-------
5 ; , ••	-
Number of trees. The existing data describing trees planted in Seattle's GSI inventory record only those
trees planted for code-triggered projects (for stormwater control purposes). In total, about 1,640 trees
have been used through 2012 to manage stormwater on these projects. Trees will be installed on an
annual basis through 2050 under each build-out scenario, both on code-triggered projects and within
the public right-of-way. The Low Build-Out Scenario will install about 4,800 trees each year, the Medium
Build-Out Scenario will install about 5,500 trees each year, and the High Build-Out Scenario will install
about 6,300 trees each year (see Appendix B for details).
Energy savings. Research in western Washington and Oregon has estimated the household energy
savings attributable to nearby trees (within about 60 feet).23 Energy savings are described in terms of
cooling energy (kWh) and heating energy (kBtu). Energy savings depend on the size of the tree and on its
location relative to the household. For this analysis, it was assumed that all trees are medium in size and
that they are 10 years old when they begin managing stormwater. In general, larger trees offer more
energy-related savings, so future GSI planning should consider selecting tree sizes to maximize these
tree-related benefits. Furthermore, the research has offered different savings factors depending on the
tree's location relative to the household (east side, south side, west side, and public). For this analysis,
the average of these factors was taken because data are not sufficient to precisely locate each tree
relative to the nearest household. Table 8 summarizes the energy savings factors used in the analysis
(per tree, per year).
Table 8. Tree-Related Energy Savings (per tree per year)
Age
Energy Savings (cooling | heating)
Age
Energy Savings (cooling | heating)
Year 5
3.75 kWh | (53.50) kBtu
Year 25
68.25 kWh | (83.75) kBtu
Year 10
20.25 kWh | (161.00) kBtu
Year 30
76.50 kWh | (39.00) kBtu
Year 15
41.50 kWh | (165.25) kBtu
Year 35
82.25 kWh | (0.25) kBtu
Year 20
57.25 kWh | (130.50) kBtu
Year 40 and beyond
86.25 kWh | (30.00) kBtu
Source: McPherson, G. et al. 2002. Western Washington and Oregon Community Tree Guide: Benefits, Costs and Strategic Planning. Center for
Urban Forest Research, U.S. Forest Service, Pacific Southwest Research Station. March.
Notes: A linear distribution of energy savings is assumed between points identified in this table.
Results. Table 9 summarizes the results of the analysis. The existing inventory of trees has the capacity
to reduce household energy consumption for cooling by about 13 million kWh over the next 100 years.
The existing inventory of trees would potentially increase household energy consumption for heating by
about 2.1 million kBtu over the next 100 years. All three build-out scenarios demonstrate a similar
pattern in cooling benefits and heating costs.
Because these benefits accrue at the household scale, household energy costs are used to estimate the
value of these energy savings. In 2013, the average residential rate for electricity from Seattle City Lights
ranged from $0.0466-$0.1071 per kWh, and the average residential rate for natural gas from Puget
Sound Energy was $0.0097 per kBtu.24 These energy prices are multiplied by the annual changes in
23	McPherson, G. et al. 2002. Western Washington and Oregon Community Tree Guide: Benefits, Costs and Strategic Planning.
Center for Urban Forest Research, U.S. Forest Service, Pacific Southwest Research Station. March.
24	Seattle City Light. 2013. Electric Rates and Provisions: Schedule RSC - Residential City. Retrieved on June 17, 2013 from
http://www.seattle.aov/liaht/accounts/rates/docs/2013/Jan/Januarv%202013%20-%20rsc.Ddf', Puget Sound Energy. 2013. Gas
Summary Sheet No. S-3. Retrieved on June 17, 2013 from
http://ase.com/aboutpse/Rates/Documents/summ a as prices 2013 05 Ol.pdf.
21

-------
energy consumption and discount all future values to estimate the 100-year NPV of energy savings (the
final column in the table). The 100-year NPV of the energy savings the existing inventory of trees
provides is about $0.2-$0.5 million. The 100-year NPV of the energy savings provided by the trees
associated with the build-out scenarios ranges from $15-$37 million under the Low Build-Out Scenario
up to $20-$49 million under the High Build-Out Scenario.
Table 9. Summary of Tree-related Energy Benefits to Households25

Number of Trees
100-year Energy Savings
100-year NPV
Code-Triggered Inventory
1,640
13.0 million kWh | (2.1 million) kBtu
$0.2-$0.5 million
Low Build-Out Scenario
4,800/year
1.1 billion kWh | (323 million) kBtu
$15-$37 million
Medium Build-Out
Scenario
5,500/year
1.3 billion kWh | (373 million) kBtu
$17-$43 million
High Build-Out Scenario
6,300/year
1.5 billion kWh | (424 million) kBtu
$20-$49 million
While these trees have the potential to decrease cooling costs during summer months, individuals that
do not use air conditioning or fans to cool their homes would not realize this benefit. To the extent that
households in Seattle use cooling systems less frequently than assumed in the literature applied in this
analysis (which is based on households in Portland, OR), these results may overstate the benefits.
Furthermore, this analysis assumes that the energy rates households face will remain constant, in real
terms over the next 100 years. The U.S. Energy Information Administration has identified five factors
that influence energy prices: (1) fuel source, (2) power plants, (3) distribution network, (4) weather
conditions, and (5) regulations.26 Climate change, efforts to prevent/reduce the impacts of climate
change, and other energy-related efforts have the capacity to influence all five of these factors, which
could increase or decrease energy rates, in real terms, in the future.
As of this writing, inventory data on green roofs do not provide sufficient detail on the building uses or
green roof characteristics to allow estimation of the energy savings from reduced heating and cooling
with green roofs in Seattle. Seattle has 1.8 million square feet of green roofs and roof gardens (Table
15). A study of avoided energy costs for a 40,000 square foot green roof in Portland found annual energy
savings of $1480.27 This equates to $66,600 annually in energy benefits, and a 100-year NPV at 2 percent
discounting of $2.87 million. Both green roofs and trees would also contribute to reduced heat island
effects, although the natural temperature buffering of Puget Sound reduces the importance of avoiding
extreme high temperatures. The increasing potential for the importance of these benefits with climate
change is discussed later.
25	While the annual number of trees planted appears high relative to the documented code-triggered inventory, it is based on
existing proportions for share of GSI management in the inventory, and relationships between number of trees and volumes
managed. See Appendix B for greater detail.
26	U.S. Energy Information Administration. 2012. Many Factors Impact Electricity Prices. Retrieved on June 26, 2013 from
http://www.eia.gov/energyexplained/index.cfm?page=electricitv factors affecting prices.
27	MacMullan, E., Reich, S., Puttman, T., & Rodgers, K. 2008. Cost-Benefit Evaluation of Ecoroofs. In Low Impact Development
for Urban Ecosystem and Habitat Protection (pp. 1-10). ASCE. November.
22

-------
Tree Size and Location Matters
As stated in this section, the size of trees and their location on a property play important roles in determining
their potential effect on household energy use. To demonstrate the magnitude of this effect, the potential
energy savings are modeled for small, medium, and large trees planted on west-facing, south-facing, and east-
facing yards as well as trees planted in public areas. While trees planted in public areas may not provide any
direct shading benefits to homes, they do provide neighborhood-wide benefits associated with reduced
temperatures and wind speeds. The tables below show the 100-year NPV of cooling benefits, heating benefits,
and the combination of cooling and heating benefits, per tree. Energy rates of $0.0769/kWh are assumed for
cooling benefits and $0.0097/kBtu for heating benefits. All values are discounted at a rate of 2%.
100-year NPV of Cooling Benefits (per tree)
100-year NPV of Heating Benefits (per tree)

West
South
East
Public

West
South
East
Public
Small
$1,897
$1,048
$788
$365
Small
$(338)
$(1,099)
$(827)
$462
Medium
$3,011
$2,418
$1,640
$934
Medium
$213
$(832)
$(426)
$1,190
Large
$4,340
$3,230
$2,639
$2,004
Large
$1,521
$185
$739
$2,549
100-year NPV of Cooling Benefits and Heating Benefits (per tree)

West
South
East
Public
Small
$1,559
$(51)
$(38)
$826
Medium
$3,224
$1,586
$1,215
$2,124
Large
$5,861
$3,415
$3,378
$4,553
The best options are highlighted in the tables. Large trees on the west side of homes offer the most cooling
benefits, while large trees on public lands offer the most heating benefits. After combining cooling and heating
benefits, large trees on the west side of homes offer the greatest NPV of benefits over 100 years. In general,
however, trees planted on public land offer both cooling and heating benefits to households and likely
represent a good opportunity for public efforts aimed at reducing household energy consumption.
5.3.3 Summary and Distribution
In this section, the economic value of GSI's impact on household energy use is discussed and calculated.
Energy metrics from the literature are applied to the number of trees in SPU's GSI inventory as well as
under the three build-out scenarios to quantify total changes in household energy demand. The existing
inventory supports about 1,600 trees, and all three build-out scenarios would support about 1,000 trees
on single-family properties as part of the code-triggered mechanism for GSI implementation.
Existing trees have the potential to reduce household energy use for cooling by about 13 million kWh,
but they may increase household energy use for heating by 2.1 million kBtu over the next 100 years. The
100-year NPV of this benefit totals about $0.2-$0.5 million. Similarly, trees associated with the build-out
scenarios have the potential to reduce household energy use for cooling, but they may increase
household energy use for heating. The 100-year NPVs of this benefit range from $15 to $50 million
across the build-out scenarios. The benefits of reducing household energy use accrue primarily to
Seattle residents with trees near their homes, although some residual benefit extends beyond
immediately adjacent buildings. Additional benefits associated with reduced energy consumption are
discussed elsewhere in this report.
23

-------
ihouse Gas Emissions and Sequestration
Extensive research shows that Seattle and other parts of the Pacific Northwest already have experienced
noticeable changes in climate, and predicts that more change will occur in the future.28 Research has
identified several types of anthropogenic greenhouse gas (GHG) emissions that contribute to climate
change; chief among them is the emissions of C02. In 2004, C02 accounted for approximately 77% of
greenhouse gases emitted into the atmosphere. Since 1850, the concentration of C02 in the atmosphere
has increased from 280 to 379 parts per million, and has grown by an average of 1.9 parts per million
per year since 1995.29 Airborne pollutants such as C02 and other GHGs are the primary contributors to
climate change. This section quantifies the volume of atmospheric reductions in GHGs associated with
GSI and monetize the value of the benefit.
5.4.1	\
The analysis of the value of reducing atmospheric GHG concentrations has three parts.
Step 1 - Estimate avoided GHG emissions. In previous sections, different ways that GSI can reduce
energy consumption were described. For all quantified reductions in energy consumption, the total
amount of energy is calculated in base units (e.g., kWh, kBtu), and multiplied by emission factors from
the relevant utility.
Step 2 - Estimate GHG sequestration. In previous sections, the number and growth rate of existing and
new trees in the GSI inventory and in each of the scenarios are described. These tree data are aligned
with sequestration factors from the literature that quantify the volume of tree-related GHG
sequestration.
Step 3 - Apply the value of carbon. This analysis relied on two per-unit values representing the
potential costs associated with C02 emissions: (1) an estimate of the social cost of carbon, and (2) a
range of market prices for carbon emissions from efforts to trade and tax emissions across the world.
5.4.2	f
Carbon values. Economists use the social cost of carbon to estimate the value of changes in greenhouse
gas emissions. The social cost of carbon represents "the full global cost today of emitting an incremental
unit of carbon at some point of time in the future, and it includes the sum of the global cost of the
damage it imposes on the entire time it is in the atmosphere."30 There are currently over 200 different
estimates of the social cost of carbon. One review of the literature found values ranging from about $7
to over $100 per ton of C02e (or C02 equivalent).31
28	See, for example, University of Washington, Climate Impacts Group. 2009. The Washington Climate Change Impacts
Assessment: Evaluating Washington's Future in a Changing Climate.
29	Intergovernmental Panel on Climate Change (IPCC). 2007. Climate Change 2007: Synthesis Report. Geneva, Switzerland.
30	Shaw, M., L. Pendleton, et al. 2009. The Impact of Climate Change on California's Ecosystem Services. California Climate
Change Center. CEC-500-2009-025-F.
31	Shaw, M., L. Pendleton, et al. 2009. The Impact of Climate Change on California's Ecosystem Services. California Climate
Change Center. CEC-500-2009-025-F.
24

-------
Over the past decade, several voluntary and regulatory carbon markets have emerged around the world
along with several attempts at taxing carbon. Table 10 summarizes the total volume, total value, and
per-unit value of carbon traded in voluntary and regulatory carbon markets around the world in 2011.
The average carbon price across these markets was about $15.50 per ton of C02e. In addition to these
carbon markets, many public agencies around the world have proposed or implemented carbon tax
schemes (e.g., South Africa, India, Japan, South Korea, Australia, New Zealand, Denmark, Finland, and
France). In 2008, British Columbia passed the Carbon Tax Act, which consumers pay when they purchase
fossil fuels in the Province. The carbon tax rate has increased each year, and in July 2012 was set at
about $27 per ton of C02e.32
Table 10. Voluntary and Regulatory Carbon Markets (2011 Summary)
Carbon Market
Tons of CO.
-------
Emissions reduction from home-related cooling and heating. Elsewhere in this report, the household
energy savings associated with trees was quantified. The inventory and the Low, Medium, and High
Build-Out Scenarios reduce energy consumption for cooling by about 13.0 million, 608 million, 785
million, 961 million kWh, respectively, over the next 100 years. The inventory and the Low, Medium, and
High Build-Out Scenarios increase energy consumption for heating by about 2.1 million, 174 million,
225 million, 275 million kBtu over the next 100 years. For marginal emissions from electricity generation,
Seattle City Light uses an emission factor of about 1.1 lbs. of C02e per kWh.35 For emissions from natural
gas, EPA uses an emissions factor of 0.1 lbs. of C02e per kBtu.36 Over the next 100 years, the inventory
would reduce emissions by about 7,000 tons of C02e. The Low, Medium, and High Build-Out Scenarios
would reduce emissions by about 323,000 tons, 416,000 tons, and 510,000 ton of C02e over the next
100 years, respectively.
A range of carbon values ($15-$58, increasing at a real rate of 2.5% per year) are used to quantify the
economic value of this benefit. The 100-year NPV of the avoided emissions stemming from the inventory's
impact on household energy consumption totals about $0.1-$0.5 million. The 100-year NPV of avoided
emissions is about $6.6-$25.6 million under the Low Build-Out Scenario, $8.5-$33.1 million under the
Medium Build-Out Scenario, and about $10.5-$40.5 million under the High Build-Out Scenario.
Emissions reduction from reduced treatment. King County's Wastewater Treatment Division has
estimated the volume of water managed at the West Point Wastewater Treatment Plant as well as the
total energy used to treat that water and the total energy costs. Emission factors are applied to these
energy units to quantify emissions volumes.37 Table 11 summarizes these data for 2009 and 2010. On
average, treatment-related energy emissions at West Point totaled about 1.0 tons of C02e per million
gallons of water treated.
Table 11. Treatment-Related Emissions at West Point Wastewater Treatment Plant
Year
Electricity
(kWh)
Gas and Propane
(therms)
Average Annual Flow
(millions of gallons)
MWh/million
gallons
Tons of CO 
-------
•	The Medium Build-Out Scenario would manage a total of 1.2 billion gallons of stormwater in the
combined sewer basin per year by 2050.
•	The High Build-Out Scenario would manage a total of 2.0 billion gallons of stormwater in the
combined sewer basin per year by 2050.
To calculate the 100-year NPV of this benefit, the volume of stormwater managed each year is aligned
with the avoided emissions and the range of carbon values described earlier. The 100-year NPV of the
existing inventory's treatment-related emissions reductions totals $25,000-$96,000. The 100-year NPVs
for the Low, Medium, and High Build-Out Scenarios total $1.2-$4.6 million, $1.9-$7.5 million, and $3.1-
$12.0 million, respectively.
Carbon Sequestration
Sequestration by trees. The U.S. Department of Energy has compiled data estimating carbon
sequestration rates for trees in urban and suburban areas.38 According to these data, younger trees tend
to sequester less carbon than older trees, and coniferous trees tend to sequester less carbon than
deciduous trees. A 10-year-old tree sequesters 13-48 lbs. of C02 per year, while a 50-year-old tree
sequesters 88-450 lbs. of C02 per year. It is assumed all trees, both those in the inventory and those in
the build-out scenarios, are 10 years old when they are installed, and the full range of sequestration
rates across both deciduous and coniferous varieties are used. Trees associated with the inventory could
sequester a total of about 6,800-34,700 tons of C02 over the next 100 years with a 100-year NPV of
about $0.1-$2.7 million. These values can range up to nearly 4 million tons of sequestration worth over
$300 million under the High Build-Out Scenario (Table 12).
Table 12. Summary of 100-Year Volume and NPV of GHG Benefits
100-Year Volume of GHG Emissions Reductions and Sequestration

Reduced Emissions
from Household
Energy Consumption
Reduced Emissions
from Reduced
Treatment
Tree-Related CO.<
Sequestration
Total CO.' Reductions
Inventory
7,000 tons
1,300 tons
6,800-34,700 tons
< 0.1 million tons
Low Build-Out Scenario
599,000 tons
44,000 tons
0.56-2.9 million tons
1.2-3.5 million tons
Medium Build-Out
Scenario
692,000 tons
77,000 tons
0.65-3.3 million tons
1.4-4.1 million tons
High Build-Out Scenario
786,000 tons
131,000 tons
0.74-3.8 million tons
1.7-4.7 million tons
100-Year NPV of Benefits Related to GHG Emissions Reductions and Sequestration

Reduced Emissions
from Household
Energy Consumption
Reduced Emissions
from Reduced
Treatment
Tree-Related CO'
Sequestration
Total 100-Year NPV
Inventory
$0.1-$0.5 million
< $0.1 million
$0.1-$2.7 million
$0.3-$3.3 million
Low Build-Out Scenario
$12-$48 million
$0.87-$3.4 million
$12-$233 million
$25-$284 million
Medium Build-Out
Scenario
$14-$55 million
$1.5-$6.0 million
$14-$270 million
$29-$331 million
High Build-Out Scenario
$16-$62 million
$2.6-$10 million
$15-$306 million
$34-$379 million
38 U.S. Department of Energy, Energy Information Administration. 1998. Method for Calculating Carbon Sequestration by Trees
in Urban and Suburban Settings. Retrieved on June 17, 2013 from
ftp://ftp. eia. doe. qov/pub/oiaf/1605/cdram/pdf/sequester. pdf.
27
-------
Sequestration by soil. Bioretention facilities such as rain gardens typically utilize soil, as do tree
plantings. Modeling by the U.S. EPA assumes that l/5th of the carbon content of soil amendments such
as compost are stored long-term in soil.39 Inventory data do not currently allow an estimation of the
compost or other carbon amendment shares or totals for soils associated with GSI facilities in Seattle.
Considering a ton of soil is roughly 40 cubic feet in volume, and if half the soil volume were carbon
amendments such as compost, l/5th storage of original volume would net a tenth of a ton of carbon
sequestration for each ton of soil utilized. In the future, tracking soil volumes and carbon amendment
shares will allow estimation of this potentially large contribution of GSI in Seattle to soil carbon
sequestration.
5.4.3 Si..- -¦
This section quantifies the value of avoided GHG emissions and GHG sequestration associated with GSI's
capacity to reduce energy consumption and to filter GHGs out of the atmosphere. Energy savings from
other sections of this report were used along with emissions data from local utilities to estimate the
change in emissions over time. Tree data with per-tree values of carbon sequestration were aligned to
quantify the reduction in atmospheric GHG concentrations over time. A range of economic values
associated with markets for GHG emission and the social cost of carbon were used to quantify the
economic value of this benefit ($15-$58 per ton of C02e, increasing at a real rate of 2.5% per year).
Table 12 summarizes the results of the analysis. The top half of the table shows the volume of GHGs for
each scenario and for each of the three forms of GHG reductions. The bottom half of the table shows
the values associated with these GHG reductions. To the extent that the effects of climate change are
global, reducing the concentration of GHG in the atmosphere benefits all of society. These benefits may
be different for different parts of the world, but the economic values used in this analysis reflect average
effects across the globe.
5.5 Air Quality
GSI has the capacity to improve air quality in two ways: (1) by decreasing energy demand, GSI indirectly
decreases the volume of harmful pollutants emitted into the atmosphere, and (2) some forms of GSI
(trees and vegetation) remove harmful pollutants from the atmosphere through biophysical processes.
Improvements in air quality have economic value in that they reduce the costs associated with air
pollution (e.g., health-related costs from respiratory illness and habitat destruction). In this section, local
emission factors and air filtration rates are used along with avoided emissions-related costs, to quantify
the value of air quality benefits.
The literature provides a proven method for quantifying the value of air quality-related benefits
associated with reductions in emissions and vegetative filtration. Typically, this methodology has two
steps: (1) calculating the volume of airborne pollutants that would have been emitted into the
atmosphere and the volume of pollutants filtered out of the atmosphere by vegetation, and (2)
multiplying those volumes by per-unit values (e.g., dollars per ton of S02) describing the value of the
benefits of improvements in air quality.
Step 1 - Quantify change in air quality. By reducing energy consumption, GSI has the capacity to reduce
energy production and, consequently, to improve air quality. To calculate the change in emissions
associated with GSI, emissions rates specific to Washington State are applied to changes in energy
39 U.S. EPA. 2013. Waste Reduction Model, htips://www.epa.gov/warm.
28
-------
consumption from other parts of this analysis. In addition to the indirect improvements in air quality
from reductions in energy consumption, some GSI BMPs (e.g., street trees and green roofs) have the
capacity to improve air quality by filtering harmful pollutants out of the atmosphere. To calculate the
contributions of these GSI facilities to improved air quality, average filtration rates from the literature
that describe per-unit pollutant removal in terms of trees and vegetated surfaces are applied.
Step 2 - Align changes in air quality with avoided costs. Quantifying the economic value of changes in
local air quality requires complex climate, epidemiological, and economic modeling efforts. For this
analysis, average costs associated with changes in air quality from the national perspective are used.
These costs represent the health costs individuals incur due to air pollution. By improving air quality, GSI
has the capacity to reduce air quality-related health costs in Seattle as well as other places that generate
electricity.
5 •. :•	-
This section summarizes the analysis of three forms of air quality improvements: (1) tree-related air
filtration, (2) reduced emissions from reduced household energy consumption, and (3) reduced
emissions from reduced treatment-related energy consumption. Energy data from other sections in this
report are used throughout this analysis.
Tree-related air pollution removal. Trees intercept and remove air pollutants from the atmosphere and
break them down through a series of biological processes and mechanisms. Nowak (2006) estimated
pollution removal rates in terms of canopy cover across 55 cities in the U.S., including Seattle.40
Pollutant-specific per-gram values for pollutant removal were used, which represent median externality
values (from a national perspective) associated with costs typically linked to air pollution (e.g., health
costs).41 Average canopy widths of 21-37 feet were used to convert these canopy cover figures to per-
tree estimates.42 Table 13 summarizes the conversion from pollutant removal to value per tree per year.
According to these assumptions, the value of tree-related pollutant removal by a mature tree in Seattle
totals $0.59-$7.69 per year. Not all trees included in this analysis are mature, however data are not
sufficient to extrapolate pollution removal over time as trees grow. For this analysis, it was assumed
that, on average, trees included provide air pollution benefits of $0.29-$3.85 per tree per year.
The existing inventory of GSI facilities includes a total of 1,640 trees.43 The Low, Medium, and High
Build-Out Scenarios would increase the number of trees by about 4,810, 5,560, and 6,310, respectively,
each year, through 2050. As discussed above, it is assumed that, on average, trees included in this
analysis provide air quality benefits (in terms of air pollutants removed from the atmosphere) of $0.29-
$3.85 per tree per year. The 100-year NPV of these benefits for the existing inventory is about $21,000-
$0.28 million. Not including the benefits from the existing inventory, the 100-year NPV of these air
quality benefits is about $1.5-20 million for the Low Build-Out Scenario, $1.8-$23 million for the
Medium Build-Out Scenario, and $2.0-$26 million for the High Build-Out Scenario.
40	Nowak, D., D. Crane, and J. Stevens. 2006. "Air Pollution Removal by Urban Trees and Shrubs in the United States." Urban
Forestry & Urban Greening. 4:115-123.
41	Murray, F., L. Marsh, and P. Bradford. 1994. New York State Energy Plan, Volume II. New York State Energy Office, Albany, NY.
42	Center for Neighborhood Technology. 2010. The Value of Green Infrastructure: A Guide to Recognizing its Economic,
Environmental, and Social Benefits.
43	Data were not sufficient to quantify the number of trees associated with some of the GSI mechanisms, so this likely
understates the actual number of trees currently supporting stormwater management efforts.
29
-------
Table 13. Air Pollution Removal by Urban Trees in Seattle
Pollutant
Pollutant Removal
(grams/square meter of canopy
cover/year)
Value of Pollutant Removal
($/gram)
Value of Pollutant Removal
($/tree/year)
CO
0.6
$0.0011
$0.01-$0.03
N02
0.7-1.9
$0.0077
$0.09-$0.73
PM10
1.2-4.8
$0.0051
$0.10-$1.23
S02
0.7-2.5
$0.0019
$0.02-$0.24
03
0.7-4.3
$0.0077
$0.09-$1.65
Source: Nowak, D., D. Crane, and J. Stevens. 2006. "Air Pollution Removal by Urban Trees and Shrubs in the United States." Urban Forestry &
Urban Greening. 4:115-123.
Notes: To convert from per-square meter of canopy cover units to per-tree units, a range of 32-100 square meters of canopy cover per tree
was assumed. Chemical abbreviations: CO - carbon monoxide, NO2 - nitrogen dioxide, PM10- particulate matter up to 10 urn in size, SO2 -
sulfur dioxide, O3 - ozone
Emissions reduction from home-related cooling and heating. Elsewhere in this report, the household
energy savings associated with trees was quantified. The inventory and the Low, Medium, and High
Build-Out Scenarios reduce energy consumption for cooling by about 13.0 million, 1.1 billion, 1.3 billion,
and 1.5 billion kWh, respectively, over the next 100 years. The inventory and the Low, Medium, and
High Build-Out Scenarios increase energy consumption for heating by about 2.1 million, 323 million,
373 million, 424 million kBtu over the next 100 years. A factor of 3,412 kBtu/MWh is used to convert
energy used for heating into a common metric.44
When fossil fuels such as coal and natural gas are used to generate electricity, harmful pollutants are
emitted into the atmosphere. Data describing Seattle-specific utilities and their emissions rates are not
available however EPA provides state-level emissions factors. According to EPA, Washington State's
emissions profile includes emissions factors of 0.087 lbs. of S02/MWh and 0.4236 lbs. of NOx/MWh.45
Relative to other parts of the country, Washington's emissions factors are low, primarily due to its heavy
reliance on renewable energy sources (over 70% of all of the state's energy comes from hydro facilities).
Applying the same pollutant values from above suggests that reducing energy consumption in Seattle
reduces the costs of air pollution by about $0.08/MWh for S02 and about $1.48/MWh for NOx. While
energy generation emits other harmful pollutants (e.g., PM10, S02, 03), data describing the extent to
which generation facilities in Washington emit these pollutants are not sufficient for inclusion in this
analysis.
A value of $1.55 per MWh of energy consumption avoided due to tree-related reductions in demand for
cooling and heating was applied. The 100-year NPV of these air quality benefits for the inventory total
about $6,700. The 100-year NPV of these air quality benefits total about $0.47 million for the Low Build-
Out Scenario, about $0.54 million for the Medium Build-Out Scenario, and about $0.61 million for the
High Build-Out Scenario.
Emissions reduction from reduced treatment. King County's Wastewater Treatment Division has
estimated the volume of water managed at the West Point Wastewater Treatment Plant, as well as the
total energy used to treat that water and the total energy costs. As shown in Table 11, treatment at
West Point requires an average of 2.6 MWh per million gallons. As described in previous sections and in
44	International Energy Agency. 2013. Unit Converter. Retrieved on June 26, 2013 from http://www.iea.org/stats/unitasp.
45	U.S. Environmental Protection Agency. 2013. eGRIDweb - State Level Data (Washington). Retrieved on June 24, 2013 from
http://cfDub.epa.gov/earidweb/view st.cfm.
30
-------
Appendix A, the existing inventory of GSI facilities manages 13.4 million gallons of stormwater runoff per
year within the combined sewer basin. The Low Build-Out Scenario would manage a total of 556 million
gallons of stormwater in the combined sewer basin per year by 2050. The Medium Build-Out Scenario
would manage a total of 986 million gallons of stormwater in the combined sewer basin per year by
2050. The High Build-Out Scenario would manage a total of 1.7 billion gallons of stormwater in the
combined sewer basin per year by 2050.
To calculate the 100-year NPV of this benefit, the volume of stormwater managed each year is aligned
with the avoided emissions and the range of carbon values described earlier. The 100-year NPV of the
existing inventory's treatment-related emissions reductions totals about $2,000. The 100-year NPVs for
the Low, Medium, and High Build-Out Scenarios total < $0.1 million, $0.11 million, and $0.20 million,
respectively.
Figure 5 shows the number of unhealthy and very unhealthy air quality days in Seattle from 1980-2011.
The data suggest that air quality in Seattle has improved over the past three decades. In fact, from
2007-2011, there was only one unhealthy day and no very unhealthy days. To some extent, the values
contained in this section may overstate the marginal costs avoided due to improvements in air quality.
Figure 5. Unhealthy and Very Unhealthy Air Quality Days in Seattle (1980-2011)
30
25
20
c £ «	15
5 s fli	1 °
>*—
or	10
¦Unhealthy
¦Very Unhealthy
¦
¦
11 ¦ 1111 I
±

III	I	.	¦	¦	¦	.
2012.2011 Air Quality Data Summary Appendix. November. Retrieved on July 8, 2013 from
>. asox.
\	\	'N "s
Source: Puget Sound Clean Air Agency. 2012. 2011 Air Quality Data Summary Appendix.
httD://www.Dscleanair.ora/aira/reoorts.asDX.
Heightened Importance of Air Quality in Seattle
A 2012 survey commissioned by the Puget Sound Partnership compiled data from 2,003 residents across the
Puget Sound region. One of the tasks that respondents faced was to identify the top two things about the Puget
Sound region's natural resources that they value most. About 8% of King County respondents listed clean air
and another 8% listed the scenery as one of the top two things they value.
Individuals consider a number of factors when deciding where to live, such as employment opportunities,
family, friends, and environmental quality and resources. A 2008 study used a series of variables (wage data,
housing data, and other amenities including heating and cooling degree days, sunshine, coastal proximity, air
quality, and other social variables) to rank states and metropolitan areas in terms of desirability. Washington
State ranked seventh on the list of states. The Seattle-Tacoma-Bremerton CMSA ranked 30th on the list of 241
metropolitan areas included in the analysis.
Source: Puget Sound Partnership. 2012. General Public Opinion Survey. Prepared by PRR Inc.; Albouy, D. 2008. Are Big Cities Really Bad
Places to Live? Improving Quality of Life Estimates Across Cities. NBER Working Paper Series. Working Paper 14472.
31
-------
5.5.3
In this section, the value of air quality improvements associated with GSI's capacity to reduce energy
consumption and filter pollutants out of the atmosphere is quantified. Energy savings from other
sections of this report are used, along with state-level emissions factors, to estimate the change in
emissions over time. Tree data with per-tree values of air filtration also is aligned to quantify the
reduction in air pollution over time Pollutant-specific values representing avoided costs of air pollution
are used to quantify the economic value of this benefit. Table 14 summarizes the results of the analysis.
Table 14. Summary of 100-Year NPV of Air Quality Benefits
100-Year NPV of Benefits Related to Air Quality Improvements from Emissions Reductions and Tree Filtration

Tree-Related Air
Filtration
Reduced Emissions
from Household
Energy Consumption
Reduced Emissions
from Reduced
Treatment
Total
Inventory
$21,000-$0.3 million
$6,700
$2,000
$0.3 million
Low Build-Out Scenario
$1.5-$20 million
$0.47 million
< 0.1 million
$2.1-$21 million
Medium Build-Out
Scenario
$1.8-$23 million
$0.54 million
$0.12 million
$2.4-$24 million
High Build-Out Scenario
$2.0-$26 million
$0.61 million
$0.20 million
$2.8-$27 million
Some of these benefits likely would accrue locally to residents in the Seattle area, while others would
accrue elsewhere. Specifically, Seattle-area residents would probably benefit from tree-related filtration
of pollutants out of the atmosphere because those trees would help improve/maintain local air quality.
Other air quality benefits described in this section (e.g., reduced emissions stemming from reduced
energy consumption) likely would accrue outside Seattle, closer to the areas surrounding energy
generation facilities.
5.6 Small-Scale Habitat
Some GSI BMPs (e.g., rain gardens, green roofs, and trees) provide habitat-related benefits to urban
mammals, birds and insect species/pollinators. Small-scale habitat in urban areas is valuable in that it
helps improve the health and diversity of local wildlife populations. Individuals benefit from these
habitat improvements insofar as they value the wildlife that habitat improvements and expansions
support. Recent larger-scale habitat restoration efforts in Seattle have come at a cost of $2,800-$28,000
per acre. These costs shed light on the extent to which the city and the public value functioning habitat
patches in urban areas. Data are not sufficient to quantify the habitat-specific value of these areas
isolated from other benefits considered in this analysis, but researchers are continuing to increase their
understanding of these small-scale habitat effects in urban areas, and are aligning their results with
economic factors describing the value of their benefits. For example, how do bird populations change in
Seattle as a result of GSI facilities? How do flowers and home gardens perform as a result of pollinators
and pest predators that utilize GSI facilities? The answers to these questions will improve understanding
of the habitat functions of GSI in Seattle and similar environs.
One way of considering the economic value of habitat improvements, however, is to identify demand
for habitat improvements, as evidenced by restoration efforts funded in the local area. In addition to
these terrestrial habitat benefits, GSI has the capacity to improve aquatic habitat by reducing the
volume of stormwater entering nearby waterways through untreated sewer systems. This section
32
-------
focuses primarily on terrestrial habitat improvements, but aquatic habitat is discussed elsewhere in this
report (see section on hydrologic function).
5.6. i Valuation Methodology
The literature does not provide sufficient evidence of the quantified effects of GSI on improvements in
small-scale habitat (e.g., percentage increase in urban bird populations per unit of GSI implementation)
to quantify habitat-specific economic values. Furthermore, the types of wildlife populations that GSI
tends to support (birds and insects) are not well represented in the economic literature describing
society's demand for wildlife. Past restoration efforts in Seattle, however, provide cost estimates that
reflect the public's demand for urban habitat. The approach to quantifying this illustrative value has
three steps (described below). These values may double-count some of the values described in other
sections of this report. For example, most habitat restoration efforts provide numerous benefits already
included in this report (e.g., carbon sequestration, air quality improvements, water quality
improvements).
Step 1 - Conduct literature review. The first step was to look through the literature to see if this kind of
analysis had been done before. Several instances were identified in which GSI facilities have changed
wildlife conditions in the local community. Ways in which GSI efforts in Seattle can support habitat-
related benefits are described.
Step 2 - Quantify the change in habitat area. Some forms of GSI increase the amount of small-scale
habitat in Seattle's urban environment. Inventory data, data from the build-out scenarios, and
conversion factors are used to quantify the area of new habitat that GSI efforts support. Only habitat
resulting from GSI projects is included, as opposed to all habitat in the Seattle area.
Step 3 - Quantify restoration costs and align with GSI-related habitat. The Green Seattle Partnership's
20-Year Plan offers a cost breakdown of habitat restoration efforts at a per-acre level. This range of
costs is aligned with the area of GSI-related habitat from the inventory and the build-out scenarios to
estimate a one-time value of the benefit.
5	"
This section summarizes the review and analysis in three parts: (1) a literature review describing the
effects of GSI on small-scale habitat, (2) the analysis of the area of the habitat GSI supports, and (3) the
quantification of the value of this habitat based on past restoration costs.
Literature review. Different types of habitat provide different sets of services from which individuals
derive benefits. Many GSI facilities cover too little land to provide quantifiable habitat value; however,
these sites may provide ecological benefits to habitat and biodiversity conservation to the extent that
they contribute to patches of urban green space. In terms of habitat provision, cities are highly
fragmented environments composed of a mosaic of natural patches of various sizes and types. While
this type of fragmentation does reduce the quality, quantity, and pattern of habitats, it is not necessarily
a limiting factor in population persistence of some species. In some cases, urban green spaces may
provide habitats for a rich and diverse range of plants and animals.46
46 Angold, P.G., J.P. Salder, M.O. Hill, A. Pullin, S. Rushton, K. Austin, E. Small, B. Wood, R. Wadsworth, R. Sanderson, and K.
Thompson. 2006. "Biodiversity in urban habitat patches." Science of the Total Environment 360(1-3): 196-204.
33
-------
Urban habitat, such as the kind of habitat some GSI BMPs provide, can increase overall vegetation cover,
which contributes to biological diversity.47 Urban green spaces also help conserve habitat, species, and
genetic biodiversity within ecosystems by creating interconnected networks that minimize the impacts
of habitat fragmentation and provide habitat corridors. For example, one study found small and
medium-sized mammals use urban greenways as wildlife corridors, which allow for the exchange of
individuals between populations, increasing genetic diversity and reducing instances of inbreeding.48
GSI exhibits many of the characteristics of urban habitats. In particular, GSI may provide an important
role in supporting wildlife corridors, particularly when cities adopt a systematic approach to planning
and managing the spatial distribution of these sites across local authority boundaries. GSI also may
increase linkages and connectivity between patches of open spaces and diverse types of urban wildlife
habitats. One recent paper argues that, when properly applied, green infrastructure could "bring
together a coherent network of components, such as open spaces, green corridors, and woodlands for
the benefit of people and wildlife."49
When designers in the United Kingdom addressed wildlife demands in their GSI planning, researchers
observed the return of increasingly rare species in some areas. Brown roofs provided habitat benefits
for several local bird species, thus prompting a renewed growth in their local population. In the case of
green roofs in London, researchers noticed a similar regeneration of rare spider and insect
populations.50 Rain gardens provide similar habitat-related benefits. Research shows that they attract
birds, butterflies, and insects while improving the habitat quality of downstream waterways for aquatic
organisms.51 The value associated with these benefits will likely increase over time. The literature
suggests that as urban areas encroach into rural habitats and agriculture reduces the quality of rural
habitat, these small-scale urban green spaces will become an increasingly important refuge for native
biodiversity.52
There are no data describing the extent to which GSI facilities in Seattle are providing benefits related to
small-scale habitat. Many of the GSI facilities are currently too new to provide many habitat-related
benefits, but likely will attract wildlife in the future when they are fully developed (e.g., as street trees
and vegetation in rain gardens mature).53
GSI-related habitat in Seattle. Data describing the inventory of GSI facilities and the assumptions used
to develop the three build-out scenarios do not explicitly quantify the area of habitat that GSI facilities
provide. For this analysis, available data were used describing the area of relevant GSI facilities,
conversion factors aligning the BMP area with the impervious management area and stormwater
47	Bratton, S.P. 1997. "Alternative models of ecosystem health." In: Costanza, R., B.G. Norton, B.D. Haskell. Ecosystem Health:
New Goals for Ecosystem Management. Island Press: 170-189. Washington D.C.
48	Opdam, P., E. Steingrover, S. van Rooij. 2006. "Ecological networks: a spatial concept for multi actor planning of sustainable
landscape." Landscape and Urban Planning 75: 322-332; Flores, A., S.T.A. Pickett, W.C. Zipperer, R.V. Pouyat, and R. Pirani.
1998. "Adopting a modern ecological view of the metropolitan landscape: A case of green space system for the New York City
region." Landscape and Urban Planning 39: 295-308.
49	Douglas, I. and J.P, Sadler. 2010. "Urban wildlife corridors: Conduits for movement or linear habitat?" In Douglas, I., D. Goode,
M. Houck, and R. Wang (eds). The Routledge Handbook of Urban Ecology. Taylor and Francis: 274-288.
50	Gredge, D. 2003. From Rubble to Redstarts. Proceedings from Greening Rooftops for Sustainable Communities, First North
American Green Roof Infrastructure Conference. May. Chicago.
51	Drew, B., B. Yim, D. Lo, and I. Liu. 2011. An Investigation into Rain Gardens. University of British Columbia.
52	Goddard, M., A. Dougill, T.G. Benton. 2010. "Scaling up from gardens: biodiversity conservation in urban environments."
Trends in Ecological Economics 25(2): 90-98.
53	Personal Communication. Spencer, Bob. Seattle Public Utilities. Telephone. June 3, 2013.
34
-------
management, and other
relationships to estimate the
total BMP area of bioretention,
trees, and green roofs for the
inventory and the three build-
out scenarios. Table 15
summarizes the habitat area
for the inventory and for each
of the three build-out scenarios
by implementation
mechanism.
In total, Seattle's inventory of
GSI facilities supports about 51
acres of habitat area. Most of
this area (about 42.2 acres) is
supported by green roofs and
roof gardens, which are not
included as a stand-alone
implementation mechanism for
the three build-out scenarios. The inventory data included in the table do not include habitat supported
by code-triggered GSI facilities, because data were not available describing the BMP area. Instead, the
data directly quantified the area of impervious surfaces managed by each facility. The Low, Medium,
and High Build-Out Scenarios implement an additional 45.3, 60.5, and 76.4 acres of habitat each year,
respectively, through 2050. Trees account for about half of the habitat area the build-out scenarios
support.
Table 15. Habitat Area (Acres)

Inventory
Low Build-Out
Scenario (per year)
Medium Build-Out
Scenario (per year)
High Build-Out
Scenario (per year)
RainWise
0,3
0,1
0,2
0.3
Green Roof and Roof Garden
42,2
N/A
N/A
N/A
Right-of-way
8.6
12,3
24.5
36.7
Code-triggered
-
67.4
67.4
67.4
Total
51.0
79.7
92.0
104.4
Notes: This analysis assumes that each tree supports about 710 square feet of habitat (the average of the range of canopy width per tree used
elsewhere in this report). For all other habitat area estimates, this analysis applies average conversion factors by implementation mechanism to
the total stormwater volume estimates described in Appendix B. "Code-triggered" refers to projects resulting from stormwater code
compliance.
Value of GSI-related habitat. GSI facilities could help support ongoing efforts in Seattle to improve
urban habitat quality. In 2004, the Green Seattle Partnership published its 20-Year Plan, which outlines
an actionable set of goals for improving the city's urban habitat.54 In the plan, an inventory was
conducted, which thoroughly describes the city's existing habitat. The plan prioritized parks in terms of
the existence of volunteer support, high-value forests, fish-bearing streams, and other factors. It used
54 Green Seattle Partnership. 2004. 20-year Strategic Plan. Retrieved on July 8, 2013 from http://areenseattle.ora/20-vear-
strategic-plan.
Madison Valley Stormwater Project, SPU. Photo: Mark Buckley
35
-------
the same prioritization model to rank sites within parks. The restoration was conducted in four phases:
(1) invasive removal, (2) secondary invasive removal and planting, (3) plant establishment, and (4) long-
term maintenance. The restoration costs associated with the plan's restoration efforts range from
$2,800-$28,000 per acre, depending on the "tree-iage" category, which is based on tree composition
value and threat.
For this analysis, a one-time restoration cost of $14,000 (midrange) was applied to each acre of habitat
supported by GSI. For the build-out scenarios, this per-acre value in the year the GSI is implemented is
applied and discounted to calculate the NPV through 2050 of the benefit.
Results. Using the methodology described above, the habitat value of Seattle's inventory of GSI facilities
totals about $715,000. The NPVs of the habitat supported by the Low, Medium, and High Build-Out
Scenarios total about $30 million, $34 million, $39 million, respectively. Utilizing the higher end of the
value range would correspondingly double these values. And this avoided cost approach doesn't capture
the full surplus value of these sites, but the availability of substitute opportunities does suggest that this
represents an appropriate representation of the financial tradeoff. As previously described, a number of
Seattle's existing GSI facilities that likely support habitat benefits were not included in this analysis due
to insufficient data describing their geographic extent. Furthermore, this analysis makes a number of
normalizing assumptions to quantify the habitat area GSI efforts implemented through the build-out
scenario support. Unlike other benefits discussed in this report, these habitat values represent one-time
restoration costs that society has been willing to pay, in the past, for habitat restoration actions in
Seattle. In other words, these values do not accrue annually. Rather, they accrue in the year that the GSI
effort is undertaken.
This section includes some of the mechanisms through with GSI can improve small-scale habitat in
Seattle and how individuals derive benefits from these habitat improvements. Data were not sufficient
to quantify the value of these benefits, however the literature provides evidence of GSI-related
improvements in some wildlife populations (e.g., insects and birds). Economic literature has shown that
people value healthy wildlife populations. Data, however, are not sufficient to transfer these benefits
across to those species potentially supported by GSI. Past efforts in Seattle, such as those described in
the Green Seattle Partnership's 20-Year Plan, however, demonstrate local demand for improvements in
the quality of urban habitat. These one-time restoration costs were used as a proxy for society's demand
for habitat improvements, and applied the low end of the range of restoration costs to GSI-related
habitat improvements to provide an estimate of the benefit.
Individuals living near GSI facilities likely benefit the most from improved habitat conditions because
they can observe the change in wildlife most directly. To the extent that others in Seattle and across the
country are concerned about urban habitat conditions, they too may derive benefits from GSI-related
habitat improvements in Seattle.
ologic Function
Development across the Pacific Northwest has altered the hydrology in and around Seattle from pre-
settlement conditions. Altering the region's hydrology has had a number of indirect effects (e.g.,
changes in floodplains, changes in aquatic habitat, and changes in species abundance and health).
Currently, many efforts aim to counter these alterations to the region's hydrology in hopes of improving
ecosystem-wide conditions, and shifting back toward pre-settlement conditions. Insofar as GSI helps
36
-------
improve hydrologic function, it also helps support the broad suite of benefits associated with
improvements in hydrologic function. The complexity of tracing the individual effects of GSI facilities on
nearby surface hydrology makes specific valuation estimates infeasible. It would require means to
estimate how quantitatively streams change as a result of GSI facilities, such as increases in baseflows,
less flashy peak flows, and lower water temperature. Then one could consider how these changes
provide valuable benefits to society. Investigations for Piper's Creek in northeast Seattle suggest
potential benefits of GSI to reduce discharge peaks and peak durations to better match historical or
reference rural land use conditions.55
As GSI contributes to providing improvements to aquatic ecosystems via return to the natural flow regime,
GSI also contributes to the benefits such a natural flow regime provides. Healthy and functional streams
can provide a wide array of benefits to communities via aesthetics, recreation opportunities, protection of
public health, and improvements to wildlife populations. A common focal benefit for stream restoration
benefits in the Puget Sound region is improvements to fish populations. Freshwater, saltwater, and
migratory fish populations are indicators of benefits from hydrologic health improvements.
In the remainder of this section, the extent to which existing efforts in the Puget Sound represent public
demand for the restoration of hydrologic function are described briefly. In addition, an illustrative
example of how survey data can be used to quantify the economic value of improving the health of local
fish populations is provided.
5.7.2 Review and Analysis
Existing efforts. There are currently several efforts underway specifically designed to improve the
hydrologic function within Puget Sound and waterways in the Seattle region. Below, four specific vehicles
through which public agencies are currently addressing water-related issues in and around Seattle are
identified and described. These efforts represent public demand for improvements in the region's
hydrologic function. As GSI facilities help address mutual concerns (e.g., water quality, salmon habitat),
they can support ongoing efforts and decrease the need for other restoration efforts in the future.
•	Water Resource Inventory Area (WRIA) 8 and 9. In 1971, the Washington State legislature
formalized 62 Water Resource Inventory Areas (WRIAs) that identify specific watersheds and
provide the Washington Department of Ecology an organizational framework to implement its
water-related efforts. Seattle covers parts of WRIA 8 (Cedar-Sammamish) and WRIA 9
(Duwamish-Green). Many of the recent efforts in these two WRIAs have involved habitat
restoration for salmon, as well as water quality-related efforts aimed at improving the health of
salmon populations.56
•	Puget Sound Partnership. The partnership's objective is to enact a real Action Agenda that brings
together citizens, governments, tribes, scientists, and businesses in an effort to restore and
protect Puget Sound. The agenda prioritizes cleanup and improvement projects, coordinates
resource streams, and ensures stakeholder cooperation while relying on science-based solutions
to Puget Sound's environmental problems.57 Expenditures associated with the Puget Sound
Partnership's Action Agenda are annually in the tens to hundreds of millions of dollars.
55	Tackett, T., D. Jacobs, C. Carlstad, J. Scheller, J. Zhen, and J. Riverson. Unpublished. Evaluating and Implementing Seattle's
Green Stormwater Infrastructure Approaches at a Creek Watershed Scale.
56	For more information, see http://www.ecy.wa.gov/water/wria/.
57	For more information, see http://www.psp.wa.gov/.
37
-------
•	King County Flood Control District. The District was formed in 2007 to provide a proactive,
regional approach to flooding as well as funding to improve King County's aging and inadequate
flood protection facilities. The District's resources total about $36 million per year, and its
strategy includes: (1) actions to reduce flood risks, (2) research to increase understanding of
flood risk, and (3) efforts to communicate flood risks to stakeholders and the public.58
•	Lower Duwamish Waterway Superfund Site. In February 2013, EPA released the Proposed Plan
for the Lower Duwamish Waterway Superfund Site. The plan presents EPA's preferred
alternative to clean up the waterway, which has been contaminated due to over 100 years of
industrial and urban use. The cleanup strategy has three components: (1) identification and
cleanup of the most contaminated areas, (2) control of sources of contamination, and (3)
cleanup of remaining contaminations in the waterway.59
Focus on aquatic habitat and fish populations. Seattle's stormwater runoff flows into three general
water bodies: (1) Puget Sound, (2) Lake Washington, or (3) smaller waterways (creeks and rivers) in and
around Seattle. Stormwater runoff directly entering surface waterways quickly flows into Puget Sound
or Lake Washington. Given the relatively short distance between where these waterways receive
stormwater runoff and where they flow into larger water bodies, it is assumed there is little in-channel
stormwater treatment during storm events. Consequently, delivery of untreated stormwater ultimately
to Puget Sound is assumed. Untreated surface stormwater runoff contributes to water contamination
that reduces the health of fish populations instream and downstream in Puget Sound. In addition, as GSI
promotes natural flow regime conditions that are conducive to fish population health, it can further
promote population numbers.
People derive value from fish in many ways. One method of estimating the value people place on fish is
asking them how much they would be willing to pay to increase fish population numbers. People are
willing to pay for fish recovery for several reasons, including: (1) an understanding of their function and
contribution to the greater ecosystem, (2) a future option of fishing for, eating, or viewing fish in the
wild, and (3) an appreciation for the existence of fish in the region.
Economic value of improving the health of fish populations. To shed light on the potential value of the
benefits of improving hydrologic function in the Seattle area, this section focuses on public demand for
improvements in fish populations, as they provide an indicator of the magnitude of overall hydrologic
function. This analysis relies on a 1999 economic study done in Washington State that surveyed
households regarding their willingness to pay for programs aimed at increasing fish populations in
western Washington and the Puget Sound area.60 The study (referred to as the LBP Study) asked
households how much they would be willing to pay each month, for 20 years, for a range of increases in
fish populations guaranteed by the end of the 20-year period. The LBP Study examined five types of fish
species. The results of three of these groups in western Washington and Puget Sound are summarized
here: (1) freshwater fish, (2) Pacific migratory fish, and (3) saltwater fish. GSI has the capacity to support
improvements in these three fish populations through several mechanisms, including: improving water
quality in Seattle's waterways and in Puget Sound and improving instream flows during storm events in
Seattle's waterways.
58	For more information, see http://www.kinacouritvfSoodcontroi.ora/.
59	For more information, see U.S. EPA, Region 10. 2013. Proposed Plan: Lower Duwamish Waterway Superfund Site. Retrieved
on June 18, 2013 from http://www.epa.gov/reQionlO/pdf/sites/ldw/pp/ldw pp 022513.pdf.
60	Layton, D., G. Brown, and M. Plummer. 1999. Valuing Multiple Programs to Improve Fish Populations. Washington State
Department of Ecology. April.
38
-------
Applying the results of the LBP Study requires specific information describing the increase in fish
populations tied to GSI. At this point, GSI's effect on local fish populations remains too uncertain to
quantify. Figure 6 shows ranges of annual household willingness to pay for improvements (5%, 20%, and
50%) in the three different fish populations. The blue bars represent freshwater fish, the green bars
represent migratory fish, and the orange bars represent saltwater fish. For example, the blue bar at the
bottom of the figure shows that households in Washington would be willing to pay about $100-$300
per year, for 20 years, for a 5% increase in western Washington and Puget Sound freshwater fish
populations. The low end of the range represents household willingness to pay, assuming that these fish
populations would remain steady for the next 20 years (but for the action). The high end represents
their willingness to pay assuming that these fish populations would decline over the next 20 years (but
for the action).
For this example, it is assumed that beginning in 2013, GSI efforts in Seattle would support a 5% increase
in freshwater, migratory, and saltwater fish in western Washington and Puget Sound by 2033. At this 5%
level, households across Washington would be willing to pay about $110-$200 per year for freshwater
fish, $140-$190 for migratory fish, and $140-$210 for saltwater fish. According to the U.S. Census, there
were about 2.7 million households in Washington in 2010. Using the combined annual household
willingness to pay for these population improvements (about $390-$600) to all households in the state
over a 20-year period, with a 2% discount rate, generates a net present value of about $17.5-$27.1
billion across all three fish groups.
Figure 6. Illustrative Value of Benefits from Improving Fish Population Numbers
in
o £
® 9
_B IT)
3
a
o
a.
CO
m
CD
a-
u
c
Saltwater Fish
Migratory Fish
Freshwater fish
Saltwater Fish
o Migratory Fish
Freshwater fish
Saltwater Fish
in
Migratory Fish
Freshwater fish
$200	$400	$600	$800	$1,000
WTP per Household per Year
39
-------
GSI, at sufficient level of implementation and function, could help streams attain higher levels of
beneficial uses. A study of small stream restoration found nearby households each on average willing to
spend $36-$60 annually for improvements in habitat function, scenic value, or swimmability.61 A
national 1993 study found a similar WTP annual average household value of $70 to maintain national
water quality standards at a fishable level.62 Given the iconic importance of salmon in the Seattle region
and fisheries in general, it can be assumed that households in the vicinity of streams in Seattle would be
willing-to-pay for demonstrable improvements, especially if they can be connected to fish populations.
Si,- - ,v;rr ! "V;.; ¦"
As understanding of the benefits of GSI for hydrologic function in Seattle increases, the benefits of
healthy surface waters can be attributed to GSI investments. Healthy streams provide an array of social,
cultural, and ecological benefits. As an example, the results of a 1999 economic study conducted in
Washington were applied that estimated the willingness of households to pay for policies and programs
that increase fish populations in the Puget Sound Basin. If there are real improvements to fish
populations, households across the state would be willing-to-pay for these benefits.
As scientific understanding of the hydrologic effects of GSI improves, it will be possible to align these
effects with outcomes that matter to society. It likely will be possible to generate watershed-specific
values offish population improvements as well as aesthetic, recreational, and public health benefits.
There can also be important social and environmental justice benefits accrued by low-income and
minority populations who rely upon fish for sustenance or see high cultural value.
61	Collins, A., R. Rosenberger, and J. Fletcher. 2005. The Economic Value of Stream Restoration. Water Resources Research 41.
62	Carson, R. and Mitchell, R. 1993. The Value of Clean Water: The Public's Willingness to Pay for Boatable, Fishable, and
Swimmable Quality Water. Water Resources Research, 29(7), 2445-2454.
40
-------
Stormwater Discharges in Seattle's Creeks
While most of the stormwater traveling through Seattle's sewer system is discharged into large water bodies
(e.g., rivers, lakes), some of the stormwater is discharged directly into urban creeks. Stormwater discharges
directly into small creeks require a separate discussion due to the sensitivity and vulnerability of these
ecosystems. Untreated stormwater discharged into these water bodies is not as polluted as the water
discharged during CSO events but it still transports pollutants directly into creeks, rivers, lakes, and Puget
Sound.
The figure below shows that Seattle's existing inventory of GSI facilities manage over six times more stormwater
in the direct discharge and creek basins than they manage in the combined sewer basin. The avoided treatment
costs associated with the reduction in stormwater runoff in the combined sewer basin total $110,000-$150,000
per year. The water quality benefits in the creek basins though considering no equivalent avoided treatment
cost would be diverse, and associated with benefit categories described in terms of aquatic habitat, aquatic fish
and wildlife, aesthetic views, recreation, human health, and all the other ways that Puget Sound and its
contributing waterways provide benefits.
The level of expenditure directly undertaking activities associated with enjoying Puget Sound, traveling to do so,
and purchasing property that facilitates benefits (e.g. homes on or with views of the Sound) combined with
governmental protection expenditures all reveal the magnitude of these benefits. Consequently, it can be
assumed that benefits from treating discharges to urban creek basins are likely well into the hundreds of
thousands of dollars annually.
80.0
70.0
60.0
50.0
40.0
30.0
20.0
o
CO
d>
T3
4>
U)
CD
c
CO
2
<2	(5
c	<£>
O	>
CD
O
o
(0
c
o
= 10.0
73.0 million
gallons
13.4 million
gallons
Combined Sewer
Basin
Creek and Direct
Discharge Basins
&
$160,000
$140,000
$120,000
$100,000
$80,000
$60,000
$40,000
$20,000
$-




















Not


Quantified
Combined Sewer Creek and Direct
Basin	Discharge Basins
41
-------
5.8 Mental Health
Researchers have been analyzing the link between human health and environmental factors for
decades. Much of this research has focused on how environmental factors influence physical health.
Recently, however, several researchers have shifted their attention to the links between mental health
and exposure to the natural environment. Exposure to the natural environment can directly improve
mental health by providing settings for cognitive respite and reducing stress. Natural spaces can
indirectly improve mental health by promoting activities and interactions (e.g., physical activity and
social interactions) that are known to improve mental health. Human Dimensions of Urban Forestry and
Urban Greening, a project at the University of Washington's School of Environmental and Forest Science,
has identified several of these direct and indirect effects, which are described below.63
•	Reduce stress. Research shows that exposure to natural features helps reduce stress and reduce
other physiological symptoms associated with stress.
•	Improve social capital. Natural features promote social interactions among neighbors and
among individuals visiting the community. Increasing social capital in a community can improve
relationships and help foster social ties and a stronger sense of place.
•	Decrease crime and improve public safety. Crime, vandalism, and littering are less common in
spaces with natural features than in spaces without them. Safer public spaces can reduce danger-
related stress and anxiety among community residents, as well as visitors to the community.
•	Increase physical activity. Research suggests that individuals living in communities with green
spaces are more physically active than those without green spaces. Improving physical health
can help improve mental health by reducing health-related stress and anxiety.
GSI provides the types of urban greening that can contribute to improved mental health. Mental health
protection and improvements have value shown by health and relaxation expenditures, as well as
avoided healthcare costs from effects of stress and other outcomes due to poor mental health. As
research demonstrates a link between the environment and mental health, opportunities are arising to
recognize such benefits from GSI. As described below, studies demonstrate a connection between GSI
and mental health. The next step will be quantifying degrees of change in mental health due to GSI
facilities, and possibly conducting surveys to assess what people would be willing-to-pay to achieve such
mental health improvements. Investigations might also seek examples of expenditures that people have
made for equivalent levels of mental health improvement. It might be possible to use property sales
data to estimate the premium that some people pay to live in areas that provide environmental
conditions that benefit mental health, although it is likely difficult to isolate this share of the value of an
environmental amenity. The remainder of this section briefly summarizes the literature describing some
of the mechanisms through which GSI can help support improvements in mental health, as well as some
of the economic literature describing the economic value of improving mental health.
5.:J. 0	^ .sfrvM
This section briefly summarizes the literature describing the mechanisms through which GSI can help
support improvements in mental health, as well as the economic literature describing the economic
value of improving mental health. These topics are discussed in four parts: (1) mental health benefits to
local residents, (2) mental health benefits to the workforce, (3) mental health benefits to commuters,
and (4) the economic value of improvements in mental health.
63 University of Washington, Urban Forestry/Urban Greening Research. 2013. Green Cities: Good Health. Retrieved on May 31,
2013 from htip://depts. washinaton.edu/hhwbZ.
42
-------
Benefit to local residents. This
desire for contact with nature
serves an important role in
supporting psychological
restoration and improvements in
mental health.64 These benefits
occur not only in the presence of
nature alone, but are also positively
correlated with the quality of nature
in an individual's surrounding.65
Ulrich (1986) used a survey-based
approach to show that American
and European groups have a strong
preference for nature above
human-created surroundings,
particularly when trees and
vegetation are present. These views
tend to have positive effects on
emotional and psychological states.
The study found that trees can reduce stress or anxiety, and that responses to trees are positively linked
to mental health.66 Several other studies have also found that individuals can improve their mental
health by increasing the amount of time they spend in urban green spaces.67 In another nationwide
survey among residents of the Netherlands, 95% of respondents said a visit to nature is a helpful way to
reduce stress.68
Beyond the psychological benefits, such as stress reduction, people also feel a strong emotional
response to natural spaces, even if they do not interact with them directly. This is called existence value.
Hull (1992) found that after Hurricane Hugo damaged infrastructure in Charleston, South Carolina, over
30% of those surveyed found urban forests the most significant feature that was damaged, regardless of
their past or intended use of the forested areas. Of those responses, the largest percentage of
respondents (11%) stated that the reason this feature was special to them was the "positive feelings or
emotions" it invoked. Findings like these suggest that nature evokes positive and relaxing emotions.
While the literature is not specific to GSI-related effects on mental health, it is related insofar as GSI
improves environmental conditions and increases the quantity of green space and natural features in
urban settings.
Madison Valley Stormwater Project, SPU. Photo: Mark Buckley
64	Van den Berg, A., T. Hartig, H. Staats. 2007. Preference for Nature in Urbanized Societies: Stress, Restoration, and the Pursuit
of Sustainability. Journal of Social Issues. 63.1: 79-96.
65	Peacock, J., R. Hine, G. Willis, M. Griffin, and J. Pretty. 2005. The Physical and Mental Health Benefits of Environmental
Improvements at Two Sites in London and Welshpool. Report for the Environmental Agency. March.
66	Ulrich, R. 1986. Human Responses to vegetation and landscapes. Landscape and Urban Planning. 13: 29-44.
67	See, for example, Korpela, K., M, Yien, L. Tyrvainen, and H. Silvernnoinen. 2008. "Determinents of Restorative Experiences in
Everyday Favorite Places," Health & Place. 14(4):636-652.
68	Van den Berg, A., T. Hartig, H. Staats. 2007. Preference for Nature in Urbanized Societies: Stress, Restoration, and the Pursuit
of Sustainability. Journal of Social Issues. 63.1: 79-96.
43
-------
Benefit to workforce. Views of and access to natural settings have been shown to improve worker
mood, productivity, and satisfaction. Hull (1992) found that urban green spaces produce positive moods
for visitors, even if the visits are shorter than 30 minutes.69 In addition to the effect on mood, research
shows that access to nature in the workplace is related to lower levels of perceived job stress and higher
levels of job satisfaction, and that seeing natural features, even if it is from a window, is an effective
means of relieving stress and improving well-being. Workers with a view of trees and flowers felt that
their jobs were less stressful and were more satisfied with their jobs than others who could only see
built environments from their window. In addition, employees with views of nature reported fewer
illnesses and headaches.70 GSI facilities could contribute to these work-related improvements in mental
health and productivity because they offer natural features for workers to either look at while working,
or escape to during their free time.
Benefit to commuters. GSI-related improvements in the
quality of the natural views along transportation corridors
may offer benefits to cyclists and drivers. To achieve these
benefits, planners must consciously consider how to
implement GSI facilities in ways that help support these
types of benefits. GSI facilities that help diversify
transportation options, ease traffic, and increase the quality
of natural landscapes surrounding transportation corridors
can alleviate stress and anxiety associated with travel. The
literature describing mental health benefits for cyclists and
drivers is discussed below.
Cyclists may gain increased utility as the aesthetics improve.
For example, two researchers report that being in an
attractive environment is mentioned as one of the most
positive aspects of cycling, although this statement was not
statistically confirmed.71 Improvements in aesthetics may
....	. .	Roadside bioretention. Photo: MIG |SvR
draw increased numbers of users, because some riders
switch from driving to cycling and some riders who are already bicycle commuters deviate from their
current routes to use routes with GSI. Krizek (2007) found cyclists are willing to travel an average
distance of 2.61 miles out of their way to use a high-quality off-street bicycle facility.72 Stinson (2003)
found that cyclists are willing to tolerate about 10% longer travel times to use routes on residential
streets and routes with dedicated bike lanes on bridges rather than routes on roads.73 To the extent that
69	Hull, R.B. 1992. Brief Encounters with Urban Forests Produce Moods that Matter. 322-324.
70	See, for example, Berto, R., M. Baroni, A. Zainaghi, and S. Bettella. 2010. "An Exploratory Study of the Effect of High and Low
Fascination Environments on Attentional Fatigue." Journal of Environmental Psychology. 30(4):494-500; Heinen, E., B. Wee, and
K. Maat. 2010. Commuting by Bicycle: An Overview of the Literature. Transport Reviews. January. 30.1: 59-96; Kaplan, R. and S.
Kaplan. 1989. The Experience of Nature: A Psychological Perspective. Cambridge University Press; Lohr, V., C. Pearson-Mims,
and G. Goodwin. 1996. "Interior Plants may Improve Worker Productivity and Reduce Stress in a Windowless Environment."
Journal of Environmental Horticulture. 14:97-100; Shibata, S. and N. Suzuki. 2002. "Effects of the Foliage Plant on Task
Performance and Mood." Journal of Environmental Psychology. 22:265-272.
71	Gatersleben, B. and D. Uzzell. 2007. Affective Appraisals of the Daily Commute: Comparing Perceptions of Drivers, Cyclists,
Walkers, and Users of Public Transport Environment and Behavior. 39.3: 416-431.
72	Krizek, K., A. El-Geneidy, K. Thompson. 2007. A detailed analysis of how an urban trail system affects cyclists' travel.
Transportation. 34: 611-624.
73	Stinson, M., and C. Bhat. 2003. Commuter Bicycle Route Choice: Analysis Using a Stated Preferences Model. Transportation
Research Record: Journal of the Transportation Research Board. Vol. 1828.107-115.
44
-------
existing and potential future GSI in Seattle improve the quality of cycling routes, cycling benefits would
increase through the quality and quantity of trips.
Drivers who use roads adjacent to well-positioned GSI facilities would derive mental health benefits
insofar as the increase in exposure to natural landscapes helps them recover more quickly from current
stress and immunizes them to future stress. Parsons (1998) found that survey participants who viewed
nature-dominated drives experience quicker recovery from stress and greater immunization to
subsequent stress than participants who viewed artifact-dominated drives.74 In addition to drivers,
cyclists may reap similar health benefits, though a study of this kind has not been done.
In this section, mechanisms through which GSI can improve the mental health of individuals throughout
Seattle were discussed. To date, research describing the relationship between GSI and mental health in
a Seattle-specific context is not sufficient to quantify the economic value of the potential benefits. The
literature does, however, identify several ways in which GSI can improve mental health, as well as
examples of ways to quantify the economic value of improved mental health. To the extent that GSI
reduces anxiety and stress, it also has the potential to support economic benefits in terms of avoided
mental health-related costs (e.g., medical costs) as well as improved relationships and output through
improvements in work performance.
Since exposure to GSI facilities is crucial in deriving mental health-related benefits, the main
beneficiaries are individuals in Seattle with GSI on their properties, in their neighborhoods, or near their
workplaces. Individuals that encounter GSI facilities on their commute or during other parts of their day
may also derive valuable mental health benefits from their experiences.
1 ::ologicali lutsracf and Behavioral Change
There is an extensive body of literature documenting the kinds of information and experiences that induce
environmentally responsible behavior. The general reasoning holds that when people have environmental
values, or they care about a natural resource for their own benefit, for the benefit of others (such as their
children), or because they recognize protection of natural resources to be a social responsibility, they
make choices that reduce pollution, resource use, and other means of environmental degradation.75
Environmental behavior can be tied to how closely connected people see themselves to the
environment.76 Consequently, individuals take on environmentally responsible behaviors when they:
•	See a personal or community connection to the environment.
•	Feel a responsibility to protect the environment.
•	Experience or expect real effects of environmental degradation on themselves and others.
The costlier or inconvenient the behavior, or the weaker the connection, the less likely people are to
undertake it. For low-cost behaviors, however, people might just need information about the
appropriate behavior. For example, placing small fee on plastic shopping bags has been shown to
74	Parsons, R., L. Tassinary, R. Ulrich, M. Hebl, M. Grossman-Alexander. 1998. The View from the Road: Implications for Stress
Recovery and Immunization. Journal of Environmental Psychology. 18:113-139.
75	See, for example, Hungerford, H., and T. Volk. 1990. "Changing Learner Behavior through Environmental Education." Journal
of Environmental Education. 21(3): 8-22; Karp, D. 1996. "Values and their Effect on Pro-environmental Behavior." Environment
and Behavior, 28(1): 111-133.
76	Davis, J., J. Green, and A. Reed. 2009. "Interdependence with the Environment: Commitment, Interconnectedness, and
Environmental Behavior." Journal of Environmental Psychology. 29(2): 173-180.
45
-------
dramatically decrease their use.77 The fees on plastic bags are negligible as a share of a typical grocery
bill, but the social pressure and signal on social responsibility, combined with a moral sense of self, are
responsive to the signal the bag fee provides.
GSI has the potential to contribute information, signals, and opportunities for environmentally
responsible behavior. Seattle residents and visitors are highly aware of Puget Sound, the importance of
its pristine water quality, and the charismatic fish and wildlife it supports.78 People care about
protecting Puget Sound, and communities like Seattle have made major financial and nonfinancial
commitments towards this end. Still, people don't always recognize how stormwater and individual
behavior and land use contribute to the problems, focusing on the ideas of large industrial and
municipal point-source polluters. This is reflected in surveys from Portland, Oregon, in which some
respondents see stormwater and wastewater as a government responsibility, not an individual one.79
GSI has the potential to contribute to increased environmentally responsible behaviors in three ways:
(1) by providing information about the connection between individual choices/actions and water
pollution, (2) by providing social signals that highlight responsible behavior, and (3) by providing
opportunities to engage directly in environmentally responsible behavior. GSI facilities tend to have
explanatory kiosks, and highly visible pools and vegetation that make the purpose, the operation, and
the effect of personal behavior more evident and relatable to water quality. To the extent that GSI
efforts can help people and businesses better understand how they influence stormwater pollution and
its connection to water quality, habitat, and wildlife, they can provide motivation for behaviors and
investments that provide benefits.
5.9.1 Valuation Methodoiogy
It is not possible, with the tools and information available at this time, to quantify the incremental
contribution of GSI efforts in Seattle to changes in behavior that generate environmental benefits.
Seattle residents and businesses are already aware of the high quality of their environment, and the
importance of individual and business responsibility to maintain this quality. Still, as illustrated by the
surveys in Portland, people don't always realize how stormwater and water quality are related. It seems
likely that highly visible GSI facilities that are aesthetically attractive, provide explanatory information,
and demonstrate comprehensible function are likely to teach and at least remind people of
environmental connections and responsibilities. GSI also provides opportunities for particularly
motivated individuals and businesses to increase their contribution to environmental benefits through
investments and installations on their property and with their financial and nonfinancial (labor,
materials, etc.) contributions. Surveys might elicit how likely experiences with GSI facilities are to inform
Seattle residents about ecological phenomena, and how this affects their environmental behavior.
77	Convery, F., S. McDonnell, and S. Ferreira. 2007. "The Most Popular Tax in Europe? Lessons from the Irish Plastic Bags Levy."
Environmental Resource Economics. 38:1-11.
78	The Puget Sound Partnership has conducted multiple studies demonstrating the importance of Puget Sound health to area
79	See, for example, Action Media. 2011. Keeping Polluted Waters Out of Puget Sound. August; Hansa, G. and ECONorthwest.
2008. Private Motivations to Invest in Stormwater Management Facilities: A Qualitative Exploration and Quantitative
Assessment." Retrieved on July 10, 2013 from http://www.portlandonline.com/bes/index.cfm?a=250709&c=50541] Vivek, S., A.
Nelson, et al. Tabor to the River Program: An Evaluation of Outreach Efforts and Opportunities for Engaging Residents in
Stormwater Management. City of Portland, Bureau of Environmental Services, Retrieved on July 10, 2013 from
h tip://www, portlandonline. com/bes/ in dex. cfm ?a=335473&c=505Q0.
46
-------
5
Evidence from Seattle and similar areas80 suggests that learning and social suggestion (persuasion) can
lead to real changes in behavior that improve environmental quality and reduce costs of its
maintenance. Seattle's GSI facilities, particularly visible and educational ones, and those on private
property, likely contribute to these benefits.
5.9.3 S
The benefits of changes in behavior that improve the quality of water and associated natural systems
accrue to society as a whole and particularly to those who use and experience the resources. Areas
where these actions take place, such as reduced littering, driving, or private property installations, are
likely to experience proportionally more of the benefit. The psychology and sociology literature, such as
that referenced earlier, suggests that the actors experience benefits from socially responsible, moral,
and admirable behavior directly. For businesses, this might contribute to positive marketing and
advertisement benefits.
: -	,idy: Embedded Energy
Recently, researchers have explored the embedded energy associated with specific sets of materials and
construction efforts. By estimating embedded energy, this research attempts to account for GHG
emissions at all stages of production. In this section, how this approach has been used in the context of
stormwater management is described and the application of this approach on two illustrative
stormwater management projects in Seattle is summarized. While these results are not readily
transferable across all forms of GSI or other stormwater infrastructure, they help shed light on the types
of materials typically required for large stormwater projects, and identify specific components of these
projects with particularly large impacts on embedded energy and GHG volumes.
5.10.1 Valuation Methodology
In this section, the embedded GHG emissions associated with the materials used in the construction of two
stormwater infrastructure projects is analyzed. Only construction materials are examined, and GHG
emissions associated with labor during construction are not estimated. GHG emissions are defined as the
total C02e released over the lifecycle of a material, including extraction and manufacturing. Transportation
costs associated with the purchase and delivery of construction materials are not considered. The two
stormwater infrastructure projects considered in this analysis are described below. The first represents a
gray infrastructure project, while the second uses GSI BMPs to manage stormwater.
• The Genesee Area CSO Reduction Project. This project combines additional storage and
transmission pipeline intended to manage a total of about 0.6 million gallons of stormwater
each year. The ability to install GSI BMPs in the area was limited due to geographic constraints.81
The total cost of this facility is about $19.7 million, which includes construction and labor costs.
80	See, for example, Action Media. 2011. Keeping Polluted Waters Out of Puget Sound. August; Hansa, G. and ECONorthwest.
2008. Private Motivations to Invest in Stormwater Management Facilities: A Qualitative Exploration and Quantitative
Assessment." Retrieved on July 10, 2013 from http://www.Dortlandonline.com/bes/index.cfm?a=25Q709&c=5Q541.
81	Seattle Public Utilities. 2013. Genesee Basins. Retrieved on June 12, 2013 from
http://www.seattle.aov/util/EnvironmentConservation/Proiects/DrainaaeSvstem/SewageOverflowPrevention/CSOReductionProiect
s/GeneseeBasin/in dex. h tm.
47
-------
• The Pinehurst Natural Drainage Project. This project was completed in 2006. The project
provides enhanced drainage in the Pinehurst neighborhood and improves the quality of runoff
into Thornton Creek. The project includes new sidewalks, roadways, and landscaping in addition
to bioretention cells designed to manage a total of 9.7 million gallons of stormwater.82 The total
costs of this facility were about $2.2 million, including only construction and labor costs.
The analysis had three steps:
Step 1 - Conduct literature review. The first step was to look through the literature to see if this
kind of analysis had been done before. Three relevant studies were identified that summarized
the results of these and applied several of the methods and data sources in the analysis.
Step 2 - Estimate embedded energy and associated GHG emissions. In the analysis, the GHG
emissions associated with a range of material components used in two stormwater
management projects were examined, the Genesee CSO Project and the Pinehurst Natural
Drainage Project. The available data did not allow for quantifying GHG emissions for all
materials used in each project. Rather, the GHG emissions associated with those materials with
lifecycle emission factors available in the literature were quantified.
Step 3 - Apply the value of carbon. As previously described, a range of per-unit values
representing the potential costs associated with atmospheric GHG concentrations were used.
This range of values includes estimates of the social cost of carbon, as well as market prices for
carbon emissions from efforts to trade and tax emissions across the world. A range of $15-$60
per ton of C02e was applied to the volumes quantified in Step 2.
Literature review. Three previous studies are particularly relevant to this analysis: (1) Wegst (2010)
conducted an eco-audit of seven different GSI BMPs to estimate the lifecycle GHG emissions tied to
materials and construction efforts, (2) Spatari (2011) examined the avoided energy and GHG emissions
of implementing GSI BMPs versus more traditional BMPs to manage stormwater on a one-block area in
New York City using the International Organization for Standardization (ISO) guidelines for lifecycle
assessment, and (3) DeSousa (2012) conducted a lifecycle assessment to compare the environmental
efficiency of three approaches to reducing CSOs to the Bronx River in New York City.83
82 Tackett, T. No Date. Making the Invisible Visible: Seattle's Green Stormwater Infrastructure. Retrieved on June 12, 2013 from
Seattle Public Utilities. 2013. Pinehurst Green Grid. Retrieved on June 12, 2013 from
h ftp://www. Seattle, t
83 Wegst, U., C. Barr, and F. Montalto. 2010. "Eco-audit of Seven Green Infrastructure Practices." Bridge Maintenance, Safety,
and Management. July: 264-272; Spatari, S., Z. Yu, and F. Montalto. 2011. "Life Cycle Implications of Urban Green
Infrastructure." Environmental Pollution. 154: 2174-2179; Wegst, U., C. Barr, and F. Montalto. 2010. "Eco-audit of Seven Green
Infrastructure Practices." Bridge Maintenance, Safety, and Management. July: 264-272; DeSousa, M.R., F.A. Montalto, and S.
Spatari. 2012. "Using Life Cycle Assessment to Evaluate Green and Grey Combined Sewer Overflow Control Strategies." Journal
of Industrial Ecology. 16(6): 901-913.
48
-------
Wegst (2010) used a wide range of material-specific sources to identify their embedded energy and
lifecycle GHG emissions. The study considered eight scenarios in total (six using GSI BMPs and two using
gray infrastructure), and calculated the embedded energy and associated GHG emissions in the materials
of each. The study's findings show that GSI doesn't always offer lower emissions than gray alternatives.
Spatari (2011) analyzed two approaches to stormwater management on a one-block study site in New
York City. The study applied ISO's guidelines for assessing embedded energy of materials and used
emissions data from Wegst (2010) and emissions data for transportation from Wang (2009).84 The study
found that the GSI scenario had lower lifecycle GHG emissions than the gray scenario and also
concluded that nonlinear relationships likely exist between lifecycle GHG emissions and stormwater
volumes.
DeSousa (2012) quantified lifecycle GHG emissions associated with three different options for reducing
CSOs in the Bronx River in New York City. DeSousa quantified the metric tons of C02e over the
construction, operation, and maintenance phases of each project. The three projects included: (1) a
combination of GSI BMPs including porous pavement, bioretention, infiltration planters, rain gardens,
and cisterns, (2) an end-of-pipe detention facility sized to achieve a similar reduction of CSO events and
volumes, and (3) an end-of-pipe detention facility that would physically and chemically treat stormwater
at the tank location. The decentralized GSI strategy outperformed the two gray strategies in terms of
GHG emissions, at all phases of the projects.
Results. In the analysis, the embedded GHG emissions associated with the construction materials used
for the Genesee CSO Project (a gray infrastructure project) and the Pinehurst Natural Drainage Project,
which uses a variety of GSI BMPs, were estimated. For each project, bid sheets from Seattle Public
Utilities were used, which provide line-item descriptions of the types of materials used in the projects
and their quantities. Using these line-item material data, the GHG emissions (in terms of tons of C02e)
were quantified using emissions factors from Wegst (2011) and Hammond and Jones (2008).ssTable 16
summarizes the results of the analysis. The available emissions factors from the literature were not
sufficient to quantify the GHG emissions stemming from each line-item in the construction bid sheets.
For the Pinehurst Project, lifecycle GHG emissions were quantified for a subset of the materials
accounting for 83% of total materials costs (about 435 tons of C02e). For the Genesee Project, the
materials included represent 40% of total materials costs (about 2,514 tons of C02e). Even with this
limitation, the Genesee Project's GHG emissions are about five times larger than those of the Pinehurst
Project. To quantify the total lifecycle GHG emissions associated with the Pinehurst Project, it was
assumed that the remaining materials (those for which the literature did not provide emissions factors)
had the same emissions-to-cost ratio as those materials for which the literature provided emissions
factors. Using this approach, the materials used in the construction of the Pinehurst Project had a total
lifecycle GHG emissions volume of about 524 tons of C02e. Given the tighter data limitations associated
with the Genesee Project, a similar figure for the gray infrastructure project was not estimated. Using
the range of the social cost of carbon indicated above, a range of values for lifecycle GHG emissions for
each line-item included in the analysis was estimated.
84	Wang, M. 2009. GREET 1.8c Spreadsheet Model. Center for Transportation Research, ESD, Argonne National Laboratory.
85	Wegst, U., C. Barr, and F. Montalto. 2010. "Eco-audit of seven green infrastructure practices." Bridge Maintenance, Safety,
Management and Life-Cycle Optimization. Taylor Francis Group: London. 264-272.; Hammond, G. and C. Jones. 2008. Inventory
of Carbon and Energy (ICE): Version 1.6a. Sustainable Energy Research Team, Department of Mechanical Engineering, University
of Bath, UK.
49
-------
Table 16. Lifecycle GHG Emissions of Two Projects

Lifecycle GHG Emissions
(tons of CO e)
Value of Lifecycle GHG Emissions

Pinehurst
Genesee
Pinehurst
Genesee
Asphalt Materials
111
488
$l,665-$6,659
$7,334-$29,337
Concrete Materials
112
2,277
$l,682-$6,727
$34,155-
$136,619
Filter Fabric
1
-
$14-$58
-
Granular Fill
35
-
$523-$2,091
-
Jute Matting
0.1
-
$l-$6
-
Mulch, Compost, and Soil
211
-
$3,166-$12,091
-
Pipes
9
-
$136-$545
-
Metal Fabrication Materials
-
5
-
$77-$310
Materials Costs Calculated (percent of total
costs)
83%
40%
83%
40%
Total Calculated GHG Emissions from
Construction Materials
479
2,771
$7,187-$28,749
$41,567-
$166,267
Total Approximated GHG Emissions from
Construction Materials
577
N/A
$8,659-$9,236
N/A
The summary in the table shows lifecycle GHG emissions in gross terms. As described above, however, the
Pinehurst Project was designed to manage about 9.7 million gallons of stormwater each year while the
Genesee Project was designed to manage only 0.6 million gallons. At a per-unit level, the Pinehurst
Project's lifecycle GHG emissions totaled about 49 tons of C02e per million gallons of stormwater managed
each year. Total lifecycle GHG emissions for all materials used in the Genesee Project was not estimated.
Using only the partial estimate (representing 40% of the Genesee Project's materials costs), its per-unit
emissions total about 4,600 tons of C02e per million gallons of stormwater managed each year.
5.10.3 Summary and Distribution
The results show that, for both projects, lifecycle emissions are generally highest for materials such as
asphalt, concrete, and soil. These results generally agree with Wegst (2010), who also found that these
categories have relatively large volumes of lifecycle GHG emissions. Most of the concrete used in the
Genesee Project went toward the storage tank and facility vault. Together, these two components of the
project accounted for nearly 224 tons of C02e. Regardless of approach (green or gray), stormwater
management efforts can reduce their overall impact on atmospheric GHG concentrations by decreasing
the amount of concrete and asphalt used.
As with other climate change-related benefits, the benefits of pursuing development options with lower
lifecycle GHG volumes accrue to society as a whole. While individuals in Seattle and across Washington will
certainly benefit from reducing the magnitude of climate change impacts, so too will individuals globally.
!	' ~idy: Economic Impacts of GSI
Up to this point, this report has focused on economic values and economic benefits related to GSI. In
this section, it will focus on economic impacts. The term economic impacts has a very specific definition
to economists. Economic impacts represent the number of jobs and the amount of income and tax
50
-------
revenue generated by a particular economic activity. These impacts are not additive to other economic
benefits, but rather a separate prism or metric for evaluating the same activity. An investment that
generates a benefit must do so through expenditures, but under benefit-cost analysis, such
expenditures fall as costs. Economic impact analysis provides useful information for understanding the
distribution and nature of the effects, and this information is particularly relevant to communities
seeking to promote local jobs and local demand for market goods and services.
Insofar as infrastructure projects require spending on labor, materials, and other goods and services,
they support economic activity. Stormwater infrastructure projects are no different. The tools
economists typically use to estimate economic impacts provide gross results. In other words, these
results do not necessarily reflect the share of new jobs or new earnings that would be possible from
considering how the money would have been spent and the workers employed otherwise. After all,
resources used to fund GSI could have been used to fund other infrastructure projects. Similarly, some
of the individuals employed by GSI-related spending could have worked on some other project, or may
have left an existing occupation to pursue GSI-related work. Comparing multiple scenarios though can
shed light on the net impacts.
In general, to the extent that investments can use local materials and local labor, the investments can
have greater local economic impacts than they would otherwise have. To further extend this reasoning,
if investments use emergent industries and generate effects that attract other business and highly
skilled workforces, they can further contribute to economic impact and development. GSI investments
can take on a wide array of investment plans, but they are often seen as offering more potential for such
local economic impacts than large-scale conventional projects that require large built capital and highly
specialized labor imported from elsewhere.
Regardless of these details, economic impacts have become important to decision makers while
justifying and choosing between public spending options. Despite declining unemployment rates (see
Figure 7 below), improving economic conditions remains a large concern for decision makers in the
Seattle area and across the country.
In this section, the results of an economic impact analysis of three of SPU's recent stormwater projects
are described. Two of these projects represent large-scale GSI efforts in Seattle, while the third
represents an even larger gray infrastructure effort. To some extent, comparing the economic activity
each of these efforts supports is like comparing apples and oranges. The specific design parameters for
each of these efforts were largely influenced by several feasibility factors such as spatial, technological,
legal, and cost feasibility (e.g., in some cases, GSI simply is not feasible due to existing land use,
topography, or a number of other factors).86 Rather than serving as a decision criterion, this analysis
helps demonstrate some of the similarities and differences in components of economic activity between
the three efforts, and can help guide future efforts to meet specific economic objectives (e.g., local
employment vs. nonlocal employment).
86 For additional information regarding the consideration of these feasibility factors, see Seattle Public Utilities. 2010. 2010 CSO
Reduction Plan Amendment. May.
51
-------
Figure 7. Average Annual Unemployment Rate
11%
10%
9%
8%
7%
6%
5%
4%
3%
2000 2001
2012 2013
	Washington 	Seattle-Bellevue-Everett MD 	Tacoma MD 	National
Source: Washington Employment Security Department. 2013. Local Unemployment Statistics: Current Estimates. Retrieved on June 13, 2013
from https://fortress.wa.aov/esd/emplovrnentdata/reports-publications/reaional-reports/l'ocal-unemolovment-statistics: U.S. Bureau of Labor
Statistics. 2013. Unadjusted Unemployment Rate. Retrieved on June 6, 2013 from
http://data.bls.aov/timeseries/LNU040000007vears option=all vears&periods option=specific periods&periods-Annual+Data.
Projects included in the analysis. For this analysis, bid sheets for three stormwater infrastructure
projects were used. Two of these projects were large GSI projects, and the third was a large gray
infrastructure project. The three projects are described below:
•	Ballard Roadside Rain Gardens Phase 1 Project was funded by the American Reinvestment and
Recovery Act, and included roadside rain gardens along eight blocks in the Ballard
neighborhood. The project relies on bioretention cells in the right-of-way to reduce the volume
of stormwater entering the combined sewer system, and to reduce the frequency and
magnitude of CSO events. The initial design for the project was intended to manage a total of
38,000 gallons of stormwater each year.87 The total cost for all three projects used in this
analysis (about $0.8 million) includes only construction costs, and does not include additional
costs, such as community engagement, design, and future maintenance.
•	Pinehurst Natural Drainage Solutions Project was completed in 2006. The project provides
enhanced drainage in the Pinehurst neighborhood and improves the quality of runoff into
Thornton Creek. The project includes new sidewalks, roadways, and landscaping in addition to
bioretention cells that are designed to manage a total of 9.7 million gallons of stormwater.88 The
total cost used in this analysis was $2.2 million.
87	Colwell, S. and T. Tackett. No Date. Ballard Roadside Raingardens, Phase 1 - Lessons Learned. Retrieved on June 12, 2013
from http://water.eoa.ciov/infrastructure/areeninfrastructure/upload/ai ballardDroiect.pdf.
88	Tackett, T. No Date. Making the Invisible Visible: Seattle's Green Stormwater Infrastructure. Retrieved on June 12, 2013 from
https://www.seattle.aov/Documents/Departments/UrbanForestrvCommission/2010/2010docs/StormwaterPresentation030310.pdf-.
Seattle Public Utilities. 2013. Pinehurst Green Grid. Retrieved on June 12, 2013 from
http://www. Seattle. aov/util/EnvironmentConservation/Proiects/GreenStormwaterlnfrastructure/CompletedGSIProiects/
Pin eh urstGreen Grid/in dex.htm.
52
-------
•	Genesee Area CSO Reduction Project combines additional storage and transmission pipeline
intended to manage a total of about 0.6 million gallons of stormwater each year. The ability to
install GSI BMPs in the area was limited due to geographic constraints.89 The total cost used in
this analysis (about $19.7 million) includes only construction costs, and does not include
additional costs, such as community engagement, design, and future maintenance.
Approach to conducting an economic impact analysis. SPU provided detailed bid sheets for each of the
three infrastructure projects included in the economic impact analysis. Each bid sheet included line-item
descriptions and costs for materials and labor activities required for the project. IMPLAN (Impact
Analysis for PLANning) modeling software, with 2011 data, was used to examine the economic impacts
of spending related to each of these projects. IMPLAN is an input-output model that works by tracing
how spending associated with a specific project circulates through the defined impact area. For this
impact analysis, the study area is defined as King County.
The results of this impact analysis are grouped into two economic impacts attributable to the
infrastructure projects:
•	Direct Impacts describe the economic activity directly tied to spending associated with each
infrastructure project (e.g., wages paid to local construction workers).
•	Secondary Impacts include indirect impacts and induced impacts. Indirect impacts occur as
businesses buy from other businesses. They begin with changes in economic activity for
businesses that supply directly affected businesses (e.g., the welding supply business that
supplies or rents equipment to construction contractors), and continue as those businesses
purchase the goods and services they need to operate. Induced impacts represent the economic
activity supported by changes in household incomes generated by direct and indirect impacts.
Each type of impact (direct and secondary) is described in terms of several different variables that
measure economic activity:
•	Output is the broadest measure of economic activity and represents the value of production.
Output includes intermediate goods plus the components of value added (including personal
income), so the two measures (output and personal income) are not additive.
•	Personal Income consists of wages and business income. Wages represent wages and salaries,
as well as other payroll benefits such as health and life insurance, retirement payments, and
non-cash compensation. Business income (also called proprietor's income) represents the
payments received by small-business owners or self-employed workers (doctors, accountants,
lawyers, etc.). Personal income is a subset of output.
•	Employment represents full-and part-time jobs. In some instances, this analysis refers to "job
years," which represents the equivalent of one full-time job for a year. Ten job years, for
example, could refer to one job for 10 years, five jobs for 2 years, 10 jobs for 1 year, etc. The
direct employment figure includes all work conducted in King County. Some of these workers,
however, may live outside King County.
Results of the economic impact analysis. Table 17 summarizes the direct impacts and secondary
impacts associated with each of the three projects within King County's boundaries. For each project
and impact category, the table shows total output, total personal income, and total employment. These
89 Seattle Public Utilities. 2013. Genesee Basins. Retrieved on June 12, 2013 from
http://www.seattie.gov/utii/EnvironmentConservation/Proiscts/DrainaaeSystem/SewageOverflowPrevention/CSOReductionPro
iects/Gen eseeB asin/in dex. h tm.
53
-------
impacts are not normalized in any way, and since the impacts are directly linked to project-specific
spending, the results are not surprising. The Genesee project is the most costly of the three (about $19.7
million), and as such, supports more economic activity than the other two projects.
Table 17. Summary of Economic Impacts in King County
Ballard Project
GSI
Gallons Managed: 38,000/year
Total Cost: $0.8 million
Impact Measure
Direct
Secondary
Total
Output
$0.8 million
$0.4 million
$1.2 million
Personal Income
$0.4 million
$0.1 million
$0.5 million
Employment
5.3
3 2
8.5
Pinehurst Project
GSI
Gallons Managed: 9.7 million/year
Total Cost: $2.2 million
Impact Measure
Direct
Secondary
Total
Output
$1.7 million
$1.0 million
$2.7 million
Personal Income
$1.1 million
$0.3 million
$0.3 million
Employment
11.6
7.6
19.2
Genesee Project
Gray
Gallons Managed: 0.6 million
gallons/year
Total Cost: $19.7 million
Impact Measure
Direct
Secondary
Total
Output
$16.3 million
$10.6 million
$26.9 million
Personal Income
$12.0 million
$3.8 million
$3.8 million
Employment
108.2
76.6
184.8
Given the large difference in scale across the three projects, a per-unit comparison offers more helpful
conclusions. Table 18 focuses on the direct economic impacts of the three projects within King County's
boundaries, and summarizes these impacts in per-unit terms. As the data indicate, the Genesee project
has larger impacts than the two GSI projects in absolute terms. Relative to total construction costs,
however, the three projects have similar economic impacts. Relative to the volume of stormwater
managed, the Genesee project has higher total construction costs and larger economic impacts than the
two GSI projects.
Table 18. Direct Impacts of Construction Costs in King County
Impact Measure
Ballard Project
Pinehurst Project
Genesee Project
Total Construction Cost
$0.8 million
$2.2 million
$19.7 million
Total construction cost per gallon managed
$21
<$1
$33
Output
$0.8 million
$1.7 million
$16.3 million
Output per million dollars in cost
$1.0
$0.8
$0.8
Output per gallon managed
$21
<$1
$27
Personal Income
$0.4 million
$1.1 million
$12.0 million
Personal income per dollar in cost
$0.5
$0.5
$0.6
Personal income per gallon managed
$10
<$1
$20
Employment
5.3
11.6
108.2
Employment per million dollars in cost
6.7
5.2
5.5
Employment per million gallons managed
139
1
180
54
-------
The results summarized above are not very helpful to decision makers looking to select one approach
over another based purely on economic activity. In general, the economic activity supported by the
construction of large stormwater projects is closely tied to overall construction costs, not the type of
infrastructure (e.g., GSI vs. gray infrastructure). Furthermore, since the model uses multipliers to
estimate economic activity, several small projects may support the same activity as one large project. If
decision makers select an approach with the objective of supporting local economic activity, there are
several means by which a project's local economic impacts can be strengthened:
•	Purchase local supplies. The economic impact analysis used a regional purchasing coefficient
specific to King County to estimate the percentage of supplies purchased within the county.
Increasing the percentage of supplies purchased in the local area increases local economic
activity because it increases the opportunity for project expenditure to circulate through
secondary impacts within the area.
•	Use the local labor force. Some projects require specialized labor from other parts of the
country. These nonlocal workers spend some of their earnings in the local area, but they send
most of their earnings back home. Local workers, on the other hand, spend a larger percentage
of their earnings in the local community. Increasing the use of local labor increases the
opportunity for project expenditures to circulate through secondary impacts within the area. In
addition, planners can attempt to utilize projects that have relative high labor-to-capital ratios
for overall costs.
•	Hire local firms. The economic impact analysis assumes that all profits accruing to nonlocal firms
leave the local area. All three stormwater infrastructure projects relied on nonlocal construction
firms. Had they used local construction firms, these projects would have had larger local
economic impacts.
As previously discussed, the analysis looked only at the economic impacts associated with construction
spending. All three projects, however, also had non-construction costs (e.g., planning, design, and O&M
costs). In terms of O&M, green and gray approaches differ. Insofar as GSI relies more heavily on labor-
related O&M costs, and if that labor comes from local sources, it has the potential to support more
economic activity than gray infrastructure efforts. For example, as workers spend their earnings in King
County, they will support additional secondary impacts within the county. According to data from the
IMPLAN model, every million dollars in personal income helps support an additional $550,000 in output,
$180,000 in personal income, and four jobs.
5.1 i.2 Economic impacts of SPU's RainWise Program
The previous section discussed the economic impacts of three large GSI and gray infrastructure projects.
In this section, the small-scale GSI projects implemented through SPU's RainWise Program are
examined. SPU's RainWise Program encourages residents to install rain gardens and cisterns to reduce
stormwater runoff from their properties in target CSO basins. Since it began, the program has offered
free training to contractors interested in becoming eligible to install facilities under the program. Over
400 contractors have gone through the training program.90 This section summarizes some data
describing the facilities installed through the RainWise Program, summarizes the results of a short
survey sent to RainWise contractors, and discusses the impacts of the RainWise Program on Seattle's
economy.
90 Personal Communication. Spencer, Bob. Seattle Public Utilities. Telephone. June 3, 2013.
55
-------
Summary of existing RainWise facilities. As of September 2012, a total of 162 facilities have been
installed through the RainWise Program. Table 19 summarizes the types of projects installed through
the program, the number of facilities, the volume of stormwater managed, and the total cost. Rain
gardens account for the majority of the GSI facilities installed through the program. Figure 8 summarizes
costs and management volumes for all facilities installed through the RainWise Program. The blue circles
represent facilities labeled as rain gardens and the red diamonds represent all other facilities (see Table
19 for full list of types of RainWise facilities). The data suggest a linear relationship between project
costs and volume management. At the per-gallon level, project costs decline slightly as the total volume
of stormwater management increases.91 Figure 8 also suggests that rain gardens are generally more
efficient for large volumes of stormwater than other RainWise projects.
Table 19. Summary of RainWise Projects (as of September 2012)
Project Description
Number of Facilities
Volume Managed
(gallons/year)
Total Cost
Rain Garden
108
1,712,869
$502,904
Rain Garden and Cistern
18
345,601
$114,305
Cistern Overflowing to Conveyance
Furrow
3
15,458
$12,496
Cistern Overflowing to Sewer
8
40,800
$31,164
Cistern
8
39,116
$35,134
Cistern Overflowing to Rain Garden
17
312,590
$99,533
Total
162
2,466,433
$795,536
Source: Personal Communication. Emerson, Pam. Seattle Public Utilities. E-mail. October 3, 2012.
Figure 8. Project Costs and Management Volume for RainWise Facilities
~ $12,000
o
o
g $10,000
0
T5 $8,000
1
$6,000
$4,000
$2,000
$-
5,000 10,000 15,000 20,000 25,000 30,000 35,000
Stormwater Volume Managed (gallons/year)
Source: Personal Communication. Emerson, Pam. Seattle Public Utilities. E-mail. October 3, 2012.
Rain Garden
° Other Projects

*



i
Linear (Rain Garden)
p
O

• ~
•
~



~
• *
~ ~
f •

*
/ .
1
"• t
-- " •
W%
L*	.
•

o ©
.«
T &





~~
•





91 Minimum = $0,20/gallon; Maximum = $1.21/gallon; Average = $0.37/gallon; Median = $0,28
56
-------
Contractor survey. To date, a total of 24 contractors have completed work on at least one RainWise
facility. To better understand who these contractors are, and their potential contributions to the Seattle
area's economy, each of them received a brief survey with nine questions asking them about their
business and about their labor/material spending on RainWise projects. A total of 17 contractors
completed at least a portion of the survey. Table 20 describes the firms that responded to the survey.
Many of the contractors provide both design and construction services. Most of the firms are small (1-3
employees) and located in King County.
Table 20. Summary of Contractor Characteristics
Type of firm
Design: 18%
Construction: 0%
Design and Construction: 82%
Other: 0%
Size of firm
1-3 employees: 56%
4-6 employees: 31%
7-10 employees: 13%
More than 10 employees: 0%
Firm location
In Seattle: 65%
Outside Seattle, but in King County: 12%
Outside King County: 23%
One of the main objectives of the survey was to identify how RainWise-related spending is allocated to
different components of the installation process. In the survey, contractors were presented with a
hypothetical RainWise project and were asked to allocate spending across five categories: construction
labor, construction materials, design, overhead, and other. Figure 9 summarizes the results of the
contractors' responses. Construction labor accounted for the majority of costs (about 47%), followed by
construction materials (about 33%). When asked where they purchase materials, respondents stated
that they purchase most (if not all) of their materials from local vendors. In some instances, however,
they stated that they rely on major retail chains for their supplies, which may skew the extent to which
this local spending stays in the Seattle area.
Figure 9. RainWise Project Spending
Construction
Labor
Construction
Materials
Design
Overhead
Other |
0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50%
Percent of Total Project Budget
57
-------
Economic impacts of RainWise Program. As stated above, a total of about $0.8 million has been spent
across 162 RainWise projects. These projects manage a total of about 2.5 million gallons of stormwater
each year. For this analysis, a number of assumptions are made regarding project spending to describe
these economic impacts:
•	Data from Figure 9 are used to distribute spending across spending categories. It is assumed that
design costs all go toward labor. No overhead costs and other costs are included.
•	It is assumed that all construction labor occurs in King County (because all RainWise facilities are
in the county) and that 77% of design labor occurs in King County (based on the locations of the
firms surveyed).
•	It is assumed that 90% of all materials are purchased in King County.
Table 21 summarizes some of the direct economic impacts associated with the existing inventory of
RainWise facilities. Overall, about 55% of all RainWise expenditures ($438,000) were spent on labor that
took place in King County. Using average income figures for construction and design labor, it is noted
these labor expenditures helped support about 6 direct jobs in King County (or about 2.5 jobs per million
gallons managed). Furthermore, about 30% of all expenditures ($237,000) were spent on materials
purchased in King County.
These figures reflect only direct impacts associated with design and construction. They do not include
secondary impacts (i.e., indirect impacts and induced impacts) associated with supplier expenditures
and the circulation of personal income through the economy, nor do they include period maintenance
costs.
Table 21. Summary of Direct Economic Impacts of RainWise

King County
Outside of King County
Labor Spending
$438,000
$20,000
Materials Spending
$237,000
$26,000
Direct Employment
6.3
< 1
6 Climate Change Resilience
In previous sections, climate change has been discussed primarily in terms of the avoided costs
associated with reducing atmospheric concentrations of GHGs. This section discusses how GSI at a
landscape scale can help support Puget Sound area's resilience to the potential effects of climate
change.
The effects of climate change on the Puget Sound region are becoming increasingly well understood,
and the direct importance of attempting to mitigate or adapt to these effects is gaining increasing
political and social support. In general, the changes in climate in terms of averages, extremes, and
patterns of variability, are all expected to contribute to increased scarcity of natural resources in the
region. The University of Washington's Climate Impacts Group identified 10 specific changes attributable
to climate change that likely will affect the Puget Sound area.92
92 The Climate Impacts Group, University of Washington. 2005. Uncertain Future: Climate Change and its Effects on Puget
Sound. October.
58
-------
1.	Continued increases in temperature
2.	Continued increases in water temperature
3.	Continued alteration of river and stream flows
4.	Increased flooding
5.	Accelerated rates of sea level rise
6.	Loss of nearshore habitat
7.	Salt marshes at risk
8.	Further pressures on salmon
9.	Warmer water temperatures
10.	Increased likelihood of algal blooms and low oxygen concentrations in bottom waters
As described throughout this report, GSI does, and increasingly has the potential to contribute directly
to mitigating these effects of climate change in the Seattle area through improvements to water quality
and habitat. One of GSI's greatest potential capacities to provide resilience to climate change comes in
the form of improved flood management. GSI can also mitigate climate change by helping reduce the
urban heat island effect and enhance groundwater recharge. The Climate Impacts Group states that
"with more of the region's winter precipitation falling as rain rather than snow, flooding in Puget Sound
watersheds likely would increase [and] if winter precipitation increases . . . the risk of flooding would be
compounded."93 In 2009, the Climate Impacts Group produced a statewide assessment of climate
change impacts. According to the report, "drainage infrastructure designed using mid-20th century
rainfall records may be subject to a future rainfall regime that differs from current design standards."94
Implicit in this statement is the need for municipalities to expect to manage more rainfall either from
more precipitation or more intense events in the future, and to implement plans and projects that could
curb the potential increase in frequency and magnitude of flood events. By capturing and slowing
stormwater, GSI reduces the extremes orflashiness of flood events.
GSI can provide climate change adaptation benefits to the extent that it can provide services that are
increasing demand under future climate conditions. The above list of identified effects of climate change
in Puget Sound represent areas of adaptation need and increasing scarcity of countervailing services.
Several of the benefits of GSI identified in this report can contribute to directly mitigating these
expected climate change effects, thereby likely increasing the value of these GSI services in the future as
they become more scarce (increasing demand relative to existing supply). For example, increasing
severity of storms and temperature extremes will also increase demand for services to buffer these
extremes.
At times, GSI also can be sited in locations that would not work for conventional stormwater systems, in
part because of the size scaling flexibility or willingness of private households or businesses to host such
facilities because of the co-benefits. In this way, it can provide localized benefits that might not be
available otherwise in the face of climate change, and it provides important diversification to Seattle's
overall climate strategy.
93	The Climate Impacts Group, University of Washington. 2005. Uncertain Future: Climate Change and its Effects on Puget
Sound. October. Pg. 7.
94	The Climate Impacts Group, University of Washington. 2009. The Washington Climate Change Impacts Assessment. June. Pg.
340.
59
-------
6,1 Sanit, wer Overflows
The potential benefits of GSI within the context of climate change are not limited by the range of
benefits described in this report. In particular, GSI can be used to help curb the frequency of sanitary
sewer overflows (SSOs) that likely would increase in the face of climate change. Appendix E contains a
brief memo describing the extent to which GSI can help curb SSOs as the effects of climate change
materialize. Table 22 summarizes the results. Under future conditions with climate change, including GSI
in Seattle's approach to stormwater management has the capacity to reduce the number of SSOs per
year by 1.5 in the Ballard CSO Basin.
Table 22. Summary of Simulation Scenarios and Estimated Number of SSOs per Year
Scenario
Conditions
Includes Climate
Change
Includes GSI
Estimated Number of
SSOs per Year
Scenario 1
Current
No
No
4.9
Scenario 2
Current
No
Yes
3.6
Scenario 3
Future
Yes
No
5.8
Scenario 4
Future
Yes
Yes
4.3
SSO events cause damages that require financial compensation by the City of Seattle. Based on damage
claims filed with Seattle, over 1000 claims were filed for sewer backup and surface water flooding from
late 2003 through mid 2013. Nearly $9 million was paid in total to these claimants, and litigation and
consulting costs were nearly $2 million. The City of Seattle paid on 620 of these claims, and the average
payment was $14,334. Each avoided SSO event would avoid at least tens of thousands of dollars in
damages, and likely tens of thousands of dollars in compensation payments from the city.
60
-------
7 Discussion and Summary
Just as Seattle's GSI efforts employ a diverse set of technologies at a range of scales and locations, so do
they provide an array of benefits to residents, businesses, and agencies in Seattle and beyond. Local,
state, and federal agencies are increasingly considering GSI in their approaches to stormwater
management because it offers a wide range of benefits beyond those related to stormwater, as well as
offering, in certain contexts, stormwater treatment benefits not provided by conventional approaches.
At the same time, GSI in Seattle and elsewhere is best thought of as part of a portfolio of approaches
rather than a single superior approach under all circumstances. GSI benefits accrue to government
agencies, utilities, businesses, communities, and individuals across scales within and beyond Seattle.
Table 23 summarizes the results reported in this analysis. The results demonstrate a number of different
ways in which GSI provides benefits to a number of different beneficiaries. In some instances, these
additional benefits may sway decisions regarding a general approach to stormwater management.
Decision makers should consider the extent to which these benefits apply to their own objectives when
developing stormwater policy and when funding stormwater management efforts. While these results
are specific to GSI efforts in Seattle, many of them also apply to stormwater management efforts
elsewhere in the country. GSI efforts outside Seattle might support only a subset of these benefits or a
broader range of benefits. Nonetheless, understanding and considering these additional benefits
remains an important component to thorough assessment of stormwater management opportunities.
Table 23. Summary of Results
Benefit Category
Inventory
Low Build-Out
Scenario
Medium Build-Out
Scenario
High Build-Out
Scenario
Stormwater T reatment
$66,000-$88,000
$1.8-$2.5 million
$3.3-$4.4 million
$5.5-$7.4 million
Water - Potable Water
Conservation
$5,000-$14,000 for 1000 square feet, and $50-$3000 per acre-foot (one-time value).
Energy - Household Use
$0.2-$0.5 million
$15-$37 million
$17-$43 million
$20-$49 million
Greenhouse Gas Emissions
$0.3-$3.3 million
$25—$284 million
$29—$331 million
$34—$379 million
Air Quality
< $0.3 million
$2.1-$21 million
$2.4-$24 million
$2.8-$27 million
Small-scale Habitat
$0.72 million
$30 million
$34 million
$39 million
Hydrologic Function
Improved hydrology of Seattle's waterways and promote wildlife populations that rely on those
waterways. Potential for annual benefits to nearby households, and lesser benefits to regional residents.
Mental Health
Improved mental health of residents interacting with GSI facilities and improve community cohesion
throughout Seattle. Reduced healthcare costs and improved happiness.
Ecological Literacy and Behavioral
Change
Improved environmental awareness and likely some improved environmental behavior.
Embedded Energy
Reduced lifecycle greenhouse gas emissions of GSI relative to gray stormwater infrastructure.
Economic Impacts
Increased local job and income creation from local GSI construction and operation.
Another way to consider these results is in per-unit terms (e.g., per gallon managed or per tree planted).
Table 24 summarizes the 100-year NPV of each economic benefit in the relevant units. The right side of
the table identifies instances in which each benefit is realized in terms of basin and GSI BMP. For
example, the 100-year NPV of benefits related to household energy use is about $130-$321 per tree.
This benefit is associated with all trees regardless of basin. It is not, however, applicable to non-tree GSI
BMPs.
61
-------
Table 24. Summary of Results (100-year NPV of benefits per unit)
Benefit Category
Units
Per-unit 100-
year NPV
Creek Basin
CSO Basin
Direct Discharge
m
Permeable Pavement
Green Roof
Rainwater Harvesting
Trees
Bioretention
Permeable Pavement
Green Roof
Rainwater Harvesting
Trees
m
Permeable Pavement
Green Roof
Rainwater Harvesting
Trees
Stormwater
Treatment
per million
gallons
$4,945-$6,594





X
X
X
X
X





Household Energy
Use (tree-related
energy reduction)
per tree
$130-$321




X




X




X
GHG Emissions
(home cooling and
heating)
per tree
$84—$324




X




X




X
GHG Emissions
(stormwater
treatment)
per million
gallons
$1,842-7,123





X
X
X
X
X





GHG Emissions
(Tree
sequestration)
per tree
$84—$1,656




X




X




X
Air Quality (tree
filtration)
per tree
$13—$169




X




X




X
Air Quality (home
cooling and
heating)
per tree
$4




X




X




X
Air Quality
(stormwater
treatment)
per million
gallons
$175





X
X
X
X
X





Small-scale Habitat
N/A
Qualitative
X

X

X
X

X

X
X

X

X
CSO Costs
N/A
Qualitative





X
X
X
X
X





Potable Water
N/A
Qualitative



X




X




X

Hydrologic Function
N/A
Qualitative
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Mental Health
N/A
Qualitative
X

X
X
X
X

X
X
X
X

X
X
X
Behavioral Change
N/A
Qualitative
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Climate Change
N/A
Qualitative
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Challenges still exist for quantifying the full range of benefits of GSI. Collection of GSI design, monitoring,
and implementation data will need to be pursued with intentions to better understand the types of
benefits, the level of benefits, and the contexts that generate these benefits. This continued
investigation is necessary for communities to choose the right types, quantities, and contexts for GSI in
their overall water quality management portfolio. While there is a growing body of evidence for a wide
range of benefits, the benefits are determined by biophysical conditions, such as soil, precipitation, and
hydrology, as well as socioeconomic conditions, such as land use patterns and economic sectors.
Community scarcities such as open space or street trees can contribute to the values of these benefits.
62
-------
Another way to consider these benefits is in terms of the groups that receive them. Table 25 provides a
high-level overview of the distribution of the benefits considered in this analysis. Many of these benefits
accrue directly to Seattle residents. Often residents with GSI facilities on their property are not the only
ones to benefit. Nearby residents and others in Seattle can benefit from GSI efforts even if the facilities
are not on their property and even if they do not directly interact with the facilities. Many of the
benefits permeate through the city by contributing to overall improvements in water quality in the
region, overall improvements in air quality, and rate changes from local utilities. Some of the benefits
accrue to individuals beyond the city's limits. This broad distribution of benefits reveals the importance
of community-level planning and implementation, in that individual benefits and incentives likely are
insufficient to generate efficient levels of GSI investment.
Table 25. Distribution of Benefits across Beneficiaries
Benefit Category
Beneficiaries
Seattle
Outside Seattle
GSI Property
Owners
Residents
near GSI
Other
Residents
and
Businesses
Utilities /
Agencies
Washington
National /
Global
Stormwater T reatment
X
X
X
X


Combined Sewer Overflow Costs
X
X
X
X
X

Potable Water
X


X


Energy - Household Use
X
X




Energy - Stormwater Treatment
X
X
X
X


Greenhouse Gas Emissions
X
X
X
X
X
X
Air Quality
X
X
X



Small-Scale Habitat
X
X
X



Hydrologic Function
X
X
X
X
X
X
Mental Health
X
X
X



Behavioral Change
X
X
X
X
X
X
Lifecycle Greenhouse Gas
Emissions
X
X
X
X
X
X
Economic Impacts


X



63
-------
Appendix A, GSI inventory
This appendix summarizes the approach to compiling data describing the inventory of GSI facilities in
Seattle and outlines how the volume of stormwater these facilities manage was quantified. The first
section of this appendix is organized by the six data sources relied on for describing the existing
inventory. Within that section, a series of conversion factors are applied to quantify the volume of
stormwater the inventory of GSI facilities manages. The second section summarizes the inventory in
geographic terms. The third section describes the distribution of the inventory within the socioeconomic
and demographic context. At the end of this appendix, a brief memo is provided specifying the
assumptions used to convert GSI footprints to stormwater volumes as well as a series of maps referred
to throughout the appendix. Table A-l identifies all of the maps included at the end of this appendix.
Table A-l. Maps in this Appendix
Map Title
Map Description
Map A-l. RainWise Facilities
This map shows all GSI facilities associated with the RainWise Program.
Map A-2. RainWise Facilities
This map focuses in on the neighborhoods the RainWise Program Targets
Map A-3. Rainwater Harvest Facilities
This map shows all GSI facilities from the rain harvesting dataset.
Map A-4. Green Roofs and Rooftop Gardens
This map shows all GSI facilities from the green roofs and rooftop gardens datasets.
Map A-5. Code-triggered GSI Facilities
This map shows all GSI facilities from the private GSI audit dataset. These facilities were
constructed prior to 5/4/2011. The facilities in the map do not incorporate the facilities
projected through the end of 2012.
Map A-6. Right-of-way GSI Facilities
This map shows all GSI facilities in the City's right-of-way.
Map A-7. Large Public Facilities
This map shows the one additional GSI facility included in the analysis that did not fit within
the other categories.
Map A-8. All GSI Facilities
This map shows all GSI facilities built in Seattle area to date. They are categorized by facility
type. These data do not include projected facilities extrapolated from the GSI audit dataset.
Map A-9. Volume Managed by GSI Facilities
This map shows the per-acre volume of stormwater managed by GSI facilities each year, by
US Census Tract.
Map A-10. Property Value per Acre
This map shows the distribution of per-acre property values across the City of Seattle, by
property parcel.
Map A-ll. Race and Ethnicity
This map shows the distribution of Seattle's racial and ethnic minority populations, by
US Census Block.
-.dlities by Category
The data describing Seattle's GSI facilities came from six distinct data sources: (1) those installed
through the RainWise Program, (2) rainwater harvest systems, (3) green roofs and roof gardens,
(4) those installed through code-triggered mechanisms, (5) private and public right-of-way GSI facilities,
and (6) one stand-alone GSI facility in a public park. Each category is described in this section. The
number of GSI facilities installed, by category, and the total volume of stormwater they manage is
summarized. Six best management practices (BMPs) are itemized these categories: (1) bioretention,
(2) permeable paving, (3) trees, (4) green roofs, (5) rainwater harvesting systems, and (6) biofiltration. At
the end of this appendix is a detailed table and memo describing the assumptions used to quantify
stormwater volumes.
A-l
-------
i J RainWise Program
SPU's RainWise Program encourages residents to reduce stormwater runoff from their properties in
target combined sewer basins. The program encourages the use of two BMPs (rain gardens and
rainwater harvesting systems). To comply with program guidelines, these facilities must mitigate at least
400 square feet of impervious roof area.95 The program provides rebates to eligible facilities based on
construction costs and the size of the facility (in terms of roof area mitigated). In some instances, these
rebates cover 100 percent of installation costs. To qualify for the rebate, a licensed contractor must
perform the installation and an SPU inspector must perform an infiltration test of the facility.96
A total of 162 facilities were installed under the RainWise Program as of September 2012 (see Map A-l
and Map A-2).97 As shown in Map A-2, most of these facilities are located between NW Market Street
and NW 85th Street, and 15th Avenue NW and 32nd Avenue NW in northwest Seattle. Nearly all of these
facilities are located in the combined sewer basin. Of these facilities, 143 are classified as bioretention,
and are responsible for managing an estimated 2.4 million gallons of stormwater per year from 164,000
square feet (about 3.8 acres) of impervious area. The remaining 19 facilities are classified as rainwater
harvesting, and are responsible for managing an estimated 95,000 gallons of stormwater per year from
24,000 square feet (about 0.6 acres) of impervious area.
i .2 Rainwater Harvest Systems
Through the end of 2012, a total of 22 facilities that were not included in any other GSI programs were
classified as rainwater harvesting systems (see Map A-3).98 These facilities are scattered across the City.
Many of them, however are located in or near downtown Seattle. Nearly all of these rainwater harvest
systems are located in the combined and partially separated sewer basins. Data are not sufficient to
estimate the annual volume of stormwater runoff these facilities manage.
1.3	Green Roofs
Through the end of 2012, a total of 90 facilities that were not included in any other GSI programs were
classified as green roofs or roof gardens (see Map A-4).99 For the purposes of this analysis, both BMP
types are classified as green roofs. As with the rainwater harvesting systems, these facilities are
scattered across the City, although there is a concentration of green roofs in or near downtown Seattle.
In total, these facilities are responsible for managing an estimated 8.4 million gallons of stormwater per
year from 1.8 million square feet (about 42.2 acres) of impervious area.
1.4	Code-triggered Projects - Private Property GSI Audit
To comply with Seattle's stormwater code, some types of new developments are required to
incorporate GSI BMPs to the maximum extent feasible (limited by engineering and design feasibility,
physical limitations of the site, and economic feasibility).100 In general, there are three types of new
developments that trigger Seattle's stormwater code: (1) new developments disturbing more than 7,000
95	Seattle Public Utilities. 2012. RainWise Detail Sheet 1: Facility Sizing Tables. Retrieved on January 18, 2012 from
http://www.seattle.pov/util/groups/public/@spu/@usm/dacuments/webcontent/02 008087.pdf.
96	Seattle Public Utilities. 2011. Rainwise Tools. Retrieved on January 18, 2013 from
https://rainwise.seattle.gov/citv/seattle/rainwise rebates.
97	Three BMPs were not mapped due to insufficient data describing their locations.
98	Three BMPs were not mapped due to insufficient data describing their locations.
99	Ten BMPs were not mapped due to insufficient data describing their locations.
100	City of Seattle. Seattle Municipal Code (SMC) 22.800-22.808.
A-2
-------
square feet, (2) new developments that generate new/replace over 2,000 square feet of impervious
area, and (3) new single-family residential dwellings.101
From November 30, 2009 to May 4, 2011, a total of 206 developments triggered Seattle's stormwater
code. Of these, 122 developments went on to install GSI facilities (see Map A-5). Table A-2 summarizes
these facilities. Data describing code-triggered facilities since May 4, 2011 are not available. Existing
data were used to extrapolate the likely contributions of these additional facilities. In total, code-
triggered facilities installed through the end of 2012 manage an estimated 8.7 million gallons of
stormwater per year, mitigating stormwater from about 0.7 million square feet (about 15.0 acres) of
impervious area.
Table A-2. Summary of Code-Triggered Facilities (through May 4, 2011)
BMP
Number of Installations
Total Stormwater Managed
(gallons per year)
Bioretention
30
1,988,000
Green Roofs
5
23,000
Permeable Pavement
25
805,000
Multiple BMPs (1)
21
705,000
Trees
114
495,000
Total
N/A
4.0 million
(1) A total of 114 sites installed trees. Most sites installed trees as well as other BMPs. In order to make the discussion more useful, trees were
considered alone followed by other BMPs installed at these sites. This row represents sites that had multiple BMPs in addition to trees
potentially planted on the site.
Tree-specific data are important to economic analyses described in this report. Table A-3 summarizes
the approach to quantifying the number of trees associated with code-triggered facilities. In total, 759
trees contributed to code-triggered GSI facilities from November 30, 2009 to May 4, 2011. Assuming a
linear increase in trees over time, there likely were a total of about 1,640 trees associated with code-
triggered GSI facilities through the end of 2012.
Table A-3. Summary of Tree-related Data from Code-Triggered GSI Facilities

Number of Trees
Total Canopy Area (square feet)
Area Mitigated (square feet)
Existing Evergreen (9/30/09-5/4/11)
98
9,933
10,756
Projected total through 2012
212
21,509
23,291
Existing Deciduous (9/30/09-5/4/11)
115
10,931
6,721
Projected total through 2012
249
23,670
14,554
New Evergreen (9/30/09-5/4/11)
104
N/A
5,200
Projected total through 2012
225
N/A
11,260
New Deciduous (9/30/09-5/4/11)
442
N/A
8,840
Projected total through 2012
957
N/A
19,142
Total (9/30/09-5/4/11)
759
N/A
31,517
Projected total through 2012
1,644
N/A
68,246
Source: Data from SPU
Notes: Data describe trees associated with code-triggered projects from November 30, 2009 to May 4, 2011.
101 Seattle Public Utilities. 2013. City Policies Requiring and Related to Using GSI. Retrieved on May 3, 2013 from
http://www.seattie.gov/util/EnvironmentConsen/ation/Praiects/GreenStormwaterinfrastructure/StormwaterCode/
CitvPoiiciesRequirin gRelatedto usin g GSI/in dex.htm.
A-3
-------
I
Data from SPU identified a total of 666 additional right-of-way GSI facilities not included in any of the
other datasets.102 Of these facilities SPU owns and operates 562 of them. The other 104 are privately
owned. As shown in Map A-6, these facilities are distributed across the City, but two areas in particular
contain many of the facilities (one in the northwest part of the City and one in the southwest part of the
City). Each of these is discussed in turn.103
•	High Point. This cluster of facilities in southwest Seattle is in the High Point neighborhood. The
cluster contains a total of 449 facilities, which manage a total of 36.0 million gallons of
stormwater each year from about 58 acres of impervious area. Nearly all of these facilities,
however, are connected to an underdrain, which is primarily tasked with water quality
treatment. These facilities also provide stormwater flow management by reducing the intensity
and duration of peak flows through the bioretention system, before stormwater reached the
detention pond downstream. These underdrain facilities manage a total of 34.3 million gallons
of stormwater each year.
•	The Cascades. This cluster of facilities is in northwest Seattle. The Cascades contains a total of
121 GSI facilities, which manage a total of 10.0 million gallons of stormwater each year from
about 15.4 acres of impervious area. Some of these facilities also incorporate surface storage
detention elements which manage additional stormwater volumes. A total of 44 facilities within
the Cascades have these detention elements. In total, these facilities manage an additional 6.2
million gallons of stormwater each year from about 9.3 acres of impervious area.
•	Swale on Yale (aka Capitol Hill Water Quality Improvement Project). This project, when
completed, will treat an average of 190 million gallons of stormwater annually flowing from
Capitol Hill into Lake Union, greatly reducing the amount of pollution flowing into the lake. It
does this by diverting the stormwater into a series of extra-wide biofiltration swales between
the sidewalk and the roadway. These naturalistic, biofiltration "swales" are designed to slow the
stormwater flow and remove pollutants before they reach the lake. The first phase of the
project was completed in 2013, and the portion of the runoff managed by infiltration through
the bioretention soils or evaporation are included within this analysis.
In total, these right-of-way GSI facilities manage an estimated 85million gallons of stormwater per year
from over 240 acres of impervious area. Table A-4 summarizes the distribution of these GSI facilities by
BMP.
Table A-4. Summary of Right-of-way GSI Facilities
BMP
Number of Installations
Total Impervious Area
Managed (acres)
Total Stormwater Managed
(gallons per year)
Bioretention
501
94.3
59.8 million
Biofiltration
48
10.3
5.3 million
Biofiltration - Swale on Yale Phase 1
2
Appx. 140
18 million
Permeable Pavement
117
3.7
2.0 million
Total
666
248
85.0 million
102	The number of facilities includes only facilities with a BMP footprint greater than 200 square feet. Smaller facilities are
assumed to be components of larger efforts. The total impervious area and stormwater volume managed includes all facilities
regardless of size.
103	More information on projects available at
httD://www.seattle.aav/util/MvSeivices/DrainaaeSewer/Proiects/GreenStormwaterlnfrastructure/CurrentGSiProiects/index.htm
A-4
-------
I
SPU identified one additional public project that does not fit into any of the categories discussed thus far
(see Map A-7). The Thornton Creek Water Quality Channel is located in the Northgate neighborhood.
The facility is designed to improve water quality by slowing the flow of stormwater runoff, allowing
sediments and pollutants to settle out of the flowing stormwater.104 This project manages an estimated
14,000 gallons of stormwater per year through infiltration and evaporation.
• ' ;raphic Summary of Stormwatc ' itmerit
Map A-8 shows all GSI facilities currently constructed in Seattle. The facilities are distributed by facility
type. This summary of GSI facilities shows that, by and large, the distribution of facilities is dictated by
type of facility and program type. The RainWise Program, for example, supports the large cluster of
bioretention facilities in northwest Seattle. Green roofs account for most of the GSI facilities in
downtown Seattle. The two large right-of-way GSI efforts in northwest and southwest Seattle represent
the remaining clusters of GSI facilities in the City. The remaining GSI facilities are distributed across
Seattle. Areas north of downtown Seattle have relatively more GSI facilities than areas south of
downtown Seattle, but there does not appear to be a general relationship describing the distribution
aside from the clustering already identified.
Table A-5 summarizes the GSI facilities considered in this analysis. The right-of-way GSI facilities manage
the most water, followed by code-triggered GSI facilities on private property, green roofs, and facilities
installed through the RainWise program. In total, these facilities manage a total of 86.6 million gallons of
stormwater each year.
Table A-5. Summary of Existing GSI Facilities in Seattle, by Data Source

Impervious Area Mitigated (acres)
Total Stormwater Managed (gallons per year)
RainWise Program
4.3
2.5 million
Rainwater Harvest Systems
N/A
N/A
Green Roofs and Roof Gardens
42.2
8.4 million
Private Property GSI Audit*
15.0
8.7 million
Right-of-way GSI Facilities
248
85 million
Additional GSI Facilities
0.2
< 0.1 million
Total
309
104.6 million
* The values in this table for the private property GSI audit project stormwater volumes through the end of 2012 based on historical data of
implementation rates. These projected volumes are not, however, reflected in Map A-8 due to uncertainty regarding their specific spatial
distribution across the City.
Table A-6 summarizes the GSI facilities by facility type. The table shows the number of facilities and the
volume of stormwater the facilities manage each year. For reasons described earlier, the number of GSI
facilities cannot be summed due to complexities involved in the code-triggered GSI projects. The volume
of stormwater managed, however, can be summed without concerns of double-counting.
104 SvR Design Company. 2009. Thornton Creek Water Quality Channel. October 28. Retrieved on May 24, 2013 from
http://www.seattle.gov/util/cs/aroups/puhlic/documents/webcontent/spu01 006146.pdf.
A-5
-------
Table A-6. Summary of GSI Facilities in Seattle, by Facility Type*

Number of Facilities
Total Stormwater Managed (gallons per year)
Bioretention
674
64.1 million
Biofi It ration
52
23.3 million
Green Roof
95
8.4 million
Permeable Paving
142
2.8 million
Rainwater Harvesting
31
0.1 million
Trees
114
0.5 million
Multiple BMPs
21
0.7 million
Total
N/A
99.9 million
* The number of facilities and the volumes presented in the table do not incorporate the projections of code-triggered facilities through the end
of 2012.
The volumes presented in the table do not incorporate the projected volume of stormwater managed at
private, code-triggered facilities. Table A-7 summarizes the distribution of the existing inventory's
stormwater management by basin and implementation mechanism.
Table A-7. Summary of GSI Facilities in Seattle, by Basin (gallons managed per year)
Basin
Right-of-way
Code-triggered
RainWise
Other
Total
Combined Sewer Basin
4.4 million
3.3 million
2.3 million
3.4 million
13.4 million
Creek Basin
62.1 million
0.8 million
-
< 0.1 million
63.0 million
Direct Discharge Basin
18.3 million
4.6 million
0.2 million
4.9 million
28.0 million
Total
66.9 million
8.7 million
2.4 million
8.4 million
104.4 million
Notes: Values elsewhere in this report may differ slightly from other values in this appendix due to rounding. Code-triggered values include
projections from mid-2011 to the end of 2012 as described assuming a weighted distribution across the three basins based on the distribution
of facilities installed prior to mid-2011.
Another way to think about the distribution of these GSI facilities is to look at the average volume of
stormwater they manage in per-acre terms, by census tract. Map A-9 shows this distribution. GSI
facilities in the dark blue census tracts manage the most stormwater in per-acre terms (over 10,000
gallons per acre per year). Yellow census tracts have no GSI facilities. As the map shows, while there are
several census tracts north of downtown Seattle with no GSI facilities, the census tracts with GSI
facilities generally manage more stormwater than census tracts south of downtown.
iatt :• • " ; - jects in a Socioeconomic Context
The final two maps in Appendix A reflect Seattle's socioeconomic profile. The first map shows per-acre
property values across the City. As expected, property values are highest in downtown Seattle and in the
more affluent areas on the north side of the City. The second map shows the distribution of the City's
population in terms of race and ethnicity. Areas in south and southeast Seattle have the largest minority
populations, while areas in north and west Seattle tend to have smaller minority populations. Aligning
these data with data describing stormwater facilities suggest that, in general, areas with high property
values and small minority populations tend to have more GSI facilities and more stormwater managed
by GSI facilities than areas with lower property values and larger minority populations.
A-6
-------
ersion Factors
The table below summarizes the data used and the assumptions applied to calculate the annual volume
of stormwater that GSI facilities in Seattle manage each year. A copy of the memo SPU staff provided is
at the end of this appendix.
RainWise Program
GSI BMP from File
Impervious Footprint Factor
(square feet)
Annual Volume Factor
(cubic feet)
Final GSI BMP used on Maps
Cistern
20.83
0.53
Rainwater Harvesting
Cistern Overflowing to Back to
Sewer
20.83
0.53
Rainwater Harvesting
Cistern Overflowing to
Conveyance Furrow
20.83
0.53
Rainwater Harvesting
Cistern Overflowing to Rain
Garden
35.71
2.00
Bioretention
Rain Garden
10.75
1.91
Bioretention
Rain Garden & Cistern
35.71
2.00
Bioretention
Green Roofs and Roof Gardens
GSI BMP from File
Impervious Footprint Factor
(square feet)
Annual Volume Factor
(cubic feet)
Final GSI BMP used on Maps
Green Roof
1.00
0.61
Green Roof
Roof Garden
1.00
0.61
Green Roof
Private Property GSI Audit
GSI BMP from File
Impervious Footprint Factor
(square feet)!
Annual Volume Factor
(cubic feet)
Final GSI BMP used on Maps
Trees
N/A
2.10
Trees
Downspout
N/A
1.91
Bioretention
Bioretention Cell
N/A
1.91
Bioretention
Permeable Pavement
N/A
1.63
Porous Pavement
Green Roof
N/A
0.61
Green Roof
Bioretention Planter
N/A
1.91
Bioretention
Roadside GSI Facilities
GSI BMP from File
Impervious Footprint Factor
(square feet)
Annual Volume Factor
(cubic feet)
Final GSI BMP used on Maps
BIO
SPU: 21.74 | Other: 10.75
SPU: 2.00 | Other: 1.91
Bioretention**
BSB
SPU: 21.74 | Other: 10.75
SPU: 2.00 | Other: 1.91
Bioretention**
BSW
1.00
0.21
Biofiltration
PP
1.00
1.63
Porous Pavement
BIO in High Point
38.46
1.91
Bioretention**
BSB /BSW in Cascades (additive)
26.20
2.06
Bioretention** and Biofiltration
Additional GSI Facilities
GSI BMP from File
Impervious Footprint Factor
(square feet)
Annual Volume Factor
(cubic feet)
Final GSI BMP used on Maps
Biofiltration
1.00
0.21
Biofiltration
* There is no impervious footprint factor for facilities associated with the private property GSI audit because the stormwater code database
includes the specific impervious area each facility manages.
"BMP square footage was divided by 1.8 to convert from top of BMP area to bottom of BMP area.
A-7
-------
City of Seattle
Seattle Publiic Utilities
April 23, 21013
MEMORANDUM
To: Mark Buckley and Tom Souhlas, ECONorthwest
Fr: Tracy Tackett, PE, GSI Program Manager
Re: Valuing the stormwater benefit for GSI technologies within Seattle study
This memo is to provide guidance on the following question: "Given X square feet of a GSI best
management practice, what is the average annual volume of stormwater runoff managed in the
Seattle area?"
Seattle has three types of drainage systems, Creeks, non-creek separated systems, and
combined sewer systems. Each system has different stormwater management objectives,
including reducing stormwater rates and volumes for creek biota protection, deductions of
peak intensities and durations for pipe capacity preservation and/or reducing combined or
sanitary sewer backups, and removing pollutants prior to discharge to our waterbodies. As a
way of providing Citywide evaluation as well as tracking Citywide implementation of GSI,
Seattle has begun to quantify the average annual volume managed by GSI approaches.
"Managed" flow equates to the volume of stormwater runoff that is reused (in the case of
rainwater harvesting), removed by evaporation or infiltrating into native soils, or slowed
through engineered soil media. The methodology for quantifying these different management
approaches are provided below.
Two primary data sources provide the technical basis of the recommendations. These sources
were generated during the development of Seattle's stormwater manual, which requires GSI to
the Maximum Extent Feasible. To facilitate installation of GSI practices, we had technologies
presized, thereby reducing the need for complicated stormwater modeling for project with less
than 10,000 square feet impervious surfaces. When a stormwater code applicant is using these
presizing factors to manage runoff from impervious surfaces, we do not require additional
management for pervious areas discharging to the BMP. This is because the stormwater code
also requires landscaped areas to be amended with compost. For simplicity I recommend we
use the same approaches for this economic valuing exercise. Although there is some runoff
from pervious areas, it is small relative to the contribution from impervious surfaces. Also
through detailed SWMM v5.022 modeling within the Ballard CSO basin, we have found the
sizing factors developed for the small projects (<10,000SF) to be representative of larger
Ray Hoffman, Director
Seattle Public Utilities	Tel (206] 684-5851
700 5th Avenue, Suite 4900	Fax (206] 684-4631
PO Box 34018	TDD (206] 233-7241
Seattle, WA 98124-4018	rav.hoffman@seattle.gov
htto: //www.seattle.gov /util
An equal employment opportunity, affirmative action employer. Accommodations for people with disabilities provided on request.
-------
projects where both impervious and pervious surfaces are managed. Note that the calculations
provided for impervious area managed assume runoff from an equivalent or larger drainage
area is contributing to the facility; if less area is contributing calculations would need to be
modified accordingly. If further information is desired, the data sources are as follows:
•	"SPU 201.1 / DPD 15-2012 Requirements for Green Stormwater Infrastructure to the
Maximum Extent Feasible for Single-Family Residential and Parcel-Based Projects "
(available at seattle.gov/util/greeninfrastructure see "stormwater Code" then "City
Policies Requiring and Related to using GSI. The majority of GSI to MEF Credits are
based on infiltrating 91%-95% of the total runoff volume produced by the runoff file
(per Section 6.5.4.6 of the Stormwater Manual Volume 3). For non-infiltrating
technologies, the GSI to MEF Credits for impervious surface reduction methods are
based on achieving a 91% reduction of the 1-year recurrence interval flow.
•	2009, Herrera. Memorandum "Average Annual Runoff Volume from Impervious Surface
in the City of Seattle". This was subsequently validated through evaluation of calibrated
models in our Long Term Control Plan. Provides basis for this calculation. Average
annual volume runoff generated from the impervious surface can be calculated by
multiplying the impervious surface are (square feet) by 2.1 (cubic feet/square foot),
resulting in cubic feet of stormwater runoff managed.
Table 1: BMPS managing runoff through volume reduction
GSI Technology/ BMP
Flow management
approach
Impervious area
managed (SF)
Average annual runoff
volume managed (CF)
Bioretention, infiltrating.
Facilities installed by SPU
(1)
Removed by infiltrating
into native soils
BMP bottom area t-
4.6%
Impervious area
managed 2.1x95%
Bioretention, infiltrating.
Facilities installed by others
(1)
Removed by infiltrating
into native soils
BMP bottom area t-
9.3%
Impervious area
managed x 2.1 x 91%
Permeable paving surface
(2)
Removed by infiltrating
into native soils
BMP area x 1
Impervious area
managed x 2.1 x
((100% + 55%)/2)
Permeable pavement
facility
Removed by infiltrating
into native soils
BMP area x 2.5
(3)
Impervious area
managed x 2.1 91%
Trees, deciduous, newly
planted or retained (4)
Removed by
evaporation,
evapotranspiration or
infiltrating into native
soils
Canopy area x 11%
Impervious area
managed x 2.1
Trees, evergreen, newly
planted or retained (4)
Removed by evaporation
or infiltrating into native
soils
Canopy area x 22.5%
Impervious area
managed x 2.1
April 26, 2013
Page | 2
-------
Rain water harvesting
Reused
Facility not presized
Project specific, refer
to project data
Biofiltration swale without
underdrain
Credit for portion
removed by infiltrating
into native soils
Facility not presized
Impervious area
managed x 2.1 x 10%
Greenroofs (5)
Evapotransporation
component
Green roof area x 1
Impervious area
managed x 2.1 x 29%
Cisterns (6)
Reuse component
Cistern area t- 4.8%
Impervious area
managed x 2.1 x 25%
Cistern to raingarden (7)
Removed by infiltrating
into native soils
Cistern area t- 2.8%
Impervious area
managed x 2.1 x 95%
(1)	SPU installed facilities are assumed to be typical NDS systems designed to infiltrate 95% runoff. The sizing
factor used is 4.6% based on the majority of City retrofit project conditions with design infiltration rates
between 0.5 and 0.9in/hr, and ponding depths between 9-inches and 12-inches.
For facilities installed by private entities or other agencies, the data based provide by the City already
included impervious area managed for all installations except those in the right-of-way For right-of-way
installations, a sizing factor of 9.3% is used because the projects were predominantly bioretention cells
with 2-inches of ponding and 0.25inch/hr infiltration rates, designed to infiltrate 91% runoff.
(2)	Average of flow control credits for low slope and high slope installations.
(3)	This assumes 150% run-on to the facility (i.e., run-on area is 1 and % times the facility area)
(4)	Source: 2008, Herrera. "The Effects of Trees on Stormwater Runoff". Assumed half canopy within 10-feet
of impervious and half greater than 10-feet from ground level impervious. These credits are more
generous than those adopted in the City of Seattle Stormwater Manual.
(5)	2012, She. Memorandum "Seattle Green Roof Modeling Results for Emergency Operation Center and Fire
Station 10". Based on the volume delayed for larger storms in Table 2
(6)	Assuming average contributing roof size 1,200 square feet and 25% reuse of stored volume.
(7)	Assuming average contributing roof size 1,200 square feet, 1 cistern, and a bioretention cell with 6-inch
ponding and 0.25in/hr native soil infiltration rate. Sizing is based on temporary storage in the cistern,
which is then metered into the raingarden; use of the cistern for delaying stormwater peaks allows a
smaller rain garden footprint than a system without a cistern. For this table, the focus is the volume
infiltrated; the calculation is representative of the total volume infiltrated by the cistern/raingarden
combination.
Some GSI practices do not remove significant stormwater volumes, but provide stormwater
management through either water quality treatment or flow delay. Note that these practices
should not be included in the valuation calculations for categories such as reduced energy use
as these volumes do enter the downstream piping system.
April 26, 2013
Page | 3
-------
Table 2: BMPS managing runoff without volume reduction
GSI Technology/ BMP
Flow management
Impervious area
Average annual

approach
managed (SF)
runoff volume
managed (CF)
Bioretention with underdrain,
Water quality treatment
BMP bottom area t-
Impervious area
non-CSO basin (8)

2.6%
managed x 2.1 x
91%
Bioretention with underdrain,
Slowed through
BMP bottom area t-
Impervious area
CSO basin (8)
engineered soil media
2.6%
managed x 2.1 x
46%
Green roofs
Slowed through
engineered soil media
Green roof area x 1
Impervious area
managed x 2.1 x
55%
Cisterns o single family
Reused
Cistern area 4.8%
Impervious area
properties


managed x 2.1x 95%
Biofiltration swale with
Credit for portion
Swale bottom area
Impervious area
underdrain (9)
slowed through
engineered soil media
h- 0.5%
managed x 2.1 x
19%
(8)	Facilities assumed to have 6-inch ponding depth. (Note, High Point is predominantly bioretention with
underdrain, non-CSO basin)
(9)	Biofiltration sizing factor of 0.5% is size estimated to conform to Department of Ecology criteria for
biofiltration (per communication with Jason Sharpley-SPU based on modification of Swale on Yale
project calculations). Flow treated through soil and drained through underdrains or infiltration into native
soil is therefore estimated to be equal to the proportional sizing factor relative to an equivalent
bioretention facility, i.e. 0.5%/2.6% = 19%.
April 26, 2013
Page | 4
-------
Map A-l.
RainWise Facilities
BMPs
~	Rainwater Harvesting
•	Bioretention
Sewer Classification
Combined
Partially Separated
Separated
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
3 mi
J
ECONorthwest
ECONOMICS • FINANCE • PLANNING


-j
-------
Map A-2.
RainWise Facilities
<
$

CD
Q)
CO
/
54tf\ St
BMPs
~	Rainwater Harvesting
•	Bioretention
Data from Seattle Public Utilities
and King County GIS Center.
July 2013.
0	0.25
	1	I	
ECONorthwest
Sewer Classification
Combined
Partially Separated
Separated
0.5 mi
	I
ECONOMICS • FINANCE ¦ PLANNING
-------
Map A-3.
Rainwater Harvest
Facilities
BMPs
~ Rainwater Harvesting
Sewer Classification
Combined
Partially Separated
Separated
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
3 mi
J
ECONorthwest
ECONOMICS • FINANCE • PLANNING
V
-------
Map A-4.
Green Roofs and
Rooftop Gardens
BMPs
¦ Green Roof
Sewer Classification
Combined
Partially Separated
Separated
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
ECONorthwest
ECONOMICS • FINANCE • PLANNING

\
V

-------
BMPs
*	Bioretention
¦ Green Roof
~ Permeable Pavement
*	Trees Only
*	Multiple BMPs
Sewer Classification
Combined
Partially Separated
Separated
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
0	1	2	3 mi
	1	I	I	I
ECONorthwest
ECONOMICS • FINANCE • PLANNING
Map A-5.
Code-triggered
GSI Facilities
-------
Map A-6.
Right-of-way GSI
Facilities
BMPs
a Biofiltration
• Bioretention
~ Permeable Pavement
Sewer Classification
Combined
Partially Separated
Separated
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
3 mi
J
ECONorthwest
ECONOMICS • FINANCE • PLANNING

rC
-------
Map A-7.
Large Public
Facilities
BMPs
± Biofiltration
Sewer Classification
Combined
Partially Separated
Separated
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
3 mi
J

Thornton Creek
Water Quality
Channel
ECONorthwest
ECONOMICS • FINANCE • PLANNING


V
-------
Map A-8.
All GSI Facilities
BMPs
a	Biofiltration
~	Rainwater Harvesting
•	Bioretention
¦	Green Roof
~	Permeable Pavement
*	Trees Only
*	Multiple BMPs
Sewer Classification
Combined
Partially Separated
Separated
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
0	1	2	3 mi
	1	I	I	I
ECONorthwest
ECONOMICS • FINANCE • PLANNING
-------
Map A-9.
Volume Managed
by GSI Facilities
Gallons managed, per acre, per year
0
0.01-49.9
H 50-99.9
100 - 999.9
IH 1,000-9,999.9
10,000 -162,168
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
3 mi
J
ECONorthwest
ECONOMICS • FINANCE • PLANNING
-------
Map A-10.
Property Value
per Acre

'• t.'T
- j: .
'Sr
Mfi. " it
•V t
,. * sail
¦"fji'sispi s':i
MfrvW r
•-"Hv.- . - is. ' u
'v	ri2
^ siisirkrin z,r- 8S^ IS iif:
•sH^KjgeBaisS	? ¦ i
L<^«9EiraBra»ife JBHBHHMI

i^lpaiisici!!fag«aii ¦¦wBsliiSsa? iilfHMli
riPMBMP
LwpSM!p'S"'«|
Total appraised value per acre
Less than $1,000,000
$1,000,000-$1,999,999
$2,000,000 - $2,999,999
HI $3,000,000 - $3,999,999
IHI $4,000,000 - $4,999,999
$5,000,000 and above
Total appraised value is equal to the
market value of the land plus market value
of improvements. Data from Seattle Public
Utilities, King County Assessors, and King
County GIS Center. July 2013.
0
ECONorthwest
ECONOMICS • FINANCE • PLANNING
«K.5f
i:T?« 'liijijii
aw =«*¦
¦n»n
IKUIMgftll CHniiEirr' , J
¦uh MKMnasfassau&lfl i
lwiii

jM m£f
»«» S 'ijSlfel
\
HMI
*£«£%¥
~wm
if %m
I'iifl
l!	- t
J i- ' +
)l fi ~r>
*\y.
IFI. Mitw
jrv *'

W-.E?" £ l*
- ¦
¦» ¦
¦3 *&¦
'•m
-------
ECONOMICS • FINANCE • PLANNING
0	1	2	3 mi
	1	I	I	I
ECONorthwest
Map A-ll.
Race and Ethnicity
Minority Population (% of total)
Less than 10%
10% -19.9%
20% - 39.9%
40% - 59.9%
| 60% or more
Population of 5 or fewer
Map shows the percent of total population
that is non-white, two or more races, or
Hispanic/Latino by 2010 census block.
Data from King County GIS and 2010 US
Census SF1. July 2013.
-------
Appendix B. Developing Build-Out Scenarios
In this analysis, three build-out scenarios (low, medium, and high) were considered, which represent
different rates of future expansions of GSI facilities in Seattle. SPU staff provided all assumptions used to
develop the build-out scenarios. These assumptions are basin-specific (three basins), mechanism-
specific (three mechanisms), and BMP-specific (five BMPs). In all cases, a linear increase in GSI
installation from 2014 to 2050 was assumed.
The three basins used to project future GSI efforts are the combined sewer basin, the creek basin, and
the direct discharge basin. By using these three basins, future GSI efforts are projected based on the
system through which stormwater is conveyed as well as the end location of that stormwater. The three
mechanisms used to project future GSI efforts are right-of-way projects, code-triggered projects, and
RainWise projects. The five BMPs used to project future GSI efforts are trees, porous pavement,
bioretention, green roofs, and rainwater harvesting.
I. Summary of Buik	sumptions
Table B-l summarizes the build-out assumptions used for right-of-way projects. Tree projections are
based on GIS data that SPU provided that indicate all areas in the City with the potential to support
trees. Bioretention projections are based on GIS data that SPU provided that indicate all portions of the
right-of-way with the capacity to support bioretention. Porous pavement projections are based on GIS
data that SPU provided showing all the alleys in the City. There are no green roofs or stormwater
harvesting facilities associated with right-of-way projects.
Table B-l. Build-Out Assumptions for Right-of-Way Projects
Trees
Low
Medium
High
Combined Sewer
Basin
10% of planting potential
20% of planting potential
30% of planting potential
Creek Basin
10% of planting potential
20% of planting potential
30% of planting potential
Direct Discharge Basin
10% of planting potential
20% of planting potential
30% of planting potential
Porous Pavement
Low
Medium
High
Combined Sewer
Basin
None
None
None
Creek Basin
None
None
10% of alleys
Direct Discharge Basin
None
None
None
Bioretention
Low
Medium
High
Combined Sewer
Basin
10% of technically feasible blocks
25% of technically feasible blocks
50% of technically feasible blocks
Creek Basin
10% of technically feasible blocks
25% of technically feasible blocks
and
50% of technically feasible blocks
and
Direct Discharge Basin
5% of technically feasible blocks
10% of technically feasible blocks
10% of technically feasible blocks
Table B-2 summarizes build-out assumptions for code-triggered projects. For these projects, there are
no basin-specific assumptions or scenario-specific assumptions. These build-out scenarios are based on
the assumption that 1% of impervious area in Seattle is redeveloped each year, and that this area must
B-l
-------
comply with GSI-related redevelopment codes. For the area facing redevelopment each year, it was
assumed that GSI would manage stormwater from 19% of the impervious parcel area.105
Table B-2. Build-Out Assumptions for Code-triggered Projects
Trees - 10% of area managed
Bioretention - 50% of area managed
Porous Pavement - 27% of area managed
Green Roof-3% of area managed
Downspout Disconnects - 7% of area managed
Rainwater Harvesting - 3% of area managed
Table B-3 summarizes build-out assumptions for RainWise projects. RainWise projections are based on
GIS data that SPU provided that indicates all parcels in the City with the potential to support RainWise
Projects. In terms of BMP distribution, it is assumed that 88% of all future facilities installed through the
RainWise Program provide bioretention, while the remaining 12% provide rainwater harvesting. These
percentages are based on the distribution of existing RainWise facilities across the two BMPs from the
inventory data.
Table B-3. Build-Out Assumptions for RainWise Projects
Parcel Selection
Low
Medium
High
Combined Sewer
Basin
10% of feasible single family
parcels, commercial parcels, and
schools.
20% of feasible single family
parcels, commercial parcels, and
schools.
30% of feasible single family
parcels, commercial parcels, and
schools.
Creek Basin
10% of feasible single family
parcels, commercial parcels, and
schools.
20% of feasible single family
parcels, commercial parcels, and
schools.
30% of feasible single family
parcels, commercial parcels, and
schools.
Direct Discharge Basin
None
None
None
jild-Out Scenarios for Right-of-Way Projects
Trees, porous pavement, and bioretention are the three GSI BMPs installed through the right-of-way
mechanism. In this section, the assumptions (described above) are applied to quantify the amount of
each BMP installed through this mechanism.
Trees. SPU staff provided data showing all the existing trees in the City of Seattle. These data also show
all the places trees could potentially be planted in the future. Table B-4 summarizes the number of trees
that can be planted within the right-of-way in Seattle, by zone and sewer basin. Across all zones and
sewer basin, data suggest a total of 277,533 additional trees can be planted in Seattle.
For this analysis, the following is assumed (1) under the Low Build-Out Scenario, 10% of all potential
trees, in each of the three basins, are planted, (2) under the Medium Build-Out Scenario, 20% of all
potential trees, in each of the three basins, are planted, and (3) under the High Build-Out Scenario, 30%
of all potential trees, in each of the three basins, are planted. Table B-5 summarizes the total number of
trees planted under each build-out scenario, by basin. It also summarizes the number of trees planted
each year, assuming a linear distribution of planting from 2013 to 2050.
105 Data from SPU show that, on average, parcels that have already installed code-triggered GSI facilities are managing
stormwater from about 19% of the total parcel impervious area with GSI.
B-2
-------
Table B-4. Total Tree Potential in the Right-of-Way (number of trees)
Zoning
Combined
Creek
Direct Discharge
Total
Commercial/Mixed Use
2,511
1,843
4,613
8,967
Developed Park or Boulevard
1,851
211
4,762
6,824
Downtown
213

200
413
Major Institutions
375
369
1,257
2,001
Manufacturing/Industrial
2,156
15
8,843
11,014
Multi-Family
9,384
2,805
15,259
27,448
Parks Natural Area
431
86
755
1,272
Single Family
70,375
37,501
111,718
219,594
Total
87,296
42,830
147,407
277,533
Table B-5. Number of Trees Planted in the Right-of-Way (through 2050 | each year)
Scenario
Combined
Creek
Direct Discharge
Total
Low Build-Out Scenario
8,730 | 236/year
4,283 | 116/year
14,741 | 398/year
27,753 | 750/year
Medium Build-Out Scenario
17,459 | 472/year
8,566 | 232/year
29,481 | 797/year
55,507 | 1,500/year
High Build-Out Scenario
26,189 | 708/year
12,849 | 347/year
44,222 | 1,195/year
83,260 | 2,250/year
On average, trees used to manage stormwater on existing code-triggered parcels manage about 652
gallons of water per year per tree.106 This conversion factor is used to quantify the volume of
stormwater trees planted through the right-of-way mechanism manage each year. Table B-6
summarizes the volume of stormwater these trees manage each year through 2050, as well as each year
from 2013-2050.
Table B-6. Tree-related Stormwater Management in the Right-of-Way (millions of gallons through
2050 | each year)
Scenario
Combined
Creek
Direct Discharge
Total
Low Build-Out Scenario
5.7 | 0.2/year
2.8 | 0.1/year
9.6 | 0.3/year
18.1 | 0.5 /year
Medium Build-Out Scenario
11.4 | 0.3/year
5.6 | 0.2/year
19.2 | 0.5/year
36.2 | 1.0/year
High Build-Out Scenario
17.1 | 0.5/year
8.4 | 0.2/year
28.8 | 0.8/year
54.3 | 1.5/year
Porous pavement. For this analysis, it is assumed that no porous pavement is installed in the combined
basin or the direct discharge basin right-of-ways under any of the build-out scenarios, or in the creek
basin right-of-way under the Low and Medium Build-Out Scenarios. For the creek basin's High Build-Out
Scenario, it is assumed that porous pavement is installed on 10% of the alleys in the creek basin by 2050.
Data from SPU show that there are a total of 64.56 acres of alleys in Seattle's creek basin. About 0.17
acres of porous pavement will be installed each year, for a total of 6.46 acres by 2050 in the creek basin
under the High Build-Out Scenario. According to data used for the inventory, each acre of porous
pavement manages a total of 0.53 million gallons of stormwater each year. This conversion factor is
used to quantify the volume of stormwater porous pavement installed through the right-of-way
106 Data from the analysis of the existing inventory show that 759 trees on code-triggered parcels manage a total of 495,000
gallons of stormwater each year, which is used to calculate an average per tree.
B-3
-------
mechanism manage each year. Table B-6 summarizes the volume of stormwater these areas manage
each year through 2050, as well as each year from 2013-2050.
Table B-7. Porous Pavement-related Stormwater Management in the Right-of-Way (millions of gallons
through 2050 | each year)
Scenario
Combined
Creek
Direct Discharge
Total
Low Build-Out Scenario



—
Medium Build-Out Scenario
—
—
—
—
High Build-Out Scenario
—
3.4 | 0.1 /year
—
3.4 | 0.1 /year
Bioretention. SPU staff provided data showing all of the City's right-of-way areas, as well as all
technically feasible areas for bioretention efforts. The two datasets were aligned to identify technically
feasible portions of the City's right-of-way system available for bioretention efforts. In total, there are
about 2,260 acres of right-of-way technically feasible for bioretention in the combined sewer basin,
about 940 acres in the creek basin, and about 3,060 acres in the direct discharge basin. For each basin
and each scenario, the total right-of-way area was multiplied by 1.83 to quantify the total imperious
area managed. In order to convert the impervious area managed into stormwater volume, it is assumed
that each acre of impervious area managed by bioretention equates to 0.62 million gallons of
stormwater each year. Table B-8 summarizes the volume of stormwater these bioretention facilities
manage per year in 2050 as well as the volume they manage per year in annual increments building up
to 2050.
Table B-8. Bioretention-related Stormwater Management in the Right-of-Way (millions of gallons
through 2050 | each year)
Scenario
Combined
Creek
Direct Discharge
Total
Low Build-Out Scenario
258 | 7.0/year
107 | 2.9/year
175 | 4.7/year
539 | 14.6/year
Medium Build-Out Scenario
644 | 17.4/year
267 | 7.2/year
349 | 9.4/year
1,260 | 34.0/year
High Build-Out Scenario
1,287 | 34.8/year
534 | 14.4/year
349 | 9.4/year
2,170 | 58.7/year
jild-Out Scenarios for Code-triggered Projects
Trees, porous pavement, bioretention, green roofs, downspout disconnects, and rainwater harvesting
are the six GSI BMPs installed through the code-triggered mechanism.107 In this section, the assumptions
(described above) are applied to quantify the amount of each BMP installed through this mechanism.
Projecting code-triggered projects. In total, there are about 38,500 acres of parcel land in Seattle, and
an estimated 20,700 acres of which are impervious.108 As stated in the assumptions, 1% of all impervious
area is redeveloped each year under each of the build-out scenarios. Using this assumption, it is
assumed that a total of about 207 acres of impervious area is redeveloped each year. Data from SPU
show that, on average, parcels that have already installed GSI facilities through this mechanism are
managing stormwater from about 19% of the total impervious area. Combining the total amount of
impervious redeveloped each year with the average mitigation factor suggests that, each year, code-
107	This is based on the data generated by the audit of code-triggered projects. Bioretention cells and bioretention planters are
group together here based on common performance assumptions.
108	Impervious calculation is based on typical percent impervious by zoning category.
B-4
-------
triggered GSI facilities will manage stormwater from about 38.7 acres of impervious surface area). The
current breakdown by BMP type summarized in Table B-2 for the code-triggered build-out is assumed
(see Table B-9.)109
Table B-9. Quantifying Annual Distribution of New Code-triggered Projects
Category/BMP
Area Managed
Total Parcel Area
38,508 acres
Total Impervious Area
20,680 acres
Annual Redevelopment
207 acres
Annual Impervious Area Mitigated
38.7 acres
Trees
3.9 acres
Bioretention
19.3 acres
Porous Pavement
10.4 acres
Green Roofs
1.2 acres
Downspout Disconnects
2.7 acres
Rainwater Harvesting
1.2 acres
Notes: The conversion factors are based on the factors presented in Appendix A.
They have been converted from square feet to cubic feet terms to acres to gallons terms.
Totals may not sum due to rounding.
Trees. The number of trees is based on the share of managed area managed by trees in the existing
inventory, and the proportion of number of trees to area in this inventory. Based on assumed conditions
described above, the build-out scenarios include enough tree plantings to manage a total of an
additional 1.9 million gallons of stormwater each year from a total of about 2.8 acres of impervious area
mitigated each year. To determine the associated number of trees planted, it is assumed that each tree
manages about 652 gallons of water per year.110 Using this conversion factor (from gallons managed per
year to number of trees per year), the build-out scenarios include a total of 2,908 trees per year (about
2,046 in the combined sewer basin, 485 in the creek basin, and 378 in the direct discharge basin).
Bioretention. As described above, the build-out scenarios include enough bioretention facilities to
manage a total of about 8.6 million gallons of stormwater each year from a total of about 13.9 acres of
impervious area mitigated each year.
Table B-10 summarizes the volume of stormwater managed, each year, by GSI facilities installed through
the code-triggered mechanism. The table distinguishes between different GSI BMPs and the three
basins. The first number indicated the total volume managed per year in 2050 and the second number
indicates the annual increase in volume managed from 2013-2050.
109	Approximately 33 percent of the area is in the combined basins, 24 percent in the creek basin, and 43 percent in the direct
discharge basin.
110	Data from the analysis of the existing inventory show that 759 trees on code-triggered parcels manage a total of 495,000
gallons of stormwater each year, which is used to calculate an average per tree.
B-5
-------
Table B-10. Stormwater Management by Code-triggered Projects
(millions of gallons through 2050 | each year)
BMP
Volume Managed
Trees
98 | 2.6/year
Bioretention
445 | 12/year
Porous Pavement
205 | 5.5/year
Green Roofs
8.5 | 0.2/year
Downspout Disconnects
62 | 1.7/year
Rainwater Harvesting
14 | 0.4/year
4, Icouild-Out Scenarios for RainWise Projects
Bioretention and rainwater harvesting are the two GSI BMPs installed through the RainWise Program. In
this section, the assumptions (described above) are applied to quantify the amount of each BMP
installed through this mechanism. Bioretention accounts for about 88% of the RainWise projects in the
existing inventory; rainwater harvesting accounts for the other 12%.
Projecting RainWise projects. In total, there are 25,599 parcels that align with the Seattle or King
County RainWise Basins as well as the area identified as suitable for infiltration.111 Table B-11
distributes these parcels across the basins, and identifies the type of parcel, according to data from the
Urban Forest Management Plan. According to the build-out assumptions described earlier, 10% of the
parcels in the combined sewer basin and the creek basin (a total of 2,560 parcels) would join the
RainWise Program under the Low Build-Out Scenario, 20% (a total of 5,120 parcels) would join the
RainWise Program under the Medium Build-Out Scenario, and 30% (a total of 7,680 parcels) would join
under the High Build-Out Scenario.
Table B-11. Number of Parcels Aligning with RainWise Requirements
Basin
Single Family
Commercial
Other
Total
Combined Sewer Basin
23,903
1,326
10
25,239
Creek Basin
355
5

360
Direct Discharge Basin




Total
25,136
1,381
10
25,599
Bioretention. Of all the RainWise projects in the inventory, about 88% of them use bioretention to
manage stormwater. The average parcel using bioretention in the inventory collects stormwater from
about 1,150 square feet of impervious surface, and manages a total of 16,580 gallons of stormwater per
year. For each build-out scenario, the annual and cumulative additions of bioretention facilities through
the RainWise program are described below. Table B-12 summarizes the volume these facilities manage
each year in 2050 and each year from 2013-2050.
111 SPU does not offer RainWise in direct discharge basins.
B-6
-------
•	The Low Build-Out Scenario would add bioretention facilities to a total of 61 parcels per year (60
in the combined sewer basin and one more in the creek basin). These facilities would manage a
total of 1.6 acres of impermeable surface and about 1.0 million gallons of stormwater each year.
By 2050, the Low Build-Out Scenario would add bioretention to a total of about 2,250 parcels,
managing a total of about 59.5 acres of impermeable surface and about 37.3 million gallons of
stormwater each year.
•	The Medium Build-Out Scenario would add bioretention facilities to a total of 122 parcels per
year (120 in the combined sewer basin and two more in the creek basin). These facilities would
manage a total of 3.2 acres of impermeable surface and about 2.0 million gallons of stormwater
each year. By 2050, the Medium Build-Out Scenario would add bioretention to a total of about
4,500 parcels, managing a total of about 118.9 acres of impermeable surface and about 74.7
million gallons of stormwater each year.
•	The High Build-Out Scenario would add bioretention facilities to a total of 183 parcels per year
(180 in the combined sewer basin and three more in the creek basin). These facilities would
manage a total of 4.8 acres of impermeable surface and about 3.0 million gallons of stormwater
each year. By 2050, the High Build-Out Scenario would add bioretention to a total of about
6,760 parcels, managing a total of about 178.4 acres of impermeable surface and about 112.0
million gallons of stormwater each year.
Table B-12. Bioretention-related Stormwater Management from RainWise Projects (millions of gallons
through 2050 | each year)
Scenario
Combined
Creek
Direct Discharge
Total
Low Build-Out Scenario
36.8 | 1.0/year
0.5 | < 0.1/year
—
37.3 | 1.0/year
Medium Build-Out Scenario
73.6 | 2.0/year
1.0 | < 0.1/year
—
74.7 | 2.0/year
High Build-Out Scenario
110.5 | 3.0/year
1.5 | < 0.1/year
—
112.0 | 3.0/year
Rainwater harvesting. Of all the RainWise projects in the inventory, about 12% of them use rainwater
harvesting to manage stormwater. The average parcel using rainwater harvesting in the inventory
collects stormwater from about 1,280 square feet of impervious surface, and manages a total of 5,020
gallons of stormwater per year. For each build-out scenario, the annual and cumulative additions of
rainwater harvesting facilities through the RainWise program are described below. Table B-13
summarizes the volume these facilities manage each year in 2050 and each year from 2013-2050.
•	The Low Build-Out Scenario would add rainwater harvesting facilities to a total of 8 parcels per
year (only a fraction of the annual build-out would occur in the creek basin). These facilities
would manage a total of 0.2 acres of impermeable surface and about 42,000 gallons of
stormwater each year. By 2050, the Low Build-Out Scenario would add rainwater harvesting
facilities to a total of about 310 parcels, managing a total of about 9.0 acres of impermeable
surface and about 1.5 million gallons of stormwater each year.
•	The Medium Build-Out Scenario would add rainwater harvesting facilities to a total of 17 parcels
per year (only a fraction of the annual build-out would occur in the creek basin). These facilities
would manage a total of 0.5 acres of impermeable surface and about 83,000 gallons of
stormwater each year. By 2050, the Medium Build-Out Scenario would add rainwater harvesting
facilities to a total of about 614 parcels, managing a total of about 18.1 acres of impermeable
surface and about 3.1 million gallons of stormwater each year.
B-7
-------
• The High Build-Out Scenario would add rainwater harvesting facilities to a total of 25 parcels per
year (only a fraction of the annual build-out would occur in the creek basin). These facilities
would manage a total of 0.7 acres of impermeable surface and about 0.1 million gallons of
stormwater each year. By 2050, the High Build-Out Scenario would add rainwater harvesting
facilities to a total of about 920 parcels, managing a total of about 27.1 acres of impermeable
surface and about 4.6 million gallons of stormwater each year.
Table B-13. Rainwater Harvest-related Stormwater Management from RainWise Projects (millions of
gallons through 2050 | each year)
Scenario
Combined
Creek
Direct Discharge
Total
Low Build-Out Scenario
1.5 | < 0.1/year
< 0.11 < 0.1/year
—
1.5 | < 0.1/year
Medium Build-Out Scenario
3.0 | < 0.1/year
< 0.1 | < 0.1/year
—
3.1 | < 0.1/year
High Build-Out Scenario
4.6 | 0.1/year
< 0.1 | < 0.1/year
—
4.6 | 0.1/year
immary of Build-Out Scenarios
This section summarizes each of the three build-out scenarios in terms of the quantify of GSI facilities
installed each year, the volume of stormwater they manage each year in 2050, and the incremental
increase in volume from 2013-2050. At the end of the section, a series of maps demonstrates some of
the spatial assumptions used in developing the build-out scenarios.
The following three figures summarize some of the quantified GSI efforts used in the analysis of the
build-out scenarios.
•	Figure B-l summarizes the amount of each GSI BMP installed, annually, under each build-out
scenario (Low, Medium, and High) and within each of the three basins (combined, creek, and
direct discharge).
•	Figure B-2 summarizes the volume of stormwater managed, per year, for the GSI BMPs installed
each year, under each build-out scenario (Low, Medium, and High) and within each of the three
basins (combined, creek, and direct discharge).Figure B-3 summarizes the volume of stormwater
managed, per year, in 2050, after all build-out scenarios have been fully implemented, under
each build-out scenario (Low, Medium, and High) and within each of the three basins
(combined, creek, and direct discharge).
B-8
-------
Figure B-l. GSI BMPs installed, per year, for each build-out scenario, in each basin

GSI Installed each year 2013-20S0
Low
Medium
High
ROW
Code
RainWise
Subtotal
ROW
Code
RainWise
Subtotal
ROW
Code
RainWise
Subtotal
Trees (Number of
Trees)
Combined
235J9
1.338.6
-
1,574.6
471.9
1.338.6
-
1$10.5
707.8
1.338.6
-
2,046.4
Creek
115.8
973.6

1,089.3
231.5
973.6

1,205.1
347.3
973.6
-
1,320£
Direct Discharge
398.4
1,744.3
-
2,142.7
796.8
1,744.3
-
2 £41.1
1^95u2
1,7443
-
2,9395
Subtotal
750.1
4,056.5

4,806.6
1500.2
4,056.5

5,556.7
2,250.3
4,056.5

6,306.7
Porous Pavement
(Acres of Impervious
Surface Managed
Combined
-
3.4
-
3.4
-
3.4
-
3.4
-
3.4
-
3.4
Creek
-
Z5
-
2.5
-
Z5
-
25
0.2
Z5
-
2.7
Direct Discharge
-
4.5
-
4.5
-
4.5
-
45
-
4.5
-
4.5
Subtotal

10.4

10.4

10.4
-
10.4
0.2
10.4

10.6
Bioretention (Acres of
Impervious Surface
Managed)
Combined
11.2
6.4
1.6
19.1
28.0
6.4
3.2
375
55.9
6.4
4.8
67.0
Creek
4.6
4.6
0.0
9.3
11.6
4.6
ao
163
23.2
4.6
ai
27.9
Direct Discharge
7£
8.3
-
15.9
15.2
&3

235
15.2
8.3

23.5
Subtotal
23.4
19.3
1.6
44.3
54.7
19.3
3.2
77.3
94.3
19.3
4.8
118.4
Green Roof (Acres of
Impervious Surface
Managed)
Combined
-
0.4
-
0.4
-
0.4
-
0.4
-
a4
-
0.4
Creek
-
0.3
-
0.3
-
a3
-
03
-
a3
-
0.3
Direct Discharge
-
0.5
-
0.5
-
as
-
05
-
as
-
0.5
Subtotal

1.2

1.2

1.2
-
12

1.2

1.2
Downspout Disconnects
(Acres of Impervious
Surface Managed)
Combined
-
10.3

10.3
-
103

10.3
-
103
-
10.3
Creek
-
18.5

18.5
-
185
'
185
-
183

18.5
Direct Discharge

4Z9

42.9

429

42.9
-
42 3
-
42.9
Subtotal

71.7

71.7

71.7

71.7

71.7

71.7
Rainwater Harvesting
(Acres of Impervious
Surface Managed)
Combined
-
0.4
0.2
0.6
-
0.4
as
0.9
-
0.4
a7
1.1
Creek
-
0.3
ao
0.3
-
0.3
ao
0.3
-
0.3
ao
0.3
Direct Discharge
-
0.5
-
0.5
-
as
-
05
-
0.5
-
0.5
Subtotal

1.2
0.2
1.4

1.2
0.5
1.6

1.2
0.7
1.9
Figure B-2. Volume of Stormwater Managed (per year) for Annual Increases in GSI (through 2050) for
Each Build-Out Scenario, in Each Basin

Millions of Gallons Managed per Year by Build-out GSI 2013-2050
Low
Medium
High
ROW
Code
RainWise
Subtotal
ROW
Code
RainWise
Subtotal
ROW
Code
RainWise
Subtotal
Trees
Combined
0.2
0.9
-
1.0
03
0.9
-
12
OS
0.9
-
1.3
Creek
0.1
0.6
-
0.7
02
0.6
-
0.8
0.2
0.6
-
0.9
Direct Discharge
0.3
1JL
-
1.4
OS
13
-
1.7
0.8
:li
-
1.9
Subtotal
0.5
2.6

3.1
1.0
2.6

3.6
1.5
2.6

4.1
Porous Pavement
Combined
-
12
-
1.8
-
1£
-
1.8
-
1.8
-
1.8
Creek
-
13
-
1.3
-
13
-
13
03
1.3
-
1.4
Direct Discharge
-
2.4
-
2.4
-
2.4
-
2.4
-
2.4
-
2.4
Subtotal

5.5

5.5

5.5
-
5.5
0.1
5.5

5.6
Bioretention
Combined
7.0
4.0
LO
11.9
17.4
4.0
2-0
23.4
345
4.0
3.0
41.8
Creek
Z9
23
on
5.8
72
23
0.0
10.1
14.4
2.9
0.0
17.4
Direct Discharge
4.7
5.2
-
9.9
9.4
S2
-
14.6
9.4
5.2
-
14.6
Subtotal
14.6
12.0
1.0
27.6
34.0
12.0
2.0
48.1
58.7
12.0
3.0
73.7
Green Roof
Combined
-
OJ.
-
0.1
-
03
-
0.1
-
0.1
-
0.1
Creek
-
0.1
-
0.1
-
OJL
-
0.1
-
0.1
-
0.1
Direct Discharge
-
OJL
-
0.1
-
OJl
-
0.1
-
0.1
-
0.1
Subtotal

0.2

02

0.2

0.2

0.2

0.2

Combined
-
0£
-
0.6
-
0£
-
0.6
-
0.6
-
0.6
Creek

0.4

0.4

0.4

0.4
-
0.4

0.4

Direct Discharge
-
0.7

0.7
-
0.7
'
0.7
-
0.7
-
0.7
Subtotal

1.7

1.7

1.7
-
1.7

1.7

1.7
Rainwater Harvesting
Combined
-
OA
0.0
02
-
OJL
03
02
-
0.1
0.1
0.2
Creek
-
OA
0.0
0.1
-
03
OD
0.1
-
0.1
0.0
0.1
Direct Discharge
-
02
-
0.2
-
0.2
-
0.2
-
0.2
-
0.2
Subtotal

0.4
0.0
0.4

0.4
0.1
0.5

0.4
0.1
0.5

Total
15.0
20.8
1A
36.9
35.0
20.8
2.1
58.0
60.2
20.8
3.2
84.2
B-9
-------
Figure B-3. Volume of Stormwater Managed (per year) in 2050 for Each Build-Out Scenario, in Each
Basin




1
I
I
1



LOW
MfJdiurr
H;g


ROW Corf?
RainWiw
5r!SfeK>T3i
ROW Cod? rUsnWifip 1 Subttta:

AinWUf Subto*3!

«,./ iA.i


11.4
is.i
1 /.I





f
/.{.¦»
8.4 ?A.'t


'i.i, 4A 1

116.0
•I /
4A1 vil.=
28.8
4 P.]
r ;!j,,

1,/./

.... ,
-
G7.7

67.7


| 49.2



49.2

a a
49.2


88.2



SS.2

SS.2 !


r 205.2

?c.-.p
-
2Q5.2
r • i

f 20.1 b

146.9
;«.*
4--.1 2
643.7
146.9
,'.•

UX..8
It*,.9
:ia.

267-0
MG.9
1.1 i v1
¦kMD 1065 1
1.6 I i. :

1 /1 | 191.5


3435
mis

:m«.9 131-5 :
540.4



i,021.3
1259.6
445.2
L779.fi

7 7".

1 2£



?»

¦¦1 2.S

Green Roof I;--"-"- -,.--- ,	-

2 JO


...
2Q


ZO


3.7


-
3.7


3.7



3 8.5

S.5

S.5
• r $. r>

¦S.5 :


20.6



20.6


28.6



15.0



15.0


15.0


26.8


-
26.8


26.8



* 62.3

f 623

62.3
r ¦¦ r 62 3

' 623 :
"



i.s i y..i
-
4.6
<;i

4.6
* i.

3.4

3,4

\ *


.1.4 !
0.1

r :i

6,0

6.0 - | 6.0
1 6.0 1
6.0





,
1 US.O j
•
" T=--
-r.G.8 770.1!
J*.
1.266.6


2,228.0 77(18 !
i-ii.7 3.125.6
The following maps summarize some of the spatial components of the assumptions used to develop the
three build-out scenarios. Table B-14 identifies all of the maps included at the end of this appendix.
Table B-14. Maps in this Appendix
Map Title
Map Description
Map B-l. Tree Potential
This map shows all of the areas with the capacity to support new trees in the future.
Map B-2. Seattle's Alleys
This map shows all of the City's alleys. Some of the build-out scenarios rely on effort to install
permeable pavement in some of these alleys.
Map B-3. Seattle's Right-of-way
This map shows all of the City's right-of-way areas.
Map B-4. Areas Suitable for Infiltration
This map shows all of the land in the City that is suitable for infiltration.
Map B-5. Development Rates
This map provides an illustrative example of redevelopment to shed light on the extent of the 1%
redevelopment per year assumption used in the analysis. The map shows all parcels in Seattle. It
also shows an area that represents 1% of those parcels (the amount assumed to be redeveloped
each year), as well as an area that represents 37% of those parcels (the amount assumed to be
redeveloped by 2050). This map is for illustrative purposes only and does not represent actual
anticipated redevelopment in any specific areas within Seattle.
Map B-6. RainWise Basins
This map shows the RainWise Basins used in the analysis.
B-10
-------
Map B-l.
Tree Potential



:" vTT •
Potential trees
. 1 dot = 50 trees
Sewer Classification
Combined
Partially Separated
Separated
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
>	
H4 . \ ' ' T".?- v-'
C- \ I	'-"»3 /vj	t'- I

M
.V
•* *C '




*;
'>i. \ •
Sir
If):.



¦ w:.v-
ffiSSlPI
lilSf
¦ '::yzy;<-:h ¦-'i'-f
' ' i ' vl£Jv .C-V
•	..inifeiA-ii
•	• •


:;V> •"•*v
I. JSJ fefete
¦U
" X'~ v. V-
¦ >•'"
l'- svfry-.-i.. -/A t'.
* -')•//¦ • '•'••• ' •• •A*.'-

- > ' v-;»: ? J
i
"V- \

3 mi
J
... .:• j

i ¦ i!
-------
Map B-2.
r	ii»,
Alleys Located in
Creek Basins	ii!
i iIL. Er
mmma h
Ifi ;
*1I1m
¦
I Hi
- 11
{ [
11 '
	1 \ 1
L
ft n f /

-MM
1 7 \MfM/ K
] f I \
V//A i

I Alleys in creek basins
1J pin,
Sewer Classification
1i?~*
Combined

Partially Separated

Separated
in '
in i
i i
Creek Basins
Ml J
s 11 1, 'H
Data from Seattle Public Utilities and
I ii
King County GIS Center. July 2013.
i 1_
~——- i
0	1 2 3 mi
1	1 1 1
i
-------
kiuaiifuminiEi1 iaar"»"iii*ii||i|||Mflai^>aiaiiiii(9u sinRP^uimiii«ii|M^&lli
MsatSRSframis^
^SsiiKstSSiSinas
Map B-3.
Seattle's
Right-of-Way
I Right-of-Way
Sewer Classification
Combined
Partially Separated
Separated
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
0	1	2	3 mi
	1	I	I	I
-------
Map B-4.
Areas Suitable for
Infiltration
Areas suitable for infiltration
Data from Seattle Public Utilities and
King County GIS Center. July 2013.


3 mi
J
-------
Map B-5.
Development
Rates
385 acres (-1%)
14,245 acres (-37%)
Data from Seattle Public Utilities and
King County GIS Center. July 2013.
3 mi
J
-------
Map B-6.
RainWise Basins
K
1
n

RainWise Basins
Data from Seattle Public Utilities and
King County GIS Center. July 2013.

-------
Appendix C. Seattle's Sewer System
This report refers to several different sewer basins, and the economic analysis relies heavily on tracking
stormwater as it travels through Seattle's sewer system. In this appendix, Seattle's sewer system is
described in order to provide a thorough context within which to consider the results and to interpret
the analysis and methods. Figure C-l describes the evolution of Seattle's sewer system. Put simply, the
system evolved from a system relying on combined sewer pipes (which manage stormwater and
wastewater) to a system relying on combined sewer pipes and stormwater pipes (which still combine
stormwater and wastewater in some instances, but manage stormwater from certain locations
separately).
Figure C-l. Evolution of Seattle's Sewer System
Combined Sewer System

Storm
Drain

Like most cities across the country, Seattle's original sewer system
was a combined system in which stormwater and wastewater
both flow to a combined sewer pipe. Currently several
neighborhoods in Seattle rely on this combined sewer system for
all their wastewater and stormwater management. After they
combine in the combined sewer pipe, the system conveys this
water to a wastewater treatment plant, where it is treated before
entering nearby waterways. During large storm events, the
system's capacity reaches its limits. In these instances, some of
the untreated water in the combined sewer pipe is discharged
directly into nearby waterways (these discharges are referred to
as CSO events).
Roof
Drain
1

Sanitary
Drain
*
To Waste Water
Treatment Plant
: Overflow
, Combined
Y Sewer Pipe
Partially Separated Sewer System

Storm
Drain

In the 1960s and 1970s, Seattle and other cities across the
country became concerned with how their combined sewer
systems were affecting water quality. In an attempt to reduce the
stress on the combined sewer system, Seattle implemented
efforts to divert stormwater flow from storm drains to a separate
sewer system. This separate sewer system does not flow to the
wastewater treatment plant, but rather flows directly into nearby
waterways. Wastewater and stormwater from roof drains still
flow to the combined sewer pipe, but all stormwater flowing into
storm drains flows to stormwater pipes.
Roof
Drain
fg

Sanitary
Drain
To Creek
or Lake
w
To Waste Water
Treatment Plant
Overflow
Partially Separated
Combined
Sewer Pipe
Fully Separated Sewer System

Storm
Drain

Recently, Seattle and other cities across the country have
expanded the concept of the partially separated sewer system. In
the fully separated sewer system, all stormwater (i.e., stormwater
flowing into storm drains as well as stormwater flowing into roof
drains) is directed toward stormwater pipes, and is completely
diverted from the combined sewer system. This process further
reduces stress on the combined sewer system, and helps reduce
the frequency and magnitude of CSO events. By diverting this
stormwater, however, the fully separated sewer system increases
the volume of untreated stormwater runoff entering nearby
waterways.
Roof
Drain
f B
1 Sanitary
To Creek
or Lake
w
To Waste Water
Treatment Plant
U _
Fully Separated
Sewer Pipe
c-i
-------
To demonstrate how these three approaches to stormwater management interact, it is useful to look at
the system in spatial terms. Map C-l shows the City's sewer system in terms of where stormwater flows.
The solid yellow areas represent parts of the City that still rely entirely on the combined sewer system
for their stormwater and wastewater management. The solid blue and purple areas represent parts of
the City in which the fully separated approach has been implemented. Wastewater in these areas enters
the combined sewer system, and stormwater enters the separate sewer pipes that discharge
stormwater directly into nearby waterways. Some areas on the map have a yellow base along with a
blue or purple grid. These areas represent parts of the City in which the partially separated approach has
been implemented. Wastewater in these areas enters the combined sewer system. Stormwater falling
on parcels (e.g., residential and commercial roofs) flows into the combined sewer system as well.
Stormwater falling on the right-of-way (e.g., roads) flows into stormwater pipes and is discharged
directly into nearby waterways. Map C-2 shows a higher-level representation of these three approaches
to sewer system design.
Depending on where GSI facilities are located, they will have different effects on stormwater
management. GSI facilities placed in the combined sewer basin will reduce stormwater flows into the
combined system. GSI facilities placed in the fully separated sewer basin will reduce the volume of
untreated stormwater flowing directly into nearby waterways through stormwater pipes. GSI facilities in
the partially separated sewer basin will do one or the other depending on whether they are place on
parcels or in the right-of-way.
C-2
-------
Map C-1.
Sewer System,
by Discharge
Basin
Discharge Basin
Combined
Creek
Direct Discharge
Data from Seattle Publ ic Utili ti esand
K ing County Gl S C alter. July 2013.
0	1	2
	1	I	1	

Ma i w
C-3
-------
Map C-2.
Sewer System,
by Management
Basin
Management Basin
Combined
Partally Separated
Separated
Data from Seattle Public Utilitiesand
King County GIS Center. July 2013.
0	1	2
	1	I	I	
•lr
ft
C-4
-------
Appendix D. Combined Sewer Overflows
Combined sewer overflows (CSOs) occur when the volume of water entering the combined sewer
system exceeds the system's capacity. When this happens, untreated water from the combined sewer
system is removed from the system through one of 92 CSO outfalls. This untreated water then flows into
nearby waterways. According to the 2010 CSO Reduction Plan, Seattle's CSO volumes have declined
since the 1980s, when they averaged about 400 million gallons (and about 2,800 CSO events) per
year.112 The goal of the plan, however, is to achieve an average of no more than one CSO event per
outfall per year.113 Figure D-l demonstrates the frequency, duration, and volume of Seattle's CSO events
from 2008 to 2012. The figure does not reflect the decline in frequency, duration, and volume of CSO
events since the 1980s. It does, however, show that more action is needed to meet SPU's goal of no
more than one CSO event per outfall per year.
Figure D-l. Seattle's CSO Events (2008-2012)
2008	2009	2010	2011	2012
¦Frequency	"Duration	"Volume
(number of CSOs per year) (number of 24-hour periods) (millions of gallons per year)
Source: Seattle Public Utilities. 2013. 2012 Annual Report CSO Reduction Program. March.
GSI can help reduce the frequency and volume of CSO events by managing some of the stormwater that
would have entered the combined sewer system. Because it can reduce the frequency and volume of
CSO events, GSI provides two types of benefits. One of these benefits represents the avoided costs of
dealing with a CSO event once it has happened. The other benefit represents the avoided costs of
relying solely on gray infrastructure techniques for managing stormwater.
Valuation Methodology
Data are not sufficient to quantify the extent to which the existing inventory or potential build-out of
GSI facilities reduce the frequency and magnitude of CSO events. Furthermore, the potential costs of
future CSOs (and the avoided costs of preventing those CSOs) are context specific, and are difficult to
accurately quantify. The methodological approach to considering the benefits of GSI-related CSO
reductions relies on an avoided cost approach. Several potential costs associated with CSO events in the
112	Seattle Public Utilities. 2010. 2010 CSO Reduction Plan Amendment. May. Pg. 3-2.
113	Seattle Public Utilities. 2010. 2010 CSO Reduction Plan Amendment. May. Pg. 1-1.
D-l
-------
future are identified and used as a proxy to estimate the value of the benefits of reducing future CSO
events.
Review and Analysis
This section presents a brief outline of how to consider two types of costs associated with reducing the
frequency and magnitude of CSO events: (1) ex post costs of CSO events, and (2) costs of preventing CSOs.
Ex post costs of CSO events
The 2010 CSO Reduction Plan contains a goal of reducing the frequency of CSO events in Seattle to no
more than one per CSO outfall per year. The reason that Seattle and other communities across the
country are trying to reduce the frequency of CSO events is because of the costs that materialize in their
aftermath. CSO events release untreated wastewater and stormwater into waterways. Described below
are two mechanisms through which these biophysical effects turn into economic costs.
•	Stormwater management agencies often face fines or penalties for allowing CSO events to
occur or for failing to comply with the relevant permits and procedures. For example, the
Washington Department of Ecology fined King County $46,000 in 2010 for failing to comply
with CSO-related water quality permits. 114 Ecology also fined Seattle $12,000 in 2007 for a CSO
event caused by a pump failure.115
•	By diverting untreated wastewater and stormwater into nearby waterways, CSO events
decrease water quality in those waterways. Decreasing water quality is costly for a number of
reasons. Most directly, it can restrict access to certain uses (e.g., swimming or fishing). More
indirectly, it can harm the aquatic ecosystem, from which people derive a number of valuable
benefits.
Past penalties. Earlier in 2013, after settlement discussions with the U.S. Department of Justice and EPA,
the city of Seattle and King County agreed to provide funding for major upgrades to their combined
sewer system.116 Between 2006 and 2010, King County discharged about 900 million gallons of raw
sewage into nearby waterways, and between 2007 and 2010, the Seattle discharged another 200 million
gallons of raw sewage into nearby waterways. These discharges violated section 301 of the Clean Water
Act as well as other agreements and regulations. As a result, the county was fined a civil penalty of
$400,000 and the city was fined a civil penalty of $350,000. The county and city also agreed to
implement long-term control plans for controlling CSO discharges in the coming years.
Potential future penalties. The city of Seattle's Consent Decree also identifies a number of penalties it
may face if it fails to meet all of its conditions. For example, the city is liable to pay a penalty of $7,500
per day for each dry-weather CSO event and a penalty of $2,500 per day for each sewer overflow. There
are a number of other per-day and per-violation penalties the city faces if it fails to meet the
requirements set forth in the Consent Decree.117 To some extent, implementing GSI efforts will help
114	Washington Department of Ecology. 2010. King County Fined for Combined Sewer Overflow Violation. Retrieved on June 5,
2013 from http://www.ecv.wa.aov/news/2010news/2010-139.htmi.
115	Washington Department of Ecology. 2007. Seattle Fined for Sewage Discharge. Retrieved on June 5, 2013 from
http://www.ecv.wa. gov/news/2007news/2007-212.html.
116	U.S. Environmental Protection Agency. 2013. Seattle, Washington and King County, Washington Settlement. Retrieved on
June 7, 2013 from http://www.epa.aov/enforcement/water/cases/washinaton.htm/.
117	Consent Decree, United States of America and the State of Washington v. the City of Seattle, Washington. April 16, 2013.
Civil Action No. 2:13-cv-678. Retrieved on June 7, 2013 from
http://www.epa.aov/enforcement/water/documents/decrees/citvofseatt/ewashinaton-cd.pdf.
D-2
-------
prevent the city and county from violating the requirements of the Consent Decree, and in doing so, will
provide valuable benefit equal to the costs of the penalties avoided.
Total cost of potential future penalties. From 2008 to 2012, there was an average of 294 CSO events
per year. Each CSO event lasted an average of 10 hours. Combined, these CSO events lasted a total of
125, 24-hour periods per year, and discharged an average of 111 million gallons of untreated water per
year. As previously stated, the Consent Decree will institute a penalty of $2,500 per day for wet-weather
CSO events and $7,500 per day for dry-weather CSO events. Assuming that these penalties can serve as
a proxy for the costs (or forgone benefits) society incurs due to CSO events, they can be applied to
average annual frequency of CSO events, and project those values into the future. The full range of
these costs is about $0.3-$2.2 million per year.118 Assuming no change in CSO frequency or duration,
the 100-year NPV of these CSO costs, discounted at a rate of 2%, is about $13.8-$96.9 million. To the
extent that GSI can be used to decrease the frequency and duration of these future CSO events, it can
provide a valuable benefit equal to the value of the avoided penalty costs.
In reality, however, if SPU implements no additional stormwater management efforts, the frequency and
duration of CSO events will increase in the future when faced with increased urbanization and increased
precipitation related to climate change. This potential increase in frequency and duration of CSO events
suggests that the 100-year NPVs presented above likely understate the total costs. Furthermore, since the
estimates are based on penalty values, they likely understate the actual costs associated with CSO events.
To the extent that GSI efforts undertaken under the build-out scenarios decrease the frequency and
duration of CSO events in the future, they can reduce the total value of these costs.
As mentioned above, the 2010 CSO Reduction Plan contains a goal of reducing the frequency of CSO
events in Seattle to no more than one per CSO outfall per year. To meet that challenge, agencies will
implement a broad range of BMPs (both GSI and gray infrastructure). The 2010 CSO Reduction Plan
identified and described 16 CSO control projects that could support the objective of reducing CSO
frequency. These projects are listed below.
Windermere
Genesee
Henderson
Ballard
N. Union Bay
Interbay
Central Waterfront	•	Leschi
Fremont/Wallingford	•	Union Bay
Duwamish	•	East Waterway
Longfellow/Delridge	•	Lake Union/Portage Bay
West Seattle
Montlake
These CSO control projects vary in terms of complexity, and many of them included multiple BMPs.
Table D-l summarizes these projects by the BMPs they implement. For each BMP, the table shows the
number of projects implemented, the control volume, and the estimated cost. In total, these CSO
control projects would cost about $330 million and would control a total of about 20.5 million gallons of
stormwater. Given the large amount of uncertainty regarding these costs, the 2010 CSO Reduction Plan
also provided a range of potential costs from about $182 million to $627 million. The plan relies heavily
on offline storage. This BMP accounts for about 79% of total control volume and about 85% of total cost.
118 The low end of the range represents the wet-weather CSO penalty ($2,500) times the average annual number of 24-hour
CSO periods (125). The high end of the range represents the dry-weather CSO penalty ($7,500) times the average annual
number of CSO events (294).
D-3
-------
Table D-l. Summary of CSO Control Projects from 2010 CSO Reduction Plan
BMP
Number of Projects
Control Volume (gallons) | % of
Total Volume
Estimated Cost | % of Total Cost
Downspout
20
831,000 | 4%
$3,270,000 | 1%
Cisterns
3
255,000 | 1%
$2,062,000 | < 1%
Bioretention
1
4,000 | < 1%
$74,000 | < 1%
Rain Gardens
9
460,000 | 2%
$4,718,000 | 1%
Permeable Pavement
2
138,000 | < 1%
$2,033,000 | < 1%
Inline Storage
8
1,677,000 | 8%
$23,881,000 | 7%
l/l Reduction
8
399,000 | 2%
$5,030,000 | 2%
Offline Storage
19
16,257,000 | 79%
$279,888,000 | 85%
Retrofit
4
484,000 | 2%
9,504,000 | 3%
Total
74
20,505,000 | 100%
$330,460,000 | 100%
Source: Seattle Public Utilities. 2010. 2010 CSO Reduction Plan Amendment. May. Pg. 5-14-5-18.
Average cost of BMP installation. Figure D-2 shows the average costs used to develop the CSO
Reduction Plan and to estimate the costs of implementation. Each line shows the per-gallon cost of
installing different BMPs in terms of the total size of the installation. The costs represent the 100-year
lifecycle costs, which include construction costs and ongoing operations and maintenance costs. For
example, a 100,000-gallon inline storage facility has a 100-year lifecycle cost of about $32 per gallon,
while a 200,000-gallon inline storage facility has a 100-year lifecycle cost of about $18 per gallon.
In general, per-gallon costs decrease as the control volume increases. The main exception is
disconnecting downspouts, which has a lifecycle cost of about $3 per gallon regardless of control
volume. For smaller projects (less than 0.1 million gallons), several GSI BMPs offer lower lifecycle costs
than gray alternatives. For example, rain gardens, permeable pavement, and bioretention cost between
$18 and $23 per gallon for projects controlling less than 0.1 million gallons while other BMPs (e.g., street
storage, inline storage, offline storage, and l/l reductions) are either costlier or not feasible at such small
scales. For larger projects (around 1 million gallons), the lifecycle costs of implementing many different
BMPs converge. In addition to cost, however, a number of factors influence these kinds of planning
efforts. One of the most influential factors is feasibility. If it were possible to meet management
objectives by relying solely on downspout disconnects, then agencies would not consider the other,
costlier BMPs. In reality, however, the control potential for downspout disconnects is limited.
BMP selection. The process of selecting a portfolio of BMPs is best demonstrated through an example.
Assuming that the objective is to control 10 million gallons of stormwater in a particular area, planning
efforts must first evaluate the extent of each BMP's feasibility (e.g., What is the total feasible control
volume supported by downspout disconnects, offline storage, or other measures?). Since downspout
disconnects are always the least expensive option, they should be implemented wherever feasible (i.e.,
wherever there are downspouts to disconnect). After that, it comes down to feasibility. From a cost
perspective, planners should implement the cheapest combination of BMPs that align with the
feasibility of their installation within the project area. This approach may mean one large gray facility,
several small GSI facilities, or something in between.
D-4
-------
Figure D-2. Average Cost Curves for GSI and Gray Infrastructure BMPs
$50
	Downspout
Cisterns
	Bioretention
$45 ¦
InlineStorage
	Raingarden
Permeable Pavement
$40 ¦
y C^ffline Storage
	Inline Storage
	I/I Reduction
Street Storage
$35 ¦
>
Offline Storage
Wet Weather
Reductions
$30 ¦
	Street Storage
Wet Weather
$25
Bioretention
Cisterns
$20
Permeable Pavement
Rain Garden
$15
$10
Downspouts
0.01
0.1
10
100
Control Volume (millions of gallons)
Source: Seattle Public Utilities. 2010. 2010 CSO Reduction Plan Amendment. May. Pg. 5-12.
Notes: This figure shows the average cost for installing different types of BMPs based on the size of the BMP. The average costs are for single
facilities, so rather than accumulating along the x-axis, new projects must begin at the left side of the figure.
Summary and Distribution
In this section, two types of benefits associated with reducing the frequency and magnitude of CSO
events were discussed. First discussed are CSO-related penalties that agencies will face in the future
associated with the recent Consent Decree outlining future goals for stormwater management in
Seattle. Data were not sufficient to quantify the extent to which the existing inventory of GSI or the GSI
installed in the build-out scenarios can decrease the frequency or magnitude of CSO events. It is clear,
however, that if GSI does decrease the frequency or magnitude of CSO events, there are real benefits in
the form of avoided penalties. Second discussed are the costs associated with reducing/preventing
future CSO events. Data from SPU offered average per-unit costs for both GSI BMPs and gray
infrastructure techniques. In some instances, GSI BMPs offer cheaper solutions to increasing the
capacity of the city's stormwater system. In these instances, GSI offers a benefit equal to the avoided
cost of implementing costlier gray infrastructure. GSI may not be feasible in all cases, however, and a
balanced approach (as indicated in SPU's 2010 CSO Reduction Plan) will be crucial in meeting future
CSO-related objectives.
D-5
-------
These benefits affect several groups in and out of Seattle. Within Seattle, agencies and utilities can
benefit by reducing future CSO-related penalties and by minimizing infrastructure costs. These benefits
have the capacity to filter through to ratepayers in the form of decreased utility rates. These benefits,
however, go beyond those tied to financial transactions. To the extent other individuals in Seattle,
across Washington, and across the country derive benefits from improved water quality, they too derive
valuable benefits from reductions in the frequency and severity of CSO events.
Managing Interest Payments
Efforts aimed at curbing CSOs oftentimes require large capital costs, which are typically covered by bond
revenues. SPU must pay interest on the value of these bonds. Currently, SPU has high ratings for its Water and
Drainage and Wastewater bonds by both Standard and Poor's (AA+) and Moody's (Aal), which means that the
interest rates SPU pays are relatively low. In its 2013 proposed budget, SPU indicates that interest payments will
account for 15% of its expenditures, and that capital costs will account for another 20% of expenditures.
Typically, these interest payments filter through to property owners or ratepayers depending on how the bonds
are issued. The higher the interest rates, the larger the impact on property owners and ratepayers.
Minimizing the need for large infrastructure projects can reduce the burden of interest payments on SPU's
annual budget and can lead to real savings for property owners and ratepayers. The Federal Reserve Bank of St.
Louis tracks state and local bond rates through its Bond Buyer Go 20-Bond Municipal Bond Index (WSLB20). The
figure below shows how the index has changed since the 1960s. Current bond rates are on the low end of the
range over the past 50 years.
14.0
12.0
10.0
g 8.0
0
1	6.0
4.0
2.0
0.0

Sources: Hoffman, R. No Date. Seattle Public Utilities Proposed Budget 2013. Retrieved on July 3, 2013 from
www.seattle.aov/financedeDartment/13DroDosedbudaet/documents/SPU 373 376.pdf: Federal Reserve Bank of St. Louis. 2013. State and
Local Bonds - Bond Buyer Go 20-Bond Municipal Bond Index. Retrieved on July 3, 2013 from
http://research.stlouisfed.ora/fred2/araph/?sfllfidJ=WSLB20.
D-6
-------
Appendix E, Sanitary Sewer Overflows
E-l
-------
TECHNICAL MEMORANDUM	CH2MHILL
Potential of Green Stormwater Infrastructure to Reduce Sanitary
Sewer Overflows in the Ballard CSO Basin
prepared FOR: Dave Jacobs, PE/SPU
PREPARED BY:	Dustin Atchison, PE/CH2M HILL
Mason Throneburg, PE/CH2M HILL
Santtu Winter, PE/CH2M HILL
DATE:	August 7, 2013
PROJECT NUMBER: 464422
1.0 Purpose
The purpose of this technical memorandum is to document the methodology and results of the analysis to
estimate the effectiveness of Green Stormwater Infrastructure (GSI) in reducing Sanitary Sewer Overflows (SSOs)
in the Ballard CSO Basin, both for current conditions and for future climate change conditions. This analysis was
completed as part of the larger SPU GSI Program, one task of which is to quantify the indirect benefits of GSI.
2.0 Methodology
The first step in the analysis was to establish baseline conditions and alternative conditions in order to be able to
compare the impacts of GSI on SSOs. The baseline conditions consisted of two scenarios; one with GSI
implemented, and one without GSI implemented. The alternative conditions also consisted of two scenarios; one
with GSI and climate change, and one without GSI but with climate change. These four scenarios made it possible
to estimate the impact that climate change may have on SSOs, and GSI's ability to mitigate those impacts.
A 32-year long simulation was completed for each scenario in the calibrated Ballard CSO model. Table 1 presents a
summary of the four scenarios. Table 2 presents the amounts of GSI included in each scenario. Because the model
does not include all of the basin's side sewers and basement elevations, a surrogate for measuring potential SSO
events was used. If the level in a maintenance hole reached within 6-feet of the ground elevation, the event was
flagged as a potential SSO for further analysis. A 6-ft threshold was selected based on a previous study in the
Broadview neighborhood that found the 6-foot assumption to be good estimate for basement depths within the
basin.
Table 1. Summary of Modeled Scenarios
Scenario No.
Condition
Includes Climate Change?
Includes GSI?
Rainfall Scaling Factor
1
Baseline
No
No
1.0000
2
Baseline
No
Yes
1.0000
3
Future
Yes
No
1.0609
4
Future
Yes
Yes
1.0609
Table 2. Amount of GSI Modeled
Scenario No.
Roadside
Green
Rainwise
Total
GSI Cost
Raingardens (ac)
Alleys (ac)
Raingardens
(ac)
Raingardens in Partially
Separated Areas (ac)
Cisterns
(ac)
(ac)
($M)
1
0
0
0
0
0
0
$0
2
41.1
5.8
7.8
4.1
9.1
67.8
$18.9
CITY OF SEATTLE CSO REDUCTION PROGRAM
-------
POTENTIAL OF GREEN STORMWATER INFRASTRUCTURE TO REDUCE SANITARY SEWER OVERFLOWS IN THE BALLARD CSO BASIN
3
0
0
0
0
0
0
$0
4
41.1
5.8
7.8
4.1
9.1
67.8
$18.9
In order to reduce model run times and to focus the results of the model, a subset of maintenance holes (MHs)
and pipes were selected for inclusion in this analysis. These maintenance holes and pipes were selected based
pipe surcharging from previous model simulations of the Ballard CSO Basin, which consisted of running 5-year
simulations instead of 32-year simulations. Figure 1 presents an overview of the Ballard CSO Basin, and the
maintenance holes and pipes included in the analysis.
3.0 Results
The result of each 32-year simulation was a list of MHs included in the analysis and the number of times the 6-ft
threshold was exceeded. This list represents the number of potential SSO events at each MH during the 32-year
period. Table 3 presents the average number of times per year the 6-ft threshold was exceeded for each of the
four scenarios modeled. Figures 2 through 5 show the locations of threshold exceedances in the Ballard CSO
Basin.
Table 3. Summary of Threshold Exceedances
Scenario No.
Average Number of Threshold
Exceedances per Year
1
48.9
2
36.4
3
58.4
4
42.7
The number of threshold exceedances was then compared with actual reported SSO events in the Ballard basin in
order to come up with a calibration factor that relates the number of threshold exceedances to the actual number
of SSO events. This factor was needed because not all of the modeled threshold exceedances are actual SSO
events. Based on conversations with SPU staff, it is estimated that approximately 5 SSO events are reported (not
verified) in the Ballard basin per year (CH2M HILL, 2013). The discrepancy between the number of modeled
threshold exceedances and actual reported SSO events is likely caused by many factors. For example, not all
homes have basements, and those that do may have daylight basements or shallower basements, backflow
preventers or elevated side sewers served by sump pumps, etc.. Other possible reasons include inherent
limitations of the model, which was built and calibrated mainly to predict CSO events, and may not represent the
individual contributions of blocks with as much accuracy. Because Scenario 1 represents current conditions (no
GSI and no climate change), the calibration factor was calculated as follows based on Scenario 1:
5 SSO events/48.9 modeled threshold exceedances = 0.10 SSO events/modeled threshold exceedance
This calibration factor was then applied to the modeled threshold exceedances in order to come up with an
estimated number of SSOs per year, as presented in Table 4.
	Table 4. Estimated Number of SSOs per Year	
2
Scenario No.
Estimated No. of SSO Events per Year
1
4.9
2
3.6
3
5.8
4
4.3
HISTORICAL_ANALYSIS_MEMO_V1.DOCX
-------
4.0 Discussion
Table 5 presents the estimated reduction in the number of SSO events for the two modeled conditions. The
results indicate the GSI may reduce the number of SSO events in the Ballard CSO Basin by approximately 1.5 per
year, or approximately 30%.
The next steps in this analysis will include quantifying the cost-savings of the reduction in SSO events due to
implementing GSI.
Table 5. Reduction in Number of SSO Events
Condition
Scenarios Included
Reduction in No. of SSOs
% Reduction in No. of SSOs
Baseline
1 & 2
1.3
27%
Future
3 & 4
1.5
26%
5.0 References
CH2M HILL, 2013. Number of SSOs per Year in Ballard CSO Basin. 2013. Personal communication with Dave
Jacobs/SPU. July 25, 2013.
HISTORICAL ANALYSIS MEMO V1.DOCX
-------
2,000
M Feet
Figure 1
Ballard CSO Basin Overview
C:\Users\swinter\Desktop\Projects\GIS\LTCP GIS\GSI\Fig_1_Overview.mxd
7/30/2013
-------


:NW,WRKEiTfe^
mmmM
Estimated # of MH's Exceeding Threshold per Year: 48.9
500 1,000
2,000
I Feet
No. of Exceedances in 32 Years
Figure 2
Scenario 1: No GSI, No Climate Change
C:\Users\swinter\Desktop\Projects\GIS\LTCP GIS\GSI\Ballard_6-ft_NoGI_NoCC.mxd
7/30/2013
o
32+
o
15 to 32
o
5 to 15
o
2 to 5
o
1 to 2
-------
iNW/80iTiflrsirt
INWi[65 JIT hrl fSjTd
^WWRKEitfe^
500 1,000
2,000
I Feet
Figure 3
Scenario 2: GSI, No Climate Change
C:\Users\swinter\Desktop\Projects\GIS\LTCP GIS\GSI\Ballard_6-ft_GI_NoCC.mxd
8/1/2013
No. of Exceedances in 32 Years
O 1 to 2
O 2 to 5
O 5 to 15
O 15 to 32
32+
-------
WWWRKEffifeg
Estimated # of Threshold Exceedance Events per Year: 58.4
500 1,000
2,000
I Feet
Figure 4
Scenario 3: No GSI, Climate Change
C:\Users\swinter\Desktop\Projects\GIS\LTCP GIS\GSI\Ballard_6-ft_NoGI_CC.mxd
8/1/2013
No. of Exceedances in 32 Years
O 1 to 2
O 2 to 5
O 5 to 15
O 15 to 32
A 32+
-------
iNW/80iTiflrsirt
INWi[65 JIT hrl fSjTd
;NWimRK5Tfeljj
500 1,000
2,000
I Feet
Figure 5
Scenario 4: GSI, Climate Change
C:\Users\swinter\Desktop\Projects\GIS\LTCP GIS\GSI\Ballard_6-ft_GI_CC.mxd
8/1/2013
GI_CC
O 1 to 2
o
O
O
2 to 5
5 to 15
15 to 32
32+
-------