PHA United States
Environmental Protection
Mm. Agency
EPA/600/R-16/146 | August 2016 | www.epa.gov/research
Arid Green Infrastructure
for Water Control and
Conservation:
State of the Science and Research
Needs for Arid/Semi-Arid Regions
RESEARCH AND DEVELOPMENT
-------
Arid Green Infrastructure
for Water Control and Conservation
State of the Science and Research Needs
for Arid/Semi-Arid Regions
Revised Draft
June 28, 2016
by
Jennifer Lee, Ph.D.
Carolyn Fisher, Ph.D.
The Scientific Consulting Group, Inc.
Gaithersburg, MD 20878
EPA Contract No. EPA-C-15-001
Project Officer
Brian Schumacher
U.S. Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
Exposure Methods and Measurements Division
Las Vegas, NV 89193
-------
Notice
This report has been funded by the U.S. Environmental Protection Agency (EPA) under contract
EP-C-15-001 to The Scientific Consulting Group, Inc. This document has been reviewed in
accordance with U.S. Environmental Protection Agency policy and approved for publication.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
Abstract
Green infrastructure is an approach to managing wet weather flows using systems and practices
that mimic natural processes. It is designed to manage stormwater as close to its source as
possible and protect the quality of receiving waters. Although most green infrastructure practices
were first developed in temperate climates, green infrastructure also can be a cost-effective
approach to stormwater management and water conservation in arid and semi-arid regions, such
as those found in the western and southwestern United States. Green infrastructure practices can
be applied at the site, neighborhood and watershed scales. In addition to water management and
conservation, implementing green infrastructure confers many social and economic benefits and
can address issues of environmental justice.
The U.S. Environmental Protection Agency (EPA) provides strong support for and promotes the
benefits of using green infrastructure in protecting drinking water supplies and public health,
mitigating overflows from combined and separate sewers, and reducing stormwater pollution.
EPA has developed tools and resources to guide the design, implementation and maintenance of
green infrastructure best management practices (BMPs) and has issued guidance to encourage the
use of green infrastructure to help manage stormwater. As a member of the Green Infrastructure
Collaborative, EPA also actively partners with federal and nonfederal organizations to foster the
adoption of green infrastructure. EPA regional offices in the West and Southwest also take an
active role in fostering adoption of green infrastructure BMPs within their regions.
Addressing drought and water sustainability through green infrastructure is the subject of policy
initiatives and guidance at the federal, state and local levels. Increasing the use of green
infrastructure to manage stormwater is one of the goals of Executive Order 13693, Planning for
Federal Sustainability in the Next Decade. Arizona, Washington and Texas all have issued
guidance on the use of green infrastructure to manage or conserve stormwater. In addition,
municipalities and counties located in arid and semi-arid climate regions provide guidance on
green infrastructure design and implementation tailored to local conditions, including local
climate, topography, hydrology and soil types. Citizens also have taken an active role in
promoting and executing the adoption of green infrastructure in their local communities.
A survey of current literature was conducted to characterize the current state of the science for the
application of green infrastructure to arid and semi-arid climates and to identify future research
opportunities. Stormwater management BMPs (e.g., bioswales; green roofs; permeable pavement;
planter boxes; rain gardens/bioretention cells; vegetated filter strips; integrated, multi-
11
-------
BMP systems; land conservation; riparian buffers; urban tree canopies) have been evaluated in
arid, semi-arid and Mediterranean climates for management of stormwater quantity and quality.
Practices have been tested for their ability to improve the water quality of stormwater and to
sequester carbon and nitrogen. For practices that use soil substrates, leaching of nutrients and
other pollutants has been quantified. The ability of BMPs to reduce runoff is a key performance
parameter that has been evaluated under different precipitation regimes. Optimal BMP design
and maintenance needs—primarily irrigation—for arid and semi-arid regions have been
explored. BMP siting considerations and approaches have been studied, primarily on a watershed
scale. Research also has been conducted relevant to the use of green infrastructure to conserve
water in arid and semi-arid regions of the United States. Studies have assessed the effectiveness
of different types of in-field rainwater harvesting systems in increasing agricultural productivity
and of roof-top water harvesting systems in meeting nonpotable domestic water needs, including
indoor use and landscape irrigation. Possible impacts of rainwater harvesting on the local water
balance have been evaluated. Design elements to maximize effectiveness of rainwater harvesting
practices also have been studied.
EPA supports an active intramural and extramural research program on green infrastructure
practices. Researchers and stakeholders have identified arid green infrastructure research
opportunities that are relevant to EPA's mission. Stakeholders who participated in a workshop
sponsored by AridLID.org identified the following research and data needs for EPA to address: a
review of institutions' codes and ordinances for treating soil as a resource; a database of existing
findings; a process for prioritizing projects within a watershed; and the research topics of
pretreatment and treatment needs before infiltration, plant/soil interactions and pollutant
removal, the impact of green infrastructure/low-impact development (LID) on flood frequency
and volume, the impact of green infrastructure/LID on floods as a downstream resource, and
ground water impacts resulting from water infiltration.
Researchers in the surveyed literature recommend support for further research at the site scale on
the following topics: replicating field studies; investigating the mechanisms behind effects on
stormwater quality; optimizing design criteria; exploring sustainable solutions to irrigation
needs; improving models of effectiveness by refining model parameters, validating models with
field studies, conducting sensitivity and uncertainty analyses, and incorporating high-resolution
data; developing a better understanding of maintenance needs; and optimizing siting of BMPs.
On a watershed scale, researchers recommend prioritizing sites for installing BMPs, assessing
economic viability, replicating results in other locations, incorporating high-resolution data in
models, improving the mechanistic understanding of biogeochemical processes, gathering more
environmental data relevant to successful installation of practices, and improving model
parameterization. For stormwater conservation research, support is needed for replicating
research studies under a variety of conditions, including geography, climate, soil type and
hydrology; conducting more systematic and comparable studies; and evaluating the cost-
effectiveness of BMPs at a variety of scales.
Further research on effectiveness is needed in general for some of the common BMPs. In
addition, the field would benefit from more research on maintenance, research at longer
timescales and larger geographic scales, refinement of models of arid and semi-arid conditions,
and more research conducted within the United States under locally relevant conditions. Such
111
-------
investigations would facilitate the use of green infrastructure to meet stormwater regulatory
goals.
iv
-------
Contents
Notice ii
Abstract ii
List of Figures viii
List of Tables viii
Acronyms and Abbreviations ix
Chapter 1. Introduction 1
1.1. Definitions of low-impact development and green infrastructure 1
1.2. Definitions of arid and semi-arid climate types 2
1.2.1. Geographic distribution of arid and semi-arid regions 2
1.3. Green infrastructure for arid and semi-arid regions of the United States 3
1.3.1. Site-scale practices—stormwater management 4
1.3.2. Site-scale practices—stormwater conservation 6
1.3.3. Site-scale practices—other purposes 6
1.3.4. Integrated systems 6
1.3.5. Watershed-scale practices—stormwater management and conservation 7
1.4. Green infrastructure in developing countries 8
1.4.1. Drinking water security 8
1.4.2. De-desertification 9
1.5. Benefits of green infrastructure in arid and semi-arid regions of the United States 9
1.5.1. Stormwater management 9
1.5.2. Stormwater conservation 10
1.5.3. S oci al b enefits of green i nfrastructure 10
1.5.4. Economic benefits of green infrastructure 12
1.6. Unique considerations of arid green infrastructure 13
1.6.1. Causes of water stress in arid and semi-arid regions 13
1.6.2. Stormwater management laws and the navigation of water rights 14
1.6.3. Financial incentives for stormwater conservation 14
1.6.4. Characteristics of precipitation and runoff in the arid and semi-arid regions
of the United States 15
Chapter 2. Arid Green Infrastructure Resources at EPA 16
2.1. Green infrastructure program 16
2.1.1. Plan and maintain green infrastructure 16
v
-------
2.1.2. Learn about green infrastructure 18
2.1.3. Collaborate with green infrastructure partners 21
2.2. Regional programs 22
2.2.1. Region 6 22
2.2.2. Region 8 23
2.2.3. Region 9 24
Chapter 3. Policy Initiatives and Guidance to Address Drought and Water Sustainability
Through Green Infrastructure 26
3.1. Federal 26
3.1.1. Executive Order 13693 26
3.1.2. Council on Environmental Quality 26
3.2. State guidance 27
3.2.1. Arizona 27
3.2.2. Washington 28
3.2.3. Texas 29
3.3. Municipal/county guidance 29
3.3.1. Denver, Colorado 30
3.3.2. Los Angeles, California 30
3.3.3. Pima County, Arizona 30
3.3.4. San Diego, California 31
3.4. Collaborative guidance 31
3.5. Nongovernmental agency guidance 31
Chapter 4. Current Research in the Application of Green Infrastructure for Stormwater
Management and Conservation in Arid and Semi-Arid Regions 33
4.1. Methodol ogy 33
4.2. Stormwater management—site-scale research 33
4.2.1. Bioswales 33
4.2.2. Green roofs 34
4.2.3. Permeable pavement 39
4.2.4. Planter boxes 41
4.2.5. Rain gardens/bioretention cells 43
4.2.6. Vegetative filter strips 46
4.2.7. Integrated systems 47
4.3. Stormwater management—watershed-scale research 49
vi
-------
4.3.1. Land conservation 49
4.3.2. Riparian buffers 51
4.3.3. Urban tree canopies 53
4.4. Stormwater conservation 54
4.4.1. Agricultural rainwater harvesting 54
4.4.2. Rain barrels and cisterns 57
Chapter 5. Proposed Areas of Research for EPA 61
5.1. AridLID 2012 research agenda 61
5.1.1. EPA and nonfederal stakeholders 61
5.1.2. Cross-federal agency research 62
5.2. Stormwater management research needs—site scale 63
5.2.1. Bioswales 63
5.2.2. Green roofs 63
5.2.3. Permeable pavement 64
5.2.4. Planter boxes 65
5.2.5. Rain gardens/bioretention cells 65
5.2.6. Vegetative filter strips 66
5.2.7. Integrated systems 66
5.3. Stormwater management research needs—watershed scale 67
5.3.1. Land conservation 67
5.3.2. Riparian buffers 67
5.3.3. Urban tree canopies 68
5.4. Stormwater conservation research needs 68
5.4.1. Agricultural water harvesting 68
5.4.2. Rain barrels and cisterns 69
5.5. Research themes 69
Chapter 6. Conclusions and Recommendations 71
Chapter 7. Acknowledgments 72
Chapter 8. References 73
vii
-------
List of Figures
Figure 1-1. World map of hyper-arid, arid, semi-arid and dry subhumid regions 2
Figure 1-2. United States humidity zones 3
Figure 1-3. Green roofs installed in semi-arid regions of the United States 5
Figure 1-4. Rain garden at the New Belgium Brewery in Fort Collins, Colorado 6
Figure 1-5. Rainwater cistern installed at a residence in Tucson, Arizona 6
Figure 1-6. Demonstration street median in Santa Fe, New Mexico 7
Figure 1-7. Dune stabilization in Mauritania 9
Figure 1-8. Sidewalk and bioswale in the completed public right of way after the Elmer
Avenue Neighborhood Retrofit in Los Angeles, California 11
Figure 3-1. A curb cut draws stormwater from the street into a bioretention basin in the
right of way 27
Figure 3-2. The climate regions of eastern Washington 28
Figure 3-3. Average annual precipitation in Texas, in inches 29
Figure 4-1. Comparison of water saving efficiency and runoff volume reduction for
U.S. cities by annual precipitation levels 59
List of Tables
Table 1-1. Green infrastructure in arid and semi-arid regions of the United States 4
Table 1-2. Agricultural water harvesting techniques 8
Table 2-1. EPA technical assistance projects in arid and semi-arid regions 17
Table 2-2. EPA Region 6 green infrastructure demonstration projects in the desert
Southwest 23
Table 2-3. EPA Region 8 green infrastructure demonstration projects in the semi-arid
west 24
Table 4-1. Mean influent and effluent concentrations and event-based concentration
reduction rates from a rain garden 44
viii
-------
Acronyms and Abbreviations
BMP
best management practice
CEQ
Council on Environmental Quality
COD
chemical oxygen demand
CSO
combined sewer overflow
DEM
digital elevation model
DOC
dissolved organic carbon
EISA
Energy Independence and Security Act of 2007
EPA
U.S. Environmental Protection Agency
EO
Executive Order
FAO
Food and Agriculture Organization
GIS
geographic information system
LID
low-impact development
LiDAR
light detection and ranging
LTCP-EZ
Long-Term Control Plan-EZ
MS4
Municipal Separate Storm Sewer System
NCAT
National Center for Asphalt Technology
NPDES
National Pollutant Discharge Elimination System
NRCS
Natural Resources Conservation Service
ORD
Office of Research and Development
PAH
polycyclic aromatic hydrocarbon
SWAT
Soil and Water Assessment Tool
SWMM
Storm Water Management Model
TKN
total Kjeldahl nitrogen
TMDL
total maximum daily load
TSS
total suspended solids
USD A
U.S. Department of Agriculture
USGS
U.S. Geological Survey
IX
-------
Chapter 1.
Introduction
The U.S. Environmental Protection Agency (EPA) commissioned a literature review to identify
the state-of-the science practices dealing with water control and conservation in arid and semi-
arid regions, with emphasis on these regions in the United States. The search focused on
stormwater control measures or practices that slow, capture, treat, infiltrate and/or store runoff at
its source (i.e., green infrastructure). The material in Chapters 1 through 3 provides background
to EPA's current activities related to the application of green infrastructure practices in arid and
semi-arid regions. An introduction to the topic of green infrastructure in arid and semi-arid
regions is presented in Chapter 1, including definitions of terms used in this document;
descriptions of green infrastructure practices applicable to arid and semi-arid regions, both in
developed and developing countries; benefits of green infrastructure in arid and semi-arid
regions; and unique aspects of green infrastructure in arid and semi-arid regions of the United
States. Chapter 2 focuses on green infrastructure resources that have been developed by EPA at
the program and regional office level. Policy initiatives and guidance to address drought and
water sustainability through green infrastructure in arid and semi-arid regions of the United
States that have been formulated at the state, regional and municipal/county levels, as well as by
nongovernmental agencies and collaboratively, are presented in Chapter 3. Chapter 3 also
includes federal actions related to green infrastructure that apply across the United States.
Chapter 4 presents the results of the literature review, organized by practice. Based on the
research needs identified in the literature, as well as topics identified by experts in a recent
conference focused on developing a research agenda, areas of research for applying green
infrastructure in arid and semi-arid regions that are relevant to EPA's mission are presented in
Chapter 5. Varying levels of detail on different practices are available in the literature, which is
reflected in Chapters 4 and 5. Chapter 6 is a summary of the findings of the literature review.
1.1. Definitions of low-impact development and green infrastructure
EPA defines low-impact development (LID) as "systems and practices that use or mimic natural
processes that result in the infiltration, evapotranspiration or use of stormwater in order to
protect water quality and associated aquatic habitat." Development or re-development using the
LID approach involves managing stormwater as close to its source as possible (USEPA 2016).
Green infrastructure as defined by EPA refers to managing wet weather flows using LID
practices (USEPA 2016i). Green infrastructure is contrasted with conventional gray stormwater
infrastructure, which is designed to move stormwater away from the built environment. In urban
areas, green infrastructure "uses vegetation, soils, and other elements and practices to restore
some of the natural processes required to manage water and create healthier urban
environments." Traditional stormwater management practices (i.e., so-called "gray
infrastructure"), such as separate storm sewers and concrete river channels, in contrast, are
designed to transport stormwater away from its source. Green infrastructure elements and
practices can be applied at the site, neighborhood, city and county scales (USEPA 2015s).
1
-------
1.2. Definitions of arid and semi-arid climate types
Deserts are arid regions that support limited vegetation. The most widely used climate
classification system is the Koppen-Geiger classification system, as updated by Peel, Finlayson
and Mcmahon (2007), which is derived from data on the mean annual precipitation, mean annual
temperature and seasonality of precipitation. Climate type Group B is arid, with subtypes
indicating arid (W) or steppe (i.e., semi-arid, S) and hot (h) or cold (k). Mediterranean climates
(Koppen-Geiger climate symbols Csa and Csb) are temperate with hot, dry summers. With
respect to vegetation, dry climates can be divided into arid and semi-arid based on the ratio
between annual precipitation and potential evapotranspiration rate (the aridity index), which is a
measure of the adequacy of moisture relative to the needs of plants and is estimated as a function
of precipitation, temperature and other factors (Meigs 1953; Thornthwaite 1948). The general
term "xeric" is used in this document to refer to arid and semi-arid, as well as seasonally dry,
climates.
1.2.1. Geographic distribution of arid and semi-arid regions
Deserts comprise approximately one-third of the Earth's land mass. Deserts are classified by
location and dominant weather pattern. Trade wind deserts are formed by the dissipation of cloud
cover by the dry trade winds and include the Sahara. Midlatitude deserts, such as the Sonoran
Desert, form between 30 and 50 degrees north and south latitude and are far inland of oceans.
Rain shadow deserts form in the lees of high mountain ranges. Coastal deserts such as the
Atacama are affected by cold ocean currents that flow parallel to the coast and generally form on
western edges of continents. Monsoon deserts form as a result of monsoons losing water in
heavy, seasonal rains as they move inland. Polar deserts such as the Dry Valleys of Antarctica
receive little rainfall annually and have temperatures that do not exceed 10°C (Walker 1996). A
world map of the Koppen-Geiger climate classification types, i ncluding arid and semi-arid
regions (Group B) and Mediterranean climates (types Csa and Csb), is presented in Figure 1-1.
The three major hot U.S. deserts
are the Chihuahuan Desert
(Arizona and New Mexico),
Sonoran Desert (southwestern
Arizona and southeastern
California), and Mojave Desert
(southeastern California and
portions of Nevada and Arizona).
The Great Basin Desert (parts of
Utah, Oregon, Idaho, Wyoming,
Colorado, Nevada and Arizona) is
the largest cold desert in the
United States (Lee et al. 2011).
The arid and semi-arid regions of
the United States are shown in
Figure 1-2. The different humidity
zones—including hyper-arid, arid
Figure 1-1. World map of hyper-arid, arid, semi-arid and
dry subhumid regions.
Source: FAO (2015). Used by permission of FAO. Copyright 2016
FAO.
2
-------
0 Ł=
Hawaii
I I Hyper-Arid
I I Arid
l l Semi-Arid
I I Dry-Subhurnid
I I Humid
I I Cold
Figure 1-2. United States humidity zones.
Source: OALS (2002).
and semi-arid—follow the designations of the United Nations Environmental Programme,
reflecting variations in the aridity index.
1.3. Green infrastructure for arid and semi-arid regions of the United States
Most green infrastructure practices were first developed in temperate climates, but most also are
applicable to arid and semi-arid regions, although they may require modification. Green
infrastructure can be a cost-effective approach to storm water management and water
conservation in arid and semi-arid regions, reducing runoff, conserving water, recharging ground
water, conserving energy and improving air quality (USEPA 2010b). Design elements for green
infrastructure that are applicable to arid and semi-arid regions of the United States are listed in
Table 1-1. The primary application for each design element is indicated, either for managing
stormwater close to its source or conserving stormwater to reduce potable water demand (or
both). Design elements can be implemented at a range of scales, from the scale of local sites by
individual property owners to larger scales that affect entire watersheds or portions of
watersheds, and some design elements can be applied both at the site and multisite scales.
The design elements in this section are described first at the site-scale, and then watershed-scale
practices are described. Within those site-scale practices, applications for stormwater
management and conservation are presented. Site-scale design elements for related purposes of
3
-------
Table 1-1. Green infrastructure in arid and semi-arid regions of the United States
Design Element Stormwater Application Scale
Management Conservation Site Watershed
Bioswales x x
Green roofs x x
Permeable pavement x x
Planter boxes x x
Rain gardens/bioretention cells x x
Integrated systems (e.g., green streets) x x x
Land conservation x x
Riparian buffers x x
Urban tree canopies x x x
Rain barrels or cisterns x x x
Agricultural water harvesting x x x
Other design elements: Green walls (primary application: energy conservation), vegetative filter strips (primary
application: pollution control).
energy conservation and pollution control also are included. Integrated systems comprise
combinations of different practices that can be implemented at the site or larger scales.
1.3.1. Site-scale practices—stormwater management
1.3.1.1. Bioswales
Bioswales are vegetated, xeriscaped or mulched channels that capture stormwater runoff, slow its
flow and enhance infiltration as the water flows downslope through the channels. Bioswales are
used to treat and retain runoff. They are linear features and often are installed along streets and
parking lots (USEPA 2015s). In arid and semi-arid regions, the vegetation selected should be
appropriate to the climate, and climate-specific maintenance techniques are needed (USEPA
2010b).
1.3.1.2. Green roofs
Green roofs, also known as vegetative roofs, are flat or sloping roofs covered with growing
media that support vegetation. The media and vegetation are designed to allow infiltration and
storage of rainfall and evapotranspiration of stored water, as well as to treat stormwater. Green
roofs also can provide recreational space in urban areas. Green roofs are designed as extensive or
intensive systems. Extensive green roofs are shallow (6 inches or less) and tend to be designed
for specific engineering or performance goals, whereas intensive green roofs can be much deeper
and can include landscaping elements such as walkways, lawns, large perennial plants and trees
(Miller 2016). In arid and semi-arid regions, green roofs require irrigation, but appropriate design
approaches—including using a greater media depth, planting native and drought-adapted species
and applying drip irrigation—can increase water efficiency. Green roofs also can be designed for
irrigation by sources other than municipal water supplies (USEPA 2010b). Figure l-3a shows a
view of EPA Region 8's extensive green roof installed in Denver, Colorado. An example of an
intensive green roof installed in a cool desert climate is shown in Figure l-3b.
4
-------
HHBHMi
Figure 1-3. Green roofs installed in semi-arid regions of the United States.
Extensive green roof installed at the EPA regional office building in Denver, Colorado (a), and
intensive green roof installed on the Church of Jesus Christ of the Latter-Day Saints Conference
Center in Salt Lake City, Utah (b),
Source: USEPA Region 8 (2016).
1.3.1.3. Permeable pavement
Permeable pavement includes a range of technologies that allow stormwater to infiltrate and be
stored where it falls. Permeable pavement can reduce stormwater runoff and improve water
quality. Permeable pavement technologies include pervious concrete, porous asphalt and
permeable interlocking pavers. Permeable pavement systems generally overlay several layers of
bedding into which water drains and from which it infiltrates into the soil below. Permeable
friction course is a type of permeable pavement where a layer of porous asphalt pavement is
placed on top of a regular impermeable roadway (USEPA 2010b, 2015s).
1.3.1.4. Planter boxes
Planter boxes are rain gardens with vertical walls and open or closed bottoms. Like rain gardens,
they collect runoff from surrounding impervious surfaces and allow it to soak into the soil.
Planter boxes are used primarily in streetscaping in space-limited locations such as dense urban
areas (USEPA 2015s).
1.3.1.5. Rain gardens and bioretention cells
A rain garden (see Figure 1-4) is a depressed area that collects precipitation that runs off from
surrounding impervious surfaces (e.g., roof, driveway, parking lot, street) and allows it to soak
into the ground, where it infiltrates into the soil and/or is evapotranspired back into the
atmosphere. Rain gardens often are distinguished from more complex systems with enhanced
drainage layers and amended soils, referred to as bioretention cells. In addition to stormwater
management, rain gardens can play an important role in water conservation in arid or semi-arid
regions because they rely on precipitation for their water needs. In these regions, the limitations
of the water supply must be considered in design, particularly in selecting vegetation, as well as
maintenance planning (USEPA 2010b, 2015s).
5
-------
Figure 1-4. Rain garden at the New Belgium
Brewery in Fort Collins, Colorado.
1.3.2. Site-scale practices—stormwater
conservation
1.3.2.1. Rain barrels, cisterns and storage
tanks
Rain barrels, cisterns and storage tanks store
rainwater from rooftops or other impervious
areas for later use, thereby reducing stormwater
runoff and reducing irrigation demand. Smaller
rain barrels can reduce runoff and irrigation
demand on a site scale (see Figure 1-5),
whereas large cisterns or storage tanks can
capture larger proportions of stormwater and
address more irrigation demand. Sizing
depends on roof area; rainfall patterns, which
are specific to climate; available space; and
costs, which increase with capacity (USEPA
2010b).
1.3.3. Site-scale practices—other purposes
1.3.3.1. Green walls
Green walls are sustainable construction
practices that cover a building envelope with
vegetation. The term "green facade" refers to
Source: USEPA Region 8 (2016). green walls in which climbing or hanging
plants are trained to cover a wall using special
support structures. "Living walls" refer to more complex systems in which pre-vegetated panels,
vertical modules or planted blankets are fixed vertically to a structural wall or frame. The main
benefits of living walls are insulation and improved outdoor air quality (Feng and Hewage 2014).
1.3.3.2. Vegetated filter strips
A vegetated filter strip is a strip or area of
vegetation intended to remove contaminants
from overland flow. The purpose of installing
vegetated filter strips is to remove suspended
solids and reduce dissolved contaminant
loadings in runoff Vegetated filter strips are
planted preferentially with permanent
herbaceous plants (NRCS 2010).
1.3.4. Integrated systems
Integrated systems are green infrastructure
comprised of multiple design elements
conceived of as a system. A green street is an example of an integrated system designed on a
neighborhood or watershed scale to manage and treat stormwater. Green streets also can beautify
Figure 1-5. Rainwater cistern installed at a
residence in Tucson, Arizona.
Source: USEPA (2010b).
6
-------
streets and slow traffic. Examples of green streets (see Figure 1-6) are rain gardens installed in
rights-of-way, medians, traffic circles and chicanes, with rainwater directed into the rain gardens
and bioswales by adding curb cuts or installing curbs
flush with the ground (USEPA 2010b). Construction
of green streets can include features to increase their
water storage capacity, such as dry wells and
infiltration galleries.
1.3.5. Watershed-scale practices—stormwater
management and conservation
1.3.5.1. Land conservation
Land conservation through protecting open spaces
and sensitive natural areas can improve water quality
and reduce flooding. Land conservation approaches
are particularly applicable to stormwater
management in urban areas, where they also provide
social benefits by increasing recreational
opportunities. Sensitive natural areas in arid and
semi-arid environments include riparian areas and
steep hillsides (USEPA 2015s).
1.3.5.2. Riparian buffers
Riparian buffers restrict development in the land adjacent to washes, arroyos, creeks or streams.
They are intended to reduce erosion and preserve the riparian channel. They can provide
environmental and social benefits (e.g., recreational trails within riparian buffers). By providing
a network of habitats, they also increase wildlife diversity (USEPA 2010b).
1.3.5.3. Urban tree canopies
Restoration of the urban tree canopy is a green infrastructure approach that cities can take to
manage stormwater. Trees reduce and slow stormwater flow by intercepting precipitation in their
leaves and branches (USEPA 2015s).
1.3.5.4. Agricultural water harvesting
Agricultural water harvesting is defined as the "collection of runoff for its productive use."
Agricultural water harvesting can be implemented at the site scale by individual farmers as well
as at larger scales. The term includes within-field microcatchments (catchment length 1 to 30 m),
external catchment systems (catchments 30 to 200 m in length), and floodwater farming
(e.g., permeable rock dams, water-spreading bunds). Agricultural water harvesting offers a low-
cost alternative to irrigation (Critchley and Siegert 1991). Major techniques of agricultural water
harvesting most applicable to a range of situations and geographic areas are presented in
Table 1-2.
Figure 1-6. Demonstration street
median in Santa Fe, New Mexico.
Source: USEPA Region 6 (2014).
7
-------
Table 1-2. Agricultural water harvesting techniques
Technique Classification Main Uses Description
Negarim
microcatchments
Microcatchment
Trees and grass
Closed grid of diamond shapes or open-ended "V"s
formed by small earth ridges, with infiltration pits
Contour bunds
Microcatchment
Trees and grass
Earth bunds on contour spaced at 5-10 m apart
with furrow upslope and cross-ties
Semicircular bunds
Microcatchment
Rangeland and
fodder (also
trees)
Semi-circular shaped earth bunds with tips on
contour. In a series with bunds in staggered
formation
Contour ridges
Microcatchment
Crops
Small earth ridges on contour at 1.5-5 m apart
with furrow upslope and cross-ties Uncultivated
catchment between ridges
Trapezoidal bunds
External
catchment
Crops
Trapezoidal-shaped earth bunds capturing runoff
from external catchment and overflowing around
wingtips
Contour stone
bunds
External
catchment
Crops
Small stone bunds constructed on the contour at
spacing of 15-35 m apart slowing and filtering
runoff
Permeable rock
Dams
Floodwater
farming
Crops
Long, low rock dams across valleys slowing and
spreading floodwater as well as healing gullies
Water-spreading
bunds
Floodwater
farming
Crops and
rangeland
Earth bunds set at a gradient, with a "dogleg"
shape, spreading diverted floodwater
Source: Adapted from Critchley and Siegert (1991). Used by permission of FAO. Copyright 2016 FAO.
1.4. Green infrastructure in developing countries
Green infrastructure techniques can be used to address domestic, livestock, agricultural and
environmental water needs in developing countries. Water harvesting and storage for domestic
and agricultural use can increase food production and sustain human habitation in arid and semi-
arid regions, mitigating effects of seasonal dry spells. De-desertification also is a green
infrastructure approach that uses vegetation to conserve water and reduce erosion.
1.4.1. Drinking water security
One of the 17 United Nations Sustainable Development Goals for 2015 to 2030 is to "ensure the
availability and sustainable management of water and sanitation for all" (United Nations 2015).
Water harvesting technologies applicable to developing countries are those that are easy to
construct, use local labor and do not require external funding. Technologies for container and in-
soil storage of rainwater that can be used to address drinking water security in arid and semi-arid
regions of developing countries include small storage in containers (e.g., rain jars, ferro-cement
tanks, stone masonry tanks), larger storage in containers (e.g., open reservoirs, cisterns), small
in-soil storage measures (e.g., contour trenches or ridges, terraces), and large in-soil storage
measures (e.g., spate irrigation, subsurface dams, sand dams). To be sustainable, water
harvesting projects must be applicable to local physical, cultural and economic circumstances
(Lasage and Verburg 2015).
8
-------
1.4.2. De-desertification
Desertification threatens large areas of arid and semi-arid lands worldwide. Desertification
results from various factors, including climatic variations and human activities. The Food and
Agriculture Organization (FAO) of the United Nations conducted a comprehensive analysis of
afforestation, reforestation and restoration projects and initiatives in arid, semi-arid and dry
sub humid climates (referred to as "drylands" by the FAO). Forests and trees are important in de-
desertification efforts and provide such environmental services in drylands as food for humans
and livestock, products for generating income, increased water infiltration, reduced soil erosion,
moderation of local climates, increased soil fertility, habitat for fl ora and fauna, and cultural
services. Restoration actions include habitat protection, assisted regeneration, sand dune
stabilization and tree planting (see Figure 1-7). Sustainable de-desertification of drylands
requires a landscape approach. Practitioners
need to choose the most cost-effective
restoration strategies, protect and manage
restoration initiatives, promote natural
regeneration, and plant when necessary.
Monitoring and evaluation provide feedback on
restoration activities. The FAO's report
provides case studies of dryland restoration
experiences (FAO 2015).
1.5. Benefits of green infrastructure in arid
and semi-arid regions of the United States
Green infrastructure has the potential to provide
stormwater management and conservation
benefits, as well as social and economic benefits, in arid and semi-arid regions of the United
States. Stormwater management benefits include reduced flooding, reduced erosion and
improved surface water quality. Stormwater conservation benefits include increased ground
water recharge and reduced water imports.
1.5.1. Stormwater management
Green infrastructure may represent a more cost-effective approach to stormwater management
than traditional practices. Green infrastructure reduces flooding by increasing infiltration,
evapotranspiration and water storage where precipitation falls. Increasing infiltration also
recharges ground water reserves and can benefit aquatic habitats. Another environmental benefit
of green infrastructure for stormwater management is that it improves water quality by reducing
runoff and allowing runoff to be treated by soils and vegetation (USEPA 2010b). Reducing
runoff can provide benefits for mitigating soil erosion, which causes upstream and downstream
problems. Soil erosion from agricultural lands causes topsoil loss, resulting in increased
susceptibility to drought. Downstream, sediment loads lead to operational costs for desilting such
infrastructure as irrigation canals and hydroelectric power dams. Vegetative filter strips and
reforestation are two green infrastructure approaches used to address this problem (Betrie et al.
2011).
Figure 1-7. Dune stabilization in
Mauritania.
Source: FAO (2015). Used by permission of FAO.
Copyright 2016 FAO.
9
-------
1.5.2. Stormwater conservation
Green infrastructure also may be a cost-effective approach for conserving stormwater to address
potable water needs. In cities and towns in arid and semi-arid regions of the United States,
impervious surfaces and engineered conveyance systems reduce infiltration of precipitation into
the ground water. Green infrastructure practices that increase infiltration can recharge ground
water resources, a potential source of potable water. Some green infrastructure practices that
conserve stormwater for reuse also can remove pollutants from stormwater, improving water
quality, making it suitable for reuse to reduce demands on potable water. Green infrastructure
has the potential to reduce landscape irrigation. Where water is imported to meet local demands,
as is done in many cities and towns in arid and semi-arid parts of the United States, rainwater
harvesting for nonpotable uses such as landscape irrigation reduces demand on the potable water
supply and the need for costly water imports (USEPA 2010b). Water rights laws can affect the
legality of water harvesting, however, as described in Section 1.6.2.
1.5.3. Social benefits of green infrastructure
Green infrastructure has many social benefits. It can improve public health by reducing the urban
heat island effect, improving air quality, providing recreational opportunities and mitigating
carbon dioxide emissions. Green infrastructure also beautifies neighborhoods, calms traffic and
builds communities, improving the urban environment.
1.5.3.1. Improve public health
Extreme heat, poor air quality and lack of access to pedestrian-friendly landscapes are public
health concerns that are linked to issues of environmental justice. Socially disadvantaged
neighborhoods often experience greater negative health impacts from extreme heat. Populations
such as the elderly, people with physical or mental illnesses, the very young, individuals living
alone, and people with low socio-economic status are more vulnerable to extreme heat.
Removing pavement and planting vegetation can reduce the urban heat island effect by cooling
and shading urban neighborhoods (USEPA 2010b). Four types of green infrastructure typically
are used for cooling in urban areas: green open spaces, shade trees, green roofs, and green walls
and fa9ades (Norton et al. 2015). In a modeling study using data from Phoenix, Arizona,
xerophytic shade trees were predicted to reduce urban heat island effects, particularly at the site
scale (i.e., the residential lot and adjacent buildings) and at night. Compared to mesic
landscaping, however, xeriscaping appeared to increase temperatures at all spatial scales and
temporal periods (Chow and Brazel 2012). Installation of permeable pavement also might be
expected to influence urban temperatures because of the insulating properties of its high air void
content and high albedo relative to black asphalt. Modeling studies have indicated that porous
asphalt pavement would have higher daytime surface temperatures but lower nighttime
temperatures compared to materials with similar albedos (i.e., traditional dense-graded asphalt
and Portland cement concrete pavement), indicating that assessing the effect of different types of
pavement on the urban heat island effect is complex (Stempihar et al. 2012). Permeable
pavement under wet conditions cooled surface temperatures by 15 to 35°C compared with
impermeable pavement under summer conditions in semi-arid Davis, California (Li et al. 2013).
10
-------
Green infrastructure also has the potential to improve public health by mitigating global climate
change and improving air quality. Soils and plant biomass sequester carbon, reducing
atmospheric carbon dioxide concentrations. For example, green roofs installed in semi-arid
regions have been shown to sequester carbon (Ondono, Martinez-Sanchez and Moreno 2016b).
Poor air quality is a particular problem for children and the elderly. Vegetation removes air
pollutants and can mitigate the formation of smog, which is a particular health problem in urban
areas (USEPA 2010b).
Many green infrastructure projects incorporate pedestrian- and bicycle-friendly designs. For
example, intensive green roofs can provide recreational space, which can be particularly valuable
in dense urban settings (Jiang, Yuan and Piza 2015). The green street design in Los Angeles
shown in Figure 1-8 incorporated new sidewalks. Pedestrian- and bicycle-friendly recreational
spaces can promote public health by fostering physical activity (USEPA 2010b).
Figure 1-8. Sidewalk and bioswale in the completed public right of way after the Elmer
Avenue Neighborhood Retrofit in Los Angeles, California.
Source: Council for Watershed Health (2015a). Photo credit: Council for Watershed Health,
11
-------
1.5.3.2. Beautify neighborhoods
Green infrastructure provides a sustainable approach to irrigating private gardens and public
green spaces. Landscapes maintained by passive and active rainwater harvesting beautify
neighborhoods and urban areas (USEPA 2010b).
1.5.3.3. Calm traffic
Many green infrastructure design elements implemented in streets and alleys reduce street widths
and introduce curves. The green infrastructure techniques can slow traffic (USEPA 2010b).
1.5.3.4. Build communities
Beautifying neighborhoods and creating a unique sense of place with green infrastructure can
increase interactions among neighbors. Many green infrastructure projects involve neighbors
working together to beautify their communities and make them more livable (USEPA 2010b).
1.5.4. Economic benefits of green infrastructure
The economic benefits of green infrastructure include reduced landscape and building
maintenance costs, increased ground water resources, reduced water imports, lower energy costs
and reduced stormwater management costs.
1.5.4.1. Reduce landscape and building maintenance costs
Landscapes planted with drought-adapted plants and irrigated by rainwater harvesting cost less to
maintain (USEPA 2010b). Buildings that include elements of green infrastructure can have lower
maintenance costs as well. Green roofs can extend the lifetime of roofs, requiring replacement
only every 40 to 50 years, compared with 10 to 20 years for conventional roofs (Jiang, Yuan and
Piza 2015). A lifecycle costs-versus-benefits comparative analysis between a green roof, black
roof and reflective roof in Utah showed that the green roof alternative was a better investment
despite larger initial capital costs (Wu and Smith 2011).
1.5.4.2. Increase ground water resources and reduce water imports
Using ground water resources for public water supplies is costly. Costs associated with pumping
ground water and importing ground water can be reduced by green infrastructure practices that
conserve water and increase ground water recharge. Rainwater harvesting decreases landscape
irrigation demands on the public water supply (USEPA 2010b).
1.5.4.3. Lower energy use
Green infrastructure can reduce energy used to import, treat and distribute municipal water, as
well as reduce energy use by buildings for heating and cooling. The transportation and treatment
of water can represent a significant fraction of electricity consumed by municipalities in arid and
semi-arid regions of the United States (USEPA 2010b).
By providing insulation and shade and dissipating heat through evapotranspiration, green roofs
have the potential to reduce energy use for building cooling. Thermal and energy simulations
have shown that green roofs potentially reduce cooling energy consumption significantly in
Santiago, Chile, which has a Mediterranean climate (Vera et al. 2015). Modeling of electricity
and heating energy costs and consumption in the semi-arid United States, however, has shown
12
-------
that savings in cooling costs as compared with white roofs may derive primarily from reduced
peak demand rather than reduced overall consumption (Sailor, Elley and Gibson 2012). The
summer and winter energy savings from green roofs in hot and dry climates with mild winters
(e.g., Los Angeles, California—a Mediterranean climate) can be enhanced by installing a
variable insulating system, an insulated smart plenum coupled or decoupled to the indoor
temperature (La Roche and Berardi 2014).
The shading, insulating and heat dissipating properties of green walls have the potential of
reducing energy use for heating and cooling. A modeling study of three types of living walls in a
Mediterranean climate (Los Angeles) showed substantial energy savings in cooling but little
effect on energy use for heating. The cooling effects are mainly from shade and
evapotranspiration (Feng and Hewage 2014). A green fa9ade lowered temperatures by 5°C
compared to a bare wall during the peak summer month of July for a building in the United Arab
Emirates (Haggag, Hassan and Elmasry 2014). In a modeling study of the effects of green walls
in a semi-arid climate in China, indoor air temperatures were lowered by as much as 15°C
compared with ordinary construction (Di, Lin and Wang 2014).
1.5.4.4. Reduce stormwater management costs
Green infrastructure also can reduce the costs of building stormwater management infrastructure.
For example, green roofs can reduce or eliminate the need for stormwater detention vaults or
ponds, reducing stormwater management costs (Jiang, Yuan and Piza 2015).
1.6. Unique considerations of arid green infrastructure
Arid and semi-arid climates are defined by their precipitation and temperature. Geographically,
they occur at a wide range of latitudes and from a range of interactions between weather and
geography. Water stress in arid and semi-arid regions of the United States, however, is caused by
other factors in addition to climate (e.g., growing population). Another unique aspect of
implementing green infrastructure in arid and semi-arid regions of the United States is the need
to navigate federal and local legislation. Finally, the characteristics of precipitation and runoff
unique to desert and semi-arid regions in the United States complicate the management of
stormwater and conservation of precipitation.
1.6.1. Causes of water stress in arid and semi-arid regions
Water stress in the arid and semi-arid regions of the United States is caused by a range of factors.
These include droughts, growing population, ground water depletion, irrigation and climate
change. Natural variation in precipitation—as well as effects on precipitation patterns from
anthropogenic climate change—can lead to drought. Green infrastructure, which increases
infiltration of stormwater and ground water recharge, can play a role in adaptation to drought.
Increasing human population in the arid and semi-arid regions of the United States, particularly
in urban areas, places increasing stress on water resources. The total population of the West is
growing at a faster rate than the total U.S. population. Population growth is greatest in
incorporated areas, which also are growing faster in the West than in the United States as a
whole. As a result, 76.4 percent of the population of the West now lives in incorporated places
(Cohen, Hatchard and Wilson 2015).
13
-------
In the arid and semi-arid West and Southwest, ground water is a significant source of water for
domestic and agricultural use. Changes in freshwater reserves are being monitored by new
satellite technologies. Data from California's Central Valley show the depletion in ground water
during the period from 2003 to 2009 (AMNH 2015).
Irrigation for residential landscapes and agriculture adds to water stress in arid and semi-arid
regions. Approximately one-third of all freshwater withdrawals in the United States are used for
irrigation (USEPA 2016j).
Climate change is predicted to have a significant effect on water supplies throughout the United
States. An analysis performed for the Natural Resources Defense Council (2010) found that by
2050, more than one-third of all U.S. counties will be at risk for water shortages because of
global warming. The arid and semi-arid regions of the Southwest and West are at particular risk
from water shortages because of climate change.
1.6.2. Stormwater management laws and the navigation of water rights
EPA policy encourages the integration of green infrastructure into National Pollutant Discharge
Elimination System (NPDES) permits and combined sewer overflow (CSO) remedies. Municipal
Separate Storm Sewer System (MS4) operators, who are required to obtain an NPDES permit
and develop a stormwater management program, are increasingly integrating green infrastructure
into their MS4 permits. EPA supports the use of green infrastructure for long-term control and
remediation of noncompliant CSOs (USEPA 2016f). Green infrastructure practices also can be
incorporated into total maximum daily load (TMDL) reports to plan for impaired waters meeting
water quality standards (USEPA 2008a).
Local mandates for meeting a portion of landscaping requirements with harvested rainwater,
requirements for water conservation or rainwater harvesting in new construction, and
requirements for controlling runoff using green infrastructure practices also have been
implemented in arid and semi-arid regions of the United States (USEPA 2010b).
Water rights laws comprise a complex legal landscape in the arid and semi-arid West. The prior
appropriation system, which grants individuals the right to put water to beneficial use based on
the priority of the user (i.e., "first in time, first in right"), applies in much of the West. State
water laws can affect green infrastructure practices (USEPA 2010b). In some states, precipitation
belongs to existing water-rights owners. In others, rainwater harvesting is restricted based on
water quality and public health concerns. Rainwater harvesting has been viewed legally in some
states (e.g., Colorado) as a potential injury to senior water rights and has been prohibited.
Colorado currently is the only state to ban the use of rain barrels. Among states with arid and
semi-arid regions, pending legislation in California will require the State Water Resources
Control Board to address the potential for storm-induced overflow from an impoundment storing
recycled water (NCSL 2016).
1.6.3. Financial incentives for stormwater conservation
At the same time, many states and municipalities are promoting rainwater harvesting through
financial incentives. State tax credits, rebates for installing rainwater harvesting systems, and
14
-------
reductions in stormwater management fees in return for using rainwater harvesting systems are
approaches that have been used in the West and Southwest (USEPA 2010b).
1.6.4. Characteristics of precipitation and runoff in the arid and semi-arid regions of the
United States
Runoff from soils in arid and semi-arid regions is affected by vegetation, the surface properties
of nonvegetated soils, and creation of impervious surfaces through development. Intact
nonvegetated soils usually are covered by biological soil crusts, communities of cyanobacteria,
lichens and mosses, which reduce runoff on a local scale. When the crusts are disrupted by
anthropogenic activity, such as hoof action by livestock, the transport of nutrients by water, soil
and organic matter within the desert ecosystem is altered (Belnap et al. 2005). Intact biological
soil crusts resist sediment transport (Rodriguez-Caballero et al. 2014). Soil type also can affect
infiltration, with less infiltrative soils requiring design modifications for green infrastructure
implementation. In urban arid and semi-arid areas, such impervious surfaces as parking lots,
roads and rooftops decrease the amount of precipitation absorbed by soil, increasing runoff
(USEPA 2010b).
The problem for stormwater management in arid and semi-arid regions posed by disturbed soils
and impervious surfaces is exacerbated by precipitation patterns. Rainfall in deserts can be of
short duration but high in intensity (Walker 1996). The precipitation patterns of the arid and
semi-arid regions of the United States vary. In the Great Basin desert, precipitation is relatively
uniform throughout the year compared to the other North American deserts. The Mojave Desert
has a winter rainy season. In the Chihuahuan Desert, rainfall is predominantly in the summer
(June to September) with occasional winter rains. The Sonoran Desert experiences extremely
rare precipitation at low elevations but more precipitation at higher elevations (Lee et al. 2011).
In addition to increased runoff, human activity can impair the quality of stormwater. During dry
periods, oils, pesticides and other organic pollutants; sediments; animal waste; and trash can
accumulate. When precipitation occurs, runoff flows across the land, carrying a pulse of wastes
and pollutants into receiving waters, which is a particularly acute problem when precipitation
events are rare, as in hot desert climates (USEPA 2010b).
15
-------
Chapter 2.
Arid Green Infrastructure Resources at EPA
EPA provided resources at the program and regional levels to policymakers and practitioners to
support the application of green infrastructure for stormwater management and conservation. This
chapter focuses on EPA's support for green infrastructure implementation in arid and semi-arid
regions of the United States. Technical assistance projects, modeling and decision support tools,
operations and maintenance guidance, design and implementation guidance, and funding
opportunities developed by EPA program offices are described, including resources that focus on
arid and semi-arid regions, where applicable. This chapter also surveys resources offered by EPA
program offices that provide information about green infrastructure performance, contributions to
climate resiliency, current intramural and extramural research efforts, benefits, cost-benefit
analyses, policy guides and tools, and integration into federal regulatory programs. An overview
is provided of EPA collaborations supporting green infrastructure. Finally, projects and resources
implemented and developed by EPA regional offices in regions that include areas with arid and
semi-arid climates are described in this chapter.
2.1. Green infrastructure program
EPA provides strong support for and promotes the benefits of using green infrastructure for cities
and wastewater treatment plants in protecting drinking water supplies and public health,
mitigating overflows from combined and separate sewers, and reducing stormwater pollution
(USEPA 2007). EPA has collaborated with other agencies and organizations to develop the
Managing Wet Weather With Green Infrastructure: Action Strategy (USEPA 2008b). A Web-
based green infrastructure resource center to assist communities with building, designing and
implementing green infrastructure practices that are in compliance with regulatory guidelines also
has been established at EPA (USEPA 2016b). Resources available on this website include
educational materials; modeling tools; design, implementation, and operations and maintenance
guidelines; and information about collaborations between EPA and its partners to advance the
implementation of green infrastructure practices. EPA's Green Infrastructure Wizard (GlWiz) is
a Web application that provides access to EPA green infrastructure tools and resources (USEPA
2015n). Also posted on the website is Tools, Strategies and Lessons Learned From EPA Green
Infrastructure Technical Assistance Projects, which summarizes green infrastructure solutions for
stormwater management challenges of municipalities (USEPA 2015r). Technical assistance
products produced by communities located in arid and semi-arid regions and for which EPA
provided technical assistance to address barriers to using green infrastructure and share lessons
learned are presented in Table 2-1.
2.1.1. Plan and maintain green infrastructure
Building sufficient green infrastructure requires planning, and EPA has several resources for
implementing green infrastructure in the community, including modeling tools that support
planning and design decisions, guidance for operations and maintenance, details on design and
implementation, and information on available funding sources.
16
-------
Table 2-1. EPA technical assistance projects in arid and semi-arid regions
Project Type
EPA
Region
Project Name
City
State
Conceptual design
6
Imperial Building Site Design
Albuquerque
NM
Conceptual design
6
Pueblo de Cochiti Green Infrastructure Concept
Design
Pueblo de Cochiti
NM
Conceptual design
8
Conceptual Green Infrastructure Design for the
Blake Street Transit-Oriented Development Site,
City of Denver (EPA-830-R-13-002)
Denver
CO
Conceptual design
9
Building Resilience to Drought in Ozone Park (EPA-
832-R-15-010)
Santa Monica
CA
Conceptual design
10
Fairview Avenue Green Street Conceptual Design
(EPA-832-R-15-011)
Boise
ID
Guidance
8
Green Infrastructure Checklists and Renderings
Denver
CO
development
Guidance
9
Tools to Promote Green Infrastructure
Pima County
AZ
development
Implementation in Arid and Semi-Arid Regions
Policy review/
9
Green Infrastructure Barriers and Opportunities in
Phoenix
AZ
recommendations
Phoenix, Arizona (EPA-830-R-13-005)
Policy review/
9
Green Infrastructure Barriers and Opportunities in
Los Angeles
CA
recommendations
the Greater Los Angeles Region (EPA-833-R-13-001)
Source: Adapted from USEPA (2015r).
2.1.1.1. Modeling and decision support tools
The modeling tools provided by EPA allow users to predict environmental outcomes of different
green infrastructure design and management approaches. The models range from site to
watershed scale. Outputs include runoff volume, runoff rate, pollutant loading and cost. Models
range from simple sizing and cost spreadsheets that produce quick estimates of cost and
performance to simple models for use as screening- and planning-level tools (e.g., EPA National
Stormwater Calculator) to complex models (e.g., EPA Stormwater Management Model
[SWMM] with Low-Impact Development [LID] Controls). These models require such inputs as
soil conditions, vegetation characteristics, topography, impervious cover, precipitation,
evaporation, land use, land cover and water costs, allowing them to be applied to specific
locations in the arid and semi-arid West and Southwest (USEPA 2015j).
Regarding cost-effectiveness, EPA's Watershed Management Optimization Support Tool
(WMOST) is a tool for local water resources managers and planners to screen potential water
resources management options, including green infrastructure, across their watershed or
jurisdiction for cost-effectiveness as well as environmental and economic sustainability
(Detenbeck et al. 2016).
For land use change, EPA's Automated Geospatial Watershed Assessment (AGWA) tool is
designed to help manage and analyze watershed water quantity and quality. AGWA provides
qualitative estimates of runoff and erosion relative to landscape change (USEPA 2016a). AGWA
was tested for its ability to simulate stormwater runoff responses in semi-arid landscapes
(Korgaonkar et al. 2014).
17
-------
2.1.1.2. Operations and maintenance
Recognizing that green infrastructure requires regular inspections and maintenance for effective
operation, EPA also provides resources about what to look for when inspecting green
infrastructure and how frequently to conduct maintenance activities. These resources provide
general guidance rather than being targeted for arid and semi-arid conditions (USEPA 20151).
2.1.1.3. Design and implementation
EPA has compiled resources for design and implementation of green infrastructure, including
design manuals, information about addressing common design challenges, implementation
information and homeowner resources. One of the design manuals, Green Infrastructure for
Southwestern Neighborhoods (MacAdam 2012), is directly applicable to the arid and semi-arid
United States. Some of the design challenges also are applicable to the West and Southwest.
Information is provided about addressing problems with poor-quality urban soils (applicable to
cities in arid and semi-arid regions) and limited water resources for irrigation. For limited water
resources, possible solutions include planning a water budget, integrating low-water-use plants,
using efficient irrigation systems, considering soil amendments, using mulches, and maintaining
the xeriscape. For implementation, references on lessons learned from construction mistakes and
ensuring practices are built as designed are provided. In addition, EPA cites examples of guides
for homeowners that have been produced by state and local governments, although none of the
examples are specific to arid or semi-arid regions (USEPA 2016c).
2.1.1.4. Funding opportunities
Federal funding sources and tools to understand available funding opportunities from local,
government and nonprofit organizations are provided by EPA. Federal agencies with grant and
assistance programs include the U.S. Departments of Agriculture, Energy, Housing and Urban
Development, Interior, and Transportation; the National Oceanic and Atmospheric
Administration; the U.S. Economic Development Administration; and EPA (USEPA 2015i).
General funding tools with recent information that may be relevant to arid and semi-arid regions
of the United States include Getting to Green: Paying for Green Infrastructure, Finance Options
and Resources for Local Decision-Makers (USEPA 2014) and a guide on community-based
public-private partnerships for green infrastructure (USEPA Region 3 2015). In addition, the
Water Infrastructure Finance and Innovation Act (WIFIA) program provides low interest rate
financing for the construction of water and wastewater infrastructure under the Water Resources
Reform and Development Act (WRRDA) of 2014.1
2.1.2. Learn about green infrastructure
In addition to construction resources, EPA's green infrastructure website provides information
for learning more about implementation. These resources include information about what green
infrastructure is and overcoming barriers to green infrastructure, performance of green
infrastructure practices, the role of green infrastructure in climate resiliency, research on green
infrastructure, the benefits of green infrastructure, cost-benefit resources, policy guides,
1 Water Resources Reform and Development Act of 2014, Pub. L. No. 113-121, 128 Stat. 1193 (2014).
18
-------
integrating green infrastructure into federal regulatory programs, and a green infrastructure
webcast series.
2.1.2.1. Basics
EPA provides information about the types of green infrastructure elements that can be used by
communities. Green infrastructure elements include downspout disconnection, rainwater
harvesting, rain gardens, planter boxes, bioswales, permeable pavements, green streets and
alleys, green parking, green roofs, urban tree canopies, and land conservation (USEPA 2015s).
Examples of communities in arid and semi-arid regions implementing these elements include the
Los Angeles Downspout Disconnection Program (City of Los Angeles 2009) and Elmer Avenue,
a green street project in Los Angeles (Green 2010).
EPA also provides information on overcoming the barriers that municipalities and developers
might face when determining whether green infrastructure is appropriate in a particular context.
For municipalities, some of the barriers are perceptions that performance is unknown, costs of
green infrastructure are high, the regulatory community is resistant to green infrastructure, green
infrastructure conflicts with principles of smart growth, and green infrastructure conflicts with
water rights laws; unfamiliarity with maintenance requirements and costs; conflicting codes and
ordinances; and lack of staff and resources. EPA offers general strategies and resources to
address these barriers. Regarding costs, EPA has compiled resources on cost-benefit analyses for
green infrastructure (USEPA 2015e).
Developers may face skepticism about long-term performance and a perception of higher costs.
Some of the same strategies suggested for municipalities to address perceptions that performance
is unknown and costs are high apply to developers.
2.1.2.2. Performance
With respect to performance, EPA provides resources on databases and summary reports on
green infrastructure practices in general, as well as resources on the performance of particular
green infrastructure design elements (green roofs, permeable pavements, rainwater harvesting,
rain gardens and planter boxes, bioswales, urban tree canopies and constructed wetlands) and
watershed-scale studies of green infrastructure performance. Current databases include the
International Stormwater Best Management Practices (BMP) Database, which provides BMP
performance summaries (WERF 2016). EPA also summarized research on the performance of
several green infrastructure practices in a report titled Green Infrastructure for Stormwater
Control: Gauging Its Effectiveness With Community Partners, which includes data on costs and
performance conducted in Denver and Phoenix (USEPA 2015h).
2.1.2.3. Climate resiliency
EPA provides information on using green infrastructure to improve climate resiliency. Green
infrastructure practices can help communities manage flooding, prepare for drought, reduce the
urban heat island effect, lower building energy demands, spend less energy managing water, and
protect coastal areas. All of these effects of climate change will be relevant to arid and semi-arid
parts of the United States (USEPA 2015g).
19
-------
2.1.2.4. Research
EPA provides resources on research on green infrastructure conducted by the scientific,
regulatory and development communities. Topics include performance, urban stormwater
impacts, surface water impacts, ground water impacts, air quality impacts, green infrastructure
and climate, green infrastructure and wildlife conservation, and the economics of green
infrastructure (USEPA 2015p).
EPA scientists and engineers conduct research on topics including assessments of green
infrastructure impacts on watersheds; best practices for design, operation and maintenance of
green infrastructure; decision-support guidance for sustainable communities; and technical
assistance with green infrastructure (USEPA 2015m).
EPA's Office of Research and Development (ORD) supports research on green infrastructure.
ORD's Safe and Sustainable Water Resources (SSWR) research program includes in its strategic
plan (2016-2019) a project focusing on green infrastructure models and tools and a project on
green infrastructure information and guidance based on community partnerships (USEPA
2015q). Through its Science To Achieve Results (STAR) grant program, ORD also is supporting
research on a decision-support tool for life-cycle cost assessment and optimization of green, grey
and hybrid stormwater infrastructure (USEPA 2016e) and the creation of a National Center for
Sustainable Water Infrastructure Modeling Research, which will facilitate sharing of green
infrastructure tools and research advancements with local communities and stakeholders
(USEPA 2016g).
2.1.2.5. Benefits
Green infrastructure benefits are described by EPA (USEPA 2015b). These include benefits for
water quality and quantity, air quality, climate resiliency, habitat and wildlife, and communities.
General resources on these topics may apply to arid and semi-arid regions. An example of a
project on water supply and green infrastructure in the semi-arid West is the Colorado Climate
Preparedness Project (Western Water Assessment 2011).
2.1.2.6. Cost-benefit resources
Cost-benefit resources for green infrastructure compiled by EPA include resources on cost
analysis, cost-benefit analysis and tools. Resources include case studies and literature reviews.
General tools that estimate the economic benefits of green infrastructure also are provided
(USEPA 2015e). One case study includes locations in the arid and semi-arid United States
(Cheyenne, Wyoming; Fort Collins, Colorado; and Glendale, Arizona) (McPherson et al. 2005).
2.1.2.7. Policy
Policy guides and tools are available to help municipalities in implementing green infrastructure.
EPA has produced an online municipal handbook that describes funding options, retrofit policies,
green street programs and policies, rainwater harvesting policies, and incentive mechanisms
(USEPA 2015k). Other policy guides include guides on sustainability and case studies of policies
adopted by municipalities to promote green infrastructure use, including those of San Jose, Santa
Monica and Emeryville in California; Olympia in Washington; and Wilsonville in Oregon
(USEPA 2010a). Additional policy tools cited are EPA's Water Quality Scorecard and a toolbox
20
-------
created by the Water Environment Research Foundation on effectively communicating and
implementing green infrastructure concepts (USEPA 2015o).
2.1.2.8. Regulations
EPA provides information related to integrating green infrastructure into federal regulatory
programs. These programs include MS4s, CSOs and TMDLs. EPA supports incorporating green
infrastructure approaches into NPDES permits and CSO remedies and has produced fact sheets
on green infrastructure permitting and enforcement. California is an example of a state that has
integrated green infrastructure into its MS4 permits (USEPA 2016f). For CSO control plans and
remedies, EPA has compiled documents to quantify green infrastructure contributions to CSO
control plans (USEPA 2015c); created an index of EPA enforcement actions incorporating green
infrastructure; and developed a template, Long-Term Control Plan-EZ (commonly known as
LTCP-EZ), for small municipalities to use to assess the potential for green infrastructure controls
to eliminate or reduce CSOs. Regarding TMDLs, EPA has developed a fact sheet on
incorporating green infrastructure concepts into TMDLs and provides information on a case
study in EPA Region 1 in which green infrastructure was used to retain and treat stormwater to
reduce phosphorus loads (USEPA 2016f).
2.1.2.9. Webcasts
EPA is producing a series of webcasts aimed at public officials and practitioners who want to
begin implementing green infrastructure or enhance existing green infrastructure programs
(USEPA 2016d). The library of past webcasts includes Green Infrastructure for Arid
Communities, which discusses green infrastructure implementation strategies in southern
California and Tucson, Arizona (USEPA 2015f). EPA also has produced webcasts on green
infrastructure as part of its Watershed Academy series (USEPA 2016d).
2.1.3. Collaborate with green infrastructure partners
EPA partners with organizations and communities to foster adoption of green infrastructure.
Collaborations include issuing the Campus RainWorks Challenge to undergraduate and graduate
students, participating in the Green Infrastructure Collaborative, recognizing community partners
in each of EPA's regions, providing technical assistance to communities, and engaging citizens
and municipalities through the Soak Up the Rain program.
2.1.3.1. Campus RainWorks Challenge
EPA's Office of Water sponsors an annual competition for undergraduate and graduate students
to design innovative green infrastructure projects for their campuses. Students compete in two
categories: Master Plan and Demonstration Project. A past winner from a semi-arid region is the
University of Arizona (Second Prize, 2012). The American Society of Landscape Architects,
American Society of Civil Engineers and Water Environment Federation assisted in the judging
and outreach for the 2015 Challenge (USEPA 2015a).
2.1.3.2. Green Infrastructure Collaborative
The Green Infrastructure Collaborative is a network-based learning alliance formed in 2014 to
help communities implement green infrastructure. Federal members include EPA and the
U.S. Departments of Agriculture, Defense, Energy, Housing and Urban Development, Interior,
21
-------
and Transportation. Nonfederal members include more than 20 academic, nongovernmental and
private sector organizations. EPA has compiled resources and tools produced by member
organizations to advance green infrastructure implementation, including the Natural Resources
Defense Council's issue paper on how implementing green infrastructure in California cities can
address emerging water resource and climate challenges (NRDC 2009; USEPA 2015d).
2.1.3.3. EPA is supporting green infrastructure
In support of green infrastructure, EPA has produced policy memoranda encouraging the use of
green infrastructure to meet regulatory requirements. EPA's Green Infrastructure Strategic
Agenda 2013 describes the actions that the Agency intends to take to promote green
infrastructure implementation, focusing on federal coordination, Clean Water Act (CWA)
regulatory support, research and information exchange, funding and financing, and capacity
building (USEPA 2013). EPA also supports a voluntary integrated planning approach by
municipalities to propose to meet multiple CWA requirements by identifying efficiencies from
separate wastewater and stormwater programs and prioritizing projects to address the most
serious water quality issues first. Implementing green infrastructure can lead to more sustainable
and comprehensive solutions to these issues, and EPA has developed a framework to provide
guidance on implementing an integrated planning approach (USEPA 2012). The Agency also has
recognized community partners in each of EPA's regions for their commitment to green
infrastructure, including Denver (Region 8), Los Angeles (Region 9) and Puyallup, Washington
(Region 10).
In addition, EPA provides technical assistance to communities to advance the adoption of green
infrastructure locally and develop knowledge and tools for a national audience. The program
focuses on overcoming technical, regulatory and institutional barriers to green infrastructure and
sharing lessons learned. Western and Southwestern jurisdictions that received technical
assistance include Ada County, Idaho; Albuquerque, New Mexico; Denver; Pueblo de Cochiti,
New Mexico; Santa Monica; Phoenix and Los Angeles (see Table 2-1). The results of the
program are summarized in Tools, Strategies and Lessons Learned From EPA Green
Infrastructure Technical Assistance Projects (USEPA 2015r, 2016k).
2.1.3.4. Soak Up the Rain program
EPA's Soak Up the Rain program offers citizens and municipalities access to information on
green infrastructure practices, communication tools, and resources on green infrastructure
benefits and funding as well as opportunities to share their stories with others (USEPA 2016h).
2.2. Regional programs
EPA Regions 6, 8 and 9 have developed their own materials on green infrastructure relevant to
their respective locales. Resources include demonstration projects, region-specific guidance,
funding opportunities and training events.
2.2.1. Region 6
EPA Region 6 has published information on its website on the basics of green infrastructure,
green infrastructure projects within the region, funding opportunities for green infrastructure
projects, and training events (USEPA Region 6 2016a). Projects in New Mexico are described in
22
-------
Table 2-2. Other projects in Texas and Oklahoma are not located in the semi-arid parts of the
states (USEPA Region 6 2016c). Funding opportunities, including local loans and grants, are
provided on the EPA Region 6 website (USEPA Region 6 2016b). Training events also are listed
on Region 6's website (USEPA Region 6 2016d).
Table 2-2. EPA Region 6 green infrastructure demonstration projects in the desert
Southwest
Project
Las Cruces Dam Restoration Project
Stormwater Demonstration Median
Railyard Park and Plaza
Source: USEPA Region 6 (2016c).
2.2.2. Region 8
The unique environment of the semi-arid West requires attention for implementing green
infrastructure practices. The climate is dry with intermittent and unpredictable rainfall, and
temperature differences between summer and winter are large with rapid freeze/thaw cycles.
These conditions require the use of native plants that are drought tolerant and low maintenance
(USEPA Region 8 2016).
2.2.2.1. Demonstration projects
Green infrastructure projects implemented in EPA Region 8 include green roofs, rain gardens,
bioswales, bioretention ponds, porous pavement and rainwater harvesting (see Table 2-3). When
implementing these practices, conditions specific to the region must be addressed and region-
specific benefits may be realized.
Green roofs provide unique benefits in this region. The insulating properties of green roofs
reduce energy expenses during the region's very warm summers and very cold winters. Because
of high elevations, buildings in Region 8 are exposed to intense solar radiation, which damages
the roof membrane. Vegetation helps protect the membrane from sun damage.
Flash floods from large storms are of concern in this region. Permeable pavement helps absorb
stormwater runoff. Many types of porous pavements are more durable than traditional nonporous
concrete in the face of the routine freezing and thawing cycles characteristic of Region 8 winters.
Regarding rainwater harvesting, water laws differ among the Region 8 states. Colorado and Utah
require permits for harvesting rainwater, whereas Montana, North Dakota, South Dakota and
Utah do not (USEPA Region 8 2016).
2.2.2.2. Resources
EPA Region 8 has compiled a list of resources on green infrastructure and LID. Some are
general, whereas others are specific to locales in Region 8 or semi-arid and arid climates. The
region-specific resources are described in Chapter 3 of this report under state and
municipal/county policy initiatives and guidance (USEPA Region 8 2016).
Description
Constructed wetlands
Rain harvesting
Rain storage and harvesting
City State
Las Cruces NM
Santa Fe NM
Santa Fe NM
23
-------
Table 2-3. EPA Region 8 green infrastructure demonstration projects in the semi-arid west
Project
Description
City
State
Stapleton Greenway Park
Bioretention pond
Denver
CO
Bioretention pond
Bioretention pond
Fort Carson
CO
Stapleton Quebec Square shopping center
Bioswale
Denver
CO
South Platte River
Bioswale
Denver
CO
Vegetative swale
Bioswale
Fort Carson
CO
EPA Region 8 building
Green roof
Denver
CO
Denver Museum of Contemporary Art
Green roof
Denver
CO
Denver Botanic Gardens
Green roof
Denver
CO
REI Parking Garage
Green roof
Denver
CO
Church of Jesus Christ of Latter-Day Saints Conference Center
Green roof
Salt Lake City
UT
Denver Housing Authority
Permeable pavement
Denver
CO
South Platte River path
Permeable pavement
Denver
CO
Urban Drainage and Flood Control District
Permeable pavement
Denver
CO
Odell Brewery
Permeable pavement
Fort Collins
CO
CTL Thompson
Permeable pavement
Fort Collins
CO
Northern Plains Resource Council building
Permeable pavement
Billings
MT
Denver Housing Authority
Rain garden
Denver
CO
Environmental Center for the Rockies
Rain garden
Boulder
CO
Regis University
Rain garden
Denver
CO
Stapleton soft running path
Rain garden
Denver
CO
TAXI Development
Rain garden
Denver
CO
New Belgium Brewery
Rain garden
Fort Collins
CO
University of Utah
Rain garden
Salt Lake City
UT
Antiques Central
Rain garden
Cheyenne
WY
Source: USEPA Region 8 (2016).
2.2.3. Region 9
EPA Region 9 supports the use of green infrastructure to reduce stormwater runoff, entrainment
of pollutants in runoff, and discharge of pollutants into receiving waters. Financial support for
promoting the use of green infrastructure and LID in the Pacific Southwest is available from the
Section 319 Nonpoint Source Management Program, which was established to support state,
territory and tribal efforts to address nonpoint source pollution under Section 319 of the CWA;
the Urban Waters Small Grants program; the Green Infrastructure Technical Assistance program;
the Clean Waters Act State Revolving Fund; and the National Estuary Program. Examples of
past projects funded by these opportunities in Mediterranean and semi-arid locations in EPA
Region 9 are the Los Angeles River Street Biofiltration Project (Section 319 Nonpoint Source
Management Program), the Redondo Beach Alta Vista Park Diversion and Reuse Project and
Hermosa Beach Strand Infiltration Trench (Clean Water Act State Revolving Fund), and the
Morro Bay National Estuary Program (National Estuary Program).
Regarding the use of green infrastructure and LID in MS4 permits, the cities of Long Beach and
Salinas, California, renewed MS4 permits containing provisions requiring the use of LID. The
city of Santa Monica updated its urban runoff pollution ordinance to require new development
and redevelopment projects to use LID. The San Diego and Los Angeles County regional
permits also include provisions requiring LID.
24
-------
EPA Region 9 also has compiled material on LID in the Pacific Southwest, including reports
from the Natural Resources Defense Council on rooftop rainwater harvesting (Garrison, Kloss
and Lukes 2011) and addressing water resource and climate challenges in California (NRDC
2009); the California LID Portal (CSQA 2016), which was established by the California
Stormwater Quality Association and contains tools and other resources on LID; and a video by
the California State Water Resources Control on slowing the flow of stormwater (CEPA 2016;
USEPA Region 9 2016).
25
-------
Chapter 3.
Policy Initiatives and Guidance to Address Drought and Water Sustainability
Through Green Infrastructure
This chapter outlines policy initiatives and guidance that promote and help implement green
infrastructure. At the federal level, direction to apply green infrastructure to improve federal
sustainability has been provided by executive order. States in arid and semi-arid regions of the
United States have issued guidelines on implementing green infrastructure practices.
Municipalities and counties have developed implementation guidelines as well. In addition,
implementation guidance has been developed through partnerships between academia, EPA
regional offices and municipalities. Finally, nongovernmental agencies are active in promoting
green infrastructure and LID in arid and semi-arid regions, mainly through advocacy and
education.
3.1. Federal
Implementation of green infrastructure is part of recent federal initiatives to plan for
sustainability. These activities build on the commitments made by federal agencies as part of the
Green Infrastructure Collaborative.
3.1.1. Executive Order 13693
On March 19, 2015, President Barack Obama signed Executive Order (EO) 13693, Planning for
Federal Sustainability in the Next Decade 2 As part of the water and stormwater management
goals of the EO, the head of each federal agency was directed to "improve agency water use
efficiency and management, including stormwater management by ... installing appropriate green
infrastructure features on federally owned property to help with stormwater and wastewater
management," beginning in fiscal year 2016.
3.1.2. Council on Environmental Quality
Under EO 13693, the Chair of The White House Council on Environmental Quality (CEQ) was
directed to establish temporary interagency working groups to provide recommendations on
implementing the goals of the order, including the goal of installing green infrastructure. The
CEQ issued implementation instructions for EO 13693 on June 10, 2015 (CEQ 2015). The
instructions cite Section 438 of the Energy Independence and Security Act of 2007 (EISA),
which mandates that construction projects for new federal facilities with a footprint of at least
5,000 square feet "manage stormwater and preserve and/or restore natural site hydrology."3 As a
target for achieving the green infrastructure goal of EO 13693, the CEQ identifies
implementation of green infrastructure and stormwater best practices on new federal construction
projects to the maximum extent technically feasible, per EISA Section 438 requirements. As a
second target, the CEQ encourages federal agencies to update the commitments they made to the
2 Exec. Order No. 13693, 80 Fed. Reg. 57 (Mar. 25, 2015).
3 Energy Independence and Security Act of 2007, 42 U.S.C. § 17094 (2007).
26
-------
Green Infrastructure Collaborative (GIC 2014a, 2014b) and develop plans to meet those
commitments.
The implementing instructions cite the information that EPA has made available on its green
infrastructure website (USEPA 2016b) regarding federal requirements for green infrastructure, as
well as strategies to plan and implement green infrastructure projects. This information is
described in detail in Chapter 2 of this report. In addition, the CEQ cites information and guidance
on green roofs available from the U.S. General Services Administration (GSA 2015).
3.2. State guidance
Some states in arid and semi-arid regions of the United States have issued guidelines on
implementing green infrastructure practices. These include a guideline for retrofitting streets,
right-of-ways and parking lots issued by Arizona; a guidance for implementing LID practices
allowed under an NPDES municipal stormwater permit in Washington; and rainwater haivesting
guidelines issued by Texas.
3.2.1. Arizona
The Arizona Department of Environmental Quality collaborated with EPA to fund the
development of Green Infrastructure for Southwestern Neighborhoods (MacAdam 2012). The
manual provides guidelines for retrofitting
existing neighborhood streets, right-of-ways
and parking lots with green infrastructure
practices. It describes the types of practices that
can be implemented streetside (see Figure 3-1),
in streets and in parking lots. Some of the
problems addressed in the manual are unique to
the Southwest, such as minimal shading by
vegetation in streets and parking lots, the
presence of degraded ephemeral channels called
washes or arroyos that flow only periodically,
low and variable precipitation, the tendency of
the trunks and stems of many desert plants to
rot when standing in water or where wet mulch
lays against their trunks or stems for extended
periods, and the design of many Southwest
streets to convey stormwater. The manual
includes information on creating a water budget
for bioretention areas, using Tucson as an
example. It also includes maintenance
guidelines, such as using native, drought-
adapted plants and climate-appropriate watering Figure 3-1. A curb cut draws stormwater
schedules, as well as pruning native trees and from the street into a bioretention basin in
shaibs to natural growth forms.
the right of way.
Source: MacAdam (2012). Photo credit: Watershed
Management Group.
27
-------
3.2.2. Washington
Washington State has developed an Eastern Washington Low Impact Development Guidance
Manual (Carlson et at, 2013). The manual describes the varied climate of eastern Washington,
the driest regions being the Central Basin, which averages 14 to 16 inches of precipitation
annually, and the intermountain regions of Okanogan, Spokane, and the Palouse, which average
12 to 22 inches of precipitation annually. The climate regions of eastern Washington are shown
in Figure 3-2.
CofMlle
Region 4
Northeastern and
Blue Mountains-
Region 3
Okanogan-
Spokane-
Palouse
Region 2
Centra! Basin
Mountains
Figure 3-2. The climate regions of eastern Washington
Source: Carlson et al. (2013).
Aridity and cold temperatures are design considerations in eastern Washington. The manual
points out some of the unique considerations associated with arid and semi -arid regions,
including intense, relatively infrequent storms; high evapotranspiration rates; sparse vegetation
that leaves soil prone to erosion; and development patterns characterized by low density and
large amounts of impervious surface area. Plant selection must consider tolerance for drought,
extreme heat, and winter conditions that include snow cover and freezing.
In 2012, the Washington Department of Ecology issued the Eastern Washington NPDES
Municipal Stormwater Permit, which takes steps toward implementing LID practices. The
Washington Low Impact Development Guidance Manual is a tool to provide local jurisdictions
with design guidance to implement the LTD projects allowed under the permit. The manual
describes the planning and design process for LID projects. LID BMPs—including the green
infrastructure BMPs of bioretention, trees, vegetated roofs, and permeable pavement rain
28
-------
harvesting—also are described and design criteria are presented. In an appendix, the manual
provides a list of native and nonnative trees, shrubs, grasses, perennials, wildflowers and
groundcover suitable for use in bioretention projects in eastern Washington.
3.2.3. Texas
The Texas Water Development Board has prepared the Texas Manual on Rainwater Harvesting
(TWDB 2005). The manual includes information on rainwater harvesting system components;
water quality and treatment; system sizing; guidelines on best practices, building codes, cistern
design and backflow prevention; cost estimation; and tax and other financial incentives. Rainfall
data for representative Texas cities and case studies also are included. The manual recognizes
that rainfall in some parts of Texas may not be sufficient to meet domestic needs (see
Figure 3-3).
26
48
a
,S6
48
48
20
¦44
'40
'36
22
26
Figure 3-3. Average annual precipitation in Texas, in inches.
Source: TWDB (2005).
29
-------
Municipal/county guidance
Municipalities and counties in the semi-arid and arid regions of the United States have issued
guidance on implementing LID and green infrastructure practices tailored to local climate and
conditions, including local topography and geography. Design and maintenance considerations
include the existence of wet and dry seasons, hot and arid conditions that require drought-
tolerant vegetation, and site-specific hydrology.
3.2.4. Denver, Colorado
Denver has issued guidance on aesthetically enhancing detention and water quality ponds
(Mancini et al. 2010). The guide is intended for design professionals so that aesthetic elements
may be incorporated into projects in the early stages of design and cites the advantages of green
infrastructure in reducing and delaying stormwater runoff volumes, as well as reducing
pollutants in stormwater. The guide includes tools on siting and functionality, physical character
and architectural elements, and landscape design to enhance sites.
3.2.5. Los Angeles, California
The city of Los Angeles has issued the Development Best Management Practices Handbook,
Part B: Planning Activities (City of Los Angeles 2011) to reflect the LID requirements that took
effect on May 12, 2012. The purpose of the handbook is to assist developers in complying with
the requirements of the Development Planning Program regulations of the city's stormwater
program. Its target audience is developers, designers, contractors and homeowners, as well as
city staff members who are engaged in plan checking, permitting and inspections related to land
development activities. The handbook provides background material, including the legal
framework behind incorporating LID BMPs into stormwater management; describes the project
review and permitting process; presents information on stormwater management measures;
provides guidance on BMP prioritization and selection; and describes offsite mitigation
measures.
The handbook specifies the prioritization and selection of BMPs. The prioritization of BMPs is
(1) infiltration systems, (2) stormwater capture and use, (3) high efficiency biofiltration or
bioretention systems, and (4) a combination of any of the aforementioned BMPs. Infiltration
feasibility screening and capture and use feasibility screening criteria are provided in the
handbook. Design specifications for infiltration and capture and use BMPs also are presented.
Among the design requirements is a specification that drought and flood resistant plant species
native to California be selected when possible.
3.2.6. Pima County, Arizona
Tucson and Pima County produced a technical guidance for the use of LID and green
infrastructure throughout Pima County (City of Tucson 2015). It is intended for the professional
community and provides information on the site assessment, planning and design process;
specific LID site planning practices; structural green infrastructure practices; and common green
infrastructure components. Appendices include an analysis of University of Arizona rainfall data,
sizing features to support vegetation based on University of Arizona evapotranspiration and
rainfall data, design volume calculations based on University of Arizona rainfall data to size
30
-------
green infrastructure and LID practices, and a list of plants recommended for the hot and dry
conditions of the desert Southwest.
3.2.7. San Diego, California
The San Diego Low Impact Development Design Manual (Tetra Tech 2011) provides guidance
to ensure that project designs effectively carry out the goal of stormwater regulation and are
reliable and cost-effective to maintain. The manual includes information on site assessment,
planning and design; LID selection; implementation considerations; and appendices on BMP
sizing, design guidance and templates, fact sheets, recommended plants, and inspection and
maintenance.
Region-specific guidance related to the local climate, topography, hydrology, surface water and
soils is provided. This includes suggestions to incorporate native vegetation or vegetation that is
resilient to water shortages and periodic flooding into LID practices; mimicking typical
predevelopment hydrology, which was characterized by the formation of vernal pools by smaller
storms and water from larger storms flowing through canyons; increasing dry-weather base flows
in streams and rivers to improve surface water quality; and accounting for the clayey soils of the
region when conducting analyses of infiltration for LID practices.
3.3. Collaborative guidance
A design manual for green roofs in arid and semi-arid regions is an example of a
multi stakeholder collaboration. Design Guidelines and Maintenance Manual for Green Roofs in
the Semi-Arid and Arid West (Tolderlund 2010) was produced through a collaboration of Green
Roofs for Healthy Cities, the City and County of Denver, EPA Region 8, the Urban Drainage
and Flood Control District, and Colorado State University. Some of the challenges identified for
green roofs in arid and semi-arid regions are low annual precipitation, low average relative
humidity and high solar radiation. These challenges require consideration of specific design
strategies, plant selection, growing media and supplemental irrigation requirements. The manual
provides recommendations and requirements regarding green roof design, implementation and
maintenance for arid and semi-arid regions. It is intended for use by professionals and local
jurisdictions. It presents information on design and implementation, leak detection, integration of
solar panels on green roofs, insurance and liability, maintenance, costs, and case studies.
3.4. Nongovernmental agency guidance
Nongovernmental agencies are active in promoting green infrastructure and LID in arid and
semi-arid regions. Examples are Amigos de los Rios, AridLID.org, the Council for Watershed
Health, and the Watershed Management Group.
Amigos de los Rios works in East County Los Angeles to protect and restore open spaces. It
specializes in developing parks in park-deficient neighborhoods using LID landscape practices.
Some of the green infrastructure practices used by Amigos are on-site water filtration, bioswales
and low-water-use irrigation, as well as drought-tolerant and native-plant landscaping (Amigos
de los Rios 2016).
AridLID.org is a website that provides information and resources on LID in arid environments,
particularly the southwestern United States. It is administered by the Ciudad Soil and Water
31
-------
Conservation District, a political subdivision of the state of New Mexico that has among its
responsibilities control and prevention of soil erosion, prevention of sediment and floodwater
damage, and conservation of water. Originally, it was the website of the 2010 Albuquerque Area
Green Infrastructure & Low Impact Development Workshop, and it catalogs presentations of past
Arid LID Conferences (Ciudad Soil and Water Conservation District 2016).
The Council for Watershed Health initiated the Water Augmentation Study (WAS), a long-term
research project to explore the potential for increasing local water supplies in the Los Angeles
region and reducing urban runoff pollution by increasing infiltration of stormwater runoff. The
study currently is developing a regional strategy for developing stormwater as a new source of
water for southern California (Council for Watershed Health 2015b).
The Watershed Management Group focuses on improving desert ecosystems. It is based in
Tucson and works in the Phoenix Valley, other southern Arizona communities, and the border
region between Arizona, United States, and Sonora, Mexico. It provides educational programs,
offers water-harvesting landscape services, and engages in advocacy and stewardship to restore
local rivers (Watershed Management Group 2015).
32
-------
Chapter 4.
Current Research in the Application of Green Infrastructure
for Stormwater Management and Conservation in Arid and Semi-Arid Regions
A literature survey was conducted to determine the state of the science for the application of
green infrastructure to manage and conserve stormwater in arid and semi-arid regions. In this
chapter, effects on water quality, stormwater infiltration, water conservation and habitat
preservation, as well as design optimization and economic benefits, are described. The
information is presented first for BMPs used principally for managing stormwater to increase
infiltration close to the source of precipitation, and then for BMPs that conserve stormwater so
that it can be used to reduce potable water demands. Research on the stormwater management
BMPs has been performed at the scale of individual sites and watersheds, including monitoring
and modeling studies.
4.1. Methodology
A survey of the recent literature (2010-2016) was conducted to report on the current state of the
science for the application of green infrastructure practices in stormwater management and
conservation. Terminology for green infrastructure BMPs is not standardized; therefore, the
literature search was conducted using the various terms of art used to describe each of the
practices (e.g., "permeable pavement," "pervious concrete"). The geographic scope of the
literature review included U.S. and international field studies conducted under arid or semi-arid
conditions, as well as modeling studies of arid and semi-arid conditions (Koppen-Geiger Climate
Classification Group B). The geographic scope also included investigations conducted in
Mediterranean climates (Koppen-Geiger climate classifications Csa and Csb), which are typified
by a prolonged dry summer season, making the results of such studies applicable to xeric
climates. The climate zones for field and modeling study sites were confirmed using an online
database (CantyMedia 2016).
4.2. Stormwater management—site-scale research
Site-scale BMPs have been evaluated in arid, semi-arid and Mediterranean climates for
management of stormwater quantity and quality. Practices have been tested for their ability to
improve the quality of stormwater, including sediment, nutrients, bacteria, pesticides and other
organic compounds, and metals. The ability of vegetation to sequester carbon and nitrogen also
has been assessed. In addition, the role of BMPs in preventing soil erosion has been investigated.
For practices that use soil substrates, leaching of nutrients has been quantified. The ability of
BMPs to reduce runoff is a key performance parameter that has been evaluated under different
precipitation regimes. Field and modeling studies have been conducted to optimize BMP design
for climate and hydrological conditions found in dry regions. Maintenance needs, primarily
irrigation, have been tested.
4.2.1. Bioswales
The effects of bioswales on stormwater runoff quality have been assessed under Mediterranean
climate conditions. The practice has been tested for removal of fecal bacteria and effects on
33
-------
nutrient loading. The effects of bioswales in sediment removal have been evaluated as part of an
integrated bioretention system as described in the discussion of David et al. (2015) below.
4.2.1.1. Water quality improvement
The ability of bioswales to remove bacteria was assessed in a Mediterranean climate. As part of
California's Clean Beach Initiative, a vegetated wetland swale in Pacifica, California, was
constructed and tested for its effect on the loading of fecal indicator bacteria to beaches. The
swale reduced bacterial densities post- versus pre-project for total coliforms, fecal coliforms and
enterococci at shoreline sampling stations up to fourfold, but these results were not statistically
significant (Dorsey 2010).
4.2.1.2. Negative effects on effluent water quality
The effects of a high-capacity vegetated swale on effluent water quality of highway runoff were
measured recently in Santa Barbara, California, a Mediterranean climate. Orthophosphate
concentrations frequently were higher in effluent than influent, which the authors attribute to
possible leaching from plants (Jiang, Yuan and Piza 2015).
4.2.2. Green roofs
Research on adapting green roofs for use in arid and semi-arid regions has focused on
categorizing their performance in sequestering carbon and nitrate, assessing possible negative
impacts on runoff water quality, and optimizing stormwater retention. Design challenges for dry
climates include selecting plants that will grow under harsh conditions of heat and drought,
evaluating irrigation needs, optimizing substrate depth for weight and ability to support plant
growth, selecting the appropriate growth media, adding natural or artificial amendments to the
growth medium to increase water retention and enhance plant growth, considering the thermal
properties of the growth medium, and optimizing roof slope.
4.2.2.1. Carbon and nitrogen sequestration
The ability of green roofs to affect climate change by sequestering carbon has been studied in
semi-arid regions. Carbon and nitrogen sequestration potential of green roofs was found to
depend on the inorganic component of green roof substrates, as well as plant species, and varied
widely, with some substrates acting as net exporters of carbon and nitrogen over the course of
the 10-month experiment in a semi-arid region of Spain (Ondono, Martinez-Sanchez and Moreno
2016b). In a separate study, carbon and nitrogen fixation by the substrate and aboveground plant
material was higher in soil-amended substrates in simulated green roofs in a semi-arid region of
southeast Spain (Ondono, Martinez-Sanchez and Moreno 2016a).
4.2.2.2. Leaching from substrate
Rather than acting as sinks for carbon and nitrogen, green roofs can act as sources of
contaminants to stormwater runoff. In a review of the performance of field-scale LID/green
infrastructure systems in arid and semi-arid climates, the authors noted that although green roofs
have shown potential in reducing such pollutants as nitrogen and phosphorus because of
microbial processes and plant uptake, studies have shown conflicting results, particularly for
nitrogen (Jiang, Yuan and Piza 2015). In a laboratory microlysimeter study comparing compost
and peat amendments to green roof substrates, nitrate leaching on first flush was high for some
34
-------
types of compost amendments, but peat and garden waste compost exhibited minimal nitrate
leaching (slightly above 10 mg L"1 nitrate nitrogen), and the first flush only lasted for the first 50
to 100 mL effluent volume (Ntoulas et al. 2015). Contaminant concentrations in runoff were
shown to decrease over time in a 9-month study of the water quality of outflow from intensive
and extensive green roof systems in the Mediterranean climate of Adelaide, Australia.
Parameters such as pH, turbidity, nitrate, phosphate and potassium were higher in outflows from
intensive green roofs, however, compared with outflows from extensive green roofs
(Razzaghmanesh, Beecham and Kazemi 2014b). In another study in Adelaide that focused on the
effects of vegetation on outflow quality, pilot-scale green roofs were found to be a source of
pollutants, including salt, nitrate, nitrite, ammonia and orthophosphate, although vegetated green
roofs generally had better outflow quality than nonvegetated control beds. For vegetated beds,
the outflow water quality from intensive green roofs was better than from extensive green roofs.
Growing media with less organic matter had better outflow water quality (Beecham and
Razzaghmanesh 2015).
4.2.2.3. Stormwater retention
The effects of organic material amendments to substrates have been tested, comparing different
types of organic amendments for their physicochemical properties (i.e., porosity and water
retention capacity). In laboratory tests comparing peat with several types of compost
amendments available locally in a semi-arid environment, peat-amended substrate was found to
have increased moisture retention (Ntoulas et al. 2015).
The physicochemical properties of different inorganic substrate components have been
compared. In an investigation using "cultivation tables" in a semi-arid region of southeast Spain,
substrates made from compost-amended mixtures of silica, crushed bricks and clay-loam soil
were compared for physicochemical properties (i.e., water retaining capacity and porosity).
Compost-amended substrates containing sand or soil and crushed bricks were determined to have
acceptable water retaining capacity and high porosity (Ondono, Martinez-Sanchez and Moreno
2016b).
The effect of vegetation on the water retention capacity of green roofs has been evaluated under
Mediterranean climate conditions. In a 2-year study of model green roofs in Adelaide, Australia,
vegetation increased water retention by intensive and extensive green roof beds compared with
nonvegetated controls. The authors attributed this affect to the role of evapotranspiration in
increasing the stormwater retention capacity of green roofs, particularly for longer antecedent
dry weather periods (Beecham and Razzaghmanesh 2015). In Corvallis, Oregon, vegetated roofs
had significantly higher retention capacity during the dry summer than medium-only roofs.
Irrigation significantly decreased the retention capacity of vegetated and medium-only roofs
during the summer. During the rainy season, vegetation had no effect on stormwater retention
(Schroll et al. 2011).
The hydrologic response of green roofs has been evaluated in a Mediterranean climate.
Researchers comparing intensive and extensive systems with different media type found the only
significant difference between intensive versus extensive systems was in peak attenuation and
peak runoff delay, which were higher for intensive green roofs (Razzaghmanesh and Beecham
2014).
35
-------
Longer dry periods and warmer seasons tend to increase water retention. Factors that affected
retention performance and runoff volume in a Mediterranean climate were rainfall depth,
intensity and duration, as well as the average dry weather period between rainfall events
(Razzaghmanesh and Beecham 2014).
Stormwater retention performance of green roofs has been assessed under semi-arid conditions.
A field-scale green roof in Denver with a 45 percent impervious cover had a 68.7 percent
average runoff reduction rate over 3 years. Rainfall retention performance has been shown to
decrease, however, with increasing rainfall amounts (Jiang, Yuan and Piza 2015).
4.2.2.4. Design optimization for plant establishment and growth
Green roofs have been characterized as hostile environments for plant growth because of factors
that include shallow substrate depth, high temperatures, lack of shade and wind exposure.
Different plants have different hydraulic responses to drought stress. In the Mediterranean
climate of Messina, Italy, an anisohydric species (e.g., Salvia officinalis) and an isohydric
species (e.g., Arbutus unedo) both were determined to be appropriate for a green roof
installation, but in experiments using extensive green roof modules, the water status of the two
types of plants was shown to depend on the water retention properties of the substrate, with
anisohydric species requiring an appropriate substrate (Raimondo et al. 2015). Comparing
indigenous Australian ground cover and grass species in prototype-scale green roofs in Adelaide,
Australia, a Mediterranean climate, researchers showed that the succulent species best tolerated
the hot, dry summer conditions in terms of growth and water use efficiency (Razzaghmanesh,
Beecham and Kazemi 2014a). Higher water use plants have been shown to die sooner under
simulated drought conditions than conservative water users, with survival being related to
reduced biomass under drought rather than increased leaf succulence (Farrell et al. 2012).
Facultative crassulacean acid metabolism plants, which can switch from low transpiration during
dry periods and high transpiration during rain events, or plants with a broad soil water niche have
been recommended for extensive green roofs in hot climates. Facultative mycorrhizal grasses
have been proposed as an alternative to sedum species, some of which fix carbon dioxide weakly
above 20°C because of their temperate origins. Another alternative is a "brown roof'—planted
with annual seeds, bulbs or other cryptophytes—that simulates desert conditions and is dormant
during the dry season (Simmons 2015).
The need for irrigation of green roofs in dry conditions has been investigated. Different irrigation
regimes have been tested with drought-resistant plants. Under Mediterranean conditions in
Athens, Greece, a sedum species {Sedum sediforme) was established in experimental plots with
high versus minimal irrigation during summer drought periods in the first year. The sedum was
able to survive in its second year without irrigation (Nektarios et al. 2015). An investigation in a
semi-arid region of southeast Spain found, however, that the two species of Mediterranean plants
tested required irrigation during drought conditions in a 9-month trial (Ondono, Martinez-
Sanchez and Moreno 2016a). A study of indigenous Australian species in prototype-scale
extensive and intensive green roofs in Adelaide, Australia, a Mediterranean climate, showed that
some plants required supplementary irrigation during the hot, dry summer (Razzaghmanesh,
Beecham and Kazemi 2014a). In a review of adapting green roof irrigation practices for
sustainable water management, the authors advised installing an irrigation system in arid
36
-------
climates to keep the vegetation alive during dry periods or leaving the roof unvegetated during
the summer, although the cooling benefits of vegetation then are lost (Van Mechelen, Dutoit and
Hermy 2015).
To optimize water usage, incorporating wastewater or recirculated water resources is a green
roof design option for arid and semi-arid areas. Because these water resources tend to have
higher salinity than fresh water, investigators tested the effects of salinity on growth of Lobelia
erinus, a potential species for use in cover applications in green roofs, in a hydroponic system
and determined that although growth was affected, salinities as high as 50 mM sodium chloride
did not produce toxic effects on the leaves (Escalona et al. 2013).
Substrate depth has been found to affect growth of vegetation under seasonally dry and semi-arid
conditions. Sedum growth in Athens was enhanced in deeper substrates under no-irrigation
conditions, although plants were able to survive even with a shallow substrate depth (Nektarios
et al. 2015). In a 2-year study of turf cover of an extensive green roof system by Paspalum
vaninatum turfgrass in Athens, the use of a deeper substrate (15 cm) reduced irrigation needs
(Ntoulas and Nektarios 2015). For Manilagrass (Zoysia matrella), which was tested using a
simulated green roof in Athens, greater substrate depth was found to be more important than
substrate formulation in reducing drought stress under water deficit conditions (Ntoulas et al.
2013). Deeper substrates produced greater rates of plant growth and aboveground biomass
production in experiments in a semi-arid region of southeast Spain (Ondono, Martinez-Sanchez
and Moreno 2016a). In an Australian study in a Mediterranean climate, the greater substrate
depths characteristic of intensive as opposed to extensive green roofs produced the best growth
(Razzaghmanesh, Beecham and Brien 2014). Greater substrate depth (15 cm vs. 7.5 cm) was
found to promote growth and increase leaf dry weight during a drought period in a study using
Dianthus fructicosus in a Mediterranean climate (Nektarios et al. 2011).
The applicability of soil and soil-less substrates has been compared under seasonally dry
conditions. In a study of the growth of Sedum sediforme established in extensive green roof
systems under Mediterranean climate conditions, using soil-amended versus soil-less substrate
did not affect growth or physiology after the first year (Nektarios et al. 2015). During a drought
period, substrate moisture was increased, however, in a soil-containing substrate as compared to
a soil-less substrate, and a soil substrate showed higher growth during establishment (Nektarios
et al. 2011). A study of the suitability of a native Mediterranean xerophyte for use on extensive
green roofs in a Mediterranean climate found that growth and flower number was promoted in a
soil-containing versus a lighter soil-less substrate, although the effect was not large (Tassoula et
al. 2015). The hydrolytic enzyme activity, which is linked to nutrient cycling, was highest in
soil-containing green roof substrates in simulated green roofs in a semi-arid region of southeast
Spain (Ondono, Martinez-Sanchez and Moreno 2016b). In a separate study under semi-arid
conditions, soil-amended substrate showed higher rates of microbial activity and nutrient
cycling, necessary for plant development, compared with non-amended substrate (Ondono,
Martinez-Sanchez and Moreno 2016a).
The effects of organic material amendments to substrates have been tested under seasonally dry
conditions, comparing different types of organic amendments for their ability to support growth.
In a simulated green roof system in Athens, different types of organic material amendments were
37
-------
tested using Manilagrass. Compost-amended substrates supported the most growth under
conditions of adequate irrigation but increased drought stress during water deficit periods. Peat
amendments, combined with deeper substrate, performed better than compost in supporting
growth under water-deficit conditions (Ntoulas et al. 2013). For Paspalum vaginatum turfgrass in
extensive green roofs in a Mediterranean climate, compost amendments increased turf cover
under water-sufficient but not water-deficient conditions (Ntoulas and Nektarios 2015).
Supplementing the growth medium with 50 percent organic compost produced more vigorous
growth compared to two commercially available media under dry climate conditions
(Razzaghmanesh, Beecham and Brien 2014).
Industry guidelines that specify a large fraction of porous, mineral-based materials in the
growing media (FLL 2008), although appropriate for temperate regions, may not be optimal in
dry climates. The choice of the inorganic component of green roof substrates has been shown to
affect plant growth. In an investigation of substrates made from compost-amended mixtures of
silica, crushed bricks and clay-loam soil, plant growth patterns on the different inorganic
substrates under semi-arid conditions varied by species (Ondono, Martinez-Sanchez and Moreno
2016b). A 12-month study of growth responses of Australian native plants on medium-scale
green roofs in Adelaide, Australia, showed that some commercially available media were able to
sustain little growth in a seasonally dry, Mediterranean climate (Razzaghmanesh, Beecham and
Brien 2014).
Substrate additives that increase water retention have been tested for their ability to increase
drought resistance and expand plant selection for green roof vegetation in dry climates. These
substrate additives have an advantage over soil in that they increase water retention but do not
decompose over time (Simmons 2015). In a controlled greenhouse environment, a
polyacrylamide water-absorbent gel increased the substrate water holding capacity by 24 percent,
increasing shoot growth (Young et al. 2014). During a greenhouse simulation of a 25-day
drought, polyacrylamide gel was more effective than a sedum living mulch in increasing the
tested species' drought tolerance (Young, Cameron and Phoenix 2015). In greenhouse and
laboratory studies, green waste biochar was shown to delay the time to reach the permanent
wilting point for the test plant, winter wheat {Triticum aestivum), and increase the substrate water
holding capacity (Cao et al. 2014). Silicate granules and hydrogel, however, had differing effects
on plant-available water for winter wheat and white lupin (Lupinus albus) in simulated drought
greenhouse experiments (Farrell, Ang and Rayner 2013). Substrates with higher water retention
increased plant survival times under simulated drought conditions (Farrell et al. 2012).
In addition to the water retention properties of growth medium, thermal conductivity and heat
capacity can be a concern. Summer temperatures exceeding 70°C have been recorded on roof
surfaces in Texas, which exceeds the heat tolerance of roots even for arid-adapted crassulacean
acid metabolism plants and may limit plant growth. Green roof growth media have not been
designed for hot, dry climates. A clay-based commercially available growth medium was shown
to have high heat capacity in laboratory trials (Simmons 2015).
The effects of roof slope on growth performance have been evaluated under seasonally dry
conditions. Mildly sloping roofs (1% slope) supported better plant growth than steep roofs (25%
38
-------
slope) in a study conducted in Adelaide, Australia, a Mediterranean climate (Razzaghmanesh,
Beecham and Brien 2014).
4.2.2.5. Design optimization for water retention
Optimizing the growing media depth and substrate water retaining capacity, considering the
addition of a water storage layer, and the site's precipitation patterns are factors that can affect
green roof design for arid and semi-arid climates. Green roof design has the conflicting goals of
optimizing drainage to improve stormwater retention and retaining sufficient moisture to support
plant survival. The effects of design parameters on a hypothetical green roof were modeled using
the EPA's SWMM LID module with rainfall records for semi-arid Billings, Montana. The runoff
reduction rate increased with growing media depth and water retaining capacity. The same green
roof was found to have a higher runoff reduction rate in a semi-arid area than a humid area
(Atlanta, Georgia), but it required more irrigation (Guo, Zhang and Liu 2014). In a modeling
study of green roof performance in a semi-arid environment, adding a storage layer was found to
increase the runoff reduction rate and reduce the need for irrigation (Guo, Zhang and Liu 2014).
In an evaluation of the hydrologic response of green roofs in a Mediterranean climate, a
nonlinear relationship between rainfall and runoff was found. The results of the study indicate
that continuous time series modeling is more appropriate than peak rainfall intensity for green
roof design (Razzaghmanesh and Beecham 2014). The use of a layer of hydroponic foam in
place of a standard retention layer to increase water retention while making stored water
available to vegetation has been tested successfully in Texas (Simmons 2015). In an established
extensive green roof in Athens, substrate moisture was found to depend on the slope of the roof
and presence of underground retaining walls, which increased substrate moisture. Substrate
moisture was not affected by the type of draining system (geotextile or sand-gravel bilayer) or
the substrate depth (300 mm to > 1,200 mm)—except in a relative flat area between the concert
hall and the atrium of the roof—but varied because of local inclinations and sunlight exposure.
Substrate moisture measurements taken 3 and 10 years after installation showed that the drainage
system was functioning well (Nektarios et al. 2014).
4.2.3. Permeable pavement
The ability to manage stormwater runoff using permeable pavements in semi-arid regions has
been the subject of modeling studies. Research also has focused on the improvement of the
quality of stormwater by permeable pavement in the field under semi-arid and Mediterranean
climate conditions. In addition, the effects of laboratory-simulated freeze-thaw cycling, traffic
wear and laboratory-simulated clogging on infiltration capabilities of permeable pavement have
been assessed relative to rainfall patterns typical of semi-arid areas.
4.2.3.1. Stormwater retention
The ability of permeable pavement to reduce stormwater runoff has been tested in the field and
modeled in semi-arid environments. In a study in Denver, permeable interlocking concrete
pavement reduced runoff volume by 33 percent, and pervious concrete pavement reduced runoff
volume by 38 percent, compared to a reference site (Jiang, Yuan and Piza 2015). A modeling
study evaluated the implementation of LID practices on a university campus with 40 percent
impervious surfaces in semi-arid Tianjin City, China, using 10 years of past precipitation data.
The researchers focused on smaller precipitation events (i.e., less than 1-inch rainfall depth,
39
-------
which represents approximately 77 percent of all events) and found that among the LID practices
modeled, porous pavement performed best in three measures of water balance: the change in
total runoff, the rainfall captured by LID on site, and the ratio of saved rainfall on site. It also
reduced peak flow by 29 percent, second in performance only to bioretention (Huang et al.
2014). High-resolution satellite imaging data were used to extract land cover information for
modeling the performance of porous pavement in a residential area of San Clemente, California,
a Mediterranean climate. Porous pavement was estimated to reduce runoff volume by 18 percent
(Khin et al. 2016). To evaluate the performance of permeable pavements to control runoff in a
semi-arid region, the permeability of experimental test sections of different types of permeable
pavement was measured using the ASTM CI701 method. Interlocking concrete pavers had the
highest permeability (0.5 cm s"1) and permeable asphalt pavements the lowest (0.1 cm s"1), but all
permeable pavements had permeabilities adequate to prevent surface runoff during typical rain
events in central California (Li et al. 2013).
Parameters other than pavement permeability need to be considered in designing permeable
pavement for stormwater retention. Researchers measured the hydraulic properties of subgrade
soil and permeable pavement material in the laboratory and conducted numerical simulations for
24-hour rainfall data from 2-, 50- and 100-year storms in three rainfall regions of California.
Sensitivity analyses revealed that the saturated hydraulic conductivity of subgrade soil was the
most important parameter for the design of permeable highway shoulder retrofits to capture
rainfall runoff (Chai et al. 2012).
The applicability of models to assess performance and design of permeable pavement under
semi-arid field conditions has been tested. A field test conducted in Denver of the theoretical
paved area reduction factor for porous pavement compared to measured rainfall events using
EPA's SWMM model found that the reduction factors are accurate and applicable to the Denver
area (Blackler and Guo 2014).
4.2.3.2. Water quality improvement
Permeable pavement has been tested for its ability to improve of the quality of stormwater runoff
in field studies. In Denver, runoff from a permeable interlocking concrete pavement site had
significantly lower levels of zinc, chemical oxygen demand (COD), total Kjeldahl nitrogen
(TKN) and total suspended solids (TSS) compared to a reference site; runoff from a porous
asphalt site had significantly lower nitrate plus nitrite and total selenium; and runoff from a
pervious concrete site had significantly lower TSS, total phosphorus, TKN, COD and copper
(Jiang, Yuan and Piza 2015). A study comparing permeable and conventional pavement systems
installed in a parking lot in Adelaide, Australia, a Mediterranean climate, showed that permeable
pavement improved stormwater quality significantly, reducing nutrient levels (i.e., total nitrogen
and total phosphorus), heavy metals (i.e., zinc, lead, copper, cadmium and nickel) and TSS. The
authors attributed the reduction of pollutant levels to mechanical filtration (Beecham, Pezzaniti
and Kandasamy 2012).
4.2.3.3. Performance characteristics
Performance characteristics of permeable pavement have been assessed in semi-arid regions. In
cold arid and semi-arid regions, degradation of pervious concrete by freeze-thaw cycling is a
40
-------
possible concern for soil-clogged or water-saturated conditions. An experimental pervious
concrete slab in northern Utah was damaged at a significantly faster rate when clogged
(i.e., failure at 93 freeze-thaw cycles versus 180 cycles) or saturated with water (i.e., failure at
80 freeze-thaw cycles versus 180 cycles). No significant differences in structural properties
between clogged and unclogged locations were observed, however, which was attributed to only
the upper 1 to 2 inches being filled with debris and the remaining depth of the slab being free
draining (Guthrie, DeMille and Eggett 2010).
Clogging would be expected to decrease water retention and water quality improvement
capabilities of permeable pavement as it ages. The effect of traffic volume-induced deterioration
on infiltration performance was evaluated in a case study of deteriorated pervious concrete
installed in a parking lot in Denver. The researchers found that the infiltration rate of the
pervious concrete was decreased in high-traffic areas compared with low-traffic areas.
Comparing the infiltration rates to estimated rainfall intensity in the metropolitan Denver area,
all but the most high-traffic area would be expected not to have stormwater drainage problems
up to a return period of 10 years (Kim et al. 2015). Laboratory testing of a pervious concrete
pavement system clogged with sand and clay revealed that the pervious concrete system still
would be effective for stormwater detention under conditions that might be encountered in semi-
arid regions (i.e., the 100-year, 1-hour design storm for Denver), although the flow-limiting layer
in these tests was found to be the subgrade, not the pervious concrete (Coughlin, Campbell and
Mays 2012). In a study of pervious concrete pavements in parking lots that included sites in
semi-arid regions of California, the age of the pavement was the main factor affecting measured
permeability, and the mass of fine particles less than 38 [j,m also was an important factor. The
porosity of the top surface layer of core samples generally was lower, indicating the importance
of a regular cleaning maintenance program to improve porosity. The field measurements were
conducted with a National Center for Asphalt Technology (NCAT) field permeameter
(Kayhanian et al. 2012).
4.2.4. Planter boxes
Treatment and retention of stormwater by prototype planter boxes—rain gardens with vertical
walls and open or closed bottoms—have been tested under arid and semi-arid conditions. The
technology's effects on gray water quality and simulated stormwater runoff quality have been
compared using vegetated and nonvegetated units. The effect of vegetation on stormwater
retention has been explored under semi-arid conditions.
4.2.4.1. Water quality improvement
The ability of prototype planter box systems to improve the quality of gray water (residential
wastewater without toilet and kitchen sources) was evaluated in a study conducted outdoors in
the United Arab Emirates to assess the applicability of gray water-fed planter boxes for arid
environments. In a 10-day trial, vegetative and nonvegetative systems performed almost equally
well in improving gray water quality, including turbidity and total coliform bacteria, but the
vegetative system was more effective in reducing sodium and COD (Chowdhury 2015). Pilot
experiments were conducted in hot and semi-arid Bryan, Texas, using four planter box units,
each measuring approximately 4 m3 in volume, planted with shrubs, grass species specified for
highways in Texas, native Texas grasses and Bermuda grass (Cynodon dactylon), respectively. A
41
-------
fifth unit was weeded regularly to provide a nonvegetated control. After 14 months of growth,
during which the units were regularly irrigated, synthetic highway stormwater runoff was added
to the boxes, simulating the mean 24-hour storm runoff for a drainage basin of 330 m2 All of the
units, vegetated and nonvegetated, removed lead, zinc, TSS and ammonia, although the shrub
and control units were approximately twice as effective at removing TSS than the grass units (Li
etal. 2011).
The ability of planter boxes to remove bacteria has been tested under semi-arid conditions. In a
follow-up study to Li et al. (2011), the vegetated units were allowed to develop naturally
outdoors without weed control for one summer. After a summer of vegetation succession, the
ability of each unit to remove Escherichia coli (E. coli) added to influent potable water was
assessed. The effluent was collected from drainage outlets installed at the bottom of the units.
The nonvegetated control unit had the highest removal efficiency for E. coli (97%), followed by
the shrub unit (88%) and three grass units (76%, 57% and 48% removal efficiencies). Although
removal mechanisms for E. coli are not well understood, the authors suggested that given the
relative size of the bacteria compared to the porous medium, adsorption is a more likely
mechanism than filtration. The root growth, rooting depth and nutrient metabolism of different
types of vegetation varied, which would have affected the retention times and pollutant removal
performance of the different units (Kim et al. 2012).
4.2.4.2. Negative effects on effluent water quality
In the study by Li et al. (2011) of pilot planter box units tested with simulated stormwater,
vegetated units, as well as a nonvegetated control unit, were shown to export pollutants. The
vegetated units and nonvegetated control unit leached nutrients, resulting in higher concentration
in the effluent than the influent for nitrate, total nitrogen and total phosphorus. The authors
speculated that denitrification in the rhizosphere in the vegetated units might have reduced
nitrate leaching relative to the nonvegetated control. Higher TSS concentrations in the effluent or
acidic soil conditions from microbial activity might have contributed to the greater leaching of
total phosphorus from the vegetated units. Copper also was higher in the effluent than the
influent for the vegetated units but not the nonvegetated control.
4.2.4.3. Stormwater retention
The effects of vegetation on influent and effluent hydrographs for pilot planter box units was
evaluated in a hot, semi-arid climate. The vegetation—shrubs and three types of grass seed
mixes—was allowed to grow for 14 months prior to the experiment. The vegetated pilot units
reduced peak flows (14.4% to 32.2%), but the degree of reduction was highest for the
nonvegetated control unit (74.8%). Surface ponding occurred immediately in the nonvegetated
control unit but only was evident after 1 hour of flow for the vegetated units. The detention time
was much longer for the nonvegetated control unit (118.3 minutes) compared with the vegetated
units (15.1 to 25.6 minutes) (Li et al. 2011). In the follow-up study to Li et al. (2011), the
retention times of vegetated units (planted with different vegetation and allowed to undergo
natural secession for one summer) and a nonvegetated control planter box unit were compared
again. The nonvegetated control unit had the longest retention time (141.6 minutes), followed by
the unit originally planted with shrubs (67.2 minutes) and the three units originally planted with
grasses (16.2 minutes for Unit B, 42.6 minutes for Unit H, and 18.3 minutes for Unit N).
42
-------
Changes in soil porosity and preferential flow paths from different patterns of root growth would
be reflected in the retention times of the units (Kim et al. 2012).
4.2.4.4. Design optimization
The suitability of different vegetation species for planter boxes has been assessed empirically in
a semi-arid environment. Only one of the three original shrub species planted in a pilot planter
box unit, Texas sage (Leucophylum frutescens), was thriving after 14 months. In the pilot units
seeded with different types of grasses (grass species specified for highways in Texas, native
Texas grasses and Bermuda grass), vegetation compositions were similar after 14 months,
dominated by Johnson grass (Sorghym halepense) and giant ragweed (Ambrosia trifida) (Li et al.
2011).
To reduce nitrate leaching, planter boxes might be constructed with a lower soil-to-compost ratio
in the growth medium. Creating a permanent water saturation zone in the bottom might facilitate
denitrification as well (Li et al. 2011).
4.2.5. Rain gardens/bioretention cells
Rain gardens/bioretention cells have the potential to improve of the quality of stormwater and
also mitigate runoff velocity and volume in arid and semi-arid environments. Rain gardens have
been evaluated in semi-arid climates as potential sources of nutrient pollution to runoff as well.
Specific design requirements for xeric climates—including vegetation selection, inclusion of a
storage layer, soil type, irrigation, sizing and siting—have been assessed.
4.2.5.1. Water quality improvement
A rain garden installed in a residential neighborhood in Lakewood, Colorado, and monitored for
3 years reduced mean TSS in the effluent to a mean of 51.3 mg/L from a mean of 264.3 mg/L in
the influent. The median event-based concentration reduction rate from stormwater was
91 percent for TSS but varied by event percentile (i.e., -5% for a 5th percentile event to 98% for
a 95th percentile event). The median event-based concentration reduction rates were positive for
TKN; ammonia nitrogen; and total lead, chromium and antimony but negative (i.e., the median
concentration was higher in the effluent than influent) for total phosphorus; dissolved
phosphorus; and total copper, arsenic, beryllium, cadmium and selenium. Like TSS, the event-
based concentration reduction rates for these analytes were negative for small storms
(5th percentile) and positive for large storms (95th percentile), as shown in Table 4-1 (Jiang,
Yuan and Piza 2015).
In a study in Salt Lake City, Utah, nutrient retention by bioretention cells with different
vegetation communities—an irrigated wetland, an unirrigated upland vegetation community and
no vegetation—was compared. Synthetic stormwater was used to simulate runoff to each cell
from an impervious surface. All three cells retained phosphate mass significantly (P < .01 by
analysis of variance [ANOVA]), retaining approximately 50 percent of the influent phosphate
during the 12-month study. The wetland and upland cells retained total nitrogen. The wetland
cell required irrigation by more than 12,000 L of water, however, during the dry summer. The
authors suggested that for optimal nutrient retention by bioretention cells, greater upland
vegetation density or irrigation of wetland communities by gray water would be sustainable
43
-------
Table 4-1. Mean influent and effluent concentrations and event-based concentration
reduction rates from a rain garden
Event-Based Concentration
Constituents
Mean
Influent
Mean
Effluent
5th
Reduction Rate (%)
Median Mean
95th
TSS (mg/L)
264.3
51.3
-5%
91%
63%
98%
NOs + NCh (mg/L)
0.7
2.1
-1327%
-128%
-559%
-5%
TKN (mg/L)
3.1
2.6
-363%
18%
-31%
70%
NHs-N (mg/L)
0.7
0.0
-378%
96%
-28%
100%
Tot. P (mg/L)
0.4
0.7
-1947%
-133%
-494%
76%
Ortho-P (mg/L)
0.2
0.4
-844%
-252%
-269%
-99%
Diss. P (mg/L)
0.5
1.1
-1357%
-196%
-358%
49%
Tot. sol. P (mg/L)
0.1
0.4
-560%
-350%
-317%
-104%
Tot. Cu (Mg/L)
16.6
23.4
-393%
-12%
-73%
63%
Tot. Pb (Mg/L)
8.1
5.0
-503%
98%
-30%
100%
Tot. As (Mg/L)
3.3
4.4
-139%
-100%
-112%
100%
Tot. Be (ng/L)
0.0
0.1
-100%
-100%
-33%
80%
Tot. Cd (Mg/L)
0.1
0.2
-407%
-100%
-107%
100%
Tot. Cr (pig/L)
2.9
1.4
-170%
100%
32%
100%
Tot. Sb (Mg/L)
0.4
0.5
-100%
32%
26%
100%
Tot. Se (Mg/L)
0.1
0.1
-100%
-100%
-68%
11%
Abbreviations: TKN = total Kjeldahl nitrogen, TSS = total suspended solids.
Source: Jiang, Yuan and Piza (2015). Reprinted by permission of the authors. Copyright 2015 Jiang, Yuan and
Piza.
solutions (Houdeshel et al. 2015). It was suggested that plant roots increase stormwater
infiltration rates by creating macropores through root growth and turnover (Houdeshel, Pomeroy
and Hultine 2012).
4.2.5.2. Negative effects on effluent water quality
Rain gardens have been evaluated in semi-arid areas as potential sources of nutrients to runoff. In
three bioretention cells located in semi-arid Salt Lake City, each planted with different
vegetation communities, only the wetland cell retained nitrate (38%), but the upland and control
(nonvegetated) cells exported two and nine times more nitrate, respectively, than was added as
synthetic stormwater (Houdeshel et al. 2015). A rain garden constructed in a residential
neighborhood in Lakewood, Colorado, acted consistently as a source rather than a sink for nitrate
plus nitrite, orthophosphate and total dissolved phosphorus (Jiang, Yuan and Piza 2015).
The water quality in constructed wetlands in an urban environment has been shown to be
affected by the amount of impervious surface that drains to the wetland, as was the case for
polycyclic aromatic hydrocarbons (PAHs) detected in water samples from urban wetlands in
Lubbock, Texas (Heintzman et al. 2015).
4.2.5.3. Stormwater retention
In a rain garden draining 0.77 hectares of a residential community in Lakewood, Colorado, with
an impervious area of 47 percent, the average runoff volume reduction rate ranged from 37 to
61 percent over 3 years (Jiang, Yuan and Piza 2015).
44
-------
A modeling study was conducted in a low-rise residential area located in a Mediterranean
climate (San Clemente) that used EPA's SWMM to assess performance. The results showed that
implementing bioretention in right-of-ways and grassy areas in front of buildings reduced runoff
volume by 37.1 percent (Khin et al. 2016).
4.2.5.4. Design optimization
The choice of vegetation in bioretention gardens can be tailored for xeric climates. Combining
deep-rooted shrubs—which have taproots to access deep soil water—with grasses that produce
extensive networks of shallow roots that interface with arbuscular mycorrhizal fungi—which
increase nutrient-absorbing ability—is recommended for optimal drought tolerance. Combining
warm season bunchgrasses with locally native shrubs is recommended for warm deserts
(i.e., Arizona, western Texas, New Mexico and southern Utah—typified by precipitation falling
as rain during the growing season), whereas a mixture of warm and cool season bunchgrasses
planted with locally native shrubs and evergreens is recommended for cool deserts (i.e., the
Great Basin and Intermountain West—typified by precipitation falling as snow in the winter or
spring) (Houdeshel, Pomeroy and Hultine 2012).
In urban areas, impervious surfaces prevent infiltration of precipitation where it lands. Including
a storage layer in the design of bioretention gardens allows a relatively large volume of water,
draining from surrounding impervious surfaces, to infiltrate in a small footprint. Designing the
storage layer to be oxygen-limited promotes denitrification (Houdeshel, Pomeroy and Hultine
2012). A demonstration garden in Salt Lake City is an example of a successful bioretention
garden that included a storage layer in which the vegetation was growing well without irrigation
two summers after establishment. Successful plant species included regionally native
bunchgrasses, shrubs, trees and flowers (Houdeshel and Pomeroy 2014).
To reduce nutrient leaching from growth medium, a sandy loam topsoil can be used. Many plants
native to xeric climates are adapted to soils with high infiltration rates and low nutrient content
(Houdeshel, Pomeroy and Hultine 2012).
Regarding maintenance, irrigation has been recommended during the first year of establishment
for spring or summer plantings. Irrigation helps root systems develop so that they can access
moisture deep in the growth medium. No irrigation should be required after establishment.
Trimming bunchgrasses each winter promotes new shoot growth in the spring. Mulch, which
requires upkeep and renewal, can be replaced by a layer of gravel to reduce maintenance needs.
Light-colored gravel also decreases surface temperature and, therefore, plant water demand.
Including a weed barrier below the gravel is recommended (Houdeshel, Pomeroy and Hultine
2012). The ability of properly selected native vegetation to survive a prolonged dry period
without irrigation was demonstrated in a cold desert bioretention garden in Salt Lake City in
which three different irrigation systems were compared for establishment. Almost all of the
plants were growing well without supplemental irrigation two summers after establishment. The
authors of the study suggest that based on regional hydrology, regionally native vegetation can
be established without irrigation in bioretention gardens constructed in cold desert climates in
any season other than the summer (Houdeshel and Pomeroy 2014).
45
-------
For sizing, continuous modeling in addition to single storm event modeling should be used.
Simulations using SWMM 5.0 were used to determine the garden to drainage area ratios for
bioretention gardens in cold desert (Salt Lake City) and warm desert (Phoenix) sites. EISA4
requires federal projects to manage the volume of rainfall from the 95th percentile storm, which
requires the garden-to-drainage area ratio for Phoenix to be no more than 9:1 (Houdeshel,
Pomeroy and Hultine 2012).
One suggested approach to siting bioretention cells is the use of high-resolution remote sensing
data to extract land cover information (Khin et al. 2016).
4.2.6. Vegetative filter strips
In a few studies, vegetative filter strips have been assessed for their effectiveness in stormwater
management under Mediterranean and seasonally dry conditions. Modeling and field study data
have been used to assess control of fecal coliform bacterial loading, runoff reduction, pesticide
loading reduction and soil loss from cultivated land and land used for livestock. Design criteria
assessed included filter strip width as a function of soil type and topography, vegetation type and
siting of BMP s.
4.2.6.1. Water quality improvement
The effectiveness of vegetative filter strips in reducing fecal coliform bacterial loads was
assessed in a study of stormwater runoff from manure-fertilized dairy pastures in the Tomales
Bay watershed, a region in California with a Mediterranean climate. In this observational, 1-year,
longitudinal study, pastures varied in size and slope. The vegetation of the filter strips, primarily
annual grasses with some associated forbs and perennial grasses, was dictated by the lifecycles
of the plants and rainfall. Linear mixed effects regression was used to test for associations
between management practices and logio transformed fecal coliform bacterial concentrations.
Directing runoff through the vegetative buffer was associated with a 24 percent reduction in
concentration of bacteria per 10 m of buffer length (Lewis et al. 2010).
Researchers modeled the protection of receiving water bodies and aquatic organisms from
pesticides in runoff by vegetative filter strips under the 30-year EPA scenario of dry
Mediterranean, irrigated, intensive horticulture (California tomato). Vegetative filter length and
application timing were found to be the most important input factors. The California tomato
scenario was influenced primarily by irrigation events and therefore experienced a lower average
cumulative runoff than under scenarios for other crops (Illinois corn and Oregon wheat)
(Sabbagh, Munoz-Carpena and Fox 2013). In a follow-up modeling study of the efficacy of
vegetative filter strips to limit pesticide transport from agricultural fields to receiving water using
a framework that included degradation of pesticide trapped in a vegetative filter strip between
runoff events, researchers investigated the California tomato scenario. The organic carbon
sorption coefficient of the pesticide and the aerobic/anaerobic aquatic metabolism half-life were
4 Energy Independence and Security Act of 2007, 42 U.S.C. § 17094 (2007).
46
-------
the most important input parameters determining acute (peak) absolute and percent reduction in
estimated environmental concentrations (Munoz-Carpena et al. 2015).
4.2.6.2. Stormwater management and soil erosion control
In a study combining field data from the Yakima River Basin in Washington State, a
Mediterranean climate, with mechanistic modeling, the use of vegetative filter strips as BMPs to
control sediment transport from furrow-irrigated agricultural land was assessed. Vegetative filter
strips were installed at the end of furrows, and water runoff and soil loss to the irrigation return
canal was modeled. The field experiments involved vegetative filter strip lengths ranging from
3.05 to 9.14 m. The researchers found that 5-m vegetative filter strips reduced water runoff and
soil loss, on average, by 5 and 80 percent, respectively, but the BMP of less water-consumptive
irrigation was more effective in mitigating both runoff and sediment delivery.
4.2.6.3. Design elements
A modeling study of BMPs for reducing soil erosion and sedimentation, a major problem in the
seasonally dry Blue Nile Basin in Africa, using the Soil and Water Assessment Tool (commonly
known as SWAT) showed that the effectiveness of filter strips depended on their width and the
local topography (Betrie et al. 2011).
In a study by Campo-Bescos et al. (2015) using modeling and field data from furrow-irrigated
agricultural land, the effectiveness of four types of vegetation were tested: Baronesse barley
(Hordeum vulgare), alfalfa (Medicago sativa), Bromar mountain bromegrass (Bromus
marginatus) and Rosana western wheatgrass (Pascopyrum smithii). Vegetation type did not
affect runoff or soil loss reductions by vegetative filter strips, but the authors suggested that this
insensitivity was caused by insufficient plant density to remove sediment. The optimal width of
the vegetative filter strip depended on the soil type and local topography.
Siting of filter strips also can change effectiveness. A modeling study of a Mediterranean
agricultural catchment (Roujan, southern France) showed that 70 percent of the variation of the
net erosion was explained by variations in vegetative filter density. Although the density of the
vegetative filters was the most sensitive parameter, strong interaction among the three modeling
parameters (the density of vegetative filters, their downslope/upslope location probability, and
the probability density function shape controller) when the density values are low indicated that
their location may influence their global trapping efficiency in more realistic cases where few
filters are in place (Gumiere et al. 2015).
4.2.7. Integrated systems
On a site scale, integrated systems have been studied for their efficiency in removing sediment,
metals, trace organics, bacteria and nutrients under Mediterranean climate conditions. These
studies have revealed some negative effects of integrated systems on effluent water quality. The
effectiveness of integrated systems also has been assessed for stormwater retention. The systems
contained combinations of detention basins, bioswales, bioretention cells and green street
elements.
On a watershed scale, a modeling study has been conducted for a semi-arid region to optimize
siting of LID BMPs to reduce nutrient and sediment loading.
47
-------
4.2.7.1. Water quality improvement
The pollution removal efficacy of a bioretention system composed of four rain gardens and one
bioswale was assessed in Daly City, California, a Mediterranean climate. The 427-m2 system
drained 16,200 m2 of impervious area, and the bioretention cells were constructed of a layer of
gravel mulch covering a layer of loamy sand mix above a pea gravel drainage gallery. The
observed rainfall event that exceeded system capacity was 5 mm h"1, which was consistent with
permit requirements. Water samples were collected pre- and post-installation for a variety of
storm events. The bioretention system reduced total suspended sediment loads in effluent from a
mean of 21 mg L"1 before installation to 15 mg L"1 after installation. When the majority of the
runoff was captured by the system, post-installation effluent concentrations for most trace metals
(e.g., total and dissolved mercury, copper, zinc, nickel, lead, cadmium) were lower than before
installation. Mean trace organic pollutant concentrations (i.e., total polychlorinated biphenyls,
total PAHs and octachlorodibenzodioxin) also decreased after system installation. Effluent
concentrations of metals were less variable after installation, indicating effective buffering by the
bioretention system. Mean trace organic pollutant concentrations also decreased. The researchers
concluded that pollutants such as metals and PAHs that originate from local sources and have
high concentrations in runoff were removed effectively by the bioretention system (David et al.
2015).
Effectiveness at removing nutrients and sediment from urban runoff has been assessed for an
integrated BMP system located in San Clemente Villages, California, and composed of a
detention basin, a series of low-capacity vegetated swales, and a high-capacity vegetated swale.
The 5.4 ha site treats runoff from recreational fields, parking lots and residential areas.
Researchers found that the system reduced pollutant discharge through sedimentation, vegetative
uptake and flow impoundment. The detention-based stormwater management system and low-
capacity swales were effective at removing metals (i.e., cadmium, copper, lead and zinc) from
the influent (Jiang, Yuan and Piza 2015).
In a green street project in Santa Monica, four types of BMPs were installed: subsurface plastic
concave infiltration chambers under the parking lane; wider, depressed parkways with climate-
appropriate flora and low-volume, solar-powered irrigation; gutter-oriented catch basin filters;
and pervious concrete parking lanes. The system showed significant removal of heavy metals,
mixed results for nutrients and reduced loadings for bacteria (Jiang, Yuan and Piza 2015).
4.2.7.2. Negative effects on effluent water quality
The bioretention system composed of four rain gardens and one bioswale in Daly City was found
to be a source for methyl mercury. The researchers postulated that a design error might have
resulted in anaerobic conditions at the bottom of one of the cells, creating conditions favoring
mercury methylation by bacteria (David et al. 2015).
In the integrated BMP system in San Clemente Villages described by Jiang, Yuan and Piza
(2015), the detention basin part of the system was effective in removing total nitrogen and
orthophosphate, but the low- and high-capacity swales either had little effect on nutrient
concentrations or acted as sources of nutrients.
48
-------
Effluent from a green street project in Santa Monica had elevated TSS compared to the influent,
likely because of runoff suspending particulates as it filtered through soils (Jiang, Yuan and Piza
2015).
4.2.7.3. Stormwater retention
Stormwater retention by a bioretention system has been assessed in the field under
Mediterranean climate conditions. The demonstration bioretention system installed in Daly City
was able to delay and reduce peak flow velocities and volumes. Because periods between storm
events were brief during the two wet (winter) seasons monitored, the second of which was an
abnormally high rainfall year, the soil and filter media stayed wet, resulting in only a small
decrease of approximately 10 percent in flow volume compared with pre-installation flow. The
authors indicated that longer periods between storms and maturing vegetation, which would
produce more efficient evapotranspiration, would be likely to increase flow reduction (David et
al. 2015).
4.2.7.4. Prioritization of siting
On a watershed scale, Martin-Mikle et al. (2015) developed an approach to identify priority sites
for LID, using a large (666 km2) mixed-use watershed in a semi-arid region of Oklahoma, the
Lake Thunderbird Watershed, as a case study. More than 40 percent of the watershed is
residential, with a high impervious surface coverage. Transport of phosphorus, nitrogen and
sediment by urban runoff into the Lake Thunderbird Reservoir has led to exceedances of TMDL
regulations. The researchers used geographic information system (commonly known as GIS)
data on land cover, impervious surface, digital elevation model (DEM), soil conductivity, soil
depth to restrictive layer, roads, zoning, building footprint, floodplain and waterbodies. They
derived locations for implementing LID at local- (rain barrels, green roofs and porous
pavement), intermediate- (rain gardens and bioswales), catchment- (detention and retention
ponds) and reach-scale (riparian buffers) sites for hydrologically sensitive areas. Selected sites in
subcatchments were validated by field visits, with a high rate of correctly identified sites (94%).
In one subcatchment, results indicated an ability to reduce nutrient and sediment loading to
receiving waters by 16 percent and 17 percent, respectively, by placing LID in 11 locations. On a
watershed scale, hydrologically sensitive areas were concentrated in the western third of the
watershed, an area of low infiltration capacity because of clayey soils and high rates of
impervious cover because of development.
4.3. Stormwater management—watershed-scale research
On a watershed scale, green infrastructure approaches that can be used to manage stormwater
include land conservation, preservation and restoration of riparian buffers, and enhancement of
urban tree canopies. As for site-scale research, these practices have been evaluated for their
ability to improve water quality, reduce erosion and reduce runoff. Best practices for restoration
of degraded landscape-scale features have been studied. Siting of BMPs is an important design
criterion. Irrigation is a primary maintenance consideration.
4.3.1. Land conservation
Prioritizing land conservation efforts has been an area of study. In addition, the likely effects of
land conservation on stormwater management and use of conserved land in stormwater
49
-------
management have been modeled. Land conservation affects imperviousness, grass cover and—in
the case of irrigated land—irrigation rates. Modeling studies have been undertaken in semi-arid
and arid regions of the predicted effects of variables characterizing land use on the loading of
nutrients, metals (e.g., selenium) and dissolved organic carbon (DOC) to receiving waters, as
well as on total runoff and soil erosion.
4.3.1.1. Prioritization of siting
Prioritizing sites for land conservation is an active area of research. A landscape-scale geospatial
assessment of wetlands was conducted in Wyoming, quantifying the wetlands' biological
diversity, protection status, susceptibility to climate change and proximity to sources of
impairment. The researchers determined that low-elevation wetland complexes were the least
protected, in the poorest current condition, and the most vulnerable to future land-use changes
(Copeland et al. 2010). In a study conducted in coastal California that focused on Sonoma
County (Mediterranean climate), 564 km2 were identified as being both flood-prone and of
natural resource conservation value. The authors suggest using flood mitigation grant programs
as a source of funds for property/structure buyout and habitat restoration projects (Calil et al.
2015).
4.3.1.2. Water quality improvement
The predicted effects of conservation of agricultural lands on water quality have been modeled.
Transport and chemical reaction processes were modeled for Colorado's Lower Arkansas River
and its tributaries (Koppen climate classification BSk: cold, arid steppe). In the study by Bailey,
Gates and Romero (2015), fallowing cultivated land to allow irrigation water to be leased to
municipalities was predicted to have a strong positive effect on nitrate loading to ground water.
Intermittent fallowing of 25 percent of the land resulted in a forecasted decrease of about 15
percent in nitrate ground water loading to streams (Bailey, Gates and Romero 2015). The same
watershed was modeled for the effects of BMPs on selenium loading. Land fallowing, as well as
the other water management BMPs modeled (i.e., reduced irrigation and irrigation canal sealing)
were predicted to yield an immediate, significant effect on selenium mass loading to the
Arkansas River and a significant decrease in mass loading during the 38-year period modeled,
much more than the land-management BMPs (Bailey, Romero and Gates 2015).
A modeling study was conducted of the effects of the changes in infrastructure design that
occurred from 1955 to 2010 (i.e., a shift from pipes to engineered channels and retention basins
to natural washes) in an arid city (the greater Phoenix area). Stormwater runoff was monitored at
outlets from nested watersheds, measuring discharge, dissolved nitrogen, dissolved phosphorus,
DOC and rainfall (where not already monitored). Path analysis was used to test hypotheses about
the relationships among infrastructure characteristics, energy use by buildings for heating and
cooling, land cover, storm characteristics, and nutrient (dissolved nitrogen and phosphorus) and
DOC delivery to the watershed. Imperviousness and grass cover were the most important land-
cover variables for predicting nutrient and DOC loading. Nutrient and DOC concentrations,
however, were most strongly related to antecedent and storm characteristics (Hale et al. 2015).
The adoption of water-sensitive urban design in established urban areas by using a portion of the
land in existing parks for stormwater filtration was explored in a region of South Australia with a
50
-------
Mediterranean climate. The results of the modeling study showed that allocating 10 percent of
parks that cover less than 16 percent of the landscape for bioretention devices would result in a
62 percent reduction of nitrogen from stormwater (Segaran, Lewis and Ostendorf 2014).
4.3.1.3. Soil conservation
The potential benefits of land conservation for the prevention of soil erosion were compared to
those from landscape design in a watershed with a Mediterranean climate (Languedoc-
Roussillon, France). The researchers found that land use was the major factor controlling
sediment production (David et al. 2014). In a study of a Mediterranean region in southeast Spain,
land management (i.e., seasonal set-aside land management) and land uses (i.e., urban,
agricultural, scrubland, forest and dense forest) were found to be most important in affecting
erosion and sediment yield (Rodriguez-LIoveras et al. 2015).
4.3.1.4. Stormwater retention
In the modeling study by Hale et al. (2015) of watersheds in Phoenix, effects of different
parameters on runoff reduction were assessed in addition to effects on water quality.
Imperviousness and grass cover were found to be the most important land-cover variables for
predicting runoff reduction, being more significantly correlated with runoff than connected
imperviousness or soil cover.
4.3.1.5. Reestablishment of vegetation
Reestablishment of native vegetation was studied on abandoned drill pads and infrastructure in
southwestern Wyoming, a cold, arid environment. Soil moisture retention was improved using
hollow frame snow fencing, a technology engineered to alter the snowpack without the negative
effects of traditional snow fencing on sagebrush dominance. Snow fencing significantly
increased the establishment of native sagebrush-steppe species, with fewer invasive species than
control areas (David 2013).
4.3.2. Riparian buffers
Riparian buffers have been studied for their ability to improve water quality in semi-arid
conditions. Modeling studies have examined the predicted effects of enhanced riparian buffers
on nutrient and selenium loading. The effects of buffer width on river habitat for fish also have
been studied in dry forests. Human and natural factors affecting riparian ecosystems, as well as
practices for restoring native vegetation, also have been studied under semi-arid conditions.
4.3.2.1. Water quality improvement
In a modeling study comparing different BMPs in Colorado's Lower Arkansas River and its
tributaries, a combination of reducing fertilizer application, reducing irrigation, sealing irrigation
canals and enhancing riparian buffer zones was predicted to have the greatest overall impact on
regional nitrate concentrations in ground water and mass loading to the river network compared
to other BMP combinations (Bailey, Gates and Romero 2015). In a separate study, enhanced
riparian buffers was one of the BMPs modeled in the Lower Arkansas River Valley in
southeastern Colorado to identify practices for mediating selenium toxicity in surface water.
Selenium, which is toxic at high concentrations, is a problem in many river basins in the western
United States, where selenium-bearing shales oxidized by oxygen and nitrate represent a source
51
-------
of the metal. Enhancement of riparian buffers was one of the most effective practices modeled,
resulting in a 14 percent selenium load reduction when combined with reduced irrigation (Bailey,
Romero and Gates 2015). A field study comparing nutrient uptake and immobilization by
riparian plant communities in the Sacramento Valley, California, revealed that riparian zones
with woody plant communities had lower soil nitrate and plant-available phosphorus levels.
Lower soil nutrient loading also was correlated with higher visual riparian health assessment
scores, a quantification of channel condition, access to the floodplain, bank stability, extent of
natural riparian zone vegetation, macroinvertebrate habitat, pool variability and pool substrate
(Young-Mathews et al. 2010).
4.3.2.2. Habitat preservation performance
Large woody debris and stream water temperature are important factors in providing fish habitat
in rivers. A modeling study that included dry, Douglas fir (Pseudotsuga menziesii) forest types in
southwest, central and north Idaho found that timber harvesting could be compatible with
maintaining habitat objectives for large woody debris and stream shade if only light thinning was
allowed in an inner 25-foot buffer zone with heavier thinning in the outer 50-foot zone (Teply,
McGreer and Ceder 2014).
4.3.2.3. Reestablishment
Restoration of riparian buffer zones may involve removal of invasive species and planting of
native species. Researchers have studied the best conditions for reestablishing willow in riparian
buffer zones. A study compared aerial cover, height and stem density attained by dormant coyote
willow (Salix exigua) cuttings planted along the banks of the Middle Rio Grande in central New
Mexico (semi-arid). Regression analysis of the percent of fine-textured soil material and
available water at different depth increments at planting sites revealed that cuttings attained
growth comparable to natural willow stands if the floodplain soil contained intermediate levels
of fine-textured soil material, and the maximum depth to ground water was within 1.5 m of the
ground surface. Where sites are dominated by coarse sand, growth was improved if the ground
water was within 1 m of the surface (Caplan et al. 2013). Although the dominance of non-native
species has significant ecological effects, a recent review concluded that the dominance of non-
native species is unlikely to significantly affect streamflow volume or ground water levels
(Hultine and Bush 2011).
Riparian restoration also can involve such practices as intentional water releases from dams,
effluent subsidies, water conservation measures and removal of artificial bank protection
(riprap). In a study of the Sacramento River, human pressures over time (1942-1999) were
shown to increase bank erosion, increase channel length and decrease active channel width. How
to reverse these changes is difficult to predict because of the various human (e.g., bank
protection, flow diversion, sediment starvation and land-use changes) and natural changes
(i.e., flood sequences acting throughout the period and the geological setting) acting over
different time scales. The authors highlight the important effects of the Shasta Dam, which
reduced peak flow and bedload sediment supply on channel conditions. They also note, however,
the effects on the river banks, river channel and floodplain lakes of bank protection and the
construction of flood control structures; land-use changes (especially conversion of riparian
52
-------
forest to agriculture); changes in tributary sediment delivery; local geologic controls; and the
sequence of large floods (Michalkova et al. 2011).
The use of effluent subsidies from the Nogales International Wastewater Treatment Plant on
riparian vegetation development and distribution in the Santa Cruz River Valley was
investigated. The current amount, distribution and diversity of vegetation was found to be linked
to effluent supply, but an initial rapid increase in the area of riparian forest and woodland after
receiving effluent was followed by extensive cottonwood tree die-off The sites that did not
receive effluent were dominated by riparian shrub and non-native herbaceous vegetation but
were more diverse and stable over time, indicating variable long-term effects of riparian
restoration by effluent subsidy that depended on land-use history (Villarreal et al. 2012).
The effects of anthropogenic water withdrawal on pioneer riparian forests have been studied on
the Upper San Pedro River in semi-arid Arizona. Native Populus-Salix forests increased most in
conservation areas with perennial stream flows, whereas deeply rooted, invasive Tamarix
dominated in agricultural areas, which were drier (Stromberg et al. 2010). The types of plant
communities that are established naturally in arid alluvial fan and fluvial dry wash surfaces were
found to depend on whether the surfaces were dominated by high-energy flash floods or lower
energy sedimentation processes (Dickerson, Forman and Liu 2013). In a study of an ecological
water diversion project in the lower Tarim River, which flows through a semi-arid region in
China, ecosystem restoration was observed to be in progress, but restoration of dense vegetation
in the riparian buffer needed continuous water diversion (Sun et al. 2011).
4.3.3. Urban tree canopies
The effects of urban trees on stormwater conservation and infiltration—as well as estimates of
tree canopies' irrigation needs—have been the subject of field and modeling studies under semi-
arid and Mediterranean climate conditions. The effects of turfgrass shading, tree species
selection and tree density on irrigation needs were assessed. In addition, tree canopy traits were
investigated for their effect on the funneling of precipitation from canopy to ground.
4.3.3.1. Stormwater infiltration
The effects of different canopy traits on metrics of stemflow, the portion of precipitation incident
on vegetation canopies that is funneled to the base of the plant rather than reaching ground
directly from gaps in the canopy or evaporating from leaf and wood surfaces, were studied in
deciduous trees in a semi-arid climate (Kamloops, British Columbia, Canada). Stemflow
production was positively correlated with high branch angles, low bark relief in multi-leader
trees and high bark relief for single-leader trees, and greater rain event depth. For rain depths less
than 3 mm, greater stemflow was associated with leafless canopies (Carlyle-Moses and
Schooling 2015). In another study at the same location, Schooling and Carlyle-Moses (2015)
found individual tree stemflow percentages (i.e., stemflow volume as a percentage of rain
incident on the canopy) were variable even for similar rain depths, which the authors suggested
was a result of meteorological factors. The maximum stemflow was 22.8 percent for a columnar
English oak. The results of the study indicate that site-scale water balances may be affected
greatly by isolated deciduous trees with traits conducive to stemflow production, making the
53
-------
infiltration capacity at the bases of urban trees important for the design of stormwater
management with vegetation.
4.3.3.2. Stormwater conservation
Modeling residential irrigation water demand over time in Salt Lake City, researchers
determined that as urban tree canopy increases in residential urban areas, exposed turf grass
decreases. As a result, a slight decrease in residential landscape water demand was predicted
because of variations in evapotranspiration rate with landscape type (Lowry, Ramsey and
Kjelgren 2011). A field study comparing evapotranspiration rates of unshaded urban lawns with
urban lawns composed of trees and turfgrass groundcover in Los Angeles showed that irrigated
turfgrass evaporation always was higher than plot-scale tree transpiration by up to a factor of 10.
The reduction in evapotranspiration of turfgrass was attributed to shading effects of trees, which
were more important than increased transpiration from trees. Partially shading irrigated lawns
with trees is therefore a potential water-saving measure in seasonally dry climates (Litvak, Bijoor
and Pataki 2014).
4.3.3.3. Irrigation needs
A field study of urban tree transpiration rates in Los Angeles found very large species
differences in whole-tree transpiration rates, and measured results did not necessarily support
common assumptions about high versus low-water-use trees (e.g., species native to
Mediterranean climates have low rates of water use) or reflect transpiration rates in natural
ecosystems. Plot-level transpiration rates for single species were estimated to increase with tree
density, although the model did not consider such nonlinear feedbacks to transpiration at high
canopy density as self-shading, altered tree shapes and reduced canopy-atmosphere coupling.
One million new trees, a proposed target for a large-scale tree planting in Los Angeles, would
use approximately 5 percent of the total daily municipal water use if high-water-use species were
planted (Pataki et al. 2011).
4.4. Stormwater conservation
For stormwater conservation, two primary practices are described: water harvesting for
agricultural production and roof-top water harvesting systems for nonpotable use, including
indoor use and landscape irrigation. The effectiveness of different types of rainwater harvesting
systems have been compared for their ability to support vegetation, reduce soil erosion, and meet
nonpotable and domestic water needs. The impact of rainwater harvesting on the water balance
of a watershed has been assessed. Design elements to maximize effectiveness of practices also
have been studied.
4.4.1. A gricultural rainwater harvesting
The effectiveness of rainwater harvesting technologies in dryland agriculture for decreasing
runoff, increasing infiltration rates, increasing crop yields and conserving soil has been assessed
under semi-arid conditions in Africa, Asia and the Middle East. The ability of rainwater
harvesting to decrease surface runoff, increase infiltration, reduce soil erosion and decrease soil
bulk density has been measured in the field. In addition, the effects of different techniques of
rainwater harvesting on crop yield and forage biomass have been assessed, and for some
54
-------
techniques, the efficacy of different designs has been evaluated. Possible negative effects of
rainwater harvesting on downstream surface water flow also have been modeled.
4.4.1.1. Water harvesting effectiveness for infiltration
Green infrastructure techniques have been shown to be effective in water harvesting for
agriculture in semi-arid regions. In mountainous, semi-arid regions of the Kingdom of Saudi
Arabia, farmers traditionally grew crops by constructing soil and stone terraces within juniper
forest and woodlots. In recent times, many of these terraces were abandoned or damaged. In a
study comparing plots containing abandoned or maintained terraces, maintaining terraces was
found to decrease surface runoff and increase infiltration rates (El Atta and Aref 2010).
4.4.1.2. Effect on crop yield and forage biomass
A literature search was conducted by Bouma, Hegde and Lasage (2016) to identify studies in
semi-arid regions of Africa and Asia for a meta-analysis of the effectiveness of rainwater
harvesting techniques that collect, store or conserve water (as opposed to interventions that
increase the capacity of soil to retain water, such as conservation agriculture). Included studies
were published from 1989 through 2011. The authors distinguished between soil storage
technologies (e.g., planting pits, earthen bunds, plastic-covered ridge-and-furrow, stone bunds,
terraces) and reservoir storage technologies (e.g., household ponds, small check dames,
underground water tanks). Rainfall data were included in the database to distinguish possible
effects from optimal rainfall years. Mann-Whitney significance tests were conducted to assess
whether water harvesting technologies' effects on crop yields were significant and whether crop
yield effects differed across rainfall classes. An econometric analysis was conducted to assess the
effects of other factors (e.g., rainfall, soil fertility treatment) on crop yield improvements. For the
crop with the largest number of observations, maize (n = 90), plot- and farmer-associated
characteristics were captured by including information about initial crop yields. Of the 158 peer-
reviewed studies that reported the impacts of water harvesting technologies, 29 studies reported
crop yield changes. Seventeen of the studies that reported crop yield changes reported relative
yield change excluding soil fertility treatment, and 15 of them studied maize yield change.
Although the average yield change was large (78%), the standard deviation and range were
substantial. Excluding soil fertility treatment did not affect the results significantly. Absolute
rainfall did not significantly affect the changes in yield for either in-soil or reservoir water
harvesting. Rainfall harvesting appeared to be especially effective in low-rainfall conditions
(below 330 mm). Studies in which yields were high showed less relative crop yield improvement
from water harvesting technologies. No significant difference was observed between in-soil and
reservoir water harvesting technologies.
A more recent study examined the efficacy of in-field rainwater harvesting on crop yield
compared to conventional tillage. In a rural, semi-arid region of South Africa, in-field rainwater
harvesting was found to increase maize yields slightly compared with conventional tillage
(Botha, Anderson and Van Staden 2015).
4.4.1.3. Soil conservation
In their study of terraces constructed in juniper forests in Saudi Arabia, El Atta and Aref (2010)
found that abandoning rainwater-harvesting terraces increased soil loss and soil bulk density.
55
-------
4.4.1.4. Water consumption by rainwater harvesting
The downstream effects of improving soil water availability through rainwater harvesting were
modeled in a semi-arid area of Iran. Reductions in the mean annual and mean monthly flows
were modest, ranging from 2 to 5 percent and 1 to 9 percent, respectively, and much less than
converting rain-fed areas to irrigation agriculture (Masih et al. 2011).
4.4.1.5. Design elements
Innovations in plowing techniques for water harvesting have been investigated. In an arid
environment (Muwaqqar, Jordan), a new water harvesting microcatchment technique, wide
furrow with back-placed transplanting area, was tested. The technique, which uses a new type of
inexpensive plow, was compared to deep furrow plowing. The deep furrow technique showed
higher water-harvesting efficiency and greater soil water storage, but both techniques had
comparable plant productivity. The wide furrow with back-placed transplanting area technique is
uniquely amenable to mechanized planting and maintenance, however, which would encourage
large-scale implementation (Gammoh 2013). The ridge-and-furrow system was found to increase
soil water content, soil water storage at 30 cm depth, soil surface temperature and yields of
Siberian wildrye (Elymus sibiricus) in a semi-arid region of North China (He et al. 2012). A
study was conducted optimizing ridge-and-furrow ratios and ridge mulching materials for the
growth of alfalfa (Medicago sativa L.) in a semi-arid region of Northwest China. The authors
found that ridge mulching materials and ridge widths had distinct effects on topsoil temperature
at ridge tops but not at furrow bottoms. Higher than average rainfall led to significant decreases
of forage yields for manually compacted ridge soil and significant increases of forage yields for
biodegradable mulch film and common plastic film (Wang et al. 2015).
Different mulching techniques have been compared for their effectiveness. A study conducted in
the semi-arid lands of China's Loess Plateau found that plastic-covered ridge and furrow
rainwater harvesting and furrow-applied mulching performed better than bare furrow treatment
in increasing the yield of corn. The plastic mulch performed best in water use efficiency
compared with a liquid film or a biodegradable film. Corn stover mulch performed worse than
bare farrow (Chen et al. 2013). Another study of mulches using the ridge-and-furrow rainfall
harvesting system in the Loess Plateau compared standard plastic film, biodegradable film,
maize straw and liquid film to a conventional flat, no-mulch control. Standard plastic film,
biodegradable film and maize straw significantly increased maize yields by 35, 35 and
34 percent, respectively (Li et al. 2012). A comparison of surface treatment techniques
(i.e., natural, plastic cover, stone cover, hay cover and compaction) in an arid environment in
Turkey found greatest water harvesting efficiency with plastic cover, but runoff improvements
were lost when soil was saturated with water. Average pistachio plant heights were highest with
plastic cover, followed by surface compaction, hay cover and stone cover. The authors noted
potential environmental problems, however, with plastic cover (Yazar et al. 2014).
Traditional and new rainwater harvesting techniques have been compared in the field under
semi-arid conditions. In a semi-arid region of Zimbabwe, researchers compared soil moisture and
crop yield with dead-level contoured plots, noncontoured plots and plots with traditional graded
contours. They found that dead-level contours resulted in crop yield benefits in fields with soil
type conditions that enable runoff generation (silt loam soil) but were not likely to have benefits
56
-------
in soils with low runoff generation (sandy soil) (Mhizha and Ndiritu 2013). In a study in Jordan
of arid rangeland, the effects of three water harvesting techniques—contour furrows and
crescent-shaped and v-shaped microcatchments—were studied on biomass production and
natural vegetation. The researchers found that using contour furrows gave higher shrub biomass
when compared to the crescent- and v-shaped techniques (Saoub et al. 2011). Modifying contour
ridges traditionally used for rainwater management by digging infiltration pits inside contour
ridge channels was found not to improve maize yield or soil moisture content in a study in semi-
arid Zimbabwe (Nyakudya, Stroosnijder and Nyagumbo 2014).
In addition, an integrated system of micro-flood irrigation and in-field rainwater harvesting with
alternating basin and runoff strips was optimized in a semi-arid region of the Free State Province
(South Africa). Different strip widths were tested. For a 1-m runoff strip width, crop biomass and
grain yield were 19 percent and 32 percent above average. The 1-m runoff strip and full
irrigation produced optimum yields (Mavimbela and van Rensburg 2012).
4.4.2. Rain barrels and cisterns
Modeling studies have been conducted under Mediterranean, semi-arid and arid climate
conditions to assess the effectiveness of rainwater harvesting with rain barrels and cisterns to
meet nonpotable water demand and contribute to stormwater management. The effects of storage
capacity, local precipitation data and downspout disconnection have been considered. Design
criteria including slope, roughness and tank sizing have been evaluated in Mediterranean and
arid regions.
4.4.2.1. Conservation effectiveness
A water-balance analysis using EPA's SWMM model was conducted to determine the water
supply benefits of rainwater harvesting in U.S. cities, including two in the semi-arid Mountain
West (Denver and Salt Lake City), three in the arid Southwest (Albuquerque; Phoenix; and Las
Vegas, Nevada) and three from semi-arid and Mediterranean climates on the West Coast
(Sacramento, San Diego and Los Angeles), using precipitation and water demand data from the
modeled cities. They found a wide variation in cistern size needed to achieve 80 percent rooftop
runoff capture (i.e., 757 liters for the Southwest, 946 liters for the Mountain West and
3,028 liters for the West Coast). The drier regions (Mountain West and Southwest) required
smaller cisterns for 80 percent capture but were able to supply only a fraction of their indoor
water needs (i.e., 47% and 19%, respectively). Installing a single rain barrel (150 liters) would
represent less than a 30 percent nonpotable indoor water-saving efficiency for the West Coast,
Mountain West and Southwest regions (Steffen et al. 2013).
A modeling study including cities located in both Mediterranean and arid climates in Iran
determined that residential rainwater harvesting could supply 75 percent of nonpotable water
demand in buildings with larger roof areas 40 percent of the time. In arid climates, rainwater
harvesting was predicted to be able to meet 75 percent of nonpotable water demand only
23 percent of the time (Mehrabadi, Saghafian and Fashi 2013).
A water balance model compared potential water supply savings from rainwater collected from
residential roofs and gray water generated by domestic use in a Mediterranean climate
(Cranbrook, Western Australia). Historical daily rainfall and evaporation data from 1950 to 2006
57
-------
were used in the model. The researchers found that gray water use had a greater maximum
reduction of nonpotable indoor and irrigation demand (32.5%) than rainwater harvesting (25.1%)
(Zhang et al. 2010).
4.4.2.2. Stormwater management
The ability of rainwater harvesting to reduce runoff has been modeled at the neighborhood and
watershed scales under arid, semi-arid and Mediterranean climate conditions. In an analysis by
Steffen et al. (2013), the performance of rainwater harvesting was modeled using a Salt Lake
City neighborhood as a case study with precipitation data from cities in the Mountain West (Salt
Lake City), the Southwest (Phoenix) and the West Coast (Sacramento). Rainwater harvesting
was predicted to reduce runoff volume up to 20 percent in semi-arid regions. In a watershed-
scale simulation of the Chollas Creek watershed in San Diego (Mediterranean climate), runoff
reductions increased linearly with storage capacity and the number of implementing households.
Maximum reduction ranged between 10.1 and 12.4 percent using precipitation data from 1948 to
2011. Sensitivity analyses found that long-term watershed runoff reduction potential was
affected primarily by precipitation characteristics and disconnection of rooftop runoff rather than
available cistern capacity (Walsh, Pomeroy and Burian 2014).
In the study by Zhang et al. (2010) comparing residential rainwater harvesting and gray water
reuse in West Australia, gray water harvesting reduced stormwater runoff by 54.1 percent or
88.1 m3/lot/year, and rainwater harvesting reduced stormwater runoff by 48.1 percent or
68.3 m3/lot/year.
4.4.2.3. Cost-benefit analysis
The costs and benefits of office building rainwater harvesting systems were assessed in different
locations, including cities located in arid, semi-arid and Mediterranean climates (Albuquerque;
Phoenix; Salt Lake City; San Diego; and San Francisco, California). The water-saving efficiency
(i.e., average percent of water demand substituted by rainwater yield) plateaued at lower values
for these cities located in dry climates compared to temperate cities (Atlanta, Georgia; Boston,
Massachusetts; Dallas, Texas; New York, New York; Philadelphia, Pennsylvania; Seattle,
Washington; Tampa, Florida; and Wichita, Kansas), although runoff volume reduction potentials
(i.e., average percent of rooftop runoff captured relative to that generated) were higher for cities
in dry climates (see Figure 4-1). The authors suggest that considerations of the costs (including
local water utility rates) and benefits of systems should include both direct benefits such as
stormwater management and indirect benefits such as CSO mitigation (Wang and Zimmerman
2015).
4.4.2.4. Design elements
Criteria for roof design to maximize rainwater quantity and quality were developed in a study in
Barcelona, Spain (Mediterranean climate). The researchers found that sloping, smooth roofs may
harvest as much as 50 percent more rainwater than flat, rough roofs. In general, physicochemical
runoff quality was better than the average quality as described in the literature, but sloping roofs
had significantly better results for some water quality parameters than flat, rough roofs (Farreny
et al. 2011).
58
-------
Ł 40-
9 30*
10 20
Cistern size (m3)
30
NYC
BOS
PHIL
TPA
CHI
DAL
icr
SP
SLC
SD
ABQ
PHX
0 50 too 150
Precipitation (cm/year)
0 2.2 4.4 6.6
Cistern size/Roof area Ratio (cm)
precipitation>105 cm/yr
10-
10 20
Cistern size (m3)
30
ABQ'
PHX
SD
CHI-
PHIL
DAL
SEA"
ATL
0 50 100 150
Precipitation (cm/year)
0 2.2 4.4 6.6
Cistern size/Roof area Ratio (cm)
175
-------
An environmental analysis of rainwater harvesting systems for urban areas in Mediterranean
climates quantified the environmental impacts of systems through a life cycle assessment
(i.e., materials, construction, transportation, use and deconstruction). The researchers found that
a distributed-over-roof tank had the least negative environmental impact compared to the other
storage systems modeled (i.e., an underground tank and a tank below the roof) because of better
distribution of tank weight on the building, reduced reinforcement requirements and enabled
energy savings. The storage subsystem and the materials stage contributed most significantly to
the impacts. The most efficient system was a building-scale system in a compact neighborhood
with a tank distributed over the roof. It was comparable in global warming potential (measured in
production of carbon dioxide equivalents) to water production and distribution by the existing
drinking main water supply and had no energy demand during use (Angrill et al. 2012).
60
-------
Chapter 5.
Proposed Areas of Research for EPA
In this chapter, research topics of interest to EPA on applying green infrastructure in arid and
semi-arid climates to manage and conserve stormwater are presented. Research topics were
identified by experts participating in a March 2012 workshop on creating a research agenda on
adapting green infrastructure to arid environments. Additional suggestions for further study were
drawn from the literature surveyed in Chapter 4. These suggestions are grouped by BMP, just as
the research results were organized in Chapter 4. Finally, a summary of research themes that are
general to multiple practices is included in this chapter.
5.1. AridLID 2012 research agenda
In March 2012, the third Arid Low Impact Development (AridLID) Conference was held in
Tucson, Arizona. The 2012 AridLID Conference featured a workshop titled "Co-Creating an
Arid-Adapted, Integrative Green Infrastructure Research Agenda." The report from the
workshop formulates and prioritizes research questions regarding arid green infrastructure and
LID and recommends particular research questions relevant to EPA's mission that might be
addressed by the Agency in collaboration with nonfederal and federal agency partners
(Cleveland 2013).
5.1.1. EPA and nonfederal stakeholders
The workshop participants identified research questions for EPA to address in collaboration
with nonfederal stakeholders:
• Pretreatment and treatment needs before infiltration.
• Review of institutions, codes and ordinances for treating soil as a resource.
To better understand pretreatment and treatment needs before infiltration, research should
determine watershed characteristics and constituents of runoff; study the kind of treatment that is
accomplished by native soil; and determine the life-cycle costs of maintenance, particularly those
associated with clogging. Determining the levels of metals, nutrients and sediment in stormwater
and their impact on ground water quality is needed. Possible partners include university
researchers, geotechnical engineers and science-based nongovernmental organizations.
Also needed is a comprehensive review of institutions, codes and ordinances for treating soil as a
resource. Municipalities and other MS4 permit-granting jurisdictions, as well as members of the
public, can assist EPA in developing this resource. Possible approaches include studying open
space, documenting remediation and soil restoration regulations, defining soil resource areas, and
researching soil transport needs.
61
-------
5.1.2. Cross-federal agency research
Additional research opportunities would best be addressed if EPA were to partner with other
federal agencies. The opportunities include the following:
• Plant/soil interactions and pollutant removal.
• Impact of green infrastructure/LID on flood frequency and volume.
• Impact of green infrastructure/LID on floods as a downstream resource.
• Ground water impacts resulting from water infiltration.
• Database of existing findings.
• Process for prioritizing projects within a watershed.
Potential federal partners with EPA to study plant/soil interactions and pollutant removal include
the U.S. Department of Agriculture's (USDA) Natural Resources Conservation Service (NRCS)
and the U.S. Geological Survey (USGS). The NRCS conducts the National Cooperative Soil
Survey, data from which is available online (NRCS 2013), and operates the Tucson Plant
Materials Center, which produces nursery stock and seeds for regional projects. The USGS
produces geochemical and mineralogical soil maps for soils in the conterminous United States
(Smith et al. 2014). Other potential partners include academia, state-level environmental
protection agencies, consultants and private industry. The study of plant/soil interactions and
pollutant removal can best be approached by developing small-scale models, undertaking pilot
comparative projects across the region, using lysimeters to create an evapotranspiration database,
conducting short- and long-term studies, comparing soil and vegetation maps, studying sunlight
impacts, and studying impacts at different depths.
Assessment of the impact of green infrastructure/LID on flood frequency and volume, as well as
water as a downstream resource, can be accomplished through collaborations between EPA and
flood control districts, academia, the U.S. Department of Housing and Urban Development, the
Bureau of Reclamation, the U.S. Army Corps of Engineers, the USGS, and professional
associations. Possible approaches include modeling LID practices and runoff reduction,
determining maintenance requirements and long-term costs, conducting monitoring efforts to
capture data from established testing and control sites (or before-and-after studies), and assessing
impacts on downstream vegetation and riparian habitat.
Impacts on the quantity and quality of ground water impacts resulting from stormwater
infiltration merit further study. Potential research partners with EPA include the U.S. Department
of the Interior's Bureau of Reclamation, state-level environmental protection agencies, the
USGS, USDA's Agricultural Research Service, water utilities and nongovernmental
organizations. Research approaches include studying tracers in stormwater; conducting bench-
scale tests of absorption rates of various pollutants; sampling at different locations in watersheds;
assessing the long-term viability of reclaim systems to recharge; modeling infiltration; and
conducting a comparative analysis among natural, concretized, bank-protected and sandy-bottom
washes.
62
-------
Compiling a database of existing findings, providing a process for prioritizing projects, and
identifying prioritization criteria are important research goals. Research partners with EPA could
include the Agricultural Research Service, academia and student organizations. Potential
products from this research could include technical manuals describing BMPs, an international
BMP database, a compilation of local knowledge of BMP effectiveness and documentation of
vegetation types and functions.
5.2. Stormwater management research needs—site scale
Experts in the field have identified research needs for applying site-scale green infrastructure to
stormwater management in arid and semi-arid regions. Their recommendations include
replicating field studies; investigating the mechanisms behind effects on stormwater quality;
optimizing design criteria; exploring sustainable solutions to irrigation needs; improving models
of effectiveness by refining model parameters, validating models with field studies, conducting
sensitivity and uncertainty analyses, and incorporating high-resolution data; developing a better
understanding of maintenance needs; and optimizing siting of BMPs.
5.2.1. Bioswales
In a study conducted by Dorsey (2010), a wetland swale showed promise in reducing bacterial
loads to beaches in a trial under Mediterranean climate conditions. The author suggests
conducting similar projects to assess bioswales for fecal indicator bacterial removal to better
judge their success at filtering runoff and reducing bacterial densities along beaches.
5.2.2. Green roofs
Key areas for research on the implementation of green roofs in arid and semi-arid environments
include better understanding the potential negative effects on effluent water quality, improving
planting strategies, increasing water retention, selecting substrates and applying irrigation.
5.2.2.1. Negative effects on effluent water quality
Short-term studies have shown that green roofs can act as pollutant sources. Researchers
recommend a long-term study of their effects on water quality in arid and semi-arid climates
(Beecham and Razzaghmanesh 2015).
5.2.2.2. Planting strategies
Authors of studies of green roofs in dry climates had the following recommendations for future
study of planting strategies:
• Research must focus on which plants provide desired characteristics to green roofs
(e.g., stormwater retention, building cooling) rather than which simply survive (Simmons
2015).
• It is necessary to identify plants that can adapt to a broad range of conditions because
global climate change is predicted to increase climate variability (Simmons 2015).
63
-------
• Performance differences between monoculture and heterogeneous plantings should be
assessed (Razzaghmanesh, Beecham and Brien 2014).
• A better understanding of how to select appropriate drought-tolerant plant species is
needed (Van Mechelen, Dutoit and Hermy 2015).
5.2.2.3. Water retention
The following topics regarding water retention by green roofs in arid and semi-arid environments
were recommended for further study:
• The effects of antecedent dry weather periods and evapotranspiration on water retention
by green roofs in dry climates (Beecham and Razzaghmanesh 2015).
• The effectiveness of water-retention additives for green roof design in dry climates
because of their species-specific effects on drought resistance (Farrell, Ang and Rayner
2013).
• Long-term effects of hydrogels on plant water status (Savi et al. 2014).
• Substrates with greater water-retaining capacity (Van Mechelen, Dutoit and Hermy
2015).
5.2.2.4. Substrates
Because of the vulnerability of root systems to high temperatures, more research is needed on the
thermal properties of growing media for extensive green roofs in hot climates. Additional
research also is needed on the optimal water retention and drainage characteristics for growth
media in arid and semi-arid climates (Simmons 2015).
Interactions between different polymer hydrogels and substrates also should be studied,
particularly over the long term (Savi et al. 2014).
5.2.2.5. Irrigation
Further investigation is needed on the optimal irrigation regimes for green roofs in arid and semi-
arid environments (Van Mechelen, Dutoit and Hermy 2015).
5.2.3. Permeable pavement
Further research on the accuracy of paved area reduction factors is needed by comparing larger
watersheds with various imperviousness and overland routing percentages. TMDLs also could be
compared to an incentive index developed for the volume-based paved area reduction factors
(Blackler and Guo 2014).
Modeling results alone should not be relied on for designing permeable shoulder pavement. Pilot
investigations using heavy vehicle simulators to verify the design depth and structural integrity
of pavement under realistic load and traffic conditions should be conducted (Chai et al. 2012).
64
-------
Because soil clogging can reduce the freeze-thaw durability of pervious concrete, the efficacy of
maintenance procedures for cleaning partially clogged pervious concrete slabs should be
investigated in future research (Guthrie, DeMille and Eggett 2010).
5.2.4. Planter boxes
Pilot experiments showed that a bioretention environment is favorable for the growth of common
roadside weeds in Texas, but further study is needed to determine whether similar successional
changes from planted vegetation to the predominance of weeds will occur under field conditions
(Li etal. 2011).
More studies are needed to delineate the rooting effects of different vegetation on performance of
bioretention units, because rooting has been shown to affect retention times and pollutant
removal efficiencies (Kim et al. 2012)
5.2.5. Rain gardens/bioretention cells
Additional study is needed on nutrient retention, leaching of metals into effluent, irrigation
needs, siting optimization using remote sensing, and optimal design criteria for arid and semi-
arid climates. Nutrient retention is a particular concern, and research is needed under different
time scales and water availability conditions. Possible design elements that merit further research
include improved root accessibility to deeper water sources, media type, vegetation density and
sizing appropriate to evaporation and transpiration rates characteristic of arid and semi-arid
climates.
5.2.5.1. Nutrient retention
Studies are needed of the carbon budgets of bioretention systems in xeric climates to ensure that
they are net carbon sinks and do not contribute to global warming (Houdeshel, Pomeroy and
Hultine 2012).
Long-term studies of nutrient treatment by vegetated and nonvegetated bioretention cells, as well
as studies of the effects of cell age and temperature, are needed (Houdeshel et al. 2015).
Research also is needed to determine the cause of elevated nutrients in rain garden effluent and
to improve performance (Jiang, Yuan and Piza 2015). Studying nutrient uptake efficiency under
different water availability scenarios might determine whether low plant and microbial activity
might be limiting nitrogen uptake in bioretention cells (Houdeshel et al. 2015).
5.2.5.2. Pollutant leaching
More research is needed on leaching of metals in bioretention cells (Jiang, Yuan and Piza 2015).
5.2.5.3. Irrigation
Xeric-adapted upland vegetation showed poor nutrient retention, but wetland communities were
effective in nutrient retention. Further research is needed on integrating bioretention and gray
water treatment to provide a sustainable water and nutrient source for wetland communities in
arid and semi-arid environments (Houdeshel et al. 2015).
65
-------
5.2.5.4. Siting
Remote sensing has been used to evaluate bioretention siting options. Future work should
include the use of this approach in an area with available monitored runoff data and high-
resolution DEM data (Khin et al. 2016).
5.2.5.5. Design criteria
Researchers had the following recommendations for further investigations of optimal designs:
• Future studies of bioretention cells should include designs where upland vegetation can
access soil water below the gravel storage layer, which may require unlined bioretention
cells or deeper test cells with a layer of soil underneath the gravel layer. The shrubs used
in a study of bioretention cells typically have deeper roots than the depth of the test cells.
Inclusion of a liner might have artificially reduced primary productivity and the
corresponding ability to remove nitrates (Houdeshel et al. 2015).
• A comparison of the phosphate treatment capacity of different media, such as expanded
shale, gravel and pumice, should be made to determine whether the use of expanded shale
is merited given its higher cost (Houdeshel et al. 2015).
• Vegetation density should be considered as a variable in future studies of bioretention of
nutrients using upland plants (Houdeshel et al. 2015).
• More research is needed to measure transpiration and evaporation rates under xeric
conditions so that facilities can be sized appropriately to meet water needs of plantings
(Houdeshel, Pomeroy and Hultine 2012).
5.2.6. Vegetative filter strips
The reduction of runoff and soil loss by vegetative filter strips was assessed using models of
physical processes to extrapolate from a limited empirical data set. Although this approach is
preferable to an empirical analysis, future modeling work should include stochastic variation of
input factors and global sensitivity and uncertainty analysis (Campo-Bescos et al. 2015).
Modeling of mitigation of pesticide loading by vegetative filter strips has been conducted
(Munoz-Carpena et al. 2015). The authors suggest that to optimize models, field work is needed
that considers degradation by monitoring pesticide concentrations, both within and when exiting
the vegetative filter strip, over time during a series of runoff and pesticide loading events. In
particular, data are needed to estimate the mass that is being deposited in the mixing layer of the
filter strip.
5.2.7. Integrated systems
More pilot studies conducted in arid and semi-arid climates are needed to help stormwater
managers assess the ability of integrated bioretention systems, such as combinations of rain
gardens and bioswales, to remove pollutants from stormwater (David et al. 2015).
High-resolution ground surface topography from light detection and ranging (commonly known
as LiDAR) would provide more accurate drainage network datasets for prioritizing remediation
of degraded urban streams, addressing water quality by facilitating infiltration, and reducing high
66
-------
runoff volume via water retention. Including other high-resolution data (e.g., soil, DEM,
impervious surface, stream network) would improve the ability of a remote sensing approach to
prioritize LID. Simulating underground drainage networks in urbanized areas is a challenge,
partially because of unregulated stream burial during development (Martin-Mikle et al. 2015).
5.3. Stormwater management research needs—watershed scale
To improve the use of green infrastructure for stormwater management on a watershed scale,
experts have similar research recommendations as those for site-scale practices but with a greater
emphasis on modeling. Prioritizing sites for installing BMPs, assessing economic viability,
replicating results in other locations, incorporating high-resolution data in models, improving the
mechanistic understanding of biogeochemical processes, gathering more environmental data
relevant to successful installation of practices, and improving model parameterization all are
important areas of study for arid and semi-arid applications of green infrastructure BMPs.
5.3.1. Land conservation
For conservation of agricultural land, research is needed on siting and economic viability.
Modeling of watershed-scale implementation of BMPs suggested that fallowing irrigated land
would reduce nitrate and selenium loading. For future work, researchers suggest investigating the
effects of implementing BMPs—including fallowing land and enhancing riparian buffers—on a
local, site-specific basis and evaluating the economic viability of alternative BMPs (Bailey,
Gates and Romero 2015; Bailey, Romero and Gates 2015).
Research is needed on possible beneficial effects of land conservation on water quality in arid
and semi-arid urban areas. Hale et al. (2015) modeled the effects of changing infrastructure
practices on nutrient and DOC fluxes from urban watershed systems using historical data from
Phoenix. The authors suggest further studies to determine the applicability of their model to
other arid urban areas.
More research is needed on approaches for reestablishing vegetation in cold and dry desert
environments. In a study of restoring native vegetation in a cold-arid climate, hollow frame snow
fencing, which increased soil retention, improved restoration success. Further research is needed
to understand the ecological drivers of the benefits of snow fencing. A better understanding also
is needed of the importance of the effects of snow fencing on the duration and penetration depth
of soil moisture for its beneficial impacts on revegetation. The hollow frame fence system has
possible applications for managing wind and particulates in such ecosystems as hot deserts
(David 2013).
5.3.2. Riparian buffers
A better understanding of the link between flow and channel degradation is needed. Modeling
studies could be conducted for channels with geomorphological data to determine whether
changes in flow are correlated with channel degradation. Future modeling efforts should use
finer resolution data than mean daily flow to produce more precise rainfall-runoff estimates
(Hawley and Bledsoe 2013).
The biogeochemical properties that affect the ability of riparian buffers to reduce pollution
loading to receiving water are complex. Future work in modeling the effects of riparian buffers
67
-------
might involve improving the representation of riparian areas and vegetation (Bailey, Romero and
Gates 2015). A better understanding of how to model spatiotemporal distribution of
biogeochemical properties that influence denitrification within the riparian corridor
(e.g., vegetative mix, organic carbon content, stratigraphy, hydraulic conductivity, water table
fluctuation, dissolved oxygen) would enhance the ability to predict the effects of riparian
corridors on nitrate loading to surface water (Bailey, Gates and Romero 2015).
To better plan riparian habitat protection and restoration efforts, more research is needed on
optimal revegetation conditions, site prioritization and buffer widths. For revegetation projects,
further research could involve placing piezometers directly in willow swales. It also would be
beneficial to monitor seasonal soil moisture availability in depth increments between the ground
surface and late summer water table (Caplan et al. 2013). Rivers produce substantial benefits for
wildlife habitat, and a better understanding of channel habitats and the ecosystems they support
is needed to prioritize protection of the most biodiverse river-fed lake habitats (Michalkova et al.
2011). On-the-ground monitoring is needed to further reduce uncertainty and provide a means to
adjust buffer width prescriptions on timber harvesting in riparian buffer zones to meet riparian
habitat protection objectives (Teply, McGreer and Ceder 2014).
5.3.3. Urban tree canopies
The source of water for urban trees generally is poorly constrained. Detailed studies of urban
water budgets are needed to understand the available water sources and plan for the irrigation
needs of different species of urban trees (Pataki et al. 2011).
A better understanding of modeling parameters for evapotranspiration and irrigation water
demand is needed. More research is needed on urban plant evapotranspiration rates under a range
of temporal and spatial conditions, such as for recently established plantings and under
conditions of water stress, to develop water-conserving landscaping strategies in semi-arid urban
environments (Litvak, Bijoor and Pataki 2014). Modeling efforts to predict the effects of urban
residential landscapes on irrigation water demand would be improved by better parameterization
of the irrigation water demand model, including more reliable water-loss coefficients for
horticultural and landscape plants and the Distribution Uniformity factor, which drives excess
irrigation in turfgrass landscapes (Lowry, Ramsey and Kjelgren 2011).
5.4. Stormwater conservation research needs
As for stormwater management, green infrastructure approaches to stormwater conservation in
arid and semi-arid environments would benefit from replication of research studies under a
variety of conditions, including geography, climate, soil type and plant species. More systematic
and comparable studies are needed. Cost-effectiveness of BMPs under a variety of scales needs
to be better understood.
5.4.1. A gricultural water harvesting
Given the large heterogeneity of water harvesting technologies, study sites and approaches to
assessing effectiveness, more systematic and comparable studies reporting effects on crop yields
are needed. Biophysical and hydrological factors, as well as farmer characteristics and crop
choice, should be accounted for (Bouma, Hegde and Lasage 2016).
68
-------
In a meta-analysis of rain harvesting studies conducted in Asia and Africa, Bouma, Hegde and
Lasage (2016) noted that more crop yield studies have been conducted in Asia, which generally
enjoys higher yields, than Africa. The authors suggest that additional studies are needed to assess
possible geographic and precipitation bias, particularly for low-rainfall years in Africa. Also,
studies that use a broader range of crops, especially other than maize, and under specific contexts
are needed.
Future work should investigate the optimum ridge-furrow ratio and suitable ridge-mulching
material for achieving the best environmental and economic benefit under different climatic
conditions, soil types and plant species. More research is needed as well on using biodegradable
mulching materials (Wang et al. 2015).
5.4.2. Rain barrels and cisterns
A better understanding of the scaling of the costs and benefits of rainwater harvesting systems is
needed. Evaluations of costs and effectiveness at the site, neighborhood and municipal levels
should be performed (Jiang, Yuan and Piza 2015). Different individual LID practices and
combinations of practices, including rain barrels and cisterns, should be assessed to maximize
watershed benefits with cost-effectiveness. Geographic placement and sizing of LID networks
should be based on land cover and other watershed parameters (Walsh, Pomeroy and Burian
2014). A better understanding is needed for household-, neighborhood- and regional-scale
rainwater harvesting systems in arid and semi-arid regions of nonmonetized benefits, such as
water pollution control and community amenities; energy implication of onsite alternative water
supplies; and long-term system performance and maintenance (NAS 2016).
More research is needed comparing the reliability of systems during wet and dry periods. Long-
term average annual rainfall is not adequate for assessing system reliability, given seasonal and
annual dry periods characteristic of arid and semi-arid regions (Mehrabadi, Saghafian and Fashi
2013).
5.5. Research themes
A number of themes for research needs apply to multiple BMPs. These research needs include
the following:
• A better understanding is needed of the performance in arid and semi-arid regions of
practices that have received less study under those conditions (e.g., bioswales, permeable
pavement, planter boxes).
• More research is needed about long-term performance and maintenance needs under arid
and semi-arid conditions.
• The mechanisms by which substrates, soil-containing or otherwise, might be a source of
nutrient or other pollutant loading from the breakdown of the substrate need attention.
• The spatial scale of research on BMPs in arid and semi-arid regions should be developed
more beyond pilot-scale projects.
• A better understanding is needed of how to prioritize siting of BMPs to maximize
stormwater management and conservation goals, including water quality improvement.
69
-------
More testing is needed to validate models developed for temperate regions for application
under arid and semi-arid conditions.
Inclusion of higher resolution data will be important for applying proven models to arid
and semi-arid environments.
A better understanding is needed of the economic and nonmonetized benefits of
stormwater management and conservation through green infrastructure practices
implemented in arid and semi-arid regions at a range of scales, including site,
neighborhood, municipal, watershed and regional.
More green infrastructure research needs to be conducted in the United States to ensure
that results are applicable to the specific conditions of the arid and semi-arid regions in
this country.
70
-------
Chapter 6.
Conclusions and Recommendations
The results of this literature review indicate that the body of current research on the design,
implementation, maintenance and effectiveness of green infrastructure under arid, semi-arid and
seasonally dry conditions is extensive for some practices and more limited for others. The
literature on green roofs in dry or seasonally dry environments is extensive, as is research on
rainwater harvesting for dryland agriculture, but relatively few studies have been published on
the performance of such commonly used vegetative practices as bioswales and planter boxes.
Permeable pavement, although not based on vegetation, would be expected to face unique
conditions in arid and semi-arid environments, such as rapid freeze-thaw cycles, but few studies
of this practice have been conducted under field conditions in these regions.
Research has tended to emphasize design over maintenance, as well as the management of
stormwater quantity over stormwater quality. For example, green roofs have been studied
extensively to optimize their design criteria under dry conditions, but the important question of
whether substrates, soil-containing or otherwise, might be a source of nutrient or other pollutant
loading from the breakdown of the substrate has received less attention.
Scale, both temporal and spatial, is an issue for applying research findings to practice. Most
studies have been conducted on pilot-scale systems. The short timescales of these investigations
might not be representative of the long-term performance of practices meant to last decades,
such as green roofs. Implementation of green infrastructure at watershed scales might have
significant differences in performance and cost-effectiveness compared with site-scale
implementation. In particular, a better understanding of how to prioritize siting to optimize cost-
effectiveness is needed for neighborhood- and watershed-scale implementation.
Modeling is key to planning green infrastructure implementation. More testing of models
developed for temperate regions and inclusion of higher resolution data will be important for
applying proven models to arid and semi-arid environments.
Finally, some of the lessons learned from international research might not be directly applicable
to the United States. Vegetation native to other countries might not reflect characteristics of
native U.S. vegetation adapted to conditions in arid or semi-arid regions of United States. In
addition, the United States relies primarily on mechanized agriculture, and many in-field rain
harvesting techniques have been developed and tested for subsistence farming. Therefore, more
arid and semi-arid green infrastructure research needs to be conducted in the United States.
Green infrastructure shows great promise in its ability to store and treat stormwater in arid and
semi-arid environments. Further research on effectiveness of BMPs in stormwater management,
treatment and conservation under the unique conditions of these regions and at scales relevant to
their implementation in the "real world" will be beneficial to developing more cost-effective
approaches to stormwater regulatory compliance.
71
-------
Chapter 7.
Acknowledgments
The preparation of this report was conducted under the task "Arid Green Infrastructure: State of
the Science and Research Needs for Arid/Semi-Arid Regions." The project was funded by the
EPA ORD. Barclay Inge and David Tran conducted the literature search. Beverly Campbell,
Kristen LeBaron, Margaret Christoph and Sally Paustian edited the report.
SCG would like to thank the individuals who conducted a peer review of this report, Alice
Witheridge, Water Research Foundation; Christopher Impellitteri, ORD/EPA; Harry Zhang,
Ph.D., P.E., Water Environment Research Foundation; and Steve Kraemer, ORD/EPA.
72
-------
Chapter 8.
References
Amigos de los Rios. 2016. "What We Do: Revitalizing Our Communities by Restoring Nature's
Balance." Amigos de los Rios. Last modified 2016. Accessed April 8, 2016.
www.amigosdelosrios.org/what-we-do.
AMNH. 2015. "GRACE: California Central Valley." American Museum of Natural History. Last
modified 2015. Accessed April 6, 2016. www.amnh.org/explore/curriculum-
collections/grace/grace-tracking-water-from-space/california-central-vallev.
Angrill, S., R. Farreny, C. M. Gasol, X. Gabarrell, B. Vinolas, A. Josa, and J. Rieradevall. 2012.
"Environmental analysis of rainwater harvesting infrastructures in diffuse and compact urban
models of Mediterranean climate." International Journal of Life Cycle Assessment 17 (1): 25-42.
Bailey, R. T., T. K. Gates, and E. C. Romero. 2015. "Assessing the effectiveness of land and water
management practices on nonpoint source nitrate levels in an alluvial stream-aquifer system."
Journal of Contaminant Hydrology 179: 102-15.
Bailey, R. T., E. C. Romero, and T. K. Gates. 2015. "Assessing best management practices for
remediation of selenium loading in groundwater to streams in an irrigated region." Journal of
Hydrology 521: 341-59.
Beecham, S., D. Pezzaniti, and J. Kandasamy. 2012. "Stormwater treatment using permeable pavements."
Proceedings of the Institution of Civil Engineers: Water Management 165 (3): 161-70.
Beecham, S., and M. Razzaghmanesh. 2015. "Water quality and quantity investigation of green roofs in a
dry climate." Water Research 70: 370-84.
Belnap, J., J. R. Welter, N. B. Grimm, N. Barger, and J. A. Ludwig. 2005. "Linkages between microbial
and hydrologic processes in arid and semiarid watersheds." Ecology 86 (2): 298-07.
Betrie, G. D., Y. A. Mohamed, A. Van Griensven, and R. Srinivasan. 2011. "Sediment management
modelling in the Blue Nile Basin using SWAT model." Hydrology and Earth System Sciences 15
(3): 807-18.
Blackler, G. E., and J. C. Y. Guo. 2014. "Field test of paved area reduction factors using a storm water
management model and water quality test site." Journal of Irrigation and Drainage Engineering
140 (4): 05013001.
Botha, J. J., J. J. Anderson, and P. P. Van Staden. 2015. "Rainwater harvesting and conservation tillage
increase maize yields in South Africa." Water Resources and Rural Development 6: 66-77.
Bouma, J. A., S. S. Hegde, and R. Lasage. 2016. "Assessing the returns to water harvesting: A meta-
analysis." Agricultural Water Management 163: 100-09.
73
-------
Calil, J., M. W. Beck, M. Gleason, M. Merrifield, K. Klausmeyer, and S. Newkirk. 2015. "Aligning
natural resource conservation and flood hazard mitigation in California." PloS ONE 10 (7):
e0132651.
Campo-Bescos, M. A., R. Munoz-Carpena, G. A. Kiker, B. W. Bodah, and J. L. Ullman. 2015. "Watering
or buffering? Runoff and sediment pollution control from furrow irrigated fields in arid
environments." Agriculture, Ecosystems and Environment 205: 90-101.
CantyMedia. 2016. "Weatherbase." Accessed April 15, 2016. www.weatherbase.com.
Cao, C. T. N., C. Farrell, P. E. Kristiansen, and J. P. Rayner. 2014. "Biochar makes green roof substrates
lighter and improves water supply to plants." Ecological Engineering 71: 368-74.
Caplan, T. R., K. Cothern, C. Landers, and O. C. Hummel. 2013. "Growth response of coyote willow
(Salix exigua) cuttings in relation to alluvial soil texture and water availability." Restoration
Ecology 21 (5): 627-38.
Carlson, W., E. Fitzpatrick, E. Flanagan, R. Kirschbaum, H. Williams, and L. Zickler. 2013. Eastern
Washington Low Impact Development Guidance Manual. Olympia, WA: Department of Ecology,
State of Washington.
Carlyle-Moses, D. E., and J. T. Schooling. 2015. "Tree traits and meteorological factors influencing the
initiation and rate of stemflow from isolated deciduous trees." Hydrological Processes 29 (18):
4083-99.
CEPA. 2016. Slow the Flow: A Film About Sponges, Landscapes, and Permeable Cities. California
Environmental Protection Agency, www.waterboards.ca.gov/stormfllm.
CEQ. 2015. Implementing Instructions for Executive Order 13693: Planning for Federal Sustainability in
the Next Decade. Washington, D.C.: Council on Environmental Quality, The White House.
Chai, L., M. Kayhanian, B. Givens, J. T. Harvey, and D. Jones. 2012. "Hydraulic performance of fully
permeable highway shoulder for storm water runoff management." Journal of Environmental
Engineering-ASCE 138 (7): 711-22.
Chen, X. L., P. T. Wu, X. N. Zhao, and N. Persaud. 2013. "Effect of different mulches on harvested
rainfall use efficiency for corn (Zea mays L.) in semi-arid regions of northwest China." Arid Land
Research and Management 21 (3): 272-85.
Chow, W. T. L., and A. J. Brazel. 2012. "Assessing xeriscaping as a sustainable heat island mitigation
approach for a desert city." Building and Environment 47: 170-81.
Chowdhury, R. K. 2015. "Greywater reuse through a bioretention system prototype in the arid region."
Water Science and Technology 72 (12): 2201-11.
City of Los Angeles. 2009. City of Los Angeles Rainwater Harvesting Program: A Homeowner's "How-
To" Guide. Los Angeles, CA: City of Los Angeles.
74
-------
City ofLos Angeles. 2011. Development Best Management Practices Handbook: Part B, Planning
Activites. Fourth Edition. Los Angeles, CA: Board of Public Works, City ofLos Angeles.
City of Tucson. 2015. Low Impact Development and Green Infrastructure Guidance Manual. Tucson,
AZ: City of Tucson.
Ciudad Soil and Water Conservation District. 2016. "About AridLID.org." Watershed Management
Group. Last modified 2016. Accessed April 8, 2016. www.aridlid.org/about.
Cleveland, J. 2013. Co-Creating an Arid-Adapted, Integrative Green Infrastructure Research Agenda.
Tucson, AZ: Water Resources Research Center, University of Arizona.
Cohen, D. T., G. W. Hatchard, and S. G. Wilson. 2015. Population Trends in Incorporated Places: 2000
to 2013. Washington, D.C.: U.S. Census Bureau.
Copeland, H. E., S. A. Tessman, E. H. Girvetz, L. Roberts, C. Enquist, A. Orabona, S. Patla, and J.
Kiesecker. 2010. "A geospatial assessment on the distribution, condition, and vulnerability of
Wyoming's wetlands." Ecological Indicators 10(4): 869-79.
Coughlin, J. P., C. D. Campbell, and D. C. Mays. 2012. "Infiltration and clogging by sand and clay in a
pervious concrete pavement system." Journal ofHydrologic Engineering 17 (1): 68-73.
Council for Watershed Health. 2015a. "Media Library." Last modified 2015. Accessed June 20, 2016.
http://watershedhealth.org/dataandreference/media.aspx.
Council for Watershed Health. 2015b. "Water Augmentation Study." Last modified 2015. Accessed June
20, 2016. http://watershedhealth.org/programsandprojects/was.aspx.
Critchley, W., and K. Siegert. 1991. Water Harvesting. Rome: Food and Agriculture Organization of the
United Nations. www.fao.org/docrep/U3160E/u3160e00.HTM.
CSQA. 2016. "California LID Portal." California Stormwater Quality Association. Last modified 2016.
Accessed April 8, 2016. www.casqa.org/resources/california-lid-portal.
David, E. 2013. "Innovative snow harvesting technology increases vegetation establishment success in
native sagebrush ecosystem restoration." Plant and Soil 373 (1-2): 843-56.
David, M., S. Follain, R. Ciampalini, Y. Le Bissonnais, A. Couturier, and C. Walter. 2014. "Simulation of
medium-term soil redistributions for different land use and landscape design scenarios within a
vineyard landscape in Mediterranean France." Geomorphology 214: 10-21.
David, N., J. E. Leatherbarrow, D. Yee, and L. J. McKee. 2015. "Removal efficiencies of a bioretention
system for trace metals, PCBs, PAHs, and dioxins in a semiarid environment." Journal of
Environmental Engineering 141 (6): 04014092.
Detenbeck, N., M. Tenbrink, R. Abele, J. Leclair, T. Garrigan, V. Zoltay, B. Small, A. Brown, I. Morin,
and A. Morrison. 2016. Watershed Management Optimization Support Tool (WMOST) v2: User
Manual and Case Studies. EPA-600-R-15-059. Washington, D.C.: U.S. Environmental Protection
Agency Office of Research and Development.
75
-------
Di, Y. H., P. Lin, and Z. P. Wang. 2014. "Analysis of eco-green wall building's indoor thermal
environment." Xi'an Jianzhu Keji Daxue Xuebao/Journal ofXi'an University of Architecture and
Technology 46 (4): 553-56.
Dickerson, R. P., A. Forman, and T. Z. Liu. 2013. "Co-development of alluvial fan surfaces and arid
botanical communities, Stonewall Flat, Nevada, USA." Earth Surface Processes andLandforms
38 (10): 1083-01.
Dorsey, J. H. 2010. "Improving water quality through California's Clean Beach Initiative: An assessment
of 17 projects." Environmental Monitoring and Assessment 166 (1-4): 95-111.
El Atta, H. A., and I. Aref. 2010. "Effect of terracing on rainwater harvesting and growth of Juniperus
procera Hochst. ex Endlicher." International Journal of Environmental Science and Technology
7 (1): 59-66.
Escalona, A., M. D. Salas, C. D. Coutinho, and M. Guzman. 2013. "How does salinity affect mineral ion
relations and growth of Lobelia erinus for use in urban landscaping?" Journal of Food
Agriculture & Environment 11 (2): 854-58.
FAO. 2015. Global Guidelines for the Restoration of Degraded Forests and Landscapes in Drylands.
Rome: Food and Agriculture Organization of the United Nations.
Farrell, C., X. Q. Ang, and J. P. Rayner. 2013. "Water-retention additives increase plant available water in
green roof substrates." Ecological Engineering 52: 112-18.
Farrell, C., R. E. Mitchell, C. Szota, J. P. Rayner, and N. S. G. Williams. 2012. "Green roofs for hot and
dry climates: Interacting effects of plant water use, succulence and substrate." Ecological
Engineering 49: 270-76.
Farreny, R., T. Morales-Pinzon, A. Guisasola, C. Taya, J. Rieradevall, and X. Gabarrell. 2011. "Roof
selection for rainwater harvesting: Quantity and quality assessments in Spain." Water Research
45 (10): 3245-54.
Feng, H., and K. Hewage. 2014. "Lifecycle assessment of living walls: Air purification and energy
performance." Journal of Cleaner Production 69: 91-99.
FLL. 2008. Guidelines for the Planning, Construction and Maintenance of Green Roofing: Green
Roofing Guideline. Bonn, Germany: Forschungsgesellschaft Landschaftsentwickung
Landschaftsbau.
Fulton, L. V., N. D. Bastian, F. A. Mendez, and R. M. Musal. 2013. "Rainwater harvesting system using a
non-parametric stochastic rainfall generator." Simulation-Transactions of the Society for
Modeling and Simulation International 89 (6): 693-702.
Gammoh, I. A. 2013. "An improved wide furrow micro-catchment for large-scale implementation of
water-harvesting systems in arid areas." Journal of Arid Environments 88: 50-56.
76
-------
Garrison, N., C. Kloss, and R. Lukes. 2011. Capturing Rainwater From Rooftops: An Efficient Water
Resource Management Strategy That Increases Supply and Reduces Pollution. New York:
Natural Resources Defense Council.
GIC. 2014a. Federal Agency Support for the Green Infrastructure Collaborative: July 16, 2014
(Amended October 8, 2014). Washington, D.C.: Green Infrastructure Collaborative.
GIC. 2014b. Green Infrastructure Collaborative Statement of Support: October 8, 2014. Washington,
D.C.: Green Infrastructure Collaborative.
Green, E. 2010. "The dry garden: Elmer Avenue becomes Green Street, a water-wise and solar-lighted
community effort." Los Angeles Times. Published July 23, 2010. Accessed April 8, 2016.
http://latimesblogs.latimes.com/home_blog/2010/07/elmer-avenue-sun-valley.html.
GSA. 2015. "Green Roofs." U.S. General Services Administration. Last modified July 28, 2015.
Accessed April 8, 2016. www.gsa.gov/portal/content/166443.
Gumiere, S. J., J. S. Bailly, B. Cheviron, D. Raclot, Y. L. Bissonnais, and A. N. Rousseau. 2015.
"Evaluating the impact of the spatial distribution of land management practices on water erosion:
Case study of a Mediterranean catchment." Journal ofHydrologic Engineering 20 (6): C5014004.
Guo, Y. P., S. H. Zhang, and S. G. Liu. 2014. "Runoff reduction capabilities and irrigation requirements
of green roofs." Water Resources Management 28 (5): 1363-78.
Guthrie, W. S., C. B. DeMille, and D. L. Eggett. 2010. "Effects of soil clogging and water saturation on
freeze-thaw durability of pervious concrete." Transportation Research Record (2164): 89-97.
Haggag, M., A. Hassan, and S. Elmasry. 2014. "Experimental study on reduced heat gain through green
facades in a high heat load climate." Energy and Buildings 82: 668-74.
Hajani, E., and A. Rahman. 2014. "Rainwater utilization from roof catchments in arid regions: A case
study for Australia." Journal of Arid Environments 111: 35-41.
Hale, R. L., L. Turnbull, S. R. Earl, D. L. Childers, and N. B. Grimm. 2015. "Stormwater infrastructure
controls runoff and dissolved material export from arid urban watersheds." Ecosystems 18 (1):
62-75.
Hawley, R. J., and B. P. Bledsoe. 2013. "Channel enlargement in semiarid suburbanizing watersheds: A
southern California case study." Journal of Hydrology 496: 17-30.
He, F., K. Wang, X. Li, and Z. Xu. 2012. "Effects of ridge and furrow rainfall harvesting system on soil
hydrothermal condition and yields of Elymus sibiricus L. in arid and semiarid regions." Nongye
Gongcheng Xuebao/Transactions of the Chinese Society of Agricultural Engineering 28 (12):
122-26.
Heintzman, L. J., T. A. Anderson, D. L. Carr, and N. E. Mclntyre. 2015. "Local and landscape influences
on PAH contamination in urban stormwater." Landscape and Urban Planning 142: 29-37.
77
-------
Houdeshel, C. D., K. R. Hultine, N. C. Johnson, and C. A. Porneroy. 2015. "Evaluation of three
vegetation treatments in bioretention gardens in a semi-arid climate." Landscape and Urban
Planning 135: 62-72.
Houdeshel, C. D., C. A. Pomeroy, and K. R. Hultine. 2012. "Bioretention design for xeric climates based
on ecological principles." Journal of the American Water Resources Association 48 (6): 1178-90.
Houdeshel, D., and C. Pomeroy. 2014. "Storm-water bioinfiltration as no-irrigation landscaping
alternative in semiarid climates." Journal of Irrigation and Drainage Engineering 140 (2):
06013004.
Huang, J. J., Y. Li, S. Niu, and S. H. Zhou. 2014. "Assessing the performances of low impact
development alternatives by long-term simulation for a semi-arid area in Tianjin, northern
China." Water Science and Technology 70 (11): 1740-45.
Hultine, K. R., and S. E. Bush. 2011. "Ecohydrological consequences of non-native riparian vegetation in
the southwestern United States: A review from an ecophysiological perspective." Water
Resources Research 47.
Jiang, Y., Y. Yuan, and H. A. Piza. 2015. "A review of applicability and effectiveness of low impact
development/green infrastructure practices in arid/semi-arid United States." Environments 2:
221-49.
Kayhanian, M., D. Anderson, J. T. Harvey, D. Jones, and B. Muhunthan. 2012. "Permeability
measurement and scan imaging to assess clogging of pervious concrete pavements in parking
lots." Journal of Environmental Management 95 (1): 114-23.
Khin, M. M. L., A. Shaker, D. Joksimovic, and W. Y. Yan. 2016. "The use of WorldView-2 satellite
imagery to model urban drainage system with low impact development (LID) techniques."
Geocarto International 31 (1): 92-108.
Kim, M. H., C. Y. Sung, M. H. Li, and K. H. Chu. 2012. "Bioretention for stormwater quality
improvement in Texas: Removal effectiveness of Escherichia coli." Separation and Purification
Technology 84: 120-24.
Kim, Y. J., S. A1 Wakeel, A. Gaddafi, and Y. J. Cha. 2015. "In-place performance of severely deteriorated
pervious concrete: A case study." ACIMaterials Journal 112 (2): 295-304.
Korgaonkar, Y., I. Burns, D. Guertin, D. Goodrich, C. Unkrich, J. Barlow, and W. Kepner. 2014.
Representing Green Infrastructure Management Techniques in Arid and Semi-Arid Regions:
Software Implementation and Demonstration Using the AGWA/KINEROS2 Watershed Model.
EPA-600-R-14-329. Washington, D.C.: U.S. Environmental Protection Agency.
La Roche, P., and U. Berardi. 2014. "Comfort and energy savings with active green roofs." Energy and
Buildings 82: 492-504.
Lasage, R., and P. H. Verburg. 2015. "Evaluation of small scale water harvesting techniques for semi-arid
environments " Journal of Arid Environments 118: 48-57.
78
-------
Lee, J., M. Dimmitt, C. Lee, and R. Arroyo. 2011. "Deserts of North America." The Encyclopedia of
Earth. Published October 16, 2011. Accessed April 8, 2016.
www. eoearth.org/view/article/151710.
Lewis, D. J., E. R. Atwill, M. S. Lennox, M. D. G. Pereira, W. A. Miller, P. A. Conrad, and K. W. Tate.
2010. "Management of microbial contamination in storm runoff from California coastal dairy
pastures." Journal of Environmental Quality 39 (5): 1782-89.
Li, H., J. T. Harvey, T. J. Holland, and M. Kayhanian. 2013. "The use of reflective and permeable
pavements as a potential practice for heat island mitigation and stormwater management."
Environmental Research Letters 8 (1): 015023.
Li, M. H., C. Y. Sung, M. H. Kim, and K. H. Chu. 2011. "Assessing performance of bioretention boxes in
hot and semiarid regions highway application pilot study." Transportation Research Record
(2262): 155-63.
Li, R., X. Q. Hou, Z. K. Jia, Q. F. Han, and B. P. Yang. 2012. "Effects of rainfall harvesting and
mulching technologies on soil water, temperature, and maize yield in Loess Plateau region of
China." Soil Research 50 (2): 105-13.
Litvak, E., N. S. Bijoor, and D. E. Pataki. 2014. "Adding trees to irrigated turfgrass lawns may be a
water-saving measure in semi-arid environments." Ecohydrology 1 (5): 1314-30.
Lowry, J. H., R. D. Ramsey, and R. K. Kjelgren. 2011. "Predicting urban forest growth and its impact on
residential landscape water demand in a semiarid urban environment." Urban Forestry and
Urban Greening 10 (3): 193-204.
MacAdam, J. 2012. Green Infrastructure for Southwestern Neighborhoods. Version 1.2. Tucson, AZ:
Watershed Management Group.
Mancini, M., O. Mikhailova, W. Hime, A. Sorrel, and J. Wuertz. 2010. Aesthetically Enhanced Detention
and Water Quality Ponds. Denver, CO: Development Services, City and County of Denver.
Martin-Mikle, C. J., K. M. de Beurs, J. P. Julian, and P. M. Mayer. 2015. "Identifying priority sites for
low impact development (LID) in a mixed-use watershed." Landscape and Urban Planning 140:
29-41.
Masih, I., S. Maskey, S. Uhlenbrook, and V. Smakhtin. 2011. "Impact of upstream changes in rain-fed
agriculture on downstream flow in a semi-arid basin." Agricultural Water Management 100 (1):
36-45.
Mavimbela, S. S. W., and L. D. van Rensburg. 2012. "Integrating micro-flood irrigation with in-field
rainwater harvesting: Maize yield and water use efficiency." Irrigation and Drainage 61: 70-81.
McPherson, G., J. R. Simpson, P. J. Peper, S. E. Maco, and Q. Xiao. 2005. "Municipal forest benefits and
costs in five US cities." Journal of Forestry 103 (8): 411-16.
79
-------
Mehrabadi, M. H. R., B. Saghafian, and F. H. Fashi. 2013. "Assessment of residential rainwater
harvesting efficiency for meeting non-potable water demands in three climate conditions."
Resources Conservation and Recycling 73: 86-93.
Meigs, P. 1953. "World distribution of arid and semi-arid homoclimates." In Reviews of Research on Arid
Zone Hydrology, edited by UNESCO, 203-210. Paris: United Nations.
Mhizha, A., and J. G. Ndiritu. 2013. "Assessing crop yield benefits from in situ rainwater harvesting
through contour ridges in semi-arid Zimbabwe." Physics and Chemistry of the Earth 66: 123-30.
Michalkova, M., H. Piegay, G. M. Kondolf, and S. E. Greco. 2011. "Lateral erosion of the Sacramento
River, California (1942-1999), and responses of channel and floodplain lake to human
influences." Earth Surface Processes andLandforms 36 (2): 257-72.
Miller, C. 2016. "Extensive Vegetative Roofs." National Institute of Building Sciences. Last modified
2016. Accessed March 11. 2016. www.wbdg.org/resources/greenroofs.php.
Munoz-Carpena, R., A. Ritter, G. A. Fox, and O. Perez-Ovilla. 2015. "Does mechanistic modeling of
filter strip pesticide mass balance and degradation processes affect environmental exposure
assessments?" Chemosphere 139: 410-21.
NAS. 2016. Using Graywater and Stormwater to Enhance Local Water Supplies: An Assessment of Risks,
Costs, and Benefits. Washington, D.C.: National Academies Press.
NCSL. 2016. "State Rainwater Harvesting Laws and Legislation." National Conference of State
Legislatures. Last modified April 4, 2016. Accessed April 6, 2016.
www.ncsl.org/research/environment-and-natural-resources/rainwater-harvesting.aspx.
Nektarios, P. A., I. Amountzias, I. Kokkinou, and N. Ntoulas. 2011. "Green roof substrate type and depth
affect the growth of the native species Dianthus fruticosus under reduced irrigation regimens."
HortScience 46 (8): 1208-16.
Nektarios, P. A., N. Ntoulas, E. Nydrioti, I. Kokkinou, E. M. Bali, and I. Amountzias. 2015. "Drought
stress response of Sedum sediforme grown in extensive green roof systems with different
substrate types and depths." Scientia Horticulturae 181: 52-61.
Nektarios, P. A., N. Ntoulas, A. T. Paraskevopoulou, A. Zacharopoulou, and I. Chronopoulos. 2014.
"Substrate and drainage system selection and slope stabilization in an intensive-type roof garden
park: Case study of Athens Concert Hall." Landscape and Ecological Engineering 10 (1): 29-46.
Norton, B. A., A. M. Coutts, S. J. Livesley, R. J. Harris, A. M. Hunter, and N. S. G. Williams. 2015.
"Planning for cooler cities: A framework to prioritise green infrastructure to mitigate high
temperatures in urban landscapes." Landscape and Urban Planning 134: 127-38.
NRCS. 2010. Filter Strip. Code 393. U.S. Department of Agriculture.
NRCS. 2013. "Web Soil Survey." Accessed April 28, 2016.
http://websoilsurvev.sc.egov.usda.gov/App/HomePage.htm.
80
-------
NRDC. 2009. How Greening California Cities Can Address Water Resources and Climate Challenges in
the 21st Century. New York: Natural Resources Defense Council.
NRDC. 2010. Climate Change, Water, and Risk: Current Water Demands Are Not Sustainable.
Washington, D.C.: Natural Resources Defense Council.
Ntoulas, N., and P. A. Nektarios. 2015. "Paspalum vaginatum drought tolerance and recovery in adaptive
extensive green roof systems." Ecological Engineering 82: 189-200.
Ntoulas, N., P. A. Nektarios, E. Charalambous, and A. Psaroulis. 2013. "Zoysia matrella cover rate and
drought tolerance in adaptive extensive green roof systems." Urban Forestry & Urban Greening
12 (4): 522-31.
Ntoulas, N., P. A. Nektarios, T. E. Kapsali, M. P. Kaltsidi, L. B. Han, and S. X. Yin. 2015.
"Determination of the physical, chemical, and hydraulic characteristics of locally available
materials for formulating extensive green roof substrates." HortTechnology 25 (6): 774-84.
Nyakudya, I. W., L. Stroosnijder, and I. Nyagumbo. 2014. "Infiltration and planting pits for improved
water management and maize yield in semi-arid Zimbabwe." Agricultural Water Management
141: 30-46.
OALS. 2002. "United States Humidity Zones." Office of Arid Land Studies. Last modified January 31,
2002. Accessed April 4, 2016. http ://cals .arizona.edu/oals/soils/survevs/states .html.
Ondono, S., J. J. Martinez-Sanchez, and J. L. Moreno. 2016a. "The composition and depth of green roof
substrates affect the growth of Silene vulgaris and Lagurus ovatus species and the C and N
sequestration under two irrigation conditions." Journal of Environmental Management 166: 330-
40.
Ondono, S., J. J. Martinez-Sanchez, and J. L. Moreno. 2016b. "The inorganic component of green roof
substrates impacts the growth of Mediterranean plant species as well as the C and N sequestration
potential." Ecological Indicators 61: 739-52.
Pataki, D. E., H. R. McCarthy, E. Litvak, and S. Pincetl. 2011. "Transpiration of urban forests in the Los
Angeles metropolitan area." Ecological Applications 21 (3): 661-77.
Peel, M. C., B. L. Finlayson, and T. A. Mcmahon. 2007. "Updated world map of the Koppen-Geiger
climate classification." Hydrology and Earth System Sciences Discussions 4 (2): 439-73.
Raimondo, F., P. Trifilo, M. A. Lo Gullo, S. Andri, T. Savi, and A. Nardini. 2015. "Plant performance on
Mediterranean green roofs: interaction of species-specific hydraulic strategies and substrate water
relations." Aob Plants 7: plv007.
Razzaghmanesh, M., and S. Beecham. 2014. "The hydrological behaviour of extensive and intensive
green roofs in a dry climate." Science of the Total Environment 499: 284-96.
Razzaghmanesh, M., S. Beecham, and C. J. Brien. 2014. "Developing resilient green roofs in a dry
climate." Science of the Total Environment 490: 579-89.
81
-------
Razzaghmanesh, M., S. Beecham, and F. Kazemi. 2014a. "The growth and survival of plants in urban
green roofs in a dry climate." Science of the Total Environment 476: 288-97.
Razzaghmanesh, M., S. Beecham, and F. Kazemi. 2014b. "Impact of green roofs on stormwater quality in
a South Australian urban environment." Science of the Total Environment 470: 651-59.
Rodriguez-Caballero, E., Y. Canton, R. Lazaro, and A. Sole-Benet. 2014. "Cross-scale interactions
between surface components and rainfall properties. Non-linearities in the hydrological and
erosive behavior of semiarid catchments." Journal of Hydrology 517: 815-25.
Rodriguez-Lloveras, X., G. Bussi, F. Frances, E. Rodriguez-Caballero, A. Sole-Benet, M. Calle, and G.
Benito. 2015. "Patterns of runoff and sediment production in response to land-use changes in an
ungauged Mediterranean catchment." Journal of Hydrology 531: 1054-66.
Sabbagh, G. J., R. Munoz-Carpena, and G. A. Fox. 2013. "Distinct influence of filter strips on acute and
chronic pesticide aquatic environmental exposure assessments across US EPA scenarios."
Chemosphere 90 (2): 195-202.
Sailor, D. J., T. B. Elley, and M. Gibson. 2012. "Exploring the building energy impacts of green roof
design decisions—a modeling study of buildings in four distinct climates." Journal of Building
Physics 35 (4): 372-91.
Saoub, H. M., R. A1 Tabini, K. A1 Khalidi, and J. Y. Ayad. 2011. "Effect of three water harvesting
techniques on forage shrub and natural vegetation in the Badia of Jordan." International Journal
of Botany 7 (3): 230-36.
Savi, T., M. Marin, D. Boldrin, G. Incerti, S. Andri, and A. Nardini. 2014. "Green roofs for a drier world:
Effects of hydrogel amendment on substrate and plant water status." Science of the Total
Environment 490: 467-76.
Schooling, J. T., and D. E. Carlyle-Moses. 2015. "The influence of rainfall depth class and deciduous tree
traits on stemflow production in an urban park." Urban Ecosystems 18 (4): 1261-84.
Schroll, E., J. Lambrinos, T. Righetti, and D. Sandrock. 2011. "The role of vegetation in regulating
stormwater runoff from green roofs in a winter rainfall climate." Ecological Engineering 37 (4):
595-600.
Segaran, R. R., M. Lewis, and B. Ostendorf. 2014. "Stormwater quality improvement potential of an
urbanised catchment using water sensitive retrofits into public parks." Urban Forestry and Urban
Greening 13 (2): 315-24.
Simmons, M. T. 2015. "Climates and microclimates: Challenges for extensive green roof design in hot
climates." In Green Roof Ecosystems, edited by R. K. Sutton, 63-80.
Smith, D. B., W. F. Cannon, L. G. Woodruff, F. Solano, and K. J. Ellefsen. 2014. Geochemical and
Mineralogical Maps for Soils of the Conterminous United States. Reston, VA: U.S. Geological
Survey, http://pubs.usgs.gov/of/2014/1082/.
82
-------
Steffen, J., M. Jensen, C. A. Pomeroy, and S. J. Burian. 2013. "Water supply and stormwater management
benefits of residential rainwater harvesting in U.S. cities." Journal of the American Water
Resources Association 49 (4): 810-24.
Stempihar, J. J., T. Pourshams-Manzouri, K. E. Kaloush, and M. C. Rodezno. 2012. "Porous asphalt
pavement temperature effects for urban heat island analysis." Transportation Research Record
(2293): 123-30.
Stromberg, J. C., M. G. F. Tluczek, A. F. Hazelton, and H. Ajami. 2010. "A century of riparian forest
expansion following extreme disturbance: Spatio-temporal change in Populus/Salix/Tamarix
forests along the Upper San Pedro River, Arizona, USA." Forest Ecology and Management 259
(6): 1181-89.
Sun, Z., N. B. Chang, C. Opp, and T. Hennig. 2011. "Evaluation of ecological restoration through
vegetation patterns in the lower Tarim River, China with MODIS NDVI data." Ecological
Informatics 6 (2): 156-63.
Tassoula, L., M. Papafotiou, G. Liakopoulos, and G. Kargas. 2015. "Growth of the native xerophyte
Convolvulus cneorum L. on an extensive mediterranean green roof under different substrate types
and irrigation regimens." HortScience 50 (7): 1118-24.
Teply, M., D. McGreer, and K. Ceder. 2014. "Using simulation models to develop riparian buffer strip
prescriptions." Journal of Forestry 112 (3): 302-311.
TetraTech. 2011. San Diego Low Impact Development Design Manual. Document No. PITS070177-01.
San Diego, CA: Storm Water Division, City of San Diego.
Thornthwaite, C. W. 1948. "An approach toward a rational classification of climate." Geographical
Review 38 (1): 55-94.
Tolderlund, L. 2010. Design Guidelines and Maintenance Manual for Green Roofs in the Semi-Arid and
Arid West. Denver, CO: Green Roofs for Healthy Cities.
TWDB. 2005. The Texas Manual on Rainwater Harvesting. Third Edition. Austin, TX: Texas Water
Development Board.
United Nations. 2015. Transforming Our World: the 2030 Agenda for Sustainable Development. New
York: United Nations.
USEPA. 2007. Green Infrastructure Statement of Intent. Washington, D.C.: U.S. Environmental
Protection Agency.
USEPA. 2008a. Incorporating Green Infrastructure Concepts Into Total Maximum Daily Loads
(TMDLs). Washington, D.C.: U.S. Environmental Protection Agency.
USEPA. 2008b. Managing Wet Weather With Green Infrastructure: Action Strategy. Washington, D.C.:
U.S. Environmental Protection Agency.
83
-------
USEPA. 2010a. Green Infrastructure Case Studies: Municipal Policies for Managing Stormwater With
Green Infrastructure. EPA-841-F-10-004. Washington, D.C.: U.S. Environmental Protection
Agency.
USEPA. 2010b. Green Infrastructure in Arid and Semi-Arid Climates. EPA-833-B-10-002. Washington,
D.C.: U.S. Environmental Protection Agency.
USEPA. 2012. Integrated Municipal Stormwater and Wastewater Planning Approach Framework.
Washington, D.C.: U.S. Environmental Protection Agency.
USEPA. 2013. Green Infrastructure Strategic Agenda 2013. Washington, D.C.: U.S. Environmental
Protection Agency.
USEPA. 2014. Getting to Green: Paying for Green Infrastructure. EPA-842-R-14-005. Washington,
D.C.: U.S. Environmental Protection Agency.
USEPA. 2015a. "2015 Campus RainWorks Challenge." U.S. Environmental Protection Agency. Last
modified November 2, 2015. Accessed April 7, 2016. www.epa.gov/green-infrastructure/2015-
campus-rainworks-challenge.
USEPA. 2015b. "Benefits of Green Infrastructure." U.S. Environmental Protection Agency. Last
modified November 20, 2015. Accessed April 7, 2016. www.epa.gov/green-
infrastructure/benefits-green-infrastructure.
USEPA. 2015c. "CSO Control Plans and Remedies." U.S. Environmental Protection Agency. Last
modified November 2, 2015. Accessed April 7, 2016. www.epa.gov/green-infrastructure/cso-
control-plans-and-remedies.
USEPA. 2015d. "Green Infrastructure Collaborative." U.S. Environmental Protection Agency. Last
modified November 2, 2015. Accessed April 7, 2016. www.epa.gov/green-infrastructure/green-
infrastructure-collaborative.
USEPA. 2015e. "Green Infrastructure Cost-Benefit Resources." U.S. Environmental Protection Agency.
Last modified December 16, 2015. Accessed April 7, 2016. www.epa.gov/green-
infrastructure/green-infrastructure-cost-benefit-resources.
USEPA. 2015f. "Green Infrastructure for Arid Communities." U.S. Environmental Protection Agency.
Last modified December 16, 2015. Accessed April 7, 2016. www.epa.gov/green-
infrastructure/green-infrastructure-arid-communities.
USEPA. 2015g. "Green Infrastructure for Climate Resiliency." U.S. Environmental Protection Agency.
Last modified November 2, 2015. Accessed April 7, 2016. www.epa.gov/green-
infrastructure/green-infrastructure -climate-re siliencv.
USEPA. 2015h. Green Infrastructure for Stormwater Control: Gauging Its Effectiveness With Community
Partners. EPA-600-R-15-219. Edison, NJ: National Risk Management Research Laboratory, U.S.
Environmental Protection Agency.
84
-------
USEPA. 2015i. "Green Infrastructure Funding Opportunities." U.S. Environmental Protection Agency.
Last modified December 7, 2015. Accessed April 7, 2016. www.epa.gov/green-
infrastructure/green-infrastructure-funding-opportunities.
USEPA. 2015j. "Green Infrastructure Modeling Tools." Last modified November 2, 2015. Accessed April
6, 2016. www.epa.gov/green-infrastructure/green-infrastructure-modeling-tools.
USEPA. 2015k. "Green Infrastructure Municipal Handbook." U.S. Environmental Protection Agency.
Last modified November 2, 2015. Accessed April 7, 2016. www.epa.gov/green-
infrastructure/green-infrastructure-municipal-handbook.
USEPA. 20151. "Green Infrastructure Operations and Maintenance." U.S. Environmental Protection
Agency. Last modified November 2, 2015. Accessed April 6, 2016. www.epa.gov/green-
infrastructure/green-infrastructure-operations-and-maintenance.
USEPA. 2015m. "Green Infrastructure Research." U.S. Environmental Protection Agency. Last modified
October 1, 2015. Accessed April 7, 2016. www.epa.gov/green-infrastructure/green-infrastructure-
research.
USEPA. 2015n. "Green Infrastructure Wizard." U.S. Environmental Protection Agency. Last modified
October 9, 2015. Accessed June 20, 2016. www.epa.gov/communitvhealth/green-infrastructure-
wizard.
USEPA. 2015o. "Green Infrastructure: Policy Guides." U.S. Environmental Protection Agency. Last
modified November 20, 2015. Accessed April 7, 2016. www.epa.gov/green-infrastructure/policv-
guides.
USEPA. 2015p. "Learn About Green Infrastructure: Green Infrastructure Research." U.S. Enivronmental
Protection Agency. Last modified November 2, 2015. Accessed April 7, 2016.
www.epa.gov/green-infrastructure/green-infrastructure-research.
USEPA. 2015q. Safe and Sustainable Water Resources Strategic Research Action Plan: 2016-2019.
EPA-600-K-15-004. Washington, D.C.: U.S. Environmental Protection Agency Office of
Research and Development.
USEPA. 2015r. Tools, Strategies and Lessons Learned From EPA Green Infrastructure Technical
Assistance Projects. EPA-832-R-15-016. Washington, D.C.: U.S. Environmental Protection
Agency.
USEPA. 2015s. "What Is Green Infrastructure?". U.S. Environmental Protection Agency. Last modified
November 2, 2015. Accessed March 17, 2016. www.epa.gov/green-infrastructure/what-green-
infrastructure.
USEPA. 2016a. "Automated Geospatial Watershed Assessment (AGWA) Tool for Hydrologic Modeling
and Watershed Assessment." U.S. Environmental Protection Agency. Last modified March 25,
2016. Accessed June 20, 2016. www.epa.gov/water-research/automated-geospatial-watershed-
assessment-agwa-tool-hydrologic-modeling-and-watershed.
85
-------
USEPA. 2016b. "Green Infrastructure." U.S. Environmental Protection Agency. Last modified March 10,
2016. Accessed April 8, 2016. www.epa.gov/green-infirastructure.
USEPA. 2016c. "Green Infrastructure Design and Implementation." U.S. Environmental Protection
Agency. Last modified February 23, 2016. Accessed April 7, 2016. www.epa.gov/green-
infrastructure/green-infrastructure-design-and-implementation.
USEPA. 2016d. "Green Infrastructure Webcast Series." U.S. Environmental Protection Agency. Last
modified April 6, 2016. Accessed April 7, 2016. www.epa.gov/green-infrastructure/green-
infrastructure-webcast-series.
USEPA. 2016e. "An Integrated Decision Support Tool (I-DST) for Life-Cycle Cost Assessment and
Optimization of Green, Grey, and Hybrid Stormwater Infrastructure." U.S. Environmental
Protection Agency. Last modified April 29, 2016. Accessed June 20, 2016.
https://cfpub.epa.gov/ncer abstracts/index.cfm/fuseaction/displav.abstractDetail/abstract/10617/r
eport/0.
USEPA. 2016f. "Integrating Green Infrastructure Into Federal Regulatory Programs." U.S. Environmental
Protection Agency. Last modified February 23, 2016. Accessed April 6, 2016.
www.epa.gov/green-infrastructure/integrating-green-infrastructure-federal-regulatorv-programs.
USEPA. 2016g. "National Center for Sustainable Water Infrastructure Modeling Research." U.S.
Environmental Protection Agency. Last modified June 20, 2016. Accessed June 20, 2016.
https://cfpub.epa.gov/ncer abstracts/index.cfm/fuseaction/displav.rfatext/rfa id/593.
USEPA. 2016h. "Soak Up the Rain." U.S. Environmental Protection Agency. Last modified March 8,
2016. Accessed April 7, 2016. www.epa.gov/soakuptherain.
USEPA. 2016i. "Urban Runoff: Low Impact Development." U.S. Environmental Protection Agency. Last
modified January 22, 2016. Accessed March 28, 2016. www.epa.gov/polluted-runoff-nonpoint-
source-pollution/urban-runoff-low-impact-development.
USEPA. 2016j. "Water Use Today." U.S. Environmental Protection Agency. Last modified February 23,
2016. Accessed April 6, 2016.
https://www3.epa.gov/watersense/our water/water use todav.html.
USEPA. 2016k. "What Is EPA Doing to Support Green Infrastructure?". U.S. Environmental Protection
Agency. Last modified February 25, 2016. Accessed April 7, 2016. www.epa.gov/green-
infrastructure/what-epa-doing-support-green-infrastructure.
USEPA Region 3. 2015. Community Based Public-Private Partnerships (CBP3s) and Alternative Market-
Based Tools for Integrated Green Stormwater Infrastructure. Philadelphia, PA: U.S.
Environmental Protection Agency.
USEPA Region 6. 2014. Stormwater Demonstration Median. Dallas, TX: U.S. Environmental Protection
Agency.
86
-------
USEPA Region 6. 2016a. "Green Infrastructure: Basics." U.S. Environmental Protection Agency Region
6. Last modified February 23, 2016. Accessed April 8, 2016.
https://www3.epa.gov/region6/water/npdes/greeninfrastructure/basics.html.
USEPA Region 6. 2016b. "Green Infrastructure: Funding." U.S. Environmental Protection Agency
Region 6. Last modified February 23, 2016. Accessed April 8, 2016.
https://www3.epa.gov/region6/water/npdes/greeninfrastructure/funding.html.
USEPA Region 6. 2016c. "Green Infrastructure: States." U.S. Environmental Protection Agency Region
6. Last modified February 23, 2016. Accessed April 8, 2016.
https://www3.epa.gov/region6/water/npdes/greeninfrastructure/states.html.
USEPA Region 6. 2016d. "Green Infrastructure: Training." U.S. Environmental Protection Agency
Region 6. Last modified February 23, 2016. Accessed April 8, 2016.
https://www3.epa.gov/region6/water/npdes/greeninfrastructure/training.html.
USEPA Region 8. 2016. "Green Infrastructure: Low Impact Development and Green Infrastructure in the
Semi-Arid West." U.S. Environmental Protection Agency. Last modified March 9, 2016.
Accessed March 17, 2016. www.epa.gov/region8/green-infrastructure.
USEPA Region 9. 2016. "Low Impact Development." U.S. Environmental Protection Agency Region 9.
Last modified February 23, 2016. Accessed April 8, 2016.
https://www3.epa.gov/region9/water/lid.
Van Mechelen, C., T. Dutoit, and M. Hermy. 2015. "Adapting green roof irrigation practices for a
sustainable future: A review." Sustainable Cities and Society 19: 74-90.
Vera, S., C. Pinto, F. Victorero, W. Bustamante, C. Bonilla, J. Gironas, and V. Rojas. 2015. "Influence of
plant and substrate characteristics of vegetated roofs on a supermarket energy performance
located in a semiarid climate." Energy Procedia 78: 1171-76.
Villarreal, M., S. Drake, S. Marsh, and A. McCoy. 2012. "The influence of wastewater subsidy, flood
disturbance and neighbouring land use on current and historical patterns of riparian vegetation in
a semi-arid watershed." River Research and Applications 28 (8): 1230-45.
Walker, A. S. 1996. Deserts: Geology and Resources. U.S. Geological Survey Unnumbered Series.
Washington, D.C.: Government Printing Office.
Walsh, T. C., C. A. Pomeroy, and S. J. Burian. 2014. "Hydrologic modeling analysis of a passive,
residential rainwater harvesting program in an urbanized, semi-arid watershed." Journal of
Hydrology 508: 240-53.
Wang, Q., X. Y. Song, F. C. Li, G. R. Hu, Q. L. Liu, E. H. Zhang, H. L. Wang, and R. Davies. 2015.
"Optimum ridge-furrow ratio and suitable ridge-mulching material for Alfalfa production in
rainwater harvesting in semi-arid regions of China." Field Crops Research 180: 186-96.
Wang, R. R., and J. B. Zimmerman. 2015. "Economic and environmental assessment of office building
rainwater harvesting systems in various US cities." Environmental Science & Technology 49 (3):
1768-78.
87
-------
Watershed Management Group. 2015. "Watershed Management Group: Our Mission." Watershed
Management Group. Last modified 2015. Accessed April 8, 2016.
https://watershedmg.org/mission.
WERF. 2016. "International Stormwater BMP Database." Accessed April 7, 2016.
www .bmpdatabase. org/index.htm.
Western Water Assessment. 2011. Colorado Climate Preparedness Project: Final Report. Boulder, CO:
Regents of the University of Colorado.
Wu, T., and R. E. Smith. 2011. "Economic benefits for green roofs: a case study of the Skaggs Pharmacy
Building, University of Utah." International Journal of Design and Nature and Ecodynamics 6
(2): 122-38.
Yazar, A., M. Kuzucu, I. Celik, S. M. Sezen, and S. E. Jacobsen. 2014. "Water harvesting for improved
water productivity in dry environments of the Mediterranean region case study: Pistachio in
Turkey." Journal of Agronomy and Crop Science 200 (5): 361-70.
Young-Mathews, A., S. W. Culman, S. Sanchez-Moreno, A. T. O'Geen, H. Ferris, A. D. Hollander, and
L. E. Jackson. 2010. "Plant-soil biodiversity relationships and nutrient retention in agricultural
riparian zones of the Sacramento Valley, California." Agroforestry Systems 80 (1): 41-60.
Young, T., D. D. Cameron, J. Sorrill, T. Edwards, and G. K. Phoenix. 2014. "Importance of different
components of green roof substrate on plant growth and physiological performance." Urban
Forestry and Urban Greening 13 (3): 507-16.
Young, T. M., D. D. Cameron, and G. K. Phoenix. 2015. "Increasing green roof plant drought tolerance
through substrate modification and the use of water retention gels." Urban Water Journal. 1-10.
Zhang, Y., A. Grant, A. Sharma, D. Chen, and L. Chen. 2010. "Alternative water resources for rural
residential development in Western Australia." Water Resources Management 24 (1): 25-36.
88
-------
vvEPA
United States
Environmental Protection
Agency
Please make all necessary changes on the below label, detach
or copy and return to the address in the upper left hand corner.
If you do not wish to receive these reports CHECK HERE
detach, or copy this cover, and return to the address in the
upper left hand corner.
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA PERMIT No. G-35
Office of Research
and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
EPA/600/R-16/146
August 2016
www.epa.gov
Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chlorine free
------- |