads in the future compared
to the current climate with BMPs.
Annual Average TN Load
13.03
.4 12.20
ul
2 0X3097 0.038 0.0080
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-124. Current and future performance for annual average total nitrogen
(TN) load, Portland, OR Green Infrastructure (GI) Only scenario.
B-105
-------
Annual Average TP Load
1,8
1.6
1.4
1.2
£ 0.8
0.6
0.4
0.2
1.646
1.560
0.00120
0.00149 0.00087
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-125. Current and future performance for annual average total
phosphorous (TP) load, Portland, OR Green Infrastructure (GI) Only scenario.
The flow duration curve analysis for the Portland GI Only site demonstrates that the future adapted
practices achieve an hourly outflow response that is reasonably similar to performance under current
climate conditions within the evaluated range of outflows.
¦Current with BMPs
• Future with BMPs
0.2
0.18
0.16
£
0.14
U
S
0.12
o
E
3
0.1
o
_>
0.08
3
O
X
0.06
0.04
0.02
0
1 Future, adapted BMPs
¦ 1-yr Hourly Flow
0.000%
0.002%
0.004% 0.006% 0.008% 0.010%
Percent of time flow is equaled or exceeded
0.012%
B-106
-------
Figure B-126. Flow duration curve (FDC) evaluation for current and future climate,
Portland, OR Green Infrastructure (GI) Only scenario.
Although the overall runoff volume is predicted to decrease under future climate conditions for Portland,
the increase in the intensity of precipitation during select storms results in higher peak flows in the future
compared to the current climate. With their current sizing, the practices in the GI Only scenario are
unable to maintain the current peak flow; in the future, without resizing, the hourly peak flow nearly
doubles. Recall that maximum hourly peak flow was not a target of the adaptation simulation; however,
Figure B-127 demonstrates that the increased BMP footprints in the "future, adapted BMPs" scenario
combine to reduce the hourly peak flow almost to the "current with BMPs" value.
Max Peak Flow
0.392
0369
0^72
^^3 ^^4
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-127. Current and future performance for maximum hourly peak flow,
Portland, OR Green Infrastructure (GI) Only scenario.
The monthly outflow volume comparison plot shown in Figure B-128 indicates a change in monthly
outflow between the current and future climate conditions, with outflow shifting to later in the winter. The
adapted future scenario tracks below the future scenario during all months with outflow.
0.4
0.35
0.3
0.25
CO _ _
t3 0.2
0.15
0.1
0.05
0
B-107
-------
Monthly Outflow Volume
0.0016
0.0014
0.0012
£ 0.001
£
O
0.0008
£
K 0.0006
0.0004
0.0002
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
¦ Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-128. Current and future performance for monthly outflow volume,
Portland, OR Green Infrastructure (GI) Only scenario.
Table B-21 summarizes the increases in BMP footprints for the Portland GI Only scenario that would be
required to maintain current performance under future climate conditions. Permeable pavement was not
resized as part of the adaptation. The bioretention swales provide almost all of the water quantity and
quality treatment for the site due to their large storage volumes and high infiltration capacity.
Table B-21. Comparison of current and future adapted best management practice
(BMP) footprints, Portland, OR Green Infrastructure (GI) Only scenario
Stormwater
Practice
Current
Future adapted
% increase
in footprint
management
scenario
Footprint
SF
Footprint as
% of site area
Footprint
SF
Footprint as
% of site area
GI Only
Bioretention swale
1,239
8.0
1,681
10.9
36
0
Permeable pavement
345
2.2
345
2.2
B.1.5.2. Cost Estimation
Table B-22 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for the Portland GI Only scenario. Also provided are the increase in cost, both in
dollars and percentage, and the increase in cost per acre of site.
B-108
-------
Table B-22. Comparison of the current and future estimated 20-year present value
costs for the Portland, OR Green Infrastructure (GI) Only stormwater management
scenario
Stormwater Current cost Future adapted Increase in cost
management 20-yr present cost (20-yr present 20-yr present
Location scenario value, Smillions value, Smillions) value, Smillions
Portland, GlOnly 0.20 0.27 0.07
OR
B.2. SENSITIVITY ANALYSIS
B.2.1. Harford County, MD
B.2.1.1. Conventional (Gray) Infrastructure
Intensity Change Minus 10%
As discussed in Section 4.3. of the report, the sensitivity analysis entailed modifying the current
precipitation record to represent potential future climate conditions by applying a graduated set of
percentage changes across the entire precipitation record. For this particular sensitivity scenario, the
resulting change in annual runoff volume and pollutant load was a decrease under the future climate
compared to the current climate, as illustrated in the figures below. For these reasons, this climate
scenario was not investigated for adaptation simulation.
Runoff Fate
70
60
50
40
30
20
10
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
¦ Infiltration
47.28
44.78
¦ ET
1.42
1.33
¦ Outflow
60.45
54.63
11.74
8.52
Increase per
% increase
acre of site
in cost
Smillions
35
0.20
B-109
-------
Figure B-129. Current and future partitioning of runoff fate for the Harford
County, MD Conventional (Gray) Infrastructure (intensity minus 10%) scenario.
Annual Average Sediment Load
25,000
20,000
15,000
10,000
5,000
0
21.795
4,610
3,427
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-130. Current and future performance for annual average sediment load,
Harford County, MD Conventional (Gray) Infrastructure (intensity minus 10%)
scenario.
Annual Average TN Load
474,4
500
450
400
350
300
250
200
150
100
50
0
Current
Untreated
Future
Untreated
54.9
Current with
BMPs
40.1
Future with
BMPs
Figure B-131. Current and future performance for annual average total nitrogen
(TN) load, Harford County, MD Conventional (Gray) Infrastructure (intensity
minus 10%) scenario.
13-110
-------
Annual Average TP Load
0
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-132. Current and future performance for annual average total
phosphorous (TP) load, Harford County, MD Conventional (Gray) Infrastructure
(intensity minus 10%) scenario.
Current with
8MPs
Future with
8MPs
Current
Untreated
Future
Untreated
Figure B-133. Current and future performance for maximum hourly peak flow,
Harford County, MD Conventional (Gray) Infrastructure (intensity minus 10%)
scenario.
B- 111
-------
Monthly Outflow Volume
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs
Figure B-134. Current and future performance for monthly outflow volume,
Harford County, MD Conventional (Gray) Infrastructure (intensity minus 10%)
scenario.
Intensity Change Plus 10%
For this sensitivity analysis, we applied a 10% increase to the precipitation depths in the current
conditions precipitation record. The 10% increase affects both intensity and precipitation volume,
resulting in increases in both total runoff volume and outflow volume. Infiltration also increases between
the "current with BMPs" and "future with BMPs" scenarios, suggesting that in the "intensity plus 10%"
future climate, the conventional practices (surface sand filters and extended dry detention basin) are able
to infiltrate some of the increased runoff volume prior to resizing. With increased practice footprints for
future climate adaptation, the partitioning of runoff to infiltration and ET increases due to the larger
surface areas, allowing outflow to be reduced below the current climate level.
13-112
-------
Runoff Fate
u
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
47.28
49.73
54.90
¦ ET
1.42
1.52
1.86
¦ Outflow
60.45
66.51
11.74
15.25
9.74
Figure B-135. Current and future partitioning of runoff fate for the Harford
County, MD Conventional (Gray) Infrastructure (intensity plus 10%) scenario.
The following figures present the pollutant load reduction performance for the current and future site
scenarios under the "intensity plus 10%" climate simulation, and indicate that the future adapted practices
combine to reduce the annual average loading for sediment, TN, and TP below the current climate load.
B-113
-------
Annual Average Sediment Load
24,690
25,000
20,000
15,000
10,000
5,000
4,610
5,864
2,518
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-136. Current and future performance for annual average sediment load,
Harford County, MD Conventional (Gray) Infrastructure (intensity plus 10%)
scenario.
Annual Average TN Load
495,8
500
450
400
350
300
250
200
150
100
50
0
474,4
54.9
71.0
37.7
Current
Future Current with Future with Future,
Untreated Untreated
BMPs
BMPs
adapted
BMPs
Figure B-137. Current and future performance for annual average total nitrogen
(TN) load, Harford County, MD Conventional (Gray) Infrastructure (intensity plus
10%) scenario.
13-114
-------
Annual Average TP Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-138. Current and future performance for annual average total
phosphorous (TP) load, Harford County, MD Conventional (Gray) Infrastructure
(intensity plus 10%) scenario.
When resized to adapt to the future "intensity plus 10%" climate conditions, the conventional practices
achieve a combined flow duration curve response that is very similar to the current flow response within
the range of evaluated flows.
B-115
-------
30
25
20
1
lis
O
>.
Current with BMPs Future with BMPs — — Future, adapted BMPs
¦2-yr Hourly Flow
g 10
X
5
0
0.0(
^
30% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-139. Flow duration curve (FDC) evaluation for current and future climate,
Harford County, MD Conventional (Gray) Infrastructure (intensity plus 10%)
scenario.
Comparison of the maximum hourly peak flow for the current and future (intensity plus 10%) climate
scenarios indicates that although the adapted practices are unable to maintain the current peak flow, they
do reduce the future hourly peak flow to within 3% of the current hourly peak flow.
B-116
-------
Max Peak Flow
37.62
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-140. Current and future performance for maximum hourly peak flow,
Harford County, MD Conventional (Gray) Infrastructure (intensity plus 10%)
scenario.
The future adapted practices for the Harford County Conventional (Gray) Infrastructure scenario combine
to decrease the future (intensity plus 10%) climate monthly outflow volumes below the current monthly
outflow volumes for all months except January, where the outflows are approximately the same.
B-117
-------
Monthly Outflow Volume
2,5
1.5
£
0.5
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-141. Current and future performance for monthly outflow volume,
Harford County, MD Conventional (Gray) Infrastructure (intensity plus 10%)
scenario.
Intensity Change Plus 20%
This sensitivity analysis entailed applying a 20% increase to the current climate precipitation record. This
increase affects both intensity and precipitation volume, resulting in increases in both total runoff volume
and outflow volume. Infiltration also increases between the "current with BMPs" and "future with BMPs"
scenarios, suggesting that in the "intensity plus 20%" future climate, the conventional practices (surface
sand filters and extended dry detention basin) are able to infiltrate some of the increased runoff volume
prior to resizing. With increased practice footprints for future climate adaptation, the partitioning of
runoff to infiltration and ET increases due to the larger surface areas, allowing outflow to be reduced
below the current climate level.
B-118
-------
Runoff Fate
u
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
47.28
52.09
60.06
¦ ET
1.42
1.62
2.33
¦ Outflow
60.45
72.70
11.74
19.00
10.32
Figure B-142. Current and future partitioning of runoff fate for the Harford
County, MD Conventional (Gray) Infrastructure (intensity plus 20%) scenario.
The following figures present the pollutant load reduction performance for the current and future site
scenarios under the "intensity plus 20%" climate simulation and indicate that the future adapted practices
combine to reduce the annual average loading for sediment, TN, and TP below the current climate load.
13-119
-------
Annual Average Sediment Load
30,000
25,000
20,000
15,000
10,000
5,000
27,375
7,179
4,610
3,027
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-143. Current and future performance for annual average sediment load,
Harford County, MD Conventional (Gray) Infrastructure (intensity plus 20%)
scenario.
Annual Average TN Load
600
500
400
300
200
100
520.1
Current Future
Untreated Untreated
54.9
BMPs
88.1
41.6
Future with
Future,
BMPs
adapted
BMPs
Figure B-144. Current and future performance for annual average total nitrogen
(TN) load, Harford County, MD Conventional (Gray) Infrastructure (intensity plus
20%) scenario.
B-120
-------
Annual Average TP Load
91.17
1UU i
90
80
50.3/
70
60
>¦
> 50
40
30
^ 10 13
20
10
^ J
u
Current
Future
Current with Future with Future,
Untreated
Untreated
BMPs BMPs adapted
BMPs
Figure B-145. Current and future performance for annual average total
phosphorous (TP) load, Harford County, MD Conventional (Gray) Infrastructure
(intensity plus 20%) scenario.
When resized to adapt to the future "intensity plus 20%" climate conditions, the conventional practices
achieve a combined flow duration curve response that is very similar to the current flow response within
the range of evaluated flows.
B-121
-------
30
25
'S 20
1
lis
O
_>¦
§ 10
X
5
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
¦2-yr Hourly Flow
\\
V
* ^
^ ^ _
"¦
—
~ — —
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-146. Flow duration curve (FDC) evaluation for current and future climate,
Harford County, MD Conventional (Gray) Infrastructure (intensity plus 20%)
scenario.
Comparison of the maximum hourly peak flow for the current and future (intensity plus 20%) climate
scenarios indicates that although the adapted practices are unable to maintain the current peak flow, they
do reduce the future hourly peak flow to within 4% of the current hourly peak flow.
B-122
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-147. Current and future performance for maximum hourly peak flow,
Harford County, MD Conventional (Gray) Infrastructure (intensity plus 20%)
scenario.
The future adapted practices for the Harford County Conventional (Gray) Infrastructure scenario combine
to maintain the future (intensity plus 20%) climate monthly outflow volumes at or below the current
monthly outflow volumes for all months except January, where the future adapted outflow volume is
slightly higher than the current outflow volume.
B-123
-------
Monthly Outflow Volume
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-148. Current and future performance for monthly outflow volume,
Harford County, MD Conventional (Gray) Infrastructure (intensity plus 20%)
scenario.
B.2.1.2. Green Infrastructure (Gl) with Gray Infrastructure
Intensity Change Minus 10%
This scenario was not selected for adaptation simulation; refer to discussion in Section B.2.1.1. and the
following figures, which demonstrate decreased outflow volume and pollutant loading in the '"intensity
minus 10%" climate scenario compared to the current climate.
B-124
-------
Runoff Fate
u
ra
60
50
40
30
20
10
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
¦ Infiltration
53.52
48,83
¦ ET
1.93
1.79
¦ Outflow
57.99
52.31
2.54
1.67
Figure B-149. Current and future partitioning of runoff fate for the Harford
County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity minus
10%) scenario.
Annual Average Sediment Load
25,000
20,000
15,000
10,000
5,000'
0
21,483
1,554
1,045
Current Future Current with Future with
Untreated Untreated iBMPs BMPs
Figure B-150. Current and future performance for annual average sediment load,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
minus 10%) scenario.
B-125
-------
Annual Average TN Load
0
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-151. Current and future performance for annual average total nitrogen
(TN) load, Harford County, MD Green Infrastructure (GI) with Gray
Infrastructure (intensity minus 10%) scenario.
Annual Average TP Load
0
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-152. Current and future performance for annual average total
phosphorous (TP) load, Harford County, MD Green Infrastructure (GI) with Gray
Infrastructure (intensity minus 10%) scenario.
B-126
-------
Max Peak Flow
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-153. Current and future performance for maximum hourly peak flow,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
minus 10%) scenario.
Monthly Outflow Volui
0.8
0.7
0.6
£ 0.5
o
0.4
A i
ne
A
A
/a\
A\
' \\
V \
Jan Feb Mar Apr May Jun Jul Aug
Current with BMPs Fut>
Sep Oct Nov Dec
ure with BMPs
Figure B-154. Current and future performance for monthly outflow volume,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
minus 10%) scenario.
B-127
-------
Intensity Change Plus 10%
Increasing the intensity of the current climate precipitation record by 10% affects both intensity and
precipitation volume for Harford County results in an increase both in total runoff volume and in outflow
volume, as shown in Figure B-155. However, in the GI with Gray Infrastructure scenario, the practices
are highly infiltrating such that the majority of the increase in runoff volume partitions into infiltration,
with only a relatively small increase in outflow volume. With resizing for future adaptation, the fraction
of runoff that is converted to infiltration and ET is increased further, enabling the site to maintain its
current performance for outflow volume.
Runoff Fate
u
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
53.52
58.19
59.16
¦ ET
1.93
2.07
2.36
¦ Outflow
57.99
63.91
2.54
3.64
2.38
Figure B-155. Current and future partitioning of runoff fate for the Flarford
County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity plus
10%) scenario.
The following figures indicate that the combination of practices in the GI with Gray Infrastructure
scenario achieves a high load reduction for sediment, TN, and TP, and that resizing the practices for the
future (intensity plus 10%) climate enables loads to be maintained at or below their current levels.
B-128
-------
Annual Average Sediment Load
24,503
25,000
20,000
15,000
10,000
5,000
0
1,554
2,193
1,486
Current Future Current Future with Future,
Untreated Untreated with BMPs 6MPs adapted
IB MPs
Figure B-156. Current and future performance for annual average sediment load,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 10%) scenario.
Annual Average TN Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-157. Current and future performance for annual average total nitrogen
(TN) load, Harford County, MD Green Infrastructure (GI) with Gray
Infrastructure (intensity plus 10%) scenario.
B-129
-------
Annual Average TP Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-158. Current and future performance for annual average total
phosphorous (TP) load, Harford County, MD Green Infrastructure (GI) with Gray
Infrastructure (intensity plus 10%) scenario.
The FDC evaluation for the GI with Gray Infrastructure scenario appears to indicate that the practices,
when resized for future climate (intensity plus 10%) adaptation, are able to very closely reproduce the
current climate flow duration curve response within the range of evaluated flows.
B-130
-------
^—Current with BMPs Future with BMPs — — Future, adapted BMPs 2-yr Hourly Flow
25
20
2"
u
I 15
> 10
3
O
5
0
0.000% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
A
N
N ^
Figure B-159. Flow duration curve (FDC) evaluation for current and future climate,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 10%) scenario.
With adaptation, the future (intensity plus 10%) practice footprints are also successful at reducing
maximum hourly peak flow from the site to 16.7 cfs, which is lower than the current hourly peak flow of
17.0 cfs.
B-131
-------
Max Peak Flow
37.47
3333 £
17.03 17JS 1670
iii\ I I
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-160. Current and future performance for maximum hourly peak flow,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 10%) scenario.
Resizing the Green and Gray practices for the future "intensity plus 10%" climate produces a monthly
outflow response that is nearly identical to the current BMP monthly outflow response.
40
35
30
25
•S 20
15
10
5
0
B-132
-------
1.2
Monthly Outflow Volume
0.8
j= 0.6
£
u
ra
0.4
0,2
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-161. Current and future performance for monthly outflow volume,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 10%) scenario.
Intensity Change Plus 20%
As discussed in Section B.2.1.1. , the "intensity plus 20%" climate scenario for Harford County results is
an increase both in total runoff volume and in outflow volume, as shown in Figure B 162. However, in
the GI with Gray Infrastructure scenario, the practices are highly infiltrating such that the majority of the
increase in runoff volume partitions into infiltration, with only a relatively small fraction of the increased
runoff partitioning to outflow. With resizing for future adaptation, the fraction of runoff that is converted
to infiltration and ET is increased further, enabling the site to maintain its current performance for
outflow volume.
B-133
-------
Runoff Fate
u
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
53.52
62.74
64.94
¦ ET
1.93
2.22
2,97
¦ Outflow
57.99
69.97
2.54
5.00
2.05
Figure B 162. Current and future partitioning of runoff fate for the Harford
County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity plus
20%) scenario.
The following figures indicate that the combination of practices in the GI with Gray Infrastructure
scenario achieves a high load reduction for sediment, TN, and TP, and that resizing the practices for the
future (intensity plus 20%) climate enables loads to be maintained at or below their current levels.
B-134
-------
Annual Average Sediment Load
30,000
25,000
20,000
15,000
10,000
5,000
27,366
1,554
3,013
1,377
Current Future Current Future with Future,
Untreated Untreated with BMPs 6MPs adapted
IB MPs
Figure B 163. Current and future performance for annual average sediment load,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 20%) scenario.
Annual Average TN Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-164. Current and future performance for annual average total nitrogen
(TN) load, Harford County, MD Green Infrastructure (GI) with Gray
Infrastructure (intensity plus 20%) scenario.
B-135
-------
Annual Average TP Load
84.14
87.14
80
~7f\
/u
60
jU
_q /in
— 4U
30
1A
z u
1 n
1.36 2.65
1.10
1U
u
Current
Untreated
Future
Untreated
Current with Future with
BMPs BMPs
Future,
adapted
BMPs
Figure B-165. Current and future performance for annual average total
phosphorous (TP) load, Harford County, MD Green Infrastructure (GI) with Gray
Infrastructure (intensity plus 20%) scenario.
The FDC evaluation for the GI with Gray Infrastructure scenario suggests that the practices, when resized
for future climate (intensity plus 20%) adaptation, produce a flow duration response that is nearly
identical to the current climate FDC within the evaluated range of outflows.
B-136
-------
25
20
ST
u
1 15
5
3
o
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Flourly Flow
\
Vs
>¦ 10
3
O
X
c,
0
0.0(
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-166. Flow duration curve (FDC) evaluation for current and future climate,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 20%) scenario.
With adaptation, the future (intensity plus 20%) practice footprints are also successful at reducing
maximum hourly peak flow from the site to 16.9 cfs, which is slightly lower than the current hourly peak
flow of 17.0 cfs.
B-137
-------
Max Peak Flow
41.02
33^93
21.09
17.03 16.87
11 i"l J 3
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-167. Current and future performance for maximum hourly peak flow,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 20%) scenario.
Resizing the Green and Gray practices for the future "intensity plus 20%" climate produces a monthly
outflow response that is nearly identical to the current BMP monthly outflow response.
45
40
35
30
« 25
" 20
15
10
5
0
B-138
-------
Monthly Outflow Volume
1.4
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-168. Current and future performance for monthly outflow volume,
Harford County, MD Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 20%) scenario.
B.2.1.3. Conventional (Gray) Infrastructure with Distributed Green Infrastructure
(GI)
Intensity Change Minus 10%
This scenario was not selected for adaptation simulation; refer to discussion in Section B.2.1.1. .
Intensity Change Plus 10%
As discussed in Section B.2.1.1. . the ''intensity plus 10%" climate scenario for Harford County results is
an increase both in total runoff volume and in outflow volume, as shown in Figure B-169. The objective
of the Conventional (Gray) Infrastructure with Distributed GI scenario is to implement distributed GI
practices without resizing the conventional (Gray) practice as a means of future climate adaptation. Figure
B-169 indicates that the addition of distributed infiltration trenches to the conventional site is able to
reduce future outflow below the current climate outflow by increasing the partitioning of runoff to
infiltration.
B-139
-------
Runoff Fate
u
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
47.28
49.73
53.59
¦ ET
1.42
1.52
1.45
¦ Outflow
60.45
66.51
11.74
15.25
11.47
Figure B-169. Current and future partitioning of runoff fate for the Harford
County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 10%) scenario.
Figure B-170 indicates that the distributed infiltration trench practices combined with the conventional
practices (surface sand filters and extended dry detention basin) are able to achieve high load reductions
for sediment, and allow the site to meet current loading for annual average sediment load without
requiring resizing of the current conventional practices.
B-140
-------
Annual Average Sediment Load
24,690
25,000
20,000
15,000
10,000
5,000
5,864
4,610
4,461
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-170. Current and future performance for annual average sediment load,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 10%) scenario.
The following figures suggest that the distributed green practices (infiltration trenches) combined with the
conventional practices (surface sand filters and extended dry detention basin) are able to achieve large
load reductions for TN and TP, and allow the site to meet current loading for annual average sediment
load without requiring resizing of the current conventional practices.
Annual Average TN Load
500
450
400
350
300
"> 250
~ 200
150
100
50
0
474.4
54.9 71 0 53.3
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-171. Current and future performance for annual average total nitrogen
(TN) load, Harford County, MD Conventional (Gray) Infrastructure with
Distributed Green Infrastructure (GI) (intensity plus 10%) scenario.
B-141
-------
Annual Average TP Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-172. Current and future performance for annual average total
phosphorous (TP) load, Harford County, MD Conventional (Gray) Infrastructure
with Distributed Green Infrastructure (GI) (intensity plus 10%) scenario.
The FDC evaluation for the Conventional (Gray) Infrastructure with Distributed GI scenario suggests that
when distributed infiltration trenches are added to the current conventional practices for future climate
(intensity plus 10%) adaptation, the combination produces a flow duration response that is nearly
identical to the current climate FDC within the evaluated range of outflows.
B-142
-------
40
35
30
£
u
i25
1 20
0
5-
1 15
o
X
10
5
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
¦2-yr Hourly Flow
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-173. Flow duration curve (FDC) evaluation for current and future climate,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 10%) scenario.
As demonstrated in Figure B-174, the addition of distributed GI for future (intensity plus 10%) climate
adaptation results in a future adapted maximum hourly peak flow that is slightly lower than the current
maximum hourly peak flow.
B-143
-------
Max Peak Flow
37.62
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-174. Current and future performance for maximum hourly peak flow,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 10%) scenario.
The addition of distributed GI practices for the future "intensity plus 10%" climate also produces a
monthly outflow response that is nearly identical to the current BMP monthly outflow response.
B-144
-------
Monthly Outflow Volume
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-175. Current and future performance for monthly outflow volume,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 10%) scenario.
Intensity Change Plus 20%
As discussed in Section B.2.1.1. , the "intensity plus 20%" climate scenario for Harford County results is
an increase both in total runoff volume and in outflow volume, as shown in Figure B-176. The objective
of the Conventional (Gray) Infrastructure with Distributed GI scenario is to implement distributed GI
practices without resizing the conventional (Gray) practice as a means of future climate adaptation. Figure
B-176 indicates that the addition of distributed infiltration trenches to the conventional site is able to
reduce future outflow below the current climate outflow by increasing the partitioning of runoff to
infiltration.
B-145
-------
Runoff Fate
u
Current
Untreated
Future
Untreated
Current
with BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
47.28
52.09
62,09
¦ ET
1.42
1.62
1.44
¦ Outflow
60.45
72.70
11.74
19.00
9.17
Figure B-176. Current and future partitioning of runoff fate for the Harford
County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 20%) scenario.
The following figures demonstrate that the distributed infiltration trench practices combined with the
conventional practices (surface sand filters and extended dry detention basin) are able to achieve high
load reductions for sediment, TN, and TP, and allow the site to meet current loading without requiring
resizing of the current conventional practices.
B-146
-------
Annual Average Sediment Load
30,000
25,000
20,000
15,000
10,000
5,000
27,375
7,179
4,610
3,481
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-177. Current and future performance for annual average sediment load,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 20%) scenario.
Annual Average TN Load
600
500
400
300
200
100
520.1
54.9
88.1
41.7
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-178. Current and future performance for annual average total nitrogen
(TN) load, Harford County, MD Conventional (Gray) Infrastructure with
Distributed Green Infrastructure (GI) (intensity plus 20%) scenario.
B-147
-------
Annual Average TP Load
91.17
1UU i
90
80
50.3/
70
60
>¦
> 50
40
30
20
_ 10 13
10
L J
u
Current
Future
Current with Future with Future,
Untreated
Untreated
BMPs BMPs adapted
BMPs
Figure B-179. Current and future performance for annual average total
phosphorous (TP) load, Harford County, MD Conventional (Gray) Infrastructure
with Distributed Green Infrastructure (GI) (intensity plus 20%) scenario.
The FDC evaluation for the Conventional (Gray) Infrastructure with Distributed GI scenario suggests that
when distributed infiltration trenches are added to the current conventional practices for future climate
(intensity plus 20%) adaptation, the combination produces a flow duration response that is nearly
identical to the current climate FDC within the evaluated range of outflows.
B-148
-------
•Current with BMPs
¦ Future with BMPs
Future, adapted BMPs
¦ 2-yr Hourly Flow
40
35
30
25
20
42
I
5
3
0
5-
1 15
0
1
10
0
0.000%
0.001% 0.002% 0.003% 0.004% 0.005%
Percent of time flow is equaled or exceeded
0.006%
0.007%
Figure B-180. Flow duration curve (FDC) evaluation for current and future climate,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 20%) scenario.
Although adding distributed GI to the conventional practices for future (intensity plus 20%) is unable to
reduce maximum hourly peak flow to the "current with BMPs" hourly peak flow, the future adapted
hourly peak flow is within less than 1% of the current hourly peak flow.
B-149
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-181. Current and future performance for maximum hourly peak flow,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 20%) scenario.
Figure B-182 indicates that for the "intensity plus 20%" climate scenario, the addition of distributed GI to
the current conventional site results in monthly outflow volumes that are lower for the future adapted
climate scenario than the "current with BMPs" scenario for all months.
B-150
-------
Monthly Outflow Volume
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-182. Current and future performance for monthly outflow volume,
Harford County, MD Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 20%) scenario.
B.2.1.4. Sensitivity Analysis Adaptation Summary
Table B-23 summarizes the increases in BMP footprints for all of the Harford County stormwater
management scenarios that would be required to maintain current performance under future climate
conditions for the "intensity plus 10%"' climate simulation.
B-151
-------
Table B-23. Comparison of current and future adapted best management practice
(BMP) footprints, Harford County, MD stormwater management scenarios,
intensity plus 10%
Location
Climate
scenario
Stormwater
management
scenario
Practice
Current
Future
adapted
% increase
in
footprint
Footprint
SF
Footprint
as % of
site area
Footprint
SF
Footprint
as % of
site area
Harford
County,
MD
Intensity
change
plus 10%
Conventional
(Gray)
Infrastructure
Extended
dry
detention
basin
25,000
2.9
25,000
2.9
0
Surface
sand filters
10,119
1.2
20,023
2.3
98
Gl with Gray
Infrastructure
Dry
detention
basin
10,000
1.1
10,000
1.1
0
Infiltration
basin
12,858
1.5
18,943
2.2
47
Infiltration
trench
14,800
1.7
20,435
2.3
38
Permeable
pavement
201,242
23.1
201,242
23.1
0
Conventional
(Gray)
Infrastructure
with
Distributed
Gl
Extended
dry
detention
basin
25,000
2.9
25,000
2.9
0
Surface
sand filters
10,119
1.2
10,119
1.2
0
Distributed
infiltration
trench
0
0.0
15,351
1.8
In the Conventional (Gray) Infrastructure scenario, the SUSTAIN optimization selected the extended dry
detention basin only for adaptation. This outcome is somewhat surprising and follows a different pattern
than the adaptations to the "20 Watersheds' -based future climate scenarios, where both sand filters and
the detention basin were selected for resizing during the optimization as being the most cost-effective
solution. The required basin footprint for future (intensity plus 10%) climate adaptation reflects a near
doubling of size.
In the GI with Gray Infrastructure scenario, the SUSTAIN optimization targeted resizing only the
infiltration basins and infiltration trenches, the opposite outcome as seen for the Conventional scenario.
This is likely due in part to sediment and TP loads being the limiting factor, resulting in the selection of
practices with the best infiltration capacity and load reduction. The required infiltration basin footprint
B-152
-------
reflects nearly a 50% increase, and nearly a 40% increase is required in the infiltration trench footprint.
For the Conventional (Gray) Infrastructure with Distributed GI scenario, the addition of 15,351 square
feet of distributed infiltration trenches would be needed to maintain current BMP performance. This
footprint represents approximately 1.8% of the total site area.
Table B-24 summarizes the increases in BMP footprints that would be required to maintain current
performance under future climate conditions for the "intensity plus 20%" climate simulation.
Table B-24. Comparison of current and future adapted best management practice
(BMP) footprints, Harford County, MD stormwater management scenarios,
intensity plus 20%
Location
Climate
scenario
Stormwater
management
scenario
Practice
Current
Future
adapted
% increase
in
footprint
Footprint
SF
Footprint
as % of
site area
Footprint
SF
Footprint
as % of
site area
Harford
County,
MD
Intensity
change
plus
20%
Conventional
(Gray)
Infrastructure
Extended
dry
detention
basin
25,000
2.9
25,000
2.9
0
Surface
sand filters
f0,ff9
f .2
28,043
3.2
111
Gf with Gray
fnfrastructure
Dry
detention
basin
f0,000
f.f
f0,000
f.f
0
tnfiltration
basin
f2,858
f .5
27,846
3.2
f f 7
tnfiltration
trench
f4,800
f .7
23,350
2.7
58
Permeable
pavement
20 f,242
23.f
20 f,242
23.f
0
Conventional
(Gray)
fnfrastructure
with
Distributed
Gf
Extended
dry
detention
basin
25,000
2.9
25,000
2.9
0
Surface
sand filters
f0,ff9
f .2
f0,ff9
f .2
0
Distributed
infiltration
trench
0
0.0
32,5f4
3.7
In the Conventional (Gray) Infrastructure scenario, the extended dry detention basin footprint must
increase by a factor of 2.8 for future (intensity plus 20%) climate adaptation. In the GI with Gray
Infrastructure scenario, the required infiltration basin footprint reflects a more than doubling in size, and
B-153
-------
nearly a 60% increase is required in the infiltration trench footprint. For the Conventional (Gray)
Infrastructure with Distributed GI scenario, the addition of 32,514 square feet of distributed infiltration
trenches would be required to maintain current performance. This footprint represents approximately
3.7% of the total site area.
B.2.1.5. Cost of Adaptation
Table B-25 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for all three Harford County stormwater management scenarios under future "intensity
plus 10%" climate, as well as the percentage increase in cost (current to future adapted) and increase
(millions of dollars) per acre of site.
Table B-25. Comparison of the current and future estimated 20-year present value
costs for the Harford County, MD stormwater management scenarios, Intensity
Change Plus 10%
Location
Climate
scenario
Stormwater
management
scenario
Current cost
20-yr present
value,
$millions
Future
adapted cost
20-yr present
value,
$millions
Increase in
cost (20-yr
present
value,
$millions)
% increase
in cost
Increase per
acre of site
$millions
Harford
County,
MD
plus
10%
Conventional
(Gray)
Infrastructure
5.3 f
8.00
2.69
5f
0. f 3
Gf with Gray
fnfrastructure
5. f 5
6.24
f .09
2f
0.05
Conventional
(Gray)
fnfrastructure
with Distributed
Gf
5.3 f
6.93
f .62
30
0.08
Table B-26 provides an estimate of cost for the future "intensity plus 20%" climate scenario.
B-154
-------
Table B-26. Comparison of the current and future estimated 20-year present value
costs for the Harford County, MD stormwater management scenarios, Intensity
Change Plus 20%
Location
Climate
scenario
Stormwater
management
scenario
Current cost
20-yr present
value,
$millions
Future
adapted cost
20-yr present
value,
$millions
Increase in
cost (20-yr
present value,
$millions)
% increase
in cost
Increase
per acre of
site
$millions
Harford
County,
MD
Intensity
change
plus 20%
Conventional
(Gray)
Infrastructure
5.3 f
f 0. f 7
4.86
92
0.24
GI with Gray
Infrastructure
5. f 5
7.29
2. f 3
4f
O.ff
Conventional
(Gray)
fnfrastructure
with
Distributed Gf
5.3 f
8.86
3.55
67
0.f8
B.2.2. Scott County, MN
B.2.2.1. Conventional (Gray) Infrastructure
Intensity Change Minus 10%
As discussed in Section 4.3. of the report, the sensitivity analysis entailed modifying the current
precipitation record to represent potential future climate conditions by applying a graduated set of
percentage changes to the current precipitation record. For this particular sensitivity scenario, the
resulting change in annual runoff volume and pollutant load was a decrease under future climate
compared to the current climate, as illustrated in the figures below. For these reasons, this climate
scenario was not investigated for adaptation simulation.
B-155
-------
Runoff Fate
£
i
u
ro
35
30
25
20
15
10
5
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
¦ Infiltration
0.00
0.00
¦ ET
1.46
1.36
¦ Outflow
34.10
30.32
32.59
28.92
Figure B-183. Current and future partitioning of runoff fate for the Scott County,
MN Conventional (Gray) Infrastructure (intensity minus 10%) scenario.
Annual Average Sediment Load
38.376
>
_Q
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
7,370
5,578
Current Future Current with
Untreated Untreated BMPs
Future with
BMPs
Figure B-184. Current and future performance for annual average sediment load,
Scott County, MN Conventional (Gray) Infrastructure (intensity minus 10%)
scenario.
B-156
-------
Annual Average TN Load
298.6
0
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-185. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Conventional (Gray) Infrastructure (intensity minus
10%) scenario.
Annual Average TP Load
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-186. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Conventional (Gray) Infrastructure
(intensity minus 10%) scenario.
B-157
-------
Max Peak Flow
80
70
60
50
•S 40
30
20
10
0
Figure B-187. Current and future performance for maximum hourly peak flow,
Scott County, MN Conventional (Gray) Infrastructure (intensity minus 10%)
scenario.
Monthly Outflow Volume
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs
Figure B-188. Current and future performance for monthly outflow volume, Scott
County, MN Conventional (Gray) Infrastructure (intensity minus 10%) scenario.
Current Future Current with Future with
Untreated Untreated BMPs BMPs
B-158
-------
Intensity Change Plus 10%
This sensitivity analysis of a 10% increase in intensity resulted in an increase of both total runoff volume
and outflow volume. As discussed previously, the soils of the Scott County, MN site are poorly
infiltrating. The primary function of the conventional practice (wet pond) is to provide storage and peak
flow control and to some degree ET. For these reasons, outflow is the dominant runoff fate pathway under
both current and future climate conditions. For future adaptation, increasing the wet pond footprint
increases the partitioning of runoff to ET due to the increased surface area. As a result, outflow is
decreased below the current climate outflow.
Because the practice resizing for the future "intensity plus 10%" climate was driven by the required
reduction in outflow volume, the future adapted BMP annual average loads for sediment (see Figure
B-190), TP (see Figure B-191), and TN (see Figure B-192) are all well below the "current with BMPs"
loads due to the relatively large footprint required to meet the flow performance measure.
40
35
30
25
20
15
10
5
Runoff Fate
6
ra
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
0.00
0.00
0.00
¦ ET
1.46
1.58
5.18
¦ Outflow
34.10
37.90
32.59
36.28
32.57
Figure B-189. Current and future partitioning of runoff fate for the Scott County,
MN Conventional (Gray) Infrastructure (intensity plus 10%) scenario.
B-159
-------
Annual Average Sediment Load
45,000
40,000
35,000
30,000
25,000
£ 20,000
15,000
10,000
5,000
0
38,376
4
1
4,695
1
1
1
l
l
7,370
1 | J ft
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-190. Current and future performance for annual average sediment load,
Scott County, MN Conventional (Gray) Infrastructure (intensity plus 10%)
scenario.
Annual Average TN Load
350 298,6
300
317.9
250
200
- 150
100
50
137.7
I
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-191. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Conventional (Gray) Infrastructure (intensity plus
10%) scenario.
B-160
-------
Annual Average TP Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-192. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Conventional (Gray) Infrastructure
(intensity plus 10%) scenario.
As discussed above, the increase in footprint of the wet pond for adaptation in the future "intensity plus
10%" climate scenario was primarily driven by the required reduction in outflow volume that would be
necessary to maintain the current climate outflow. As a result, the increased wet pond size produces a
"future, adapted BMPs" flow duration curve that is reduced well below the "current with BMPs" curve
for almost the entire range of flows evaluated.
B-161
-------
-Current with BMPs
Future with BMPs
Future, adapted BMPs 2-yr Hourly Flow
60
0
0.000%
0.001%
0.002% 0.003% 0.004% 0.005%
Percent of time flow is equaled or exceeded
0.006%
0.007%
Figure B-193. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Conventional (Gray) Infrastructure (intensity plus 10%)
scenario.
Figure B-194 indicates that the future adapted ("intensity plus 10%") wet pond results in a reduction in
maximum hourly peak flow due to the increased sizing of the practice to maintain current performance for
outflow.
B-162
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-194. Current and future performance for maximum hourly peak flow,
Scott County, MN Conventional (Gray) Infrastructure (intensity plus 10%)
scenario.
Due to the resizing of the wet pond being driven by the required reduction in outflow for the future
"intensity plus 10%" climate scenario, the resulting adaptation produces a monthly outflow volume
response that is very similar to the current monthly outflow response.
B-163
-------
Monthly Outflow Volume
6
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-195. Current and future performance for monthly outflow volume, Scott
County, MN Conventional (Gray) Infrastructure (intensity plus 10%) scenario.
Intensity Change Plus 20%
As discussed in Section 4.3. of the report, the sensitivity analysis entailed modifying the current
precipitation record to represent potential future climate conditions by applying a graduated set of
percentage changes across the entire record. For this particular sensitivity scenario, the result was an
increase, both in total runoff volume and in outflow volume. For future adaptation, increasing the wet
pond footprint increases the partitioning of runoff to ET due to the increased surface area. As a result,
outflow is decreased below the current climate outflow.
B-164
-------
Runoff Fate
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
0.00
0.00
0.00
¦ ET
1.46
1.70
8.93
¦ Outflow
34.10
41.73
32.59
39.99
32.56
Figure B-196. Current and future partitioning of runoff fate for the Scott County,
MN Conventional (Gray) Infrastructure (intensity plus 20%) scenario.
Because the practice resizing for the future "intensity plus 20%" climate was driven by the required
reduction in outflow volume, the future adapted BMP annual average loads for sediment, TN, and TP are
all well below the "current with BMPs" loads due to the relatively large footprint required to meet the
flow performance measure, as shown in the following figures.
B-165
-------
Annual Average Sediment Load
60,000
50,000
40,000
30,000
20,000
10,000
0
51,887
11,891
7,370
m m
1,519
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-197. Current and future performance for annual average sediment load,
Scott County, MN Conventional (Gray) Infrastructure (intensity plus 20%)
scenario.
Annual Average TN Load
336.9
350
300
250
200
- 150
100
50
110.4
a
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-198. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Conventional (Gray) Infrastructure (intensity plus
20%) scenario.
B-166
-------
Annual Average TP Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-199. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Conventional (Gray) Infrastructure
(intensity plus 20%) scenario.
The increased wet pond size for the future ("intensity plus 20%") climate adaptation produces a flow
duration curve that is reduced below the "current with BMPs" curve for the entire range of outflows
evaluated.
B-167
-------
120
100
'S 80
1
1 60
O
_>
o 40
X
20
Current with BMPs Future with BMPs — — Future, adapted BMPs
¦2-yr Hourly Flow
•s.
\
v
\
0
0.0(
W *
^
» — — — _
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-200. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Conventional (Gray) Infrastructure (intensity plus 20%)
scenario.
Figure B-201 indicates that the future adapted ("intensity plus 20%") wet pond results in a reduction in
maximum hourly peak flow due to the increased sizing of the practice to maintain current performance for
outflow.
B-168
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-201. Current and future performance for maximum hourly peak flow,
Scott County, MN Conventional (Gray) Infrastructure (intensity plus 20%)
scenario.
The future adapted conventional practice (wet pond) footprint for the "intensity plus 20%" climate
scenario results in monthly outflow volumes that are less than, very close to, or only slightly higher than
current outflows for all months.
B-169
-------
Monthly Outflow Volume
7
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-202. Current and future performance for monthly outflow volume, Scott
County, MN Conventional (Gray) Infrastructure (intensity plus 20%) scenario.
B.2.2.2. Green Infrastructure (Gl) with Gray Infrastructure
Intensity Change Minus 10%
This scenario was not selected for adaptation simulation; refer to discussion in Section B.2.2.1. and the
following figures, which demonstrate decreased outflow volume and pollutant loading in the "intensity
minus 10%" climate scenario compared to the current climate.
B-170
-------
Runoff Fate
r.
i
U
CO
40
35
30
25
20
15
10
5
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
¦ Infiltration
7.45
6.97
¦ ET
2.32
2.16
¦ Outflow
35.68
31.75
25.90
22.62
Figure B-203. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) with Gray Infrastructure (intensity minus 10%)
scenario.
Annual Average Sediment Load
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
39T78
'1 A OQC
11,749
Current
Untreated
Future Current with Future with
Untreated BMPs BMPs
Figure B-204. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity
minus 10%) scenario.
13-171
-------
Annual Average TN Load
0
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-205. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) with Gray Infrastructure
(intensity minus 10%) scenario.
Annual Average TP Load
Current with Future with
BMPs BMPs
Current
Untreated
Future
Untreated
Figure B-206. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Green Infrastructure (GI) with Gray
Infrastructure (intensity minus 10%) scenario.
B-172
-------
Max Peak Flow
80
70
60
50
40
30
20
10
0
62.69
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Figure B-207. Current and future performance for maximum hourly peak flow,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity
minus 10%) scenario.
Monthly Outflow Volume
4.5
4
3.5
3
¦C
c
o
2.5
JE
d?
2
u
ro
1.5
1
0.5
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs — Future with BMPs
Figure B-208. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity minus
10%) scenario.
B-173
-------
Intensity Change Plus 10%
As discussed in Section B.2.2.1. , the "intensity plus 10%" climate scenario for Scott County results is an
increase both in total runoff volume and in outflow volume, as shown in Figure B-209. Because the soils
in the Scott County site are poorly infiltrating, the majority of the increase in runoff volume partitions
into outflow. When the green (bioretention) and gray (dry detention basin) practices are resized for future
adaptation, the fraction of runoff that is converted to infiltration and ET is increased, enabling the site to
maintain its current performance for outflow volume.
Runoff Fate
ID]]]
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
7.45 7.70 10.90
2.32 2.49 4.11
35.68 39.64 25.90 29.45 24.64
Figure B-209. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) with Gray Infrastructure (intensity plus 10%)
scenario.
The following figures demonstrate that the adapted Green (bioretention) and Gray (dry detention basin)
practices in the Scott County GI with Gray Infrastructure scenario are able to mitigate the increases in
annual average sediment (see Figure B-210), TN (see Figure B-211), and TP (see Figure B-212) load
under future "intensity plus 10%" climate conditions.
40
35
30
l_
25
>-
d?
20
u
ro
15
10
5
0
¦ Infiltration
¦ ET
¦ Outflow
B-174
-------
Annual Average Sediment Load
46'^
45,000
40,000
35,000
30,000
25,000 _ 18,401
20,000
15,000
10,000
5,000
0
14,895 14,849
a a a
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-210. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 10%) scenario.
Annual Average TN Load
333.2
350
300
250
- 200
£ 150 109.1 12CU
100
50
312.9
lua-L 99.6
I I ¦
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-211. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) with Gray Infrastructure
(intensity plus 10%) scenario.
B-175
-------
Annual Average TP Load
47.34
50
45
40
35
30
25
20
15
10
17.48
14.84 14 47
¦ I ¦
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-212. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Green Infrastructure (GI) with Gray
Infrastructure (intensity plus 10%) scenario.
The flow duration curve analysis (see Figure B-213) for the Scott County GI with Gray Infrastructure
("intensity plus 10%") scenario suggests that the adapted practices are able to reasonably match the
outflow response of the current practices within the evaluated range of flows.
B-176
-------
90
80
70
2"
u 60
J 50
O 40
>¦
§ 30
X
20
10
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Hourly Flow
\\
v\
A
\\
x
NX
s\
>
- -¦
]0% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-213. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 10%) scenario.
Although maintaining maximum hourly peak flow at current performance is not an objective of the
adaptation simulation, Figure B-214 indicates that the adapted Green and Gray practices are able to
reduce hourly peak flow to some extent compared to the future (not adapted) practice sizing.
B-177
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-214. Current and future performance for maximum hourly peak flow,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 10%) scenario.
The future ("intensity plus 10%") climate adaptation produces a monthly outflow volume response that is
very similar to the current monthly outflow response. The greatest observed discrepancy is in March,
when the future adapted monthly outflow volume is lower than the current monthly outflow volume by
more than 1 acre-foot.
B-178
-------
Monthly Outflow Volume
5
4.5
4
3.5
-n 3
£ 3
o
£ 2 5
-----
tz
U 2
ro
1.5
1
0.5
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-215. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity plus
10%) scenario.
Intensity Change Plus 20%
As discussed in Section B.2.2.1. , the "intensity plus 20%" climate scenario for Scott County results is an
increase both in total amoff volume and in outflow volume, as shown in Figure B-216. Because the soils
in the Scott County site are poorly infiltrating, the majority of the increase in amoff volume partitions
into outflow. When the Green (bioretention) and Gray (dry detention basin) practices are resized for
future adaptation, the fraction of runoff that is converted to infiltration and ET is increased, enabling the
site to maintain its current performance for outflow volume, although outflow remains the dominant
runoff fate pathway .
B-179
-------
Runoff Fate
u
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
7.45
7.90
13.44
¦ ET
2.32
2.67
6.30
¦ Outflow
35.68
43.65
25.90
33.08
23.91
Figure B-216. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) with Gray Infrastructure (intensity plus 20%)
scenario.
The following figures demonstrate that the adapted Green (bioretention) and Gray (dry detention basin)
practices in the Scott County GI with Gray Infrastructure scenario are able to mitigate the increases in
annual average sediment (see Figure B-217), TN (see Figure B-218), and TP (see Figure B-219) load
under future "intensity plus 20%" climate conditions.
B-180
-------
Annual Average Sediment Load
60,000
50,000
40,000
30,000
20,000
10,000
0
53,496
22,633
14,895 fll 14,478
J If
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-217. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 20%) scenario.
Annual Average TN Load
400
350
300
250
200
150
100
50
353.3
109.1
131,8
I
94.5
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-218. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) with Gray Infrastructure
(intensity plus 20%) scenario.
B-181
-------
Annual Average TP Load
60 51.81
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-219. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Green Infrastructure (GI) with Gray
Infrastructure (intensity plus 20%) scenario.
The flow duration curve analysis (see Figure B-220) for the Scott County GI with Gray Infrastructure
("intensity plus 20%") scenario suggests that the adapted practices are able to reasonably match the
outflow response of the current practices within the evaluated range of flows.
B-182
-------
120
100
'S 80
1
1 60
O
_>
o 40
X
20
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Hourly Flow
V\
— —,
_
— — —
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-220. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 20%) scenario.
Although maintaining maximum hourly peak flow at current performance as not an objective of the
adaptation simulation, Figure B-221 indicates that the adapted Green and Gray practices are able to
slightly reduce hourly peak flow compared to the future (not adapted) practice sizing.
B-183
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-221. Current and future performance for maximum hourly peak flow,
Scott County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity
plus 20%) scenario.
The adapted practice footprints in the GI with Gray Infrastructure scenario are able to reduce the future
("intensity plus 20%") monthly outflow volumes so that they are approximately the same as, or lower
than, the current monthly outflow volumes for all months.
B-184
-------
Monthly Outflow Volume
6
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-222. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) with Gray Infrastructure (intensity plus
20%) scenario.
B.2.2.3. Green Infrastructure (GI) Only
Intensity Change Minus 10%
This scenario was not selected for adaptation simulation; refer to discussion in Section B.2.2.1. and the
following figures, which demonstrate decreased outflow volume and pollutant loading in the '"intensity
minus 10%" climate scenario compared to the current climate.
B-185
-------
Runoff Fate
35
30
25
l_
>-
20
ir
Q
ro
15
10
5
0
i Infiltration
IET
I Outflow
Current
Untreated
34.78
Future
Current with
Future with
Untreated
BMPs
BMPs
13.54
12.51
1.95
1.82
30.94
19.28
16.61
Figure B-223. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) Only (intensity minus 10%) scenario.
Annual Average Sediment Load
>
_Q
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
38.990
11,914
9426
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-224. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) Only (intensity minus 10%) scenario.
B-186
-------
Annual Average TN Load
350
300
250
I—
200
_Q
150
100
50
0
305.1
71.6
61.9
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Figure B-225. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) Only (intensity minus 10%)
scenario.
Annua! Average TP Load
42.43
45
40
35
30
> 25
£ 20
15
10
5
0
10.90
8.49
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Figure B-226. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Green Infrastructure (GI) Only
(intensity minus 10%) scenario.
B-187
-------
Max Peak Flow
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Figure B-227. Current and future performance for maximum hourly peak flow,
Scott County, MN Green Infrastructure (GI) Only (intensity minus 10%) scenario.
Monthly Outflow Volume
3.5
3
2.5
-C
c
o
2
.£
£
1.5
(D
1
0.5
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Current with BMPs Future with BMPs
Figure B-228. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) Only (intensity minus 10%) scenario.
B-188
-------
Intensity Change Plus 10%
As discussed in Section B.2.2.1. , the "intensity plus 10%" climate scenario for Scott County results is an
increase both in total runoff volume and in outflow volume when the green practices (bioretention,
rooftop downspout disconnection, and permeable pavement) are not resized for adaptation, as shown in
Figure B-229. With future adaptation, the increased practice footprints reduce the volume of outflow
below the "current with BMPs" scenario due to a larger proportioning of runoff to ET and infiltration.
The soils in the Scott County study site have low infiltration capacity, so outflow remains the dominant
runoff pathway, even for this GI Only scenario and even when practice sizes are increased.
Runoff Fate
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
13.54 14.26 16.88
1.95 2.09 3.79
34.78 38.65 19.28 22.30 17.99
Figure B-229. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) Only (intensity plus 10%) scenario.
The following figures demonstrate that the adapted green practices in the Scott County GI Only scenario
are able to mitigate the increases in annual average sediment (see Figure B-230), TN (see Figure B-231),
and TP (see Figure B-232) load under future 'intensity plus 10%" climate conditions.
40
35
30
l_
25
>-
£
20
u
ro
15
10
5
0
¦ Infiltration
¦ ET
¦ Outflow
B-189
-------
Annual Average Sediment Load
50,000 45i362_
45,000
40,000
35,000
30,000
25,000
20,000 11 q-id
15,000
10,000
5,000
0
11,914 11,869
J I 1
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-230. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) Only (intensity plus 10%) scenario.
Annual Average TN Load
350
300
250
- 200
^ 150
100
50
325.0
305.1
71.6
80.6
63.5
i i i
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
Current Future
Untreated Untreated
Figure B-231. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) Only (intensity plus 10%)
scenario.
B-190
-------
Annual Average TP Load
46.74
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-232. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Green Infrastructure (GI) Only
(intensity plus 10%) scenario.
The flow duration curve analysis (see Figure B-233) for the Scott County GI Only (''intensity plus 10%")
scenario suggests that the adapted practices are able to reasonably match the outflow response of the
current practices within the evaluated range of flows.
B-191
-------
90
80
70
2"
IE60
J 50
O 40
_>¦
§ 30
X
20
10
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Hourly Flow
\
\S
s \
-
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-233. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Green Infrastructure (GI) Only (intensity plus 10%) scenario.
Figure B-234 indicates that, although maximum hourly peak flow was not targeted as part of the
adaptation simulation, the green practices, when adapted to meet the other performance measures, are
only able to reduce the hourly peak flow by about 2 cfs for the "intensity plus 10%" climate scenario.
This result may suggest a lower ability of GI to mitigate peak flows compared to conventional (Gray)
infrastructure.
B-192
-------
Max Peak Flow
84.09
85
80
75
70
65
60
78.81
70.50
J
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-234. Current and future performance for maximum hourly peak flow,
Scott County, MN Green Infrastructure (GI) Only (intensity plus 10%) scenario.
The future ("intensity plus 10%") climate adaptation for the GI Only scenario produces a monthly
outflow volume response that is very similar to the current monthly outflow response.
Monthly Outflow Volume
4
3.5
3
2.5
2
£
ra 1.5
1
0.5
0
Jan Feb Mar Apr Mav Jun Jul Aug Sep Oct Nov Dec
•Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-235. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) Only (intensity plus 10%) scenario.
B-193
-------
Intensity Change Plus 20%
As discussed in Section B.2.2.1. , the "intensity plus 20%" climate scenario for Scott County results is an
increase both in total runoff volume and in outflow volume when the practices are not resized for
adaptation, as shown in Figure B-23936. Increasing the green practice (bioretention, rooftop downspout
disconnection, and permeable pavement) footprints increases the proportioning of runoff to infiltration
and ET, allowing the site to maintain the current climate outflow performance under future "intensity plus
10%" climate conditions. As discussed above, due to the poor infiltration capacity of the soils in the Scott
County study site, outflow is an important runoff fate pathway. However, the increased green practice
footprints in the future adapted scenario increase the proportioning of runoff to infiltration enough to
make infiltration the dominant fate pathway for the "intensity plus 10%" climate scenario.
Runoff Fate
Mm
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
13.54 14.88 20.55
1.95 2.24 5.87
34.78 42.57 19.28 25.45 16.15
Figure B-236. Current and future partitioning of runoff fate for the Scott County,
MN Green Infrastructure (GI) Only (intensity plus 20%) scenario.
Figure B-237, Figure B-238, and Figure B-239 demonstrate that with the increased practice sizes, the GI
Only site is able to mitigate the increased sediment, TN, and TP loads due to future climate impacts in the
"intensity plus 20%" climate scenario.
ce
45
40
35
30
25
20
15
10
5
0
. Infiltration
I ET
I Outflow
B-194
-------
Annual Average Sediment Load
60,000 52,602
50,000
40,000
30,000
10,000
0
18,906
20,000 IB ¦ 11,914
11,914
I I
10,167
0
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-237. Current and future performance for annual average sediment load,
Scott County, MN Green Infrastructure (GI) Only (intensity plus 20%) scenario.
Annual Average TN Load
344.7
350
300
250
- 200
^ 150
100
50
0
Current Future
Untreated Untreated
Figure B-238. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Green Infrastructure (GI) Only (intensity plus 20%)
scenario.
71.6
90.0
1 m
Current with
Future with
Future,
BMPs
BMPs
adapted
BMPs
B-195
-------
Annual Average TP Load
60 51.30
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-239. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Green Infrastructure (GI) Only
(intensity plus 20%) scenario.
The FDC evaluation for the Scott County GI Only ("intensity plus 20%") scenario indicates that the green
practices alone are able to achieve a reasonably close flow response to the current climate FDC within the
evaluated range of flows. However, there is a discrepancy between the highest hourly peak flows,
indicating the adapted practices may be unable to mitigate the increase in the highest flows.
B-196
-------
120
100
'S 80
1
1 60
O
_>
o 40
X
20
0
0.0(
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Hourly Flow
\\
\
__
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-240. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Green Infrastructure (GI) Only (intensity plus 20%) scenario.
Figure B-241 provides additional insight into the behavior seen in the uppermost range of flows in the
flow duration curve analysis. The GI Only practices, even with adaptation, do not appear to significantly
reduce the maximum peak flow under future climate ("intensity plus 20%") conditions compared to the
original practice sizes ("future with BMPs").
B-197
-------
Max Peak Flow
100
90
80
70
60
50
40
30
20
10
0
92.05
75.70
86.84
83.46
70.50
Current Future Current with Future with
Untreated Untreated BMPs BMPs
Future,
adapted
BMPs
Figure B-241. Current and future performance for maximum hourly peak flow,
Scott County, MN Green Infrastructure (GI) Only (intensity plus 20%) scenario.
Figure B-242 indicates that with resizing, the future adapted GI Only practices are successful at
mitigating increased monthly outflow volumes under future ("intensity plus 20%") climate. The future
adapted monthly outflows are lower than, or very close to, the current monthly outflows for all months.
4.5
4
3.5
3
_c
c
o
2.5
JE
2
o
ro
1.5
1
0.5
0
Monthly Outflow Volume
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs Future, adapted BMPs
Figure B-242. Current and future performance for monthly outflow volume, Scott
County, MN Green Infrastructure (GI) Only (intensity plus 20%) scenario.
B-198
-------
B.2.2.4. Conventional (Gray) Infrastructure with Distributed Green Infrastructure
(Gl)
Intensity Change Minus 10%
As discussed in Section B.2.2.1. , the Conventional (Gray) Infrastructure scenario was not investigated for
adaptation in the future "intensity minus 10%" climate scenario because there was no decrease in the
performance of the conventional (Gray) practice (wet pond) between the current and future climate.
Therefore, the future low intensity climate scenario was also not investigated for adaptation through the
addition of distributed GI practices because there was no performance gap to address.
Intensity Change Plus 10%
The change in flow and pollutant related performance for the Scott County Conventional (Gray) scenario
due to future climate sensitivity ("intensity plus 10%") impacts was discussed in Section B.2.2.1. . The
purpose of the Conventional (Gray) Infrastructure with Distributed GI scenario is to implement
distributed GI practices without resizing the conventional (Gray) practice as a means of future climate
adaptation. Figure B- 243 indicates that the addition of distributed bioretention to the conventional site is
able to reduce outflow below the current climate outflow by increasing the partitioning of runoff to
infiltration and ET. Due to the poorly infiltrating soils on the Scott County site, outflow remains the
dominant runoff fate pathway for this scenario, even with the addition of infiltrating bioretention.
Runoff Fate
40
35
30
1—
25
>-
d?
20
u
ro
15
10
5
0
I Infiltration
I ET
I Outflow
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
0.00
0.00
4.48
1.46
1.58
2.40
34.10
37.90
32.59
36.28
30.97
Figure B- 243. Current and future partitioning of runoff fate for the Scott County,
MN Conventional (Gray) Infrastructure with Distributed Green Infrastructure (GI)
(intensity plus 10%) scenario.
B-199
-------
Figure B-244 indicates that the distributed bioretention practices combined with the wet pond are able to
achieve high load reductions for sediment, and allow the site to meet current loading for annual average
sediment load under "intensity plus 10%" climate without requiring resizing of the existing wet pond.
Annual Average Sediment Load
44,695
45,000 38,376
40,000
35,000
30,000
% 25,000
£ 20,000
15,000 ™ 9,374
/ 7,370 g 301
I
o
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-244. Current and future performance for annual average sediment load,
Scott County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 10%) scenario.
The following figures suggest that the distributed bioretention practices combined with the wet pond are
able to achieve modest load reductions for TN and TP, allowing the site to meet current loading for
annual average sediment load under the "intensity plus 10%" climate conditions without requiring
resizing of the existing wet pond.
B-200
-------
Annual Average TN Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-245. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Conventional (Gray) Infrastructure with Distributed
Green Infrastructure (GI) (intensity plus 10%) scenario.
Annual Average TP Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-246. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Conventional (Gray) Infrastructure with
Distributed Green Infrastructure (GI) (intensity plus 10%) scenario.
The FDC analysis for the future "intensity plus 10%" scenario suggests that although the addition of
distributed bioretention practices is able to maintain the current climate outflow volume, these practices
as designed are not effective at reducing the highest peak flow rates. As a result, the flow duration curves
between the "current with BMPs"' and "future, adapted BMPs" diverge, particularly for the highest
outflows.
B-201
-------
90
80
70
2"
IE60
J 50
O 40
_>¦
§ 30
X
20
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Hourly Flow
V
\
A
— %
\
10
0
0.0(
• " « ¦
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-247. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 10%) scenario.
Figure B-248 indicates that the addition of distributed bioretention to the conventional site does not result
in a significant decrease in the maximum hourly peak flow under future ("intensity plus 10%") climate
conditions.
B-202
-------
Max Peak Flow
84.58
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-248. Current and future performance for maximum hourly peak flow,
Scott County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 10%) scenario.
The future ("intensity plus 10%") climate adaptation for the Conventional (Gray) Infrastructure with
Distributed GI scenario produces a monthly outflow volume response that is very similar to the current
monthly outflow response.
B-203
-------
Monthly Outflow Volume
£ 3
----- J
£
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-249. Current and future performance for monthly outflow volume, Scott
County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 10%) scenario.
Intensity Change Plus 20%
The change in flow and pollutant-related performance for the Scott County Conventional (Gray) scenario
due to future climate sensitivity ("intensity plus 20%") impacts was discussed in Section B.2.2.1. . The
purpose of the Conventional (Gray) Infrastructure with Distributed GI scenario is to implement
distributed GI practices without resizing the conventional (Gray) practice as a means of future climate
adaptation.
Figure B-250 indicates that the addition of distributed bioretention to the conventional site is able to
reduce outflow below the current climate outflow by increasing the partitioning of runoff to infiltration
and ET. Due to the poorly infiltrating soils on the Scott County site, outflow remains the dominant runoff
fate pathway for this scenario, even with the addition of infiltrating bioretention.
B-204
-------
Runoff Fate
0
Current
Untreated
Future
Untreated
Current with
BMPs
Future with
BMPs
Future,
adapted
BMPs
¦ Infiltration
0.00
0.00
7.03
¦ ET
1.46
1.70
3.23
¦ Outflow
34.10
41.73
32.59
39.99
31.42
Figure B-250. Current and future partitioning of runoff fate for the Scott County,
MN Conventional (Gray) Infrastructure with Distributed Green Infrastructure (GI)
(intensity plus 20%) scenario.
Figure B-251 indicates that the distributed bioretention practices combined with the wet pond are able to
achieve high load reductions for sediment, and allow the site to meet current loading for annual average
sediment load under "intensity plus 20%" climate without requiring resizing of the existing wet pond.
B-205
-------
Annual Average Sediment Load
60,000 51,887
50,000
40,000
30,000
20,000 | 11,891
7,370 6,573
10,000
0
J
Current Future Current Future with Future,
Untreated Untreated with BMPs BMPs adapted
BMPs
Figure B-251. Current and future performance for annual average sediment load,
Scott County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 20%) scenario.
The following figures suggest that the distributed bioretention practices combined with the wet pond are
able to achieve modest load reductions for I N and TP, allowing the site to meet current loading for
annual average sediment load under future "intensity plus 20%" climate conditions without requiring
resizing of the existing wet pond.
Annual Average TN Load
336.9
350 29^5 A
1AA A
l l
150
0
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-252. Current and future performance for annual average total nitrogen
(TN) load, Scott County, MN Conventional (Gray) Infrastructure with Distributed
Green Infrastructure (GI) (intensity plus 20%) scenario.
B-206
-------
Annual Average TP Load
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-253. Current and future performance for annual average total
phosphorous (TP) load, Scott County, MN Conventional (Gray) Infrastructure with
Distributed Green Infrastructure (GI) (intensity plus 20%) scenario.
The FDC analysis for the future "intensity plus 20%" scenario suggests that although the addition of
distributed bioretention practices can maintain the current climate outflow volume, these practices as
designed are not effective at reducing the highest peak flow rates. As a result, the flow duration curves
between the "current with BMPs" and "future, adapted BMPs" diverge, particularly for the highest
outflows.
B-207
-------
120
100
'S 80
1
1 60
O
_>
o 40
X
20
Current with BMPs Future with BMPs — — Future, adapted BMPs
2-yr Hourly Flow
n\
v. \
I
\ \
>
0
0.0(
X)% 0.001% 0.002% 0.003% 0.004% 0.005% 0.006% 0.007%
Percent of time flow is equaled or exceeded
Figure B-254. Flow duration curve (FDC) evaluation for current and future climate,
Scott County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 20%) scenario.
Figure B-255 indicates that the addition of distributed bioretention to the conventional site does not result
in a significant decrease in the maximum hourly peak flow under future ("intensity plus 10%") climate
conditions.
B-208
-------
Max Peak Flow
Current Future Current with Future with Future,
Untreated Untreated BMPs BMPs adapted
BMPs
Figure B-255. Current and future performance for maximum hourly peak flow,
Scott County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 20%) scenario.
The future ("intensity plus 20%") climate adaptation for the Conventional (Gray) Infrastructure with
Distributed GI scenario produces a monthly outflow volume response that is very similar to the current
monthly outflow response. Outflow volumes in the adapted scenario are approximately the same as, or
lower than, the current climate scenario for all months.
B-209
-------
Monthly Outflow Volume
§4
3
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
• Current with BMPs Future with BMPs ¦ Future, adapted BMPs
Figure B-256. Current and future performance for monthly outflow volume, Scott
County, MN Conventional (Gray) Infrastructure with Distributed Green
Infrastructure (GI) (intensity plus 20%) scenario.
B.2.2.5. Sensitivity Analysis Adaptation Summary
Table B-27 summarizes the increases in BMP footprints for all of the Scott County stormwater
management scenarios that would be required to maintain current performance under future climate
conditions for the "intensity plus 10%"' climate simulation. Adaptation for the Conventional (Gray)
Infrastructure scenario would require the wet pond footprint to increase by nearly 230%. For the GI with
Gray Infrastructure scenario, no increase in the dry detention basin footprint is required, but the
bioretention footprint would need to more than double in size. Adaptation for the GI Only scenario would
require an 86% increase in bioretention footprint. When distributed GI is added to the Conventional
(Gray) Infrastructure scenario for adaptation, the required bioretention footprint of 17,500 square feet
would comprise approximately 1.3% of the total site area.
B-210
-------
Table B-27. Comparison of current and future adapted best management practice
(BMP) footprints, Scott County, MN stormwater management scenarios, intensity
plus 10%
Location
Climate
scenario
Stormwater
management
scenario
Practice
Current
Future
adapted
% increase
in
footprint
Footprint
as % of
site area
Footprint
SF
Footprint
as % of
site area
Scott
County,
MN
Intensity
change
plus 10%
Conventional
(Gray)
Infrastructure
Wet pond
32,670
2.5
107,484
8.2
229
GI with Gray
Infrastructure
Bioretention
34,848
2.7
70,348
5.4
102
Dry detention
basin
26,136
2.0
26,136
2.0
0
GI Only
Bioretention
(modified)
43,275
3.3
80,405
6.2
86
Rooftop
downspout
disconnection
94,901
7.3
94,901
7.3
0
Permeable
pavement
39,390
3.0
39,390
3.0
0
Conventional
(Gray)
Infrastructure
with
Distributed
GI
Wet pond
32,670
2.5
32,670
2.5
0
Distributed
bioretention
0
0.0
17,500
1.3
Table B-28 summarizes the increases in BMP footprints for all of the Scott County stormwater
management scenarios that would be required to maintain current performance under future climate
conditions for the "intensity plus 20%" climate simulation. Adaptation for the Conventional (Gray)
Infrastructure scenario would require the wet pond footprint to increase by nearly 430%. For the GI with
Gray Infrastructure scenario, the dry detention basin footprint would need to increase by 163%, and the
bioretention footprint would need to increase by 139%. Adaptation for the GI Only scenario would
require a 172% increase in bioretention footprint. When distributed GI is added to the Conventional
(Gray) Infrastructure scenario for adaptation, the required bioretention footprint of 30,500 square feet
would comprise approximately 2.3% of the total site area.
B-211
-------
Table B-28. Comparison of current and future adapted best management practice
(BMP) footprints, Scott County, MN stormwater management scenarios, intensity
plus 20%
Intensity
County,
MN
change
plus 20%
Current
Future adapted
Stormwater
management
scenario
Practice
Footprint
as % of
site area
Footprint
SF
Footprint
as % of
site area
% increase
in
footprint
Conventional
(Gray)
Infrastructure
Wet pond
32,670
2.5
172,171
13.2
427
GI with Gray
Bioretention
34,848
2.7
83,348
6.4
139
Infrastructure
Dry detention
basin
26,136
2.0
68,636
5.3
163
GI Only
Bioretention
(modified)
43,275
3.3
117,601
9.0
172
Rooftop
downspout
disconnection
94,901
7.3
94,901
7.3
0
Permeable
pavement
39,390
3.0
39,390
3.0
0
Conventional
Wet pond
32,670
2.5
32,670
2.5
0
(Gray)
Infrastructure
with
Distributed
GI
Distributed
bioretention
0
0.0
30,500
2.3
B.2.2.6. Cost of Adaptation
Table B-29 provides an estimate of the 20-year present value costs for the current and future adapted
climate conditions for all four Scott County stormwater management scenarios under future "intensity
plus 10%" climate, as well as the percentage increase in cost (current to future adapted) and increase
(millions of dollars) per acre of site.
B-212
-------
Table B-29. Comparison of the current and future estimated 20-year present value
costs for the Scott County, MN stormwater management scenarios, intensity plus
10%
Location
Climate
scenario
Stormwater
management
scenario
Current
cost (20-yr
present
value,
$millions)
Future
adapted cost
20-yr present
value,
$millions
Increase in
cost (20-yr
present
value,
$millions)
% increase
in cost
Increase
per acre of
site
$millions
Scott
County,
MN
Intensity
change
plus 10%
Conventional
(Gray)
Infrastructure
3.05
8.96
5.92
194
0.30
GI with Gray
Infrastructure
4.92
8.11
3.20
65
0.16
GI Only
8.51
12.76
4.25
50
0.21
Conventional
(Gray)
Infrastructure with
Distributed GI
3.05
4.62
1.58
52
0.08
Table B-30 provides an estimate of cost for the future "intensity plus 20%" climate scenario.
Table B-30. Comparison of the current and future estimated 20-year present value
costs for the Scott County, MN stormwater management scenarios, intensity plus
20%
Location
Climate
scenario
Stormwater
management
scenario
Current
cost (20-yr
present
value,
$millions)
Future
adapted cost
20-yr present
value,
$millions
Increase in
cost (20-yr
present
value,
$millions)
% increase in
cost
Increase
per acre
of site
$millions
Scott
County,
MN
Intensity
change plus
20%
Conventional
(Gray)
Infrastructure
3.05
14.08
11.03
362
0.55
GI with Gray
Infrastructure
4.92
11.31
6.40
130
0.32
GI Only
8.51
17.12
8.61
101
0.43
Conventional
(Gray)
Infrastructure
with
Distributed GI
3.05
5.79
2.75
90
0.14
B-213
-------
SEPA
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
------- |