xvEPA
United States
Environmental Protection
Agency
Using Economic Incentives to
Manage Stormwater Runoff in the
Shepherd Creek Watershed, Part I
-------
EPA/600/R-08-129
October 2008
Using Economic Incentives to
Manage Stormwater Runoff in the
Shepherd Creek Watershed,
Parti
By
Hale W. Thurston, Allison H. Roy, William D. Shuster, Matthew A. Morrison*,
Michael A. Taylor, and Heriberto Cabezas
A portion of the work outlined in this report was done under
Contract number EP-C-05-061 Task Order #08
Project Officer Hale W. Thurston, PhD.
Sustainable Technology Division
*Land Remediation and Pollution Control Division
National Risk Management Research Laboratory
US Environmental Protection Agency
Cincinnati, OH 45268
-------
Notice
The U.S. Environmental Protection Agency (EPA) through its Office of Research and
Development performed and managed the research described in this report. It has been subjected
to the Agency's peer and administrative review and has been approved for publication as an EPA
document. Any opinions expressed in this report are those of the author and do not, necessarily,
reflect the official positions and policies of the EPA. Any mention of products or trade names
does not constitute recommendation for use by the EPA.
Acknowledgments
The U.S. Environmental Protection Agency through its Office of Research and
Development funded and collaborated in the research described here under the Scientific,
Technical, Research, Engineering and Modeling Support (STREAMS) contract EP-C-05-061,
Task Order 8: "Implementing Dispersed Best Management Practices in the Shepherd Creek Area
of Cincinnati." Support for water quality sampling and analysis was provided by Pegasus
Technical Services, Inc. under EPA Contract # EP-C-05-056. The U.S. EPAs Central Regional
Laboratory in Chicago, IL provided critical and timely analytical services for the majority of the
water quality analyses. Periphyton and macroinvertebrate identifications and chlorophyll
analyses were performed in 2003-04 by PhycoTech, Inc. under simplified acquisition 3C-R312-
TTSA and in 2005-06 by EnviroScience, Inc. under simplified acquisition EP05C00134.
Directly connected impervious area (DCIA) data entry and assessment were performed under
STREAMS Contract EP-C-05-059, task order 0009, to Eastern Research Group, Inc. Support for
some of the hydrologic monitoring was made available through an Interagency Agreement (IAG)
with the U. S. Department of Agriculture-National Resources Conservation Service (USDA-
NRCS) for "Hydroecological benefits and economic attributes of on-lot stormwater best
management practices", DW12921597-01-0, 2004 and an IAG with the US Geological Survey
(USGS) titled "Hydroecological benefits and economic attributes of on-lot stormwater best
management practices", DW-14-921600-01-0, 2004-2007.
This report would not be possible without the work of many people along the way.
Haynes C. Goddard, David Szlag, Beth Lemberg, Michael Miller, and Jamie C. Coleman had
significant intellectual input at the inception of this project. Laura Boczek of ORD's Water
Supply and Water Resources Division coordinated microbial assessments for all water samples.
Ward Wilson, Mike Valerius and David Snook provided invaluable work in the field.
-------
Foreword
The U. S. Environmental Protection Agency (USEPA) is charged by Congress with
protecting the Nation's land, air, and water resources. Under a mandate of national
environmental laws, the Agency strives to formulate and implement actions leading to a
compatible balance between human activities and the ability of natural systems to support and
nurture life. To meet this mandate, USEPA's research program is providing data and technical
support for solving environmental problems today and building a science knowledge base
necessary to manage our ecological resources wisely, understand how pollutants affect our
health, and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory (NRMRL) is the Agency's center
for investigation of technological and management approaches for preventing and reducing risks
from pollution that threaten human health and the environment. The focus of the Laboratory's
research program is on methods and their cost-effectiveness for prevention and control of
pollution to air, land, water, and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites, sediments and ground water; prevention and control
of indoor air pollution; and restoration of ecosystems. NRMRL collaborates with both public
and private sector partners to foster technologies that reduce the cost of compliance and to
anticipate emerging problems. NRMRL's research provides solutions to environmental problems
by: developing and promoting technologies that protect and improve the environment; advancing
scientific and engineering information to support regulatory and policy decisions; and providing
the technical support and information transfer to ensure implementation of environmental
regulations and strategies at the national, state, and community levels.
This publication has been produced as part of the Laboratory's strategic long-term
research plan. It is published and made available by USEPA's Office of Research and
Development to assist the user community and to link researchers with their clients.
Sally Gutierrez, Director
National Risk Management Research Laboratory
in
-------
Abstract
Communities nationwide are facing increased responsibility for controlling stormwater
runoff, and, subsequently, rising costs of stormwater management. In this report we describe and
test a methodology that can be used by communities to focus limited budgets on the most
efficient and ecologically-effective installation of stormwater management practices. The
overall project has two primary objectives: (1) to test the use of an auction to cost-effectively
allocate stormwater management practices among landowners, and (2) to determine the
effectiveness of the resulting implementation in terms of hydrological, water quality, and
ecological measures. Here, we describe the theories, methods, and criteria used to distribute rain
gardens and rain barrels to homeowners in a small, midwestern watershed. The first round of the
reverse auction in 2007 resulted in 50 rain gardens and 100 rain barrels installed at 67 of the
approximately 350 residential properties in the experimental watershed. In 2008, the auction
was repeated and we accepted bids for an additional 35 rain gardens and 74 rain barrels.
Stormwater management practices were distributed relatively evenly throughout the watershed
and are expected to result in significant improvements in stream quality. We describe our
monitoring approach, including 1) parcel-scale hydrology and water quality monitoring of
selected rain gardens, and 2) stream monitoring following before-after-control-impact approach
for assessing the hydrological, water quality, and biotic responses to stormwater management
installation. By employing a multidisciplinary approach to watershed management, the case
study offers an example of stormwater management that should be readily transferable to other
residential watersheds.
IV
-------
Table of Contents
Page
Title Page i
Notice ii
Acknowledgments ii
Foreword iii
Abstract iv
List of Tables vi
List of Figures vi
CHAPTER 1 Introduction 1
1.1 Background 1
1.2 Purpose 1
1.3 Study setting 1
CHAPTER 2 Developing a Retrofit Watershed Management Strategy 2
2.1 Characterizing the impairment 2
2.2 Assessing the source of impairment and potential for improvement 5
2.3 Evaluating economic incentive mechanisms 9
CHAPTERS Stream Monitoring Design 14
3.1 Experimental design 14
3.2 Hydrologic monitoring 16
3.3 Water quality monitoring 16
3.4 Ecological monitoring 18
CHAPTER 4 Auction Design 21
4.1 Background 21
4.2 Information to homeowners 21
4.3 Methods for assessing auction bids 25
CHAPTERS Auction Results 29
5.1 Phase 1, Year 2007 29
5.2 Phase 2, Year 2008 33
CHAPTER 6 Stormwater Management Practice Installations 36
6.1 Overview 36
6.2 Rain gardens 36
6.3 Rain barrels 38
CHAPTER 7 Rain Garden and Rain Barrel Monitoring 40
7.1 Overview 40
7.2 Hydrology and soils 40
7.3 Water quality 40
CHAPTER 8 Conclusions 42
Publications related to this project 43
References 45
APPENDIX A: Rain Garden Schematics 49
APPENDIX B: Rain Barrel Schematic 54
APPENDIX C: Quality Management Program 55
-------
List of Tables
Table 2.2.1 Total (TIA) and Directly Connected (DCIA) Impervious Area Categorized by Impervious Surface Type. 7
Table 3.1.1 Start dates for hydrology, water quality, and ecology monitoring 14
Table 3.3.1 Summary of water quality monitoring parameters and associated impairments 17
Table 3.4.1 Summary of samples collected and sampling resolution 18
Table 5.1.1 Summary of bids for rain gardens and rain barrels in 2007 and 2008 29
Table 5.1.2 Rain barrel rankings, 2007 30
Table 5.1.3 Rain garden rankings, 2007 31
Table 5.2.1 Rain garden rankings, 2008 33
Table 5.2.2 Rain barrel rankings, 2008 34
List of Figures
Figure 2.1.1 Map of impervious areas and parcels within the Shepherd Creek study watersheds 2
Figure 2.1.2 Pre-implementation data for Shepherd Creek sites, comparing baseflow and stormflow water quality data
for the years 2005 and 2006. (A) Dissolved Inorganic Nitrogen; (B) Total Phosphorus and (C) Extent of bacterial
contamination (as Escherichia coli); note log scale fory axis on these plots 4
Figure 2.2.1 Total connected and disconnected impervious area in the Shepherd Creek watershed based on property
ownership 6
Figure 2.2.2 Percent TIA given 0, 25, 50, and 100% landowner acceptance rates of rain gardens and rain barrels 8
Figure 2.3.1 Costs of runoff detention with Tradable Allowances 11
Figure 3.1.1 Location of sites for various sampling activities in the Shepherd Creek watershed 15
Figure 4.2.1 Brochure sent to eligible homeowners 22
Figure 4.2.2 Auction bid form included in second mailing 23
Figure 4.2.3 Pictures of demonstration rain barrel (A), rain garden (B), and signage (C) 24
Figure 5.1.1 Location of rain barrel and rain garden bids in Phase 1, 2007 32
Figure 5.2.1 Location of rain barrel and rain garden bids in Phase 2, 2008 35
Figure 6.2.1 Species in plant communities planted in rain gardens, 2007 37
Figure 6.2.2 Typical rain garden installation 38
VI
-------
CHAPTER 1 Introduction
1.1 Background
Traditional stormwater control policies have concentrated on solutions that build
centralized capacity to direct and hold excess runoff within a storm sewer system. However, such
centralized infrastructures do not sufficiently alleviate water quality problems for receiving waters,
and can be expensive for municipalities who must work within budget constraints. This project
tests an alternative approach to stormwater control using a decentralized approach that disperses
retrofit runoff detention practices throughout a small suburban watershed, thus reducing runoff
before it reaches the sewer system. By distributing decentralized stormwater detention through
market mechanisms, we concurrently study the hydrological, water quality, and ecological benefits
of stormwater management practice implementation and explicitly evaluate the cost of meeting
environmental quality standards through this type of approach.
1.2 Purpose
The overall project has two primary objectives: (1) to test the use of an auction to cost-
effectively allocate stormwater management practices among landowners, and (2) to determine the
effectiveness of the resulting implementation in terms of hydrological, water quality, and
ecological measures. The stormwater management practices used in this project were limited to
rain barrels and rain gardens. A voluntary, sealed-bid auction was used to allocate stormwater
management practices and determine the compensation private landowners will receive in
exchange for accepting the installation of a stormwater management practice on their property.
The winners of the auction were those who offered to accept stormwater management practices
with the greatest environmental benefits at the lowest price. The auction provided a means of
identifying willing landowners and offers a mechanism by which these stormwater management
practices can be systematically allocated to these landowners within the watershed.
1.3 Study setting
The study took place in the Shepherd Creek watershed, a tributary to the West Fork Mill
Creek, located in Mt. Airy, Cincinnati, Ohio. The watershed is 1.85 km2 (457 acres),
approximately one-third of which lies within a city park with mature deciduous forest. The other
two-thirds of the watershed represent a mix of 1960-1980s residential parcels in the headwaters,
and horse pastures at downstream locations. The residential area consists primarily of single
family homes and has a median lot size of 880 m2 (0.22 acres). Over three-quarters of the 406
houses in the catchment were built between 1950 and 1990. There are also three apartment
complexes (27 buildings) in the headwaters and several public buildings with parking lots (e.g., a
church, police station, park arboretum). The watershed sits on calcareous shale and limestone
formations with moderate slopes, and silt and silty clay loam soils dominate.
-------
CHAPTER 2 Developing a Retrofit Watershed Management Strategy
2.1 Characterizing the impairment
Cincinnati's Mill Creek is considered by the Ohio Environmental Protection Agency to
be among the most polluted waterways in the state (OEPA, 1994), with stormwater runoff being
a major direct and indirect contributor to the pollution. The Shepherd Creek is similarly
impaired by stormwater runoff, as evidenced by hydrologic, geomorphic, water quality, and
biotic assessments of the stream and its tributaries (Figure 2.1.1).
Legend
• study sites
'"X.- stream
£3 subcatchments
| | parcels
^B connected impervious area
disconnected impervious area
0 0.1 0.2 0.4 Kilometers
Figure 2.1.1 Map of impervious areas and parcels within the Shepherd Creek study watersheds.
In terms of stream geomorphology, the cobble/gravel riffles are highly embedded with silts
and other smaller-sized eroded sediments. A high percentage of total streambed area is observed
to be highly scoured and entrenched, leaving platy formations of bedrock which are in an active
process of rotting, and these surfaces are typically found to be covered with a thin layer of silt.
Since the streambeds have been down cut to bedrock, stream reaches have attempted to adjust to
-------
present stream flow patterns through lateral expansion. This process involves the erosion of soils
and wearing away of bedrock to each side of the original streambed. For areas where roads are
adjacent to the streams, some banks have been fortified with concrete, limiting lateral expansion
to one stream bank only.
Developed areas exhibit the predictable flashy behavior found in urban streams, which was
much more pronounced than the more gradual rise and fall hydrograph shape for adjacent
forested areas in Mt. Airy park (Sub5). Storm runoff from urban areas is flashy and delivered
quickly and in great volumes to stream reaches (Paul and Meyer 2001). These small urban
streams may also lack natural baseflows, resulting in streambeds that are dry in the summer
months such that stream flows that can be sporadic and limited to storm events.
Water chemistry in the sub-watersheds is characterized by neutral to alkaline pH (average
values range from 7.7 to 7.9), with high average alkalinity (225 to 295 mg CaCCvl"1) from natural
dissolution of the calcareous shale bedrock that is extant throughout the watershed. As in many
established urban watersheds, chloride (Cl) concentrations are elevated at some stream locations
mostly because of road salt application (Godwin et al. 2003, Kaushal et al. 2005). Sub4, which
receives direct street runoff and is downstream from a moderately steep hill (where salt
application would be expected to be heavy during the winter months), averages 280 mg/L year
round. This value is above the chronic toxicity limit for Cl in freshwater (250 mg/L), and much
higher than the approximately 50 mg/L average measured at Subl and Sub2. Sub5 also receives
road runoff and has an average Cl concentration of 161 mg/L that appears to cycle throughout the
year as salt is loaded into the stream during the winter and flushed out in the summer. Nitrogen
and phosphorus concentrations fall within the range of values expected for urban land uses (Figure
2.1.2 A and B) (e.g., Schoonover et al. 2005). Average baseflow values for dissolved inorganic
nitrogen (DIN) and total phosphorus (TP) ranging from 0.20 to 0.98 nig-l"1 and 0.17 to 0.37 nig-l"1
respectively for the years 2005-2006.
Since hydrology and water chemistry tend to regulate biological activity in aquatic
ecosystems, we were not surprised to find that our periphyton samples contained high
concentrations of chlorophyll a (average 4.3 to 10.2 mg/m2 across sites, 2003-2006). Average
algal ash-free-dry-mass in 2005-2006 ranged from 57.1 (Catch) to 87.9 (SubSa) g/m2. A
majority of the algal cells sampled were cyanobacteria, which reflects overall poor water quality.
Macroinvertebrate assemblages were typically dominated by isopods, amphipods, and
chironomids. In 2005-2006, there was an average richness of 18.0 (Sub4) to 24.1 (Sub2) taxa
per site. Sensitive taxa in the orders Ephemeroptera, Plecoptera, and Trichoptera (EPT)
constituted a mere 0.5% (Sub2) to 5.4% (Catch) of the samples by abundance, further reflecting
poor biotic integrity within the tributary streams. Hilsenhoff s Family Biotic Index scores for
macroinvertebrates suggest fairly poor (5.76-6.50) or poor (6.51-7.25) water quality (Hilsenhoff
1988), which is consistent with water quality and periphyton samples.
-------
a.
Baseflow Dissolved Inorganic
Nitrogen (DIN; 2005-2006)
Storm flow Dissolved Inorganic
Nitrogen (DIN; 2005-2006)
SubS Sub4 Sub3 Sub1 Sub2 Catch
Sub5 Sub4 Sub3 Sub1 Sub2 Catch
b. Baseflow Total Phosphorus (TP; 2005-2006) Stormflow Total Phosphorus (TP; 2005-2006)
Sub5 Sub4 Sub3 Sub1 Sub2 Catch
Sub5 Sub4 Sub3 Sub1 Sub2 Catch
C.
E 10000
8
^ 1000
Baseflow E. co/i (2005-2006)
Stormflow E. co/i (2005-2006)
Sub5 Sub4 Sub3 Sub1 Sub2 Catch
Sub5 Sub4 Sub3 Sub1 Sub2 Catch
Figure 2.1.2 Pre-implementation data for Shepherd Creek sites, comparing baseflow and
Stormflow water quality data for the years 2005 and 2006. (A) Dissolved Inorganic Nitrogen; (B)
Total Phosphorus and (C) Extent of bacterial contamination (as Escherichia coli)\ note log scale
for y axis on these plots.
-------
Of great interest to public health interests in stormwater and watershed management, we
observed average fecal coliform bacteria and Escherichia coli counts derived from baseflow
water samples that were 1-2 orders of magnitude higher than Ohio EPA's ambient surface water
quality criteria (e.g., mean limit 126 CPU 100 ml"1 for Escherichia coli). These high counts of
pathogens were observed at both baseflow and, more recently, in storm flow samples; Figure
2.1.2 (C) shows that median baseflow concentrations of Escherichia coli are highest at Subl
(1,400 CFU/100 mL), while median stormflow concentrations are highest at Sub2 (37,500
MPN/100 mL). After further investigation we concluded that the most likely sources are: a)
wastewater infrastructure, which is sometimes improperly connected from residences, leaking
flows to stormwater conveyances; b) exfiltration from septic fields that were hydrologically-
connected to headwater reaches in the north-central area of the watershed; and c) improper
storage of horse manure in the lower part of the watershed. Additional investigation is needed to
clearly identify and allocate bacterial loads among these, and possible wildlife sources.
2.2 Assessing the source of impairment and potential for improvement
Impervious area
Impervious surfaces are a primary source of impairment in urban and suburban areas,
resulting in increased stormwater runoff and reduced infiltration compared to more natural
settings (Arnold and Gibbons 1996). This exacerbated amount of stormwater runoff translates to
hydrologic impairments in streams such as increased volume of peak flow and storm "flashiness"
(Konrad and Booth 2005), which subsequently alters stream morphology and sediments (Booth
and Jackson 1997, Bledsoe and Watson 2001). Stormwater runoff also carries pollutants from
the landscape, resulting in altered water quality in urban streams (Hatt et al. 2004). The
combined physical effects of impervious surfaces in streams has led to impaired biotic
communities and reduced ecosystem functioning in urban streams (Paul and Meyer 2001; Walsh
et al. 2005b).
Percent urban land cover and total impervious area (TIA) are commonly used as
indicators of urban disturbance. More recently, studies have shown that the subset of impervious
surfaces that route stormwater runoff directly to streams via stormwater pipes, called directly
connected impervious area (DCIA), may be responsible for the majority of stream alteration due
to urbanization (Booth and Jackson 1997, Brabec et al. 2002, Walsh 2004, Walsh et al. 2005a).
For this project, we used a combination of field assessments, aerial photography, and GIS data to
determine both TIA and DCIA in the Shepherd Creek watershed.
Impervious (22.1 ha) and semi-impervious (1.8 ha) areas comprise 13.1% of the
Shepherd Creek catchment (Catch; Figure 2.1.1). Of the impervious area, 56.3% was connected,
although percent connectivity varied widely across parcels. Across watersheds, the lowest
percent TIA (11.2%) and DCIA (5.4%) were at Sub4, and the highest percent TIA (19.9%) and
DCIA (11.6%) were at Subl. A majority of the TIA in Shepherd Creek was on private land
(70.5%) compared to public land (29.5%). Public properties (e.g., roads, city park) encompassed
a larger proportion of the connected (37.4%) versus disconnected (19.4%) impervious area.
Conversely, single-family residential properties comprised a higher proportion of the total
impervious area found to be disconnected (68.1%) than connected (42.1%). The public parcels
and the private, multi-family residential parcels both had more than double the total amount of
-------
connected versus disconnected impervious area; whereas single-family residential parcels had
overall lower amounts of connected than disconnected impervious area (Figure 2.2.1).
Total Impervious Area
Private,
single-family
Private,
multi-family
Public
Figure 2.2.1 Total connected and disconnected impervious area in the Shepherd Creek watershed
based on property ownership.
We evaluated the primary types of impervious surfaces and percent connectivity of those
surfaces to assess types of stormwater management practices that may result in best potential for
retrofit. The highest amounts of TIA were due to buildings (i.e., rooftops; 27.6%), driveways
(24.6%), streets (22.7%), and parking areas (12.3%; Table 2.2.1). A majority (89.2%) of the
streets were connected, so that streets composed the highest percent DCIA (36.0%) in the
catchment, followed by buildings (32.9%), driveways (17.4%), and parking areas (13.5%). The
remaining impervious surface types (e.g., sidewalks, concrete patios, etc.) comprised less than
13% of TIA and 0.2% of DCIA.
-------
Table 2.2.1 Total (TIA) and Directly Connected (DCIA) Impervious Area Categorized by
Impervious Surface Type.
TIA
Surface Type
Building
Driveway
Street
Parking area
Sidewalk
Concrete
Wooden deck
Pool
Shed
Other
(m2)
66168
58918
54432
29473
13097
6963
4984
2363
882
2047
(%)
27.6
24.6
22.7
12.3
5.5
2.9
2.1
1.0
0.4
0.1
DCIA o/n
(m2)
44364
23525
48551
18144
3
211
0
0
12
4
(%) Connected
32.9
17.4
36.0
13.5
0.0
0.2
0.0
0.0
0.0
0.0
67.0
39.9
89.2
61.6
0.0
3.0
0.0
0.0
1.3
6.7
Selected stormwater management practices
Because a majority of TIA was on private property (70.5%) and in buildings and
driveways (52.2%), we targeted private properties for installation of stormwater management
practices in the form of rain gardens and rain barrels. Parcels in Shepherd Creek are of adequate
size (median lot size = 880 m2 or 0.22 acres) to permit placement of rain gardens within lawns
and rain barrels on roof gutter downspouts. Rain gardens are engineered bioretention cells that
have porous substrate and soils designed to allow rain and snowmelt to seep naturally into the
ground. Rain gardens typically have a concave surface to increase the capacity for holding rain
water, and they are planted with hearty plant species that are selected for their tolerance of local
climate and extremes in root zone water content. They are ideally located downslope of
impervious surfaces to capture stormwater runoff generated from rooftops, driveways, sidewalks,
and patios. By encouraging infiltration, rain gardens reduce flooding and pollution in local
streams and rivers, and can help to recharge local water tables. Rain barrels, or cisterns, are
tanks attached to roof gutter downspouts which are used to collect rainwater from rooftops.
Water from rain barrels can be used to water gardens and lawns and other domestic uses and,
therefore, may diminish other household demands on potable water. By storing stormwater
runoff during storm events, rain barrels reduce downstream flooding. Furthermore, if barrels are
emptied into the landscape after storms (i.e., then the soil is no longer saturated), they can
effectively recharge local water tables. The combination of stormwater management practices
were selected for their ease of implementation in an existing neighborhood and potential for
addressing the major sources of impairment in the Shepherd Creek watershed.
Potential for improvement
Previous studies have reported that biotic impairment of streams typically occurs around 15-
20% urban land cover or 8-12% impervious surface cover (see reviews by Schueler, 1994; Paul
and Meyer, 2001; Walsh etal., 2005b). In Figure 2.2.2, we report % TIA in each of the six
subcatchments (black bars), and projected % TIA (effective) with various rates of stormwater
management practice acceptance. The calculations assume that rain barrels and rain gardens
-------
effectively eliminate runoff effects from both rooftops and driveways. With 100% homeowner
acceptance rates, these stormwater management practices have the combined potential to reduce
effective impervious area in sites 2-5 from above the 8-12% impervious area threshold to well
below the threshold (Booth and Jackson 1997). If 50% of the homeowners have stormwater
management practices installed, it is still likely that some subwatersheds will exhibit
improvements in stream condition.
Some important caveats should be noted regarding the estimates of potential improvement.
First, it is unlikely that all runoff from impervious surfaces will be routed to rain gardens and
rain barrels, and there is a limited capacity for storing and infiltrating runoff. Detailed rainfall-
runoff modeling is necessary to determine actual amounts of stormwater detained or infiltrated
following various size precipitation events. Estimated reductions in DCIA may more effectively
capture the potential improvements with retrofit; however, thresholds of biotic impairment for
DCIA are limited and variable (Wang et al. 2001, Walsh et al. 2005a, Wenger et al. 2008).
Further, there are no empirical models that provide data on expected responses of hydrology and
water quality to changes in TIA or DCIA. Given the mechanism of stream impairment via
stormwater runoff, we expect that changes will first be detected in surface hydrology parameters,
followed by water quality, and, lastly, biotic assemblages. Thus, our estimates of potential
improvement may be conservative with respect to hydrology and water quality. Finally, we
A 24
20 -
16
F12
&••
8 -
4 -
0
empirical threshold
stream impairment
\
BMP
Acceptance
Rates (%)
• 0
S 25
m so
n 100
1 2
(control)
3 4
Site
5 6
(control)
Figure 2.2.2 Percent TIA given 0, 25, 50, and 100% landowner acceptance rates of rain gardens
and rain barrels.
expect the potential for detecting change to be highest in the smallest catchments (e.g., the
stormwater outfalls and Subl), as larger catchments are more likely to have additional sources of
impairment and mediating factors. Although stream sampling will be necessary to determine
actual changes associated with retrofit, it is important that our selection of stormwater
management practices addressed the primary sources of imperviousness, and that the
subcatchments do not have excessive amounts of imperviousness (e.g., >50%, Walsh et al.
2005a), such that stream improvement following retrofit is possible.
-------
2.3 Evaluating economic incentive mechanisms
Command-and-control
The economic portion of the study began with a theoretical examination of the potential cost
savings from using various market based mechanisms to provide economic incentive for
widespread participation in a dispersed storm water retention exercise. This was a necessary first
step because we wanted to see if there was any potential savings over current policy before
proceeding to the actual implementation. To evaluate the cost effectiveness of various market
mechanisms we wanted first to determine, in a case study setting, the cost of a command and
control mechanism. Command and control is a mechanism that allows for very little flexibility
among the regulated entity; the regulating agency sets a standard and the regulated must meet the
standard under penalty of fines. Below we compare, in a real setting, using realistic cost and
hydrology figures, the costs in per unit runoff detention under a command and control situation
or either of two market based incentive mechanisms: tradable credits, and fee and rebate. We
conclude by proposing a market mechanism that will fit legally with our study area, a
procurement auction.
Using the parcels from the actual Shepherd Creek area the type of stormwater management
practice, concomitant detention cost (DC) functions and inverse cost functions, were assigned to
parcels in the watershed based on land use and soil type to calculate what the cost would be, in
the absence of any market incentives, to control the excess stormwater runoff from watershed
areas with dispersed, small-scale stormwater management practices. To calculate the cost of a
dispersed set of stormwater management practices to store all of the excess runoff discharge
from a one and a half year storm event of 3.12 cm (1.23 inches) on site, we assigned the
appropriate least-cost stormwater management practice technologies on a parcel-by-parcel basis
in our small case area and solve each landowner's cost equation.
• Parcels with residential land use and hydrologic soil group (HSG) B, which are soils of
silt or loam, are assumed to employ sand filters. The cost function is:
DC^ = 26.60° 64+0.126Q
• Residential parcels with HSG C or D, soils with lower infiltration capabilities due to
existence of more clay, are assumed to use rain gardens/grassed swales, the relevant cost
function of which is £>CRexC = 4.94Q+ 0.126Q
Where Q is the quantity of stormwater runoff detained in cubic feet, and second term in each is
the log-linear estimated opportunity cost of land. We choose to not include a quadratic term
estimated from a previous paper because while it is statistically significant, it is not economically
significant at the stormwater management practice sizes we are concerned with. The cost
functions are modified from Schueler (1987) and Heaney et al (2002).
This calculation presumes that the parcel owner is responsible for all the runoff over and
above that which would result if the parcel were in its undeveloped state. This in effect is a
command-and-control regime with no rebate and the water detained with this constraint is 2344
m3 (82,767 ft3), at an average cost of $950 per homeowner. When we considered only
construction costs of stormwater management practices, we calculated an average cost of $4.62
per cubic foot of stormwater runoff detained via stormwater management practices over all
-------
properties in the study area. Finally, we recalculated the cost to include the opportunity cost of
land, which was estimated from the hedonic price function and noted in the cost functions above;
for the log-linear case this adds $0.126 per ft2. Assuming again that command-and-control
policy is implemented that causes all parcel owners to use stormwater management practices to
manage all excess stormwater runoff, the recalculated cost was $6.49 per ft3, at an average cost
per homeowner of $ 13 3 7.
Tradeable allowances
The first hypothetical study we conducted in the Shepherd Creek area was one that
envisioned the use of tradable credits for stormwater runoff, an idea first suggested by Coleman
(2000). The cost effectiveness of the tradable allowance approach to pollutant reduction in
airsheds is well established in the literature (Eheart 1980, Baumol and Gates, 1988, and
Tietenberg 2000), and the SO2 trading program in the United States has been operating
successfully for several years. Watersheds differ from airsheds, however, in key aspects, such as
confinement to a channel, non-uniform mixing and downstream accumulation, and these present
new challenges for the establishment of tradable allowance systems. Watershed trading is not a
new concept, but the specific application explored in this paper is new. The US EPA'sDraft
Framework for Watershed-Based Trading (1996) provides an overview of some twenty tradable
allowance programs across the United States. Several of these grew out of cooperative
agreements with the US EPA, the Water Environment Research Foundation, and various local
stakeholder groups. These programs focus on reducing concentrations of nutrients or toxics, and
most rely upon an organizational effort similar to US EPA's total maximum daily load (TMDL)
process to drive stakeholder involvement.
Two necessary conditions for tradable allowance regimes to be cost reducing are that: i)
transactions costs of such programs be no greater than the gains achieved and ii) there be
sufficient difference in abatement cost across parcel owners so that potential cost savings can be
realized through market exchange of runoff control. With these conditions satisfied, a tradable
allowance system can efficiently assign runoff control to dispersed locations and may avoid the
larger cost of centralized approaches.
The usefulness of inclusion of opportunity costs can hardly be overstated in this
application. We use the results of our opportunity cost estimation1 to inform a tradable
allowances system much like those currently used in water quality trading programs around the
country (USEPA 2003). Figure 2.3.1 compares, for the Shepherd Creek Pilot project, the per-
unit runoff reduction costs faced by homeowners under assumptions of different credit prices,
and inclusion or exclusion of opportunity costs. Not surprisingly when opportunity costs are
ignored all costs are lower. The cost line named "18-mile tunnel" represents the costs per unit of
detention in a proposed large infrastructural stormwater conveyance system that is under
consideration by the Metropolitan Sewer District for the city of Cincinnati2.
1 Thurston (2006) describes in detail how a Hedonic price function is estimated to approximate the opportunity cost
of dedicating land to Stormwater management practices for stormwater runoff control.
2 It should be noted that the tunnel would serve a much larger population than the Shepherd Creek Pilot project area,
which is used in the other calculations.
10
-------
Costs of Runoff Detention
Using Tradable Allowances
14
12 -
10 -
$/CF
4 -
2 -
0 J
\
o.c = $o
O.C = $0.13
O.C = $0.28
18 mile tunnel
2.5
8 15
Allowance Price
C&C
Figure 2.3.1 Costs of runoff detention with Tradable Allowances
Fee and rebate
We also considered the potential application of a fee and rebate system. An efficient tax
on pollution should be a direct tax that equals the marginal external damages (Pigou 1962). But
directly taxing pollution is complex, especially when monitoring is difficult (such as with a non-
point source) or when there are institutional barriers to imposing differing taxes on different
people in the same area. Fullerton and Wolverton (1999, 2003) note that when directly taxing
the pollution is not an option, the policy maker can exploit the relationship the optimal pollution
tax has with income tax and rebates on products that are related to, and relatively cleaner than the
polluting good.
This type of policy is already in place in many municipalities in the United States (Doll et
al 1998, Doll and Lindsey 1999), but these programs are almost exclusively for commercial
properties. Where residential fees are in place they tend to be too small to either promote abating
behavior or warrant a rebate. For example, monthly residential stormwater fees in Columbus,
OH, St. Louis, MO, and Indianapolis, IN are about $2.70, $0.24, and $1.25 respectively. Many
agree that the existing programs have not encouraged the desired behavior because the fees and
rebates are simply too low (Doll and Lindsey 1999). If the fee and the rebate are high enough to
make households reflect their true underlying preferences, based on our knowledge of the costs
facing the residential property owner, a stormwater runoff reduction goal can be met using
dispersed stormwater management practices and a two part instrument at a relatively low cost to
the utility and at a low per-unit cost to the average stakeholder. We include the opportunity costs
11
-------
of land into the calculated the fee and rebate scenario for the Shepherd Creek watershed which
makes the model more realistic. But we find that to substantially increase people's participation
in stormwater infiltration practices, the fees and rebates have to be orders of magnitude larger
than those that are currently in use.
Auction
There are a variety of legal barriers to implementation of most incentive mechanisms.
These barriers stem mostly from language in the Clean Water Act (CWA) of 1972, and "takings"
issues. The CWA provides no regulatory "stick" for policy dealing with water quantity as it is
only regulated under specific circumstances. Tradable credits are simply not tested as they have
just recently been recognized by the EPA as a means to control stormwater runoff. Furthermore,
this recognition comes in the form of relatively weak support in the EPA Water Quality Trading
Policy (EPA 2003). Imposition of a "cap" on a watershed's runoff, and the consequent
requirement for attainment through impingement on established property rights is recognized as
a takings issue. As for a fee and rebate program, although we have shown it to be economically
feasible, existing fees are not tightly tied to excess runoff, and imposition of sufficient fees
(much like the imposition of the cap in the trading scenario) would not be politically feasible.
Thus, we turned to a wholly voluntary economic auction approach designed to encourage
landholders to install stormwater management practices so as to control runoff without
necessitating a legal mandate.
There are essentially four types of auctions, English, Dutch, first-price sealed bid, and
Vickrey. The English auction is the type of auction most people are familiar with. An auctioneer
calls out prices sequentially higher and players bid until only one remains, then that person pays
the highest price called. In the Dutch auction the auctioneer calls out bids in descending order
until the price reaches that which one person is willing to pay, and that person wins the item at
that price. In a first price sealed bid auction bids are submitted without other bidders seeing
them and the highest bidder wins. The Vickrey auction (named after William Vickrey, the first
person to investigate some of the nuances of auction theory) is similar to the first price sealed bid
auction, but the winning bidder pays the price that the second highest bidder bids. Using
rigorous mathematical proofs these mechanisms have been proven to all be "revenue
equivalent," that is that they all theoretically elicit the same winning price; although
experimental and experiential evidence is mixed. Revenue equivalence and other theoretical
characteristics of auction such as efficiency conditions, revenue equivalence, or pareto efficient
allocation at a Bayesian Nash equilibrium are not in the scope of this paper to go into. The
theoretical underpinnings of auctions have been well defined in the economics and game theory
literature and are treated rigorously elsewhere. The auction we use is a variant of the first price
sealed bid auction, known as a reverse or procurement auction. In this case the auctioneer is the
one who wants to buy the item and there are many sellers. We describe this type of action in
detail below.
Regardless of the type of auction, auctions are viewed as superior to other means of
allocating public resources due to their efficiency, objectivity, transparency, and flexibility
(CSIRO 2005). They are efficient because they will allocate the resource to those who are
willing to pay the most and therefore are situated to make the best use of them. Objectivity is
12
-------
achieved because the price is not determined capriciously by a government official, rather it is
market determined. The auction process is transparent because the rules for bidding and winning
are known. Finally, auctions are flexible in that the mechanism can be altered to allow for
various contingencies, such as changing annual budgets (CSIRO 2005).
Using market mechanisms as incentive for pollution control has increased in popularity
over the years because they allow flexibility among the regulated which can decrease overall
policy cost to society. Market mechanisms also act as an organizing device through which
transactions costs to the policy maker or regulating agency are decreased. We have assumed that
the policy maker has at his disposal several mechanisms that range, in order of increasing
flexibility: command and control, tradable credits, fee and rebate and auctions. The choice of
which mechanism to employ depends heavily on the situation and the legal obstacles and
ecological goals of the program.
13
-------
CHAPTER 3 Stream Monitoring Design
3.1 Experimental design
The project uses a before-after control-impact site design (Underwood 1994), where the
"impact" is the installation of parcel-level stormwater management practices. We established six
hydrologic, water quality, and ecological monitoring sites in the watershed (Table 3.1.1), four of
which are receiving streams for the stormwater management practice installation area (Subl,
Sub2, Sub3, and Catch), and two which are control watersheds that will not receive stormwater
management practices (Sub4 and Sub5). Stream sampling sites were monitored for a minimum
of two years prior to installation of stormwater management practices (Summer 2007) and will
continue to be monitored for three years following the last installation (Summer 2008). In
addition to sampling locations along main tributaries of Shepherd Creek, there are 3
neighborhood sites (Nl, N2, and N3) at stormwater outfalls of residential areas and that are
monitored for hydrology. Sampling locations for hydrology, water quality, and/or ecology are
mapped in Figure 3.1.1.
The original study design consisted of five ecological and water quality sampling sites
(Subl, Sub2, Sub3, SubSa, and Catch). The sampling location for the control site moved from
SubSa to Sub5 in 2005 after two years of pre-implementation sampling. The site was re-located
downstream because: 1) SubSa lacked adequate baseflow for proper sampling during dry times
of the year, 2) the hydrologic monitoring station was located at SubS, and 3) the larger drainage
area at Sub5 was more comparable to the other sites. Ecological and water quality data collected
from both SubS and SubSa in 2005 compared well, although stormwater samples were not
collected at SubSa. We also added a second control site, Sub4, in spring/summer of 2005.
Table 3.1.1 Start dates for hydrology, water quality, and ecology monitoring.
Site
Subl
Sub2
Sub3
Sub4
Sub5
SubSa
Catch
Nl
N2
N3
Type
Treatment
Treatment
Treatment
Control
Control (replaced
SubSa)
Control
(decommissioned
Spring 2006)
Treatment
Treatment
Treatment
Treatment
Hydrology
Fall 2004
Fall 2004
Fall 2004
Summer 2005
Fall 2004
N/A
Fall 2004
Spring 2006
Spring 2006
Spring 2006
Water Quality
Baseflow, Spring 2004
Stormflow, Summer 2005
Baseflow, Spring 2004
Stormflow, Summer 2005
Baseflow, Spring 2004
Stormflow, Summer 2005
Baseflow, Spring 2005
Stormflow, Summer 2005
Baseflow, Summer 2005
Stormflow, Summer 2005
Baseflow, Spring 2004
through Summer 2005
Baseflow, Spring 2004
Stormflow, Summer 2005
N/A
N/A
N/A
Ecology
Spring 2003
Spring 2003
Spring 2003
Spring 2005
Summer 2005
Spring 2003
Spring 2003
N/A
N/A
N/A
14
-------
Legend
• Ecology Sites
* Hydrology Sites
• Hydrology Sites (neighborhood) \^
• Water Quality Sites S
Figure 3.1.1 Location of sites for various sampling activities in the Shepherd Creek watershed.
15
-------
3.2 Hydrologic monitoring
Hydrology is a "master" variable that drives dynamics in other aspects of watershed
structure and function (Konrad and Booth 2005), and its characterization is key to understanding
watershed dynamics and any impact that the LID retrofits might have. For this project, we set up
a gaging network to measure flows at stormwater outfalls (neighborhood sites) and within
streams at various locations within the watershed, thus capturing the spatial hydrology of the
watershed.
Stream stage is measured at intervals of 5 minutes or less, so as to better resolve storm peaks
and transient characteristics of stormflow in urbanized watersheds. Flow controls were designed
for each site and implemented as either broad-crest with v-notch weirs in natural stream reaches,
or as semi-elliptical plate weirs for the site with pipe culverts. Flow depth at each gage is
measured with bubbler-type stage measurement devices (Design Analysis H350XL; Logan UT).
Depth is converted to discharge with a site-specific stage-discharge relationship that has been
determined over multiple years with episodic manual measurements made at different water
depths during both baseflow and storm flows. Flows at the outlet of the watershed (Catch) are
realized as mean daily flows only. The cost was prohibitive to establish a proper flow control at
this large a stream cross-section, and so discharge is therefore estimated via an empirical
relationship between flows explicitly measured at Sub2, Sub3, and Sub5, and accounting for the
increase in drainage area and cross-sectional dimensions. We also used peak flow data derived
from crest-stage gages (which passively mark and record flow depth at peak storm discharge)
and converted into discharge values through modeling with USAGE HEC-RAS (River Analysis
System) to refine the empirical model for Catch.
Rainfall is measured at four locations around the watershed: 1) in the northeastern part of
the watershed (operated by Hamilton County), 2) in the centroid of the watershed near Sub3, 3)
in the eastern part of the watershed at Mt. Airy Arboretum, and 4) along the eastern edge of the
watershed (near the police station). Rainfall data is gathered with tipping-bucket-type rain gages
with 0.01" sampling resolution. Tip data is truncated to 5-minute intervals, and all time bases are
coordinated and referenced to Greenwich Mean Time (GMT). In 2008, we also installed an
integrated evapotranspiration measurement system (Campbell ET107 station; Logan UT), which
we use to generate estimates of hourly evapotranspiration data.
3.3 Water quality monitoring
Urbanization causes significant changes in water chemistry that are regulated under the
Clean Water Act, to the extent that these changes have the potential to harm human health and
impair biological condition (Paul and Meyer 2001). Table 3.3.1 lists each of the water quality
parameters sampled for the Shepherd Creek sites, and the potential sources and impairments
associated with each parameter. Water quality and chemistry sampling were conducted monthly
under baseflow conditions, on or near the 15th day of the month throughout the year. Water
quality sampling was performed in conjunction with ecological data collection when feasible,
and followed baseflow protocols defined as no significant rainfall (i.e., >1 mm) in the 72 hours
prior to sampling. The sampling dates and times were subject to modification due to inclement
weather, or other environmental or climatic conditions deemed unsafe.
16
-------
Water quality baseflow samples were collected in clean bottles (I-Chem, HDPE
Environmental Sample bottles) as grab samples or field-filtered during collection using a
peristaltic pump (Geopump Series II, Geotech Environmental Equipment), Teflon sample tubing
and in-line filter holders (47mm; with 0.45 jim Millipore Durapore membranes) for dissolved
constituents. Samples were preserved by acidification (as needed, depending on the type of
analysis) and shipped on ice, overnight, to U.S. EPA's Central Regional Laboratory for most
analyses. Alkalinity was determined using EPA Method 310.1; organic carbon analysis via EPA
Method 415.1 (high temperature combustion); anions were determined by ion chromatography;
nutrients by automated colorimetric methods (EPA Methods 353.2, 350.1, 351.2 and 365.4); and
metals and cations by ICP-AES (EPA Method 200.7 and 200.2). Suspended sediment
concentration was determined in-house following ASTM Method D 3977-97. For each baseflow
sample, in-situ water quality parameters (temperature, dissolved oxygen, etc. as indicated in
Table 3.1.1) were measured using a YSI 6600 Water Quality sonde.
In addition to monthly baseflow sampling, periodic stormflow samples were collected at a
frequency of 6-8 storms per year, with a goal of sampling 2 storm events per season. Stormflow
samples can be used to estimate pollutant loads from nonpoint source runoff and provide water
quality data at high flow conditions. For each storm event, 4-6 samples per site were collected
using automated samplers (Teledyne-ISCO Model 6712; timed or triggered from changes in
depth or turbidity signals from YSI 6600 Water Quality sondes). Stormflow samples were
analyzed using the same methods described above for baseflow samples, except that filtration
was performed in the laboratory. Samples were split for analysis as needed using a Dekaport
Teflon Cone Sample Splitter (Geotech Environmental Equipment).
Table 3.3.1 Summary of water quality monitoring parameters and associated impairments.
Water Quality
Category
In-situ water
quality monitoring
(YSI 6600)
Sediment
Nutrients
Constituents to be Monitored
Temperature, Specific
Conductance, Dissolved Oxygen,
pH, ORP, Turbidity
Suspended Sediment Concentration
(SSC); Turbidity; Particle Size
Distribution
Dissolved Inorganic Nitrogen
(DIN; NO3-N+NO2-N); Ammonia
(NH4-N); Total Kjeldahl Nitrogen
(TKN); Total and Total Dissolved
Phosphorus (IP and TOP)
Potential Sources in
the Watershed
General condition
assessment used to
support specific
parameter analyses
Erosion, both hillslope
and channel; Runoff
from impervious
surfaces
Lawn and garden
fertilizers; Human and
animal waste
Potential Impairments
pH critical for determining
metal speciation;
Conductivity is a surrogate
for chloride; Turbidity is a
surrogate for suspended
sediment
Changes to sediment regime
may degrade the physical
habitat in streams;
Sediment-associated
contaminants may impair
biological condition
Excess nutrients may cause
localized eutrophication;
Export of nutrients to more
sensitive waterbodies may
promote eutrophication
17
-------
Organic Carbon
Ions and
Alkalinity
Metals
Microbiology
Total and Dissolved Organic
Carbon (TOC and DOC)
Anions: Cl', Bf , F, SO42~, NO3",
PO43"; Cations: Na+, Mg2+, K+,
Ca2+; Alkalinity (mostly HCO3")
Total and Dissolved Fe, Mn, Al,
Cu, Zn
Enterococci, Fecal Coliforms,
E.Coli
Runoff from lawns and
gardens and inputs
from riparian
vegetation; In-stream
primary production;
Wastewater
Natural weathering;
Fertilizers; Road salt;
Wastewater
Roadway and rooftop
runoff; Wastewater
Human and animal
waste
Degradation of excess OC
may cause localized hypoxia
or anoxia
Changes in ionic strength
may alter contaminant
speciation; Decreased
alkalinity limits capacity to
buffer inputs of acidic
wastewater; Cl" can be a
chronic or acute stressor
Excess may cause
impairment of biological
condition (chronic/acute
toxicity)
Bacteria may impair human
health and present a contact
hazard
3.4 Ecological monitoring
We designated 61m (200 ft) reaches at each site for sampling periphyton, benthic
macroinvertebrates, and physical habitat. Pre-treatment sampling began in 2003 and sampling
will continue for three years following the installation of stormwater management practices.
Sites were sampled for all parameters five times per year, approximately 6 weeks from April
through October. An additional, quantitative bucket method for sampling benthic
macroinvertebrates occurred 3 times per year associated with the spring, mid-summer, and fall
sampling events. An overview of the types of samples collected is provided in Table 3.4.1, and
sampling methods are described below in more detail.
Table 3.4.1 Summary of samples collected and sampling resolution.
Sample Type
Periphyton (50 mL)
Periphyton (-400
mL)
Periphyton (2 glass
fiber filters)
Data collected
Biomass
Taxonomic ID and
abundance (cell and
natural units)
Chlorophyll a
Data resolution
5 times per year (every
6 weeks Apr. -Oct.)
5 times per year (every
6 weeks Apr. -Oct.)
5 times per year (every
6 weeks Apr. -Oct.)
Output/Indices
dry mass, ash-free dry
mass
richness, diversity,
composition, tolerance,
multimetric indices, etc.
chlorophyll a
concentration
Reference
Barbour et
al. (1999)
Barbour et
al. (1999)
Barbour et
al. (1999)
18
-------
Benthic
macroinvertebrates
(triangular dip net)
Benthic
macroinvertebrates
(bucket sampler)
Physical
characterization
and habitat
assessment data
sheets
Taxonomic ID,
abundance, and
length
Taxonomic ID,
abundance, and
length
Ranking of physical
attributes of sampling
sites
5 times per year (every
6 weeks Apr. -Oct.)
3 times per year (April,
July, October)
5 times per year (every
6 weeks Apr. -Oct.)
richness, diversity,
composition, tolerance,
multimetric indices, etc.
richness, diversity,
density, biomass,
composition, tolerance,
multimetric indices, etc.
Qualitative Habitat
Evaluation Index (QHEI),
separate habitat metrics
Barbour et
al. (1999)
Fritz et al.
(2006)
Barbour et
al. (1999)
Periphyton
Periphyton samples were collected from -12 cobbles selected randomly from pool and
riffle habitats throughout the reach. Cobbles were removed from the stream and a designated
11.4-cm2 ring (PVC circle) on each rock was brushed with a toothbrush for ~2 min. Rocks and
brushes were then rinsed with stream water into a 500-mL bottle. Periphyton from all rocks
within a reach were composited into a single bottle and placed in the dark on ice. In the lab,
samples were split for three analyses: chlorophyll a, algal biomass, and periphyton identification.
First, the total volume of the sample was measured and recorded. The sample was shaken and
homogenized prior to every 10 mL of sample removed. A minimum of 30 mL was filtered onto
each of two glass fiber filters and frozen for analysis of chlorophyll a using a multi-wavelength
spectrophotometer (Arar 1997). Both the biomass sample and the remaining sample for
periphyton identification were preserved with a 1% solution of gluteraldehyde. For biomass
analysis, 50 mL of sample was filtered onto a pre-ashed glass fiber filter (47 mm, PALL Type
A/E, l-|im pore size). Filters were dried for 24 hours at 105°C to a constant weighted and then
ashed in a muffle furnace for 1.5 hours at 500°C to obtain ash-free dry weight. Periphyton cells
(300 ±10%) were identified to genus level and densities were reported as both cells and natural
units per cm2.
Macroinvertebrates
Benthic macroinvertebrates were collected using two methods: qualitative kick samples
and quantitative bucket samples. The kick sample method was adapted from EPA's Rapid
Bioassessment Methods and designed to capture the diversity of macroinvertebrates by sampling
habitat types according to their dominance in the stream reach (Barbour et al. 1999). Qualitative
kick samples were collected by kicking or otherwise disturbing the substrate over the entire reach
and collecting invertebrates and debris in a triangular net (0.5-mm mesh). The quantitative
bucket sampler (Fritz et al. 2006) was designed to sample a fixed area and does not require flow
for adequate performance. Samples were taken from three haphazardly-selected locations within
riffle habitats (rocky pools if riffles are dry). A five-gallon bucket (0.053 m2) with the bottom
removed was driven into the stream bed 3-5 cm deep. Using a trowel, the bed sediment was
disturbed to 10 cm for 10 seconds, and a dip net (0.5-mm mesh) was used to remove the
suspended material. Material was placed into a wash basin and the procedure was repeated two
19
-------
more times. For both the qualitative and quantitative samples, organic material was elutriated in
a bucket to remove inorganic material, sieved, and preserved in 70% ethanol. Bucket samples
were kept separate for identification of macroinvertebrates. Three hundred individuals (± 10%)
per sample were counted, measured, and identified to genus level.
Physical habitat
General morphometric, geomorphic, and water quality parameters were measured within
the 61-m sample reach during ecological sampling. Physical attributes measured included visual
estimates of: % riffle, % pool, % run, average width, average depth, wetted area, surface
velocity, large wood density, % small wood, % large wood, % detritus, bed texture (% bedrock,
cobble, gravel, sand, silt), and % canopy cover. Water quality measurements were taken with a
YSI multi-probe, and included: water temperature, conductivity, dissolved oxygen, pH,
oxidation-reduction potential, and turbidity. Finally, we calculated two visual assessment habitat
evaluation scores: EPA's Rapid Bioassessment Protocols Quantitative Habitat Assessment
(QHEI), and the Primary Headwater Habitat Evaluation Form (HHEI) that is specifically
designed for streams with water depths <40 cm (Barbour et al. 1999). Visual assessments
included ten parameters: substrate, pool depth, bankfull width, riparian zone width, vegetative
protection, bank stability, flow regime, sinuosity, bed stability, embeddedness, and channel
alteration.
An additional geomorphic assessment was conducted once during the pre-treatment phase
(Fall 2004) and will be repeated in 2010 (post-treatment). Cross-section profiles were taken at
several locations along the stream reach using an electronic total station (Trimble 5600 Direct
Reflex, Dayton, OH). The number of cross-section surveys made was dependent upon the
number of breaks between reach areas with contrasting or changing geomorphic properties. The
total station was also used to generate bankfull and water surface longitudinal surveys. Bank soil
was collected at locations where cross sectional surveys had been made. Soil samples were dried
and analyzed in the laboratory for particle size distribution. Finally, a pebble count (Harrelson
et al. 1994) was conducted at 100 random locations along the thalweg to determine particle size
distribution and median particle size (D50) for the reach.
20
-------
CHAPTER 4 Auction Design
4.1 Background
Within the treatment areas of the Shepherd Creek watershed, rain barrels and rain gardens
were distributed to homeowners using an auction mechanism. There are approximately 350
parcels, each of which was offered the opportunity to receive stormwater management practices
free of charge. Apartment buildings (11 within 2 apartment complexes) were also eligible to
receive stormwater management practice free of charge; however, apartment owners were
approached separately and were not included in the auction.
We conducted the procurement auction design which is consistent with the types of
auctions currently being used by federal, state, and local agencies for the purposes of land
conservation. We determined a base amount of money to be used for the auction, and
stormwater management practices were allocated based on that limit. All homeowners within
the pilot area were provided fair and equal opportunity to participate in the auction.
We use a sealed bid, first price, discriminative price auction, where bids are further
tempered by the environmental weighting index, and are accepted up to the cumulative
reservation price of the agency. A non-uniform price auction is employed because of its
theoretical "truth-revelation" properties, which should induce an optimal bidding strategy and
will reflect the actual opportunity cost of stormwater management practice adoption. All eligible
participants received background information (described below) and those wishing to adopt
stormwater management practices submitted bids. The goal was to pay those landowners who
adopt the most effective best management practices at the lowest price. The auction was run in
the spring 2007and repeated in spring 2008, and resulting in actual payouts and installations of
rain gardens and rain barrels. The auction, installations, and maintenance were contracted to
TetraTech, Inc.; however, we assessed the auction bids and performed all monitoring.
4.2 Information to homeowners
Potential participants received two direct mailings detailing the function and uses of the
rain gardens and rain barrels, and the auction process. The first mailing included a cover letter
and informational brochure (Figure 4.2.1), which was intended to notify the landowners in the
watershed of the opportunity to participate in the Shepherd Creek project. The second mailing
was sent out two weeks after the first and contained a cover letter, a copy of the informational
brochure, an auction bid form (Figure 4.2.2), a self-addressed, stamped envelope. All recipients
also received nominal financial compensation ($5) to encourage bidding and compensate
homeowners for their time. In 2007, due to the novelty of the project, we used door hangers to
inform homeowners of the forthcoming mailing. We also extended the auction for 2-3 weeks
and sent an additional letter with another copy of the bid form. Homeowners were directed to
the project website (www.mtairyraincatchers.org) or a contact phone number for additional
information.
21
-------
Srti'infl the B*st Nru-f /or Lort
Tf € EDA" • "
U«S« £1 Hi i .
free rain gatdtm and r.iin barrel!
unl nunbci at nan f udnu
••i J i «EI Wtrh in thr Mi
p-inl hinmtii-Ui miKi bid in
mi tucoon to ir,-n. r ihr in
n Iwrtl hi* ttupbd wiili
Bid Iwmt will W COTIV n
Itu Bill next mcL. fend •
US. EIH1* ceninKW >it* tr.lv hw , md — i-.«nn Hnticutiural
M.nijnuri.1. Inc.. vOl imufl M»i ••utaln itw e«ni™. Mid Uncb
H olfel wfc toe Ml Ar, I
lunmll iguuititv mul »»in .|u>I«> inuJnnt dm the onubiud «H«:u
.ri .lufc-.-idiwI i .m f,.dn» md ram hmirk
wwtMtiidmefi
i . |.r, |U-i«l
, ..i|. .
VJinti n:Hit
LVilt a If nlli«>f iDinthiri inn (id I runcfl h.w ihc fvtngt toot vfll
r*lv fill rh* rvp«] band VChnher .hf nun «•« u mBcotd l« mt
ini. n Jr.n-i«-i
lr»vn yna |m
-------
REVERSE AUCTION FORM - MT. AIRY RAIN CATCHERS
FUNDED BY THE U.S. ENVIRONMENTAL
PROTECTION AGENCY WITH ASSISTANCE FROM
CONTRACTOR TETRA TECH EM, INC.
A few weeks ago, we sent a form to bid on a rain barrel and/or rain garden to Mt. Airy
residents. We have received many bids, but have also heard that some residents had
questions about the process, so we have extended the deadline on bidding.
If you did not respond last time because you are not interested, we understand and
thank you foryour time: please disregard this second mailing. But, if you didn't respond
because you had a question or thought your property wouldn't be eligible (all properties
in the area will be considered and would provide environmental benefits) we are glad to
give you this second opportunity.
To bid, please complete this form and send it back to us in the enclosed envelope by
May 9th. If you have any questions at all, please contact Mike at 241-0149 or
Michael .valerius(5)ttemi. com. There is more information at www.mtairvraincatchers.org.
EPA and Tetra Tech will select winners of the auction based on the amount of the bid
(lower bids are more likely to be accepted) and the potential environmental benefits
based on where you live. If you are selected, you will receive the rain catcher(s) for free
plus a one-time payment equal to your bid amount.
Bid amount for rain garden (amount you will receive if selected):
D $0
D $50
D $100
D $150
D $250
D other amount
Bid amount forrain barrels (amount you will receive if selected):
D $0 D $150
D $50 D $250
D $100 D other amount
Number of barrels requested | |
Name:
Address:
Phone: Email:
Preferred contact method (email, regular mail, phone):_
Please send us your commentson this process, even if you do not bid:
Figure 4.2.2 Auction bid form included in second mailing.
The back of the form had map of the parcel, house, driveway, etc. and asked the
homeowner to indicate their preferred rain garden location.
23
-------
In addition to the information mailed to homeowners, a demonstration rain barrel and two
rain gardens were installed in Fall 2006 at the Mt. Airy Arboretum, a public park area within the
Shepherd Creek watershed. The rain gardens had signage explaining the project and the
stormwater management practices (Figure 4.2.3).
Figure 4.2.3 Pictures of demonstration rain barrel (A), rain garden (B), and signage (C).
24
-------
4.3 Methods for assessing auction bids
To rank auction bids for stormwater management practices, we considered each eligible
parcel's total impervious area, rooftop connectivity to sewer pipes, the predominant soil type,
and distance from a stream based on GIS data. Our scoring procedure weighted bids from a
procurement auction with the above-listed environmental criteria to systematically place rain
gardens and rain barrels in the watershed. Individual auction bids were assessed for their relative
acceptability based on the cost of installation (C), the bid price (B), and the environmental
benefit index (EBI). Bids for rain gardens and rain barrels were evaluated and ranked separately.
For the purposes of ranking bids, C was assumed to be uniform across properties, and was
estimated to be $1500 for each rain garden and $250 for each rain barrel, such that:
(i) Cgarden
(ii) Cbarrei = $250 x # barrels
The overall rank (R) for rain gardens and rain barrels is designed to give weight to the bid
price, in order to appropriately reward lower bids, which is reflected in B/EBI. Equally
important is the actual cost of installing the rain gardens and rain barrels, and this is reflected in
C/EBI. Importantly, the potential environmental benefits are included in both portions. In order
to give equal weight to the two calculations, they are normalized by the maximum value across
all bids, and further scaled from 1-100. The equation for ranking, which is to be calculated
separately for rain gardens and rain barrels, is as follows:
R = (((B-HEBI) - Max(B-HEBI)) + ((OEBI) - Max(OEBI))) x 50 (iii)
Environmental valuation provides a basis for assessing the potential environmental
benefits resulting from stormwater management practice installation. The variable set was
developed based on the potential beneficial environmental impact on the receiving stream. In
addition, all variables can be evaluated using Geographic Information System (GIS) software
and do not require individual site visits, which would constitute a major undertaking. The
environmental values were determined from sum of likewise weighted and property-specific
characteristics, which are detailed below for rain gardens and rain barrels.
Beyond these objective criteria, there was also potential for bid refusal based on the
location of the stormwater management practice on the property. Because the goal of the project
is to mitigate stormwater runoff within the Shepherd Creek watershed, any barrel or garden that
would be located such that it benefits areas outside of Shepherd Creek was not accepted. This
was the case for the few bids from households on the watershed's border. For the rain barrels,
this can largely be determined on GIS (i.e., if the rooftop is outside of the watershed, the bid will
be refused). However, if there are roof downspouts inside and outside the watershed, then bids
were only be accepted if the homeowner agreed to put the barrel on a downspout inside the
watershed (determined during field visits). For the rain gardens, the homeowners indicated their
preferred location on the bid form; however, the precise location was based on field evaluations
and discussions with the homeowner. If the field crew determined that the garden would have no
25
-------
benefit or minimal benefit for the watershed, for example if the house's back yard sloped into the
adjacent watershed, the bid was refused.
Environmental Benefits Index (EBI)
The scoring process was designed to be a simple, informative, and repeatable technique
for quantifying the potential environmental value of placing stormwater management practices
on the property. This allows for objective comparison of landowner bids based on the potential
environmental benefits and utilizes available GIS information. Variables were scored for each
property, with high numbers indicating a preferred condition. The rationale and scoring for each
variable is detailed below.
Rain Gardens
The environmental value of rain gardens was based on the amount of stormwater runoff
potentially infiltrated and the proximity of property to a stream channel. Potential infiltration of
runoff was determined using percent total impervious area (TIA) on the parcel and soil drainage
characteristics. Environmental value for rain garden installation will be maximized where: there
is high % TIA on the parcel, soils have comparatively low capacity for drainage, and the
property is in close proximity to stream channels. Scoring criteria for each variable are detailed
in the next section. The following formula was used to create a linear combination of variable
scores for the environmental weighting. TIA is expected to be twice as important for influencing
runoff quantity (and hence, environmental valuation) compared to the other two variables, and is
likewise reflected in the coefficient.
EBIgarden = 2(TIA score) + (Soil score) + (Proximity score)
(iv)
Calculation of scores:
1) Percent total impervious area
Total impervious area (TIA) is the area of land that is covered by rooftops, driveways,
sidewalks, pools, or other surfaces that do not allow for water to infiltrate into the ground.
Parcels with a high percent of total area as TIA received higher scores because these areas
presently have the least opportunity for infiltration, and would therefore benefit the most from
rain gardens. Parcel-level TIA was available as GIS layers of rooftops, driveways, and
sidewalks which were digitized from 2001 aerial photography (specifically, ortho-rectified
images), and updated with recent site-specific surveys (ca. 50% of the properties).
Variable Description: TIA
>30% parcel area as TIA
15-30% parcel area as TIA
5-15% parcel area as TIA
<5% parcel area as TIA
Score
4
3
2
1
26
-------
2) Soil drainage characteristics
Soil series differ in terms of capacity to infiltrate precipitation and surface runoff.
Therefore, the dominant soil series on the parcel was used to estimate the existing capacity for
drainage for a given parcel. For parcels in this watershed the dominant soil types are
Switzerland silt loam, which has moderate drainage characteristics; and Eden silty clay loam,
which has relatively poor drainage characteristics. Where the soils have been modified due to
urban disturbance (UELC = urban Eden land complex; USLC = urban Switzerland land
complex), they are assumed to have slightly less infiltrative capacity compared to the
undisturbed soils, and are scored accordingly. Parcels where the dominant soils have poorer
drainage characteristics were given preference for receiving a rain garden because these
properties are likely to generate a greater amount of surface runoff, and would likely benefit the
most from having a rain garden. A detailed soil survey map created by the Natural Resource
Conservation Service was used to determine the dominant soil series for a given parcel. Scoring
was as follows:
Variable Description: Soils
Dominant soil type has low infiltration capacity (UELC)
Dominant soil type has med-low infiltration capacity (Eden)
Dominant soil type has med-high infiltration capacity (USLC)
Dominant soil type has high infiltration capacity (Switzerland)
Score
4
O
2
1
3) Proximity to stream channel
The proximity of a parcel to a stream channel is inversely proportional to the area of land
available to act as a buffer for upstream developed land. The closer the parcel is to the stream
channel, the fewer opportunities for infiltration of runoff, and therefore an increased potential for
runoff contributing to peak flows in the stream channel. Properties that are closer to the stream
would be expected to benefit the most from a rain garden acting to intercept and infiltrate surface
runoff. The stream network determined from 2 ft. Digital elevation maps (OEMs) was used to
calculate the distance between the centroid of the parcel and the stream channel following the
flow path. Scoring proximity was as follows:
Variable Description: Proximity
Centroid of parcel <50 m from stream channel
Centroid of parcel 50-100 m from stream channel
Centroid of parcel 100-150 m from stream channel
Centroid of parcel >150 m stream channel
Score
4
3
2
1
Rain Barrels
The environmental value of rain barrels was based on the potential amount of water that
would otherwise be lost to direct connection or conveyance to storm sewers. Environmental
value will be maximized where 1) higher numbers of barrels were requested, and 2) the roof
gutter downspouts were wholly or in-part connected to storm sewers. The environmental
weighting of rain barrels are therefore calculated accordingly:
= # barrels x Rooftop connectivity score
(v)
27
-------
The proportion of rooftop area that is directly connected to storm sewer pipes reflects the
potential for precipitation to be stored in rain barrels rather than contributing directly to peak
flows in streams. Thus, properties with higher connectivity of rooftops to storm sewers received
higher scores, because runoff from these areas are currently routed directly into streams and
could benefit the most from storage in rain barrels. Rooftop connectivity was determined using a
combination of local storm sewer pipe information, a rooftop data layer digitized from 2001
ortho-rectified aerial photography, and site-specific surveys of roof gutter downspouts (ca. 50%
of properties surveyed and 50% estimated from neighboring lots). Rooftop connectivity was
scored as follows:
Variable Description: Rooftop connectivity
75-100% connectivity of rooftops to storm sewers
50-75% connectivity of rooftops to storm sewers
25-50% connectivity of rooftops to storm sewers
0-25% connectivity of rooftops to storm sewers
Score
4
3
2
1
28
-------
CHAPTER 5 Auction Results
5.1 Phase 1, Year 2007
Of the -350 eligible homeowners, we received 57 bids for rain gardens and 61 bids for a
total of 121 rain barrels (Table 5.1.1). Most homeowners bid for both rain gardens and rain
barrels (47), although there were some bids for just gardens (10) and some for just barrels (16;
Figure 5.1.1). Bids ranged from $0 to $500, with a mean bid of $50.27 for a rain garden and
$32.06 for a rain barrel. Interestingly, a majority of bids were $0, indicating the willingness of
homeowners to receive stormwater management practices for free without any additional
compensation (Table 5.1.1).
Table 5.1.1 Summary of bids for rain gardens and rain barrels in 2007 and 2008
Rain gardens
Number of bids
Minimum bid
Maximum bid
Mean bid
Mean bid excluding max
Number of $0 bids
Percent $0 bids
Rain barrels
Number of bids
Average number of barrels per bid
Minimum bid (per barrel)
Maximum bid (per barrel)
Mean bid (per barrel)
Mean bid excluding max (per barrel)
Number of $0 bids
Percent $0 bids
2007
57
$0.00
$500.00
$58.16
$50.27
30
52.6
63
1.9
$0.00
$500.00
$32.06
$24.51
38
60.3
2008
37
$0.00
$1,000.00
$88.54
$63.22
16
43.2
45
1.7
$0.00
$250.00
$44.30
$34.74
20
44.4
Total
94
$0.00
$1,000.00
$70. 12
$60.12
46
48.9
106
1.8
$0.00
$500.00
$36.44
$32.03
58
54.7
% Change
-21.3
0.0
50.0
43.3
21.5
-30.4
-19.2
-17.0
-11.5
0.0
-50.0
33.6
31.9
-31.0
-29.0
Because the sum of all bids did not exceed the amount of money allocated for subsidies
and installations, we accepted nearly all of the bids. Two barrel bids were refused because the
roofs were located outside the Shepherd Creek watershed, and one barrel bid was refused due to
the high bid amount ($500 for 1 barrel) relative to other bidders and the cost of installation. All
but one rain garden bid were initially accepted, again excluding the extremely high bid ($500)
relative to other bids. Bidders were relatively evenly distributed throughout the treatment area
(Figure 5.1.1).
29
-------
Table 5.1.2 Rain barrel rankings, 2007
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Barrel
Bid ($)
0
100
0
0
0
0
0
50
0
0
0
0
0
0
0
0
0
0
0
0
0
50
100
100
100
100
150
0
250
500
0
0
# Weighted
Barrels
4
4
3
4
3
3
4
3
3
2
2
2
2
4
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
Score
3
4
4
4
4
4
4
5
6
6
6
6
6
6
6
6
6
6
6
6
6
7
8
8
8
8
8
8
9
13
13
13
Number of barrel bids accepted
Total
Total
number of barrels
cost (bid
amount)
Accept?
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
60
118
$2,532
Rank
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
n/a
n/a
n/a
Barrel
Bid ($)
0
0
0
0
0
0
0
5
27
50
50
50
100
100
0
0
0
50
250
100
0
0
0
0
100
150
0
0
500
100
50
# Weighted
Barrels
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3
1
1
4
3
2
2
2
2
2
2
1
1
1
1
1
Score
13
13
13
13
13
13
13
13
13
14
14
14
15
15
17
17
17
18
19
20
25
25
25
25
30
33
50
50
100
n/a
n/a
Accept?
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
no
no
no
30
-------
Table 5.1.3 Rain garden rankings, 2007
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Garden
Bid ($)
0
0
0
5
0
25
0
0
0
0
0
50
30
0
0
0
0
0
50
100
0
0
0
0
50
50
0
0
0
Weighted
Score Accept?
25
25
25
25
27
28
29
29
29
29
29
30
31
31
31
31
31
31
32
33
33
33
33
33
35
35
36
36
36
Number of garden bids
Total
cost (bid
amount)
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
accepted
Rank
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
n/a
56
$2,815
Garden Weighted
Bid ($)
0
0
0
5
50
100
100
0
0
0
0
150
50
100
250
250
250
250
100
100
50
50
0
0
100
250
250
500
Score
36
36
36
37
38
39
39
40
40
40
40
42
45
46
46
46
46
46
50
50
50
50
50
50
56
59
59
90
Accept?
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
no
31
-------
Shepherd Cieek
Wateished AiejfM
ML Ally
Rain Catcheis Pi ojett,
EPA STREAMS
Task Order 8
TTEMI P* G9020.000S
May 2007
Figure 5.1.1 Location of rain barrel and rain garden bids in Phase 1, 2007
32
-------
5.2 Phase 2, Year 2008
When the auction was repeated in spring 2008, there were 37 bids for rain gardens
averaging $88.54, 43% higher than in 2007 (Table 5.1.1). The high bid was $1000 and the
percent $0 bids dropped to 43.2%. For barrels, there were 45 bids for 76 barrels, averaging
$44.30 per barrel, a 33% increase from 2007 bids. Thirteen homeowners who bid in 2007 also
bid in 2008, reflecting their pleasure with the stormwater management practices received the
previous year and/or their continued interest in this project. Two garden bids and one barrel bid
were refused due to their high weighted score relative to other bidders, reflecting the high bid
price and/or low environmental benefit of the stormwater management practice on that property
(Table 5.2.1, 5.2.2). Again, bidders were located throughout the treatment portion of the
watershed, although there were clusters of bidders that may reflect influences from neighbors
(Figure 5.2.1).
Table 5.2.1 Rain garden rankings, 2008
Garden Weighted
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Bid ($) Score Accept?
0 19
0 20
0 20
0 20
50 20
0 21
50 22
25 22
0 23
0 23
0 23
0 23
100 23
150 23
50 25
0 25
0 25
250 27
0 27
Number of garden bids
Total
cost (bid amount)
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
accepted
Rank
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
n/a
n/a
35
$2,026
Garden
Bid ($)
100
0
0
0
0
250
150
100
50
250
100
150
1
50
50
100
250
1000
Weighted
Score
29
30
30
30
30
30
31
32
33
33
35
38
38
41
41
44
71
80
Accept?
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
no
no
33
-------
Table 5.2.2 Rain barrel rankings, 2008
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Barrel
Bid ($)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
50
50
50
50
25
50
50
50
50
# Weighted
Barrels
4
4
3
3
2
2
2
2
2
2
1
1
1
1
3
2
2
2
1
1
1
1
1
Score
13
13
13
13
13
13
13
13
13
13
13
13
13
13
14
15
15
15
15
17
17
17
17
Number of barrel bids accepted
Total
Total
number of barrels
cost (bid
amount)
Accept?
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
44
74
$2,152
Rank
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
n/a
Barrel
Bid ($)
50
0
0
150
50
25
100
150
150
150
0
1
200
250
250
0
0
0
1
100
100
300
# Weighted
Barrels
1
1
1
2
2
1
1
1
1
1
2
1
1
1
1
4
2
1
1
2
2
2
Score
17
17
17
19
19
20
21
25
25
25
25
25
29
33
33
50
50
50
51
67
67
100
Accept?
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
no
34
-------
[HI00-GMen rES-BmS
• VES • GaMon. TO - Bare!
•• VES - Gard*n and Barrel
&d Parcels^Prase II (label!
e
Figure 5.2.1 Location of rain barrel and rain garden bids in Phase 2, 2008
35
-------
CHAPTER 6 Stormwater Management Practice Installations
6.1 Overview
We installed rain barrels and rain gardens on those properties identified through the
auction and granted subsequent permission by the homeowner. Soon after the bidding process
we contacted the homeowners by mail and informed them they had been selected to participate,
and scheduled a site visit. Upon visiting the site for the first time the homeowner was asked to
sign an access agreement that gave consent to the "Mt. Airy Rain Catchers Project Team" to
access their yard to install and maintain the rain barrels and rain gardens. The agreement
specified the term of three years for the maintenance period. Both the property owner and a
representative of the Project Team signed the agreement. Also in the first site visit the
homeowner was handed a check in the amount of his or her bid. In 2007, we installed 50
gardens and 100 barrels. The lower number of installations relative to accepted bids was due to
not finding an acceptable location for the stormwater management practice on the property (e.g.,
preferred garden location above utility or lacked hydrologic benefit) and homeowners changing
their mind upon consultation (e.g., did not want their gutter downspouts cut). In 2008, we
installed 25 gardens and 50 barrels. An explanation of the installation guidelines and
maintenance protocol for rain gardens and rain barrels follows.
6.2 Rain gardens
Installation protocol
We worked with individual homeowners to determine location and design (i.e., shape) of
the rain garden stormwater management practice. Rain gardens were approximately 150 sq. ft,
although the final shape depended on the landscape and property features of each individual
parcel. Using our index of ecological effectiveness we worked with individual homeowners to
determine an optimal location and design for the rain garden, to maximize hydrological benefit
while being suitable for the homeowner. If the homeowner indicated a location on the auction
bid form, we flagged the location of the garden on the lawn and work with the homeowner to
determine shape and design. If the homeowner indicated that they were flexible, we worked
with the homeowner to determine the best location. We maintained the qualification that if we
and the homeowner could not reach agreement about placement of the rain garden, either party
could terminate the process of installing the rain garden.
Rain gardens were installed between June and September. First, the sod was removed
and a Dingo excavator was used to excavate a trench for installing a perforated underdrain pipe
at the bottom of the trench. Most rain gardens were fitted with underdrains and outlets to allow
for free-drainage of water to prevent the incidence of standing water for any extended period of
time (Appendix A, Garden A). If landscape slope was insufficient to place an underdrain, the
rain garden was instead made slightly larger (160 ± 10% sq. ft.) and 6" deeper (Appendix A,
Garden B). In 2007, a majority of the rain gardens (34 of 50) were fitted with underdrains. Rain
36
-------
gardens were developed from native soils, but amended with coarse peat moss and sand and
supplemented with organic fertilizer. In 2008, we amended the soils with peat moss and native
compost (aged 3 years), since we felt that the proportion of sand was so low that they were
unlikely to increase infiltration. Exposed soil on the berm was seeded and covered in straw held
in place with netting.
Due to drought conditions that persisted into September 2007, the gardens were planted
in the fall that year so as to maximize plant survival. The planting area was covered with mulch
to retain moisture and reduce weeds. Plants were primarily native species and cultivars that are
adapted to periodic wet conditions expected in the rain gardens (Figure 6.1.1). Homeowners
were given some choice in plants, and were asked to water the gardens periodically in the 1-2
weeks following planting. In spring 2008 we conducted a plant inventory to determine plant
success. In the second year, we used larger plant starts (8-10" pots) and planted species that
were most successful in 2008. Additional species that were expected to be more resistant to deer
browsing were also used. Figure 6.2.1 overviews the numbers of species planted while figure
6.2.2 presents a typical raingarden in the study area.
Number of Each Species Planted in Summer 2007
40 80 120 160 200
240
Great Blue Lobelia
Cardinal Flower
Fox Sedge
Side Flowering Aster
Mountain Mint
Smooth Aster
Pink Turtle Head
Monkey Flower
Smooth Beardtongue
Crested Sedge
Blue Flag Iris
Lurid Sedge
Lance-Fruited Sedge
Marsh Milkweed
W Sweet Black-eyed Susan
Bergamont
Boneset
Riddle's Goldenrod
Showy Black-eyed Susan
Foxglove Beard Tongue
Monarda
Spreading Oval Sedge
Ohio Golden Rod
Golden Alexander
Franks Sedge
Heath Aster
i/)
o>
'o
a>
a.
221
H 209
H 209
172
=1 155
=l 153
147
147
=1 139
H 138
H 137
=1 131
=1 129
=1 128
Zl 127
121
D 106
97
91
55
H 49
45
H 40
=1 33
H 30
Figure 6.2.1 Species in plant communities planted in rain gardens, 2007.
37
-------
Figure 6.2.2 Typical rain garden installation
Maintenance protocol
We took on responsibility for maintenance of rain gardens over the research project
period of three years from the date of construction. Regular maintenance was implemented into
the project plan so as to sustain good performance and improve consistency in rain garden
condition over the course of the research project. This includes assuring that the rain gardens are
draining properly and plants remain in good conditions for the duration of the pilot study.
Maintenance involves mulching and re-planting (as necessary) in the spring, weeding in the
summer, and mulching and pruning in the fall. Homeowners also received a manual that
explained our maintenance plan and asked homeowners not to add plants, fertilize, or otherwise
disturb the garden.
6.3 Rain barrels
Protocol for installation
The location of rain barrel(s) was determined by the amount of rooftop draining into a
gutter downspout, the current drainage location of the gutter, and preferences of the homeowner.
Ideally, rain barrels were located where a large portion of the roof runoff will be collected, and
38
-------
where runoff currently drains to stormwater conveyances. However, rain barrels connected to
any downspout were deemed to be acceptable.
We used green, 75-gallon rain barrels which were fitted with a spigot and hose located
near the bottom of the barrel that allowed the barrels to be emptied in accordance with
procedures set out in the owners manual. The barrels also have an internal overflow tube that is
routed to another external opening at the bottom of the barrel. A screened cap was placed on top
of the barrel to prevent debris from entering the barrel and prevent insects from breeding in the
barrel. Roof gutter downspouts were cut and fitted with a flexible pipe to route the water to the
barrel. Barrels were placed on cinder blocks and overflow tubes were routed into storm drains or
landscape areas away from the house. Up to four barrels could be linked together on a single
downspout to allow for increased detention capacity. Although the standard-issue rain barrel in
the Mt. Airy project was 75 gallons, 23% of barrels issued to homeowners in 2007 were 55
gallons due to shortfall in supplier inventory. See Appendix B for schematic representations of
the rain barrels and their typical setting for installation.
Maintenance protocol
As a part of the licensing agreement, the homeowner agreed to take responsibility for
maintenance of the rain barrels (e.g., emptying the barrel after a rainfall event). Homeowners
were instructed on how to empty their barrels and given a manual that explained recommended
maintenance. If a homeowner did not plan to use the water in their barrel, we recommended that
the overflow and spigot hoses empty into the rain garden (if applicable) or other landscape
feature that is designed to infiltrate water (e.g., a swale). Homeowners were advised to empty
their barrels in the winter to prevent freezing and cracking of the barrels.
39
-------
CHAPTER 7 Rain Garden and Rain Barrel Monitoring
7.1 Overview
In cooperation with individual homeowners, we identified a subset of rain gardens to
monitor for a period of three years from the date of construction. This monitoring will be
conducted for a period of three years, and includes hydrology, soil, and water quality
measurements. Five rain gardens are being monitored from the 2007 installations and an
additional 5 gardens will be selected in 2008 to represent a range in soil types, landscape settings
and drainage types (garden types A and B). Rain barrels are also monitored for water level and
water quality.
7.2 Hydrology and soils
At each location, we are monitoring rain gardens for infiltration and water redistribution.
We are measuring soil water content by averaging measurements taken every 30 seconds and
recorded as a datum point every 10 minutes and at 10-cm intervals to 50-cm (Sentek EasyAG;
Campbell Scientific; Logan UT), and this data is stored using an automatic logging system
(Campbell CR800). Underdrains are evaluated qualitatively for evidence of flow (trickle flow
after storms, build up of algae at outlet, erosion around outlet). Rain barrels are monitored
continuously for water level using level loggers (In-Situ; Fort Collins CO) in order to determine
patterns in the times and rates of water use by residents and number of overflow events. The
level loggers are removed each winter (November-March) as rain barrels will be left open and
freezing would damage the level logger pressure sensors.
7.3 Water quality
Monitoring for water quality will consist of episodic, storm-event based sampling of
surface runoff to rain gardens, soil water sampling using suction lysimeters installed at a depth of
12 to 18 inches in rain gardens, and underdrain flow sampling (where applicable; one rain garden
site does not have an underdrain). Stainless steel suction lysimeters (SW-071 single chamber;
Soil Measurement Systems, LLC) were installed in finished rain gardens during the months of
April and May 2008 according to ASTM standard specifications (ASTM International 1992).
Water quality sampling will also include periodic sampling of water collected in rain barrels at
the chosen site locations.
For rain garden and rain barrel performance monitoring, the results from sampling will be
aggregated for data analysis purposes in order to determine: 1) the chemical composition of roof
runoff as captured in rain barrels; 2) the chemical composition of lawn runoff to rain gardens on
a seasonal basis; and 3) the treatment provided by rain gardens with respect to changes in
nutrient concentrations (nitrogen and phosphorus), organic carbon, alkalinity, ions and metals.
Lawn runoff samples also will be analyzed for suspended sediment concentration. Treatment
will be determined by comparing soil water (lysimeter) and underdrain concentrations directly
with lawn runoff and rain barrel samples using paired sample analysis of variance methods
40
-------
(Dietz and Clausen 2006, Davis 2007). Sampling frequency will consist of 2 storm events per
season for a total of 8 events per year. The minimum acceptable number is 6 storm events, given
the element of chance and practical problems associated with storm sampling. For purposes of
replication, the goal will be to collect 3 samples of each type (lawn runoff, soil water from
lysimeters, and underdrain discharge) at each site for each storm event. This yields a minimum
of 18 samples of each type per site each year, where applicable. Only a single sample will be
collected from the rain barrel at the end of the storm event, for a total of 6 to 8 samples per year.
In the event that the replicate samples for a given storm exhibit temporal trends, they will be
treated as independent samples.
41
-------
CHAPTER 8 Conclusions
This study is a part of an ongoing effort to fully describe a realistic market-based
mechanism to alleviate the water quality and ecological problems caused by the typically large
volumes of excess stormwater runoff in urban and urbanizing areas. As impervious surface
grows relative to natural landscapes the problem will increase, and municipal authorities will
look for ways to deal with it that are both practical from a political standpoint and ecologically
effective. We hypothesized that deployment of stormwater management practices in a watershed
via market incentives would reduce stormwater runoff substantially, thus improving stream water
quality and biotic integrity. To test this hypothesis, we examined four market mechanisms:
command and control, cap and trade, fee and rebate, and auction. We modeled each policy using
realistic cost functions, including the estimated opportunity cost of residential land dedicated to
particular stormwater control practices, and we employed sound hydrologic models. In most
cases we showed the incentive schemes to be economically efficient. When considering actual
application of the market mechanism to a small Midwestern watershed, there were several legal
and regulatory obstacles to the imposition of a cap on stormwater runoff (for a tradable credit
mechanism) and the appropriately high stormwater fee (to allow for a fee and rebate policy).
Thus, we opted for a wholly voluntary approach using a reverse auction.
We are now testing our hypothesis that deployment of enough distributed, low-tech
stormwater management practices in a watershed will reduce stormwater runoff sufficiently to
effect positive hydrologic, water quality, and ecological change. We are doing this with actual
installation of stormwater management practices in the Shepherd Creek watershed. We
employed a before-after-control-impact experimental design, wherein we monitor streams for
three years before installing the stormwater management practices in the treatment watersheds,
and then continue monitoring for an additional three years. We employed a reverse auction to
determine where to install the management practices in a cost effective way. The reverse, or
procurement, auction is often used when there are many potential sellers of an item (in this case,
homeowners selling the limited use of their property) and a single buyer (in this case, a
government agency trying to buy as much stormwater retention capacity as economically
feasible). To try to improve the subscription rate among homeowners and to test our hypotheses
about iterative auctions, we conducted the auction and installation two years in a row. In 2007,
we installed 50 rain gardens and 100 rain barrels at 68 properties for a total bid payout of $5,347.
In 2008, we received acceptable bids for 35 rain gardens and 74 rain barrels on 49 properties,
including 12 properties that received some stormwater management practice(s) in 2007. After
two years of running the auction, we will have installed stormwater management practices on
about 30% of the eligible properties in the treatment watershed. Future research includes the
ongoing monitoring effort wherein we will quantify the hydrologic and ecological effectiveness
of the installed storm water management practices. Overall, this methodology will provide
defensible results about the efficiency, both economic and ecologic, of deploying rain gardens
and rain barrels in a suburban watershed in order to reduce the effects of stormwater runoff.
42
-------
Publications related to this project
Morrison M.A., W.D. Shuster and R.Webb. 2008. A streamlined monitoring framework for
sustainable and low-impact development stormwater management practices. Proceedings of the
11th International Conference on Urban Drainage, August 31-September 5, 2008, Edinburgh,
Scotland, UK.
Parikh, P., M. Taylor, T. Hoagland, H. Thurston, and W. Shuster. 2005. At the intersection of
hydrology, economics, and law: application of market mechanisms and incentives to reduce
stormwater runoff. Environmental Science and Policy 8:133-144.
Prahalad, P. P., M. P. Clagett, and N. T. Hoagland. 2007. Beyond water quality: can the Clean
Water Act be used to reduce the quantity of stormwater runoff? The Urban Lawyer 39:85-109.
Roy, A.H. and W.D. Shuster. In press. Assessing impervious surface connectivity and
applications for watershed management. Journal of the American Water Resources Association.
Roy, A.H., H. Cabezas, M.P. Clagett, N.T. Hoagland, A.L. Mayer, M.A. Morrison, W.D.
Shuster, JJ. Templeton, and H.W. Thurston, 2006. Retrofit stormwater management: navigating
multidisciplinary hurdles at the watershed scale. Stormwater 7:16-29.
Roy, A.H., A.L. Mayer, W.D. Shuster, H.W. Thurston, N.T. Hoagland, M.P. Clagett, P.K.
Parikh, and M.A. Taylor. 2005. A multidisciplinary approach to stormwater management at the
watershed scale. Proceedings of the 10th International Conference on Urban Drainage, August
21-26, 2005, Copenhagen, Denmark.
Roy, A.H., SJ. Wenger, T.D. Fletcher, CJ. Walsh, A.R. Ladson, W.D. Shuster, H.W. Thurston,
and R.R. Brown. 2008. Impediments and solutions to sustainable, watershed-scale urban
stormwater management: lessons from Australia and the United States. Environmental
Management 42:344-359.
Morrison M.A., W.D. Shuster and R.Webb. Inpress. Front-loading urban stormwater
management for success - a perspective incorporating studies of retrofit low-impact
development. Cities and the Environment.
Shuster, W.D., J. Bonta, H. Thurston, E. Warnemuende, and D.R. Smith. 2005. Impacts of
impervious surface on watershed hydrology: A review. Urban Water Journal 2:263-275.
Shuster, W.D., R. Gehring, and J. Gerkin. 2007. Prospects for enhanced groundwater recharge
via infiltration of urban storm water runoff: A case study. Journal of Soil Water and
Conservation 62(3 ): 129-13 7.
43
-------
Shuster, W.D., A.H. Roy, H.W. Thurston, M. Morrison, M. Taylor, and M. Clagett. 2008.
Implementation of retrofit best management practices in a suburban watershed (Cincinnati, OH)
via economic incentives. Proceedings of the 11th International Conference on Urban Drainage,
August 31-September 5, 2008, Edinburgh, Scotland, UK.
Shuster, W.D., Thurston, H.W. and Zhang, Y. 2006. Simulated rain Garden Effectiveness and
performance in Response to synthetic and Natural Rainfall Patterns. Proceedings of the 7th
International Conference on Urban Drainage Modelling and the 4th International Conference on
Water Sensitive Urban Design, April 2-7 2006, Melbourne, Australia.
Thurston, H.W. 2006. Opportunity costs of residential best management practices for
stormwater runoff control. Journal of Water Resource Planning and Management 132:89-96.
Thurston, H.W., H.C. Goddard, D. Szlag, and B. Lemberg. 2002. Trading stormwater abatement
credits in Cincinnati's Shepherd Creek: A proposal to spread responsibility for stormwater
control across a watershed cost-effectively and equitably. Stormwater 3(5):50-59.
Thurston, H.W., H.C. Goddard, D. Szlag, and B. Lemberg. 2003. Controlling stormwater runoff
with tradable allowances for impervious surfaces. Journal of Water Resources Planning and
Management 129:409-418.
Thurston, H.W., W.D. Shuster, M.A. Taylor and S. Stewart. 2006. Evaluation of Economic
Incentives for Decentralized Stormwater Runoff Management: The Shepherd Creek Watershed
Pilot Project. Proceedings of the 7thInternational Conference on Urban Drainage Modelling and
the 4th International Conference on Water Sensitive Urban Design, April 2-7 2006, Melbourne,
Australia.
Thurston, H.W., M.A. Taylor, A.H. Roy, M. Morrison, W.D. Shuster, J. Templeton, M. Clagett,
and H. Cabezas. 2008. Appling a reverse auction to reduce stormwater runoff. Ambio 37(4):326-
327.
44
-------
References
Arar, EJ. 1997. Method 446.0, In vitro determination of chlorophylls a, b, cl, +c2 and
pheopigments in marine and freshwater algae by visible spectrophotometry. Revision 1.2. U.S.
Environmental Protection Agency, National Exposure Research Laboratory, Office of Research
and Development.
Arnold, C.L. and CJ. Gibbons. 1996. Impervious surface coverage: the emergence of a key
environmental indicator. Journal of the American Planning Association 62:2443-258.
ASTM International. 1992. Standard Guide for Pore-Liquid Sampling from the Vadose Zone. D
4696 - 92 (Reapproved 2000) ASTM International, West Conshohocken, PA.
Barbour, M.T., J. Gerritsen, B.D. Snyder, and J.B. Stribling. 1999. Rapid bioassessment
protocols for use in streams and wadeable rivers: periphyton, benthic macroinvertebrates and
fish. Second edition. EPA 841-B-99-002. U.S. Environmental Protection Agency, Office of
Water, Washington, DC.
Baumol, W. J. and Gates, W. E. 1988. The Theory of Environmental Policy, Cambridge
University Press, Cambridge.
Bledsoe, B.P. and C.C. Watson. 2001. Effects of urbanization on channel instability. Journal of
the American Water Resources Association 37:255-270.
Booth, D.B. and C.R. Jackson. 1997. Urbanization of aquatic systems: degradation thresholds,
stormwater detention, and the limits of mitigation. Journal of the American Water Resources
Association (JAWRA) 33:1077-1090.
Brabec, E., S. Schulte, and P.L. Richards. 2002. Impervious surfaces and water quality: a review
of current literature and its implications for watershed planning. Journal of Planning Literature
16:499-514.
Davis, A.P. 2007. Field performance of bioretention: water quality. Environmental Engineering
Science 24: 1048-1064.
Dietz, M.E. and J.C. Clausen. 2006. Saturation to improve pollutant retention in a rain garden.
Environmental Science & Technology 40:1335-1340.
Coleman, J.C., II 2000. Cost Minimization Analysis as a Foundation for a Marketable Credit
Trading System for Impervious Areas in Mill Creek Watershed. Spring Research Seminar Series,
Univ.of Cincinnati, Cincinnati, OH.
45
-------
CSIRO. 2005. Catchment Care - Developing an Auction Process for Biodiversity and
Water Quality Gains
http://www.clw.csiro.au/publications/consultancy/2005/catchment_care_auction_ vol 1 .pdf
(accessed 09/09/2008).
Doll, A., P.P. Scodari, and G. Lindsey. 1998. Credits as Economic Incentives for On-Site
Stormwater Management: Issues and Examples. Proc., U.S. Environmental Protection Agency
National Conference on Retrofit Opportunities for Water Resource Protection in Urban
Environments, USEPA, Chicago, IL, 113-117.
Doll, A. and G. Lindsey. 1999. Credits Bring Economic Incentives for Onsite Stormwater
Management. Watershed and Wet Weather Technical Bulletin 4(1), 12-15.
Eheart, J. W. 1980. Cost-Efficiency of Transferable Discharge Permits for the Control of BOD
Discharges. Water Resources Research, 16(6), 980-986.
Fritz, K.M., B.R. Johnson, and D.M. Walters. 2006. Field operations manual for assessing the
hydrologic permanence and ecological condition of headwater streams. EPA/600/R-06/126. U.S.
Environmental Protection Agency, Office of Research and Development, Washington, DC.
Godwin, K.S., S.D. Hafner, M.F. Buff. 2003. Long-term trends in sodium and chloride in the
Mohawk River, New York: the effect of fifty years of road-salt application. Environmental
Pollution 124:273-281
Fullerton, D. and A. Wolverton. 1999. The Case for a Two-Part Instrument: Presumptive Tax
and Environmental Subsidy. Environmental and Public Economics: Essays in Honor of Wallace
E. Gates. A. Panagariya, P. Portney, and R. Schwab, eds., Edward Elgar, Cheltenham, UK.
Fullerton, D. and A. Wolverton. 2003. The Two-Part Instrument in a Second-Best World,"
Working paper Series, NCEE, USEPA.
Harrelson, C.C., C.L. Rawlins, and J.P. Potyondy. 1994. Stream channel reference sites: an
illustrated guide to field technique. United States Department of Agriculture, Forest Service.
General Technical Report RM-245
Hatt, B.E., T.D. Fletcher, CJ. Walsh, and S.L. Taylor, 2004. The influence of urban density and
drainage infrastructure on the concentrations and loads of pollutants in small streams.
Environmental Management 34:112-124.
Heaney, J.P., D. Sample and L.Wright. 2002. Costs of Urban StormwaterControl. EPA Report
No. EPA-600/R-02/021; NTIS No. PB2003103299. National Risk Management Research
Laboratory, U.S. Environmental Protection Agency, Cincinnati, OH.
Hilsenhoff, W.L. 1988. Rapid field assessment of organic pollution with a family-level biotic
index. Journal of the North American Benthological Society 7:65-68.
46
-------
Kaushal, S.S., P.M. Groffman, G.E. Likens, K.T. Belt, W.P. Stack, V.R. Kelly, L.E. Band, and
G.T. Fisher. 2005. Increased salinization of fresh water in the northeastern United States.
Proceedings of the National Academy of Science 102:13517-13520.
Konrad, C.P. and D.B. Booth. 2005. Hydrologic changes in urban streams and their ecological
significance. In: Effects of urbanization on stream ecosystems, L.R. Brown, R.H. Gray, R.M.
Hughes, and M.R. Meador (Editors). American Fisheries Society, Symposium 47, Bethesda,
Maryland, pp. 157-177.
Paul, MJ. and J.L. Meyer, 2001. Streams in the urban landscape. Annual Review of Ecology and
Systematics 32:333-365.
Pigou, A.C. 1962. The Economics of Welfare, 4th Ed., Macmillan, London.
Schoonover, I.E., E.G. Lockaby, and S. Pan. 2005. Changes in chemical and physical properties
of stream water across an urban-rural gradient in western Georgia. Urban Ecosystems 8:107-124.
Schueler, T.R. 1994. The importance of imperviousness. Watershed Protection Techniques
1:100-111.
Schueler, T. R. 1987. Controlling Urban Runoff: A Practical Manual for Planning Designing
Urban BMPs, Washington Metropolitan Water Resources Planning Board, Washington, DC.
Tietenberg, T. 2000. Environmental and Natural Resource Economics, Addison-Wesley,
Reading, Massachusetts.
Underwood, AJ. 1994. On Beyond BACI: Sampling Designs that Might Reliably Detect
Environmental Disturbances. Ecological Applications 4:4-15.
U.S. Department of Agriculture, Technical Release 55, Urban hydrology for small watersheds
U.S. EPA. 1996. Draft Framework for Watershed-Based Trading, Office of Water, EPA 800-R-
96-001, May.
U.S. EPA. 2003. Office of Water, Water Quality Trading Policy
http://www.epa.gov/owow/watershed/trading/fmalpolicy2003.pdf (accessed 09/09/2008).
Walsh, CJ. 2004. Protection of in-stream biota from urban impacts: minimize catchment
imperviousness or improve drainage design? Marine and Freshwater Research 55:317-326.
Walsh, C.J., T.M. Fletcher, and A.R. Ladson. 2005a. Stream restoration in urban watersheds
through re-designing stormwater systems: a hypothetical framework for experimental
watershed manipulation. Journal of the North American Benthological Society 24(3):690-705.
47
-------
Walsh, C.J., A.H. Roy, J.W. Feminella, P.O. Cottingham, P.M. Groffman, and R.P. Morgan, II.
2005b. The urban stream syndrome: current knowledge and the search for a cure. Journal of the
North American Benthological Society 24(3):706-723.
Wang, L., J. Lyons, P. Kanehl, R. Bannerman. 2001. Impacts of urbanization on stream habitat
and fish across multiple spatial scales. Environmental Management 28:255-266.
Wenger, S.J., J.T. Peterson, M.C. Freeman, BJ. Freeman, D.D. Homans. 2008. Stream fish
occurrence in response to impervious cover, historic land use, and hydrogeomorphic factors.
Canadian Journal of Fisheries and Aquatic Sciences 65:1250-1264.
48
-------
APPENDIX A: Rain Garden Schematics
49
-------
(Jl
o
Drainage pipe
or swole
Slope pipe to drain
to pop up drain or
blind gravel outlet
Subsurface dr^in
with gravel on
4" Perforated
corrugated
HPDE pipe '-'
— Berm
10' Min Distance
10' Typ. Width
from Buildings
GARDEN A
No Scale
General Notes:
1. Locate all buried utilities prior to excavation.
2. Rain Garden shape will vary by site.
3. Garden A top area 150 sq ft ± 10 sq ft.
Garden B top area 160 sq ft ± 10 sq ft.
4. Shops of rain garden, berms and drainage
will be site specific.
4 Spillway
opening —
Berm -
from Buildings
Drainage pipe
or swale
GARDEN B
No Scale
Rain Garden
Shepherd Creek Area
Cincinnati, Ohio
Figure 1
Plan View
TetraTech, Inc.
O. Nfr OKaOCCM W
-------
15' Typ, Length
Top ol Berrn
EL 100.75
Existing
Grcde
6" Peo Grovel — ' 4" Perforated
Corrugoted HOPE -
12' Mln J_engtn_
Spillwoy Crest
EL 100.00
EL 98.83
4' Win
EL 97.50
4" Solid corrugated
HPDE pipe with 2
90' elbows
Existing
Grode
- Slope pipe
to drain
GARDEN A PROFILE
No Scale
Note:
Elevations shown are not actual
elevations, but show relative
location only.
Rain Garden
Shepherd Creek Area
Cincinnati. Ohio
Figure 4
Profile
Tetra Tech, Inc.
-------
10' Typ. Width
Berm -
o.E
-
to a.
1
Bottom of
Rain Garden
Existing
Grade '
V
£
(M
S
V
V2" Win_
""-- — 4° Perforatec
Corrugated H
"'••— %" Peo Grav
Berm
Existing
Grade
'•- 12"-18" Planting Layer
GARDEN A CROSS SECTION 1-1
Note: No Scale
1. Planting layer to be mechanically tilled.
2. Planting layer may incorporate sand.
compost and other amendments to
improve permeabilty.
3. Do not compact berm soil material.
4. Existing sod can be removed, stored and
replaced on berm area.
5. Use excavated material to construct berm.
Rain Garden
Shepherd Creek Area
Cincinnati, Ohio
Figure 2
Cross Section
Tetra Tech, Inc.
-------
10' Typ. Width
Berrn -
Existing
Grode
.. Bottom of
Rain Garden
- Berm
- Existing
Grade
GARDEN B CROSS SECTION 2-2
No Scale
— 1B"-Z4" Planting Layer
Note:
1. Planting layer to be mechanically tilled.
2. Planting layer moy incorporate sand,
compost and other amendments to
improve permeabilty.
3. Do not compact berm soil material.
4. Existing sod can be removed, stored ond
replaced on berm area.
5. Use excavated material to construct berm.
Ram Garden
Shepherd Creek Area
Cincinnati. Oh to
Figure 3
Cross Section
Tetra Tech, Inc.
-------
APPENDIX B: Rain Barrel Schematic
HOUSE
DOWNSPOUT
RAIN
BARREL
OVERFLOW HOSE
SPIGOT
CINDER
BLOCKS
NOT TO SCALE
SITE NAME
ADDRESS
CITY STATE
FIGURE 1
RAIN BARREL SCHEMATIC
I^IL. I T
TETRATECH
54
-------
APPENDIX C: Quality Management Program
1.0 Purpose
The intent of this appendix is to better illustrate the sampling processes for hydrologic,
ecological, and water quality monitoring efforts, and describe important changes that have been
implemented since the project's inception. This document will be updated periodically to
document, amend, and otherwise record changes and progress in this monitoring effort.
2.0 Problem definition/Background
Successful ecosystem management of watersheds requires the integration of the ecological,
economic, and social influences that collectively determine the sustainability of human
communities. Additional impervious surface as a consequence of urbanization has altered the
partitioning of stormwater into runoff, infiltration, interflow, stream flow, and groundwater
recharge. These urbanized conditions increase the risk of downstream flooding, stream channel
degradation, and damage to both aquatic and terrestrial ecosystems.
Although off-site detention and retention are used to mitigate the hydrologic impacts of
urbanization, the large areas needed for these best management practices (BMPs) are considered
by many to be impractical. BMPs based solely on economic decisions typically lead to
inadequate BMP capacity; while BMPs may improve the stormwater retention and water quality
on the site in which they are located, there are no data to suggest that BMPs effectively improve
the ecological integrity of either the streams directly downstream, nor the integrity of the overall
watershed.
An alternate approach to stormwater runoff management addresses runoff production at smaller
scales, and involves reduction of impervious surface and increased temporary storage of runoff at
the parcel level, or on-lot measures. The idea of limiting excess runoff at the parcel level is
relatively new, and capitalizes on converting impervious surfaces to at least partially pervious
surface, as well as capturing modest runoff volumes in economical parcel-level facilities. The
implementation of this approach to storm water management has not been tested, nor has a path
to implementation been outlined. There is a critical need to determine whether the type and
storage capacity of typical on-lot BMPs, as chosen using economic constraints, are adequate to
restore or protect ecosystems.
3.0 Project organization
The Shepherd Creek Monitoring Project is primarily supported by the Office of Research and
Development, National Risk Management Laboratory, Sustainable Technology Division,
Sustainable Environments Branch with Dr. William Shuster serving as the primary point person
for the monitoring. Ongoing monitoring includes hydrology, water quality, and ecology, with Dr.
William Shuster, Dr. Matthew Morrison, and Dr. Allison Roy as the primary leads associated
with each portion, respectively. Other divisions within NRMRL, contractors, and inter-agency
agreements (IAG) with other federal organizations also support this project. The impervious
survey was a one-time survey to further characterize directly connected impervious areas in the
Shepherd Creek watershed and assist in predicting stream responses to BMPs.
55
-------
4.0 Project description
4.1 Overview
The purpose of this study is to investigate the potential for stakeholder participation in an on-lot
stormwater management program. Field monitoring is used to determine whether such a program
is effective insofar as the improvement of hydrologic (including sediment dynamics), ecologic,
and water quality conditions of downstream subcatchments. In this project, the effectiveness of
BMPs under realistic field conditions will be measured.
4.3 Experimental design
This study has a before/after and control/impact (BACI) experimental design. This approach is
popular in watershed work, as it attempts to control for the high levels of variance that are
common to experiments conducted at large-scales. Due to this experimental approach, we will be
unable to avoid a certain level of pseudoreplication, as errors will not be independent between
the before and after treatment time periods (since consequences that lead to fluctuations
accumulate over time). We will be collecting at 6 different sites (3 of which should be relatively
independent as they are disconnected hydrologically from each other) for at least one year before
and two years after BMP implementation (Table 4.1). Two of the sampling sites will not have
BMPs installed within their respective sub-watersheds, and therefore serve as controls. Because
this project will not be replicated in other watersheds, interpretation of our results will be
somewhat restricted to the results and conclusions drawn from the Shepherd Creek and similar
environmental settings.
Table 4.1. Sample sites, type (control/treatment), and initiation of monitoring by category.
Site
REF1
PWR2
DRI3
ROA4
CONS
URB6
REF7
Nl
N2
N3
Type
Control
Treatment
Treatment
Treatment
Treatment
Control
Control (to
replace REF1)
Treatment
Treatment
Treatment
Hydrology
Fall 2004
Fall 2004
Fall 2004
Fall 2004
Fall 2004
Summer 2005
Fall 2004
Spring 2006
Spring 2006
Spring 2006
Water Quality
Baseflow, Spring 2004
Baseflow, Spring 2004
Stormflow, Summer 2005
Baseflow, Spring 2004
Stormflow, Summer 2005
Baseflow, Spring 2004
Stormflow, Summer 2005
Baseflow, Spring 2004
Stormflow, Summer 2005
Baseflow, Spring 2005
Stormflow, Summer 2005
Baseflow, Summer 2005
Stormflow, Summer 2005
N/A
N/A
N/A
Ecology
Spring 2003
Spring 2003
Spring 2003
Spring 2003
Spring 2003
Spring 2005
Summer 2005
N/A
N/A
N/A
56
-------
4.4 Sampling procedure
Details regarding the sampling approach for hydrology, water quality, and ecology can be found
in the respective supporting QAPPs and SOPs. Briefly, hydrology is sampled continuously (5
min. intervals) for stage and discharge, and one rain gauge (at ROA4) services the entire
watershed. Channel cross sections, bed sediment, and bank soils will be assessed at one time
before and 3 years after BMP implementation. Water quality is sampled monthly during
baseflow conditions (grab sample), and opportunistically during stormflow conditions
(automated ISCO sampler). Periphyton and macroinvertebrates are sampled 5 times per year
(every 4-6 weeks from April to October) during baseflow conditions. Macroinvertebrates are
also sampled seasonally using a quantitative, bucket sampler. During all ecological sampling
periods, physical characteristics, water quality parameters (YSI), and habitat characteristics
(visual assessment) are measured. A summary of the sampling is in Table 4.2.
Table 4.2. Summary of sampling approach, type, and sampling resolution.
Instrument/Sample
Data collected
Data resolution
Output/Indices
Hydrology
Stream stage and discharge
via control structures (i.e.,
weir or culvert)
Tipping-bucket raingauge
located at ROA site
Geomorphic assessment
Stream stage and
discharge
Rainfall intensity,
cumulative amount
Cross-section surveys,
bankfull and water
surface longitudinal
survey, bank soil
collection, Wolman
pebble count
5 minute
Continuous logging,
truncated to Sminute
intervals
Baseline survey done Fall
2004, repeat post-BMP
implementation in 2009 or
2010
Rating curve, stream
discharge record, event
hydrographs
Rainfall intensity,
duration, and amount
D50 (pebble count),
particle size
distribution
Water Quality
YSI 6600 data sonde
Grab (baseflow) and
automated ISCO (stormflow)
water quality samples
Sediment
Temperature, ORP,
conductivity, pH,
dissolved oxygen,
turbidity
Nutrients, anions, metals
suspended and bed-load
sediment
Monthly baseflow
sampling, opportunistic
stormflow sampling, and
with ecology sampling
Monthly baseflow
sampling, opportunistic
stormflow sampling
Monthly baseflow
sampling, opportunistic
stormflow sampling
Water quality
parameters
Chemical
concentrations
Sediment flux, weight
and particle size
distribution
Ecology
Periphyton (2 50ml jars) 1 jar: taxonomic ID 5 times per year (every 4-6 Multimetric indices, %
1 jar: biomass weeks April-October) orders, ordination
57
-------
Periphyton (1 glass fiber
filter)
Benthic macroinvertebrates (1
Ljar)
Physical characterization and
habitat visual assessment
Chlorophyll a
Taxonomic ID,
abundance and richness
Ranking of physical
attributes of sampling
sites
5 times per year (every 4-6
weeks April-October)
5 times per year (every 4-6
weeks April-October);
seasonal bucket samples
5 times per year (every 4-6
weeks April-October);
seasonally with
macroinvertebrates
Multimetric indices, %
orders, ordination
Multimetric indices,
tolerance indices,
ordination
Qualitative Habitat
Evaluation Index
(QHEI)
5.0 Project changes and additions
Since the inception of the Shepherd Creek monitoring project in 2003, various aspects of the
monitoring have changed. Below is a chronological listing of the major changes and additions to
the project, including the creation of supporting documents:
1. September 2005. The baseline monitoring period was extended until spring 2007
(originally spring 2006), after which BMPs will be implemented.
2. January 2005. An additional control site (URB6) was established; the weir for
hydrological monitoring was installed in May 2005.
3. May 2005. Water quality assessment expanded to include monthly baseflow and
opportunistic stormwater samplings (previously only during ecological sampling), and
additional analytes (e.g., metals).
4. May 2005. An additional reach downstream of REF1 site was established (called REF7)
due to dry upstream conditions at REF1. Because REF7 is the location of hydrologic
monitoring, all stormflow water quality sampling will take place at REF7.
5. June 2005. A survey of directly-connected impervious areas (DCIA) in the Shepherd
Creek watershed was initiated in conjunction with USEPA summer interns.
6. January 2006. A Health and Safety Plan for field and laboratory was created for work
conducted under this QMP.
7. April 2006. Implementation of three additional hydrologic monitoring sites at
neighborhood stormwater outfalls to increase the spatial resolution of runoff and drainage
monitoring.
8. November 2007. Install equipment in 5 rain gardens to monitor soil water content and soil
water quality (suction lysimetry).
6.0 Quality objectives and criteria for measurement data
The investigators followed appropriate, established protocols so that data conforms to commonly
accepted and reasonable standards of accuracy and resolution. The standards are detailed in the
family of supporting QAPP and SOPs related to this project.
58
-------
SEPA
United States
Environmental Protection
Agency
Office of Research and Development
National Risk Management
Research Laboratory
Cincinnati, OH 45268
Official Business
Penalty for Private Use
$300
EPA/600/R-08-129
October 2008
www.epa.gov
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chlorine free
------- |